diff --git a/.gitattributes b/.gitattributes
index 5004930cb5655e4ae59473d80bd07245d91f85b6..b35593da2da780c0c3ac827a66630d73015a0af8 100644
--- a/.gitattributes
+++ b/.gitattributes
@@ -875,3 +875,12 @@ gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-
 gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_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
+gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-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-8828-sd-1/checkpoint-11048/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-0.9-num-8828-sd-1/checkpoint-1381/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-0.9-num-8828-sd-1/checkpoint-2763/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-0.9-num-8828-sd-1/checkpoint-4144/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-0.9-num-8828-sd-1/checkpoint-5526/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-0.9-num-8828-sd-1/checkpoint-6907/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-0.9-num-8828-sd-1/checkpoint-8289/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-0.9-num-8828-sd-1/checkpoint-9670/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-0.9-num-8828-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-0.9-num-8828-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-0.9-num-8828-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-0.9-num-8828-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-0.9-num-8828-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-0.9-num-8828-sd-1/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e99bbcd43df1c19d98706c7e3be95c93844c5349
--- /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-0.9-num-8828-sd-1/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-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-8828-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-0.9-num-8828-sd-1/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..c79d70858ce055c6f1bb883cb35180f5113e2f20
--- /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-0.9-num-8828-sd-1/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bdd71ca55a2af1b8facc7c698205b6599fe0e892ac0b224b8ae93a917b4ed891
+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-0.9-num-8828-sd-1/checkpoint-11048/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-0.9-num-8828-sd-1/checkpoint-11048/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-0.9-num-8828-sd-1/checkpoint-11048/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-0.9-num-8828-sd-1/checkpoint-11048/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-0.9-num-8828-sd-1/checkpoint-11048/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e99bbcd43df1c19d98706c7e3be95c93844c5349
--- /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-0.9-num-8828-sd-1/checkpoint-11048/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-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-8828-sd-1/checkpoint-11048/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-0.9-num-8828-sd-1/checkpoint-11048/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..ccfaeeea929000371803ae9ca50bb1c7cb765a56
--- /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-0.9-num-8828-sd-1/checkpoint-11048/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:452c77583957da6b4293899c95b5be9701d7480c4d5a2b0f2fa68c4fb0b9731b
+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-0.9-num-8828-sd-1/checkpoint-11048/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-0.9-num-8828-sd-1/checkpoint-11048/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..506d13db3f66eaa07cd88dde5a9dab148d81affa
--- /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-0.9-num-8828-sd-1/checkpoint-11048/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f3b9dc5674e22ec76147af6a69b9bbabc1f0e2f49564c5b73267a13abf7667b
+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-0.9-num-8828-sd-1/checkpoint-11048/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-0.9-num-8828-sd-1/checkpoint-11048/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..6e1c8f330a7ead0c006626ce05530312ba87cf1f
--- /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-0.9-num-8828-sd-1/checkpoint-11048/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5ac61707428e46dc6875dc3f37d3153002d158b4d52f392a06aae074add7ec80
+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-0.9-num-8828-sd-1/checkpoint-11048/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-0.9-num-8828-sd-1/checkpoint-11048/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..465ee81cfb83902b3701aefcc3a37d798ae3014f
--- /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-0.9-num-8828-sd-1/checkpoint-11048/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:85b06ab96c3fe86bdf5decd66ad11232808aad800149c581e0ee95fa30ae0ec6
+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-0.9-num-8828-sd-1/checkpoint-11048/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-0.9-num-8828-sd-1/checkpoint-11048/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-0.9-num-8828-sd-1/checkpoint-11048/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-0.9-num-8828-sd-1/checkpoint-11048/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-0.9-num-8828-sd-1/checkpoint-11048/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-0.9-num-8828-sd-1/checkpoint-11048/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-0.9-num-8828-sd-1/checkpoint-11048/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-0.9-num-8828-sd-1/checkpoint-11048/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-0.9-num-8828-sd-1/checkpoint-11048/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-0.9-num-8828-sd-1/checkpoint-11048/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-0.9-num-8828-sd-1/checkpoint-11048/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-0.9-num-8828-sd-1/checkpoint-11048/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-0.9-num-8828-sd-1/checkpoint-11048/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-0.9-num-8828-sd-1/checkpoint-11048/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..41f964935b2ae34fb55e28a330bf42fc43159ddb
--- /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-0.9-num-8828-sd-1/checkpoint-11048/trainer_state.json
@@ -0,0 +1,7825 @@
+{
+  "best_metric": 1.4217946529388428,
+  "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-0.9-num-8828-sd-1/checkpoint-4144",
+  "epoch": 7.997104596453131,
+  "eval_steps": 10,
+  "global_step": 11048,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007238508867173362,
+      "grad_norm": 1.2523442506790161,
+      "learning_rate": 0.0002,
+      "loss": 4.7061,
+      "step": 10
+    },
+    {
+      "epoch": 0.014477017734346724,
+      "grad_norm": 1.8887330293655396,
+      "learning_rate": 0.0002,
+      "loss": 3.3493,
+      "step": 20
+    },
+    {
+      "epoch": 0.021715526601520086,
+      "grad_norm": 0.9668035507202148,
+      "learning_rate": 0.0002,
+      "loss": 2.7585,
+      "step": 30
+    },
+    {
+      "epoch": 0.028954035468693448,
+      "grad_norm": 2.9167306423187256,
+      "learning_rate": 0.0002,
+      "loss": 2.3699,
+      "step": 40
+    },
+    {
+      "epoch": 0.036192544335866814,
+      "grad_norm": 2.649867296218872,
+      "learning_rate": 0.0002,
+      "loss": 2.2679,
+      "step": 50
+    },
+    {
+      "epoch": 0.04343105320304017,
+      "grad_norm": 1.5120655298233032,
+      "learning_rate": 0.0002,
+      "loss": 2.2202,
+      "step": 60
+    },
+    {
+      "epoch": 0.05066956207021354,
+      "grad_norm": 0.7879868149757385,
+      "learning_rate": 0.0002,
+      "loss": 2.2026,
+      "step": 70
+    },
+    {
+      "epoch": 0.057908070937386896,
+      "grad_norm": 0.7616953253746033,
+      "learning_rate": 0.0002,
+      "loss": 1.9447,
+      "step": 80
+    },
+    {
+      "epoch": 0.06514657980456026,
+      "grad_norm": 1.8809149265289307,
+      "learning_rate": 0.0002,
+      "loss": 2.0112,
+      "step": 90
+    },
+    {
+      "epoch": 0.07238508867173363,
+      "grad_norm": 0.9294016361236572,
+      "learning_rate": 0.0002,
+      "loss": 1.8337,
+      "step": 100
+    },
+    {
+      "epoch": 0.07962359753890698,
+      "grad_norm": 0.7145281434059143,
+      "learning_rate": 0.0002,
+      "loss": 1.8419,
+      "step": 110
+    },
+    {
+      "epoch": 0.08686210640608034,
+      "grad_norm": 0.7564446330070496,
+      "learning_rate": 0.0002,
+      "loss": 2.0036,
+      "step": 120
+    },
+    {
+      "epoch": 0.09410061527325371,
+      "grad_norm": 1.1681925058364868,
+      "learning_rate": 0.0002,
+      "loss": 1.9306,
+      "step": 130
+    },
+    {
+      "epoch": 0.10133912414042708,
+      "grad_norm": 0.6708641648292542,
+      "learning_rate": 0.0002,
+      "loss": 1.7875,
+      "step": 140
+    },
+    {
+      "epoch": 0.10857763300760044,
+      "grad_norm": 0.7625647783279419,
+      "learning_rate": 0.0002,
+      "loss": 1.786,
+      "step": 150
+    },
+    {
+      "epoch": 0.11581614187477379,
+      "grad_norm": 0.8463464975357056,
+      "learning_rate": 0.0002,
+      "loss": 1.6687,
+      "step": 160
+    },
+    {
+      "epoch": 0.12305465074194716,
+      "grad_norm": 0.7502335906028748,
+      "learning_rate": 0.0002,
+      "loss": 1.6214,
+      "step": 170
+    },
+    {
+      "epoch": 0.13029315960912052,
+      "grad_norm": 0.6929958462715149,
+      "learning_rate": 0.0002,
+      "loss": 1.7433,
+      "step": 180
+    },
+    {
+      "epoch": 0.1375316684762939,
+      "grad_norm": 0.6798707842826843,
+      "learning_rate": 0.0002,
+      "loss": 1.6009,
+      "step": 190
+    },
+    {
+      "epoch": 0.14477017734346725,
+      "grad_norm": 0.7566508650779724,
+      "learning_rate": 0.0002,
+      "loss": 1.6208,
+      "step": 200
+    },
+    {
+      "epoch": 0.15200868621064062,
+      "grad_norm": 0.7196869850158691,
+      "learning_rate": 0.0002,
+      "loss": 1.5823,
+      "step": 210
+    },
+    {
+      "epoch": 0.15924719507781396,
+      "grad_norm": 0.8401045799255371,
+      "learning_rate": 0.0002,
+      "loss": 1.738,
+      "step": 220
+    },
+    {
+      "epoch": 0.16648570394498732,
+      "grad_norm": 0.8503773212432861,
+      "learning_rate": 0.0002,
+      "loss": 1.7574,
+      "step": 230
+    },
+    {
+      "epoch": 0.1737242128121607,
+      "grad_norm": 0.7183733582496643,
+      "learning_rate": 0.0002,
+      "loss": 1.7861,
+      "step": 240
+    },
+    {
+      "epoch": 0.18096272167933405,
+      "grad_norm": 0.7082605957984924,
+      "learning_rate": 0.0002,
+      "loss": 1.6693,
+      "step": 250
+    },
+    {
+      "epoch": 0.18820123054650742,
+      "grad_norm": 0.9386326670646667,
+      "learning_rate": 0.0002,
+      "loss": 1.619,
+      "step": 260
+    },
+    {
+      "epoch": 0.19543973941368079,
+      "grad_norm": 0.7332451939582825,
+      "learning_rate": 0.0002,
+      "loss": 1.6511,
+      "step": 270
+    },
+    {
+      "epoch": 0.20267824828085415,
+      "grad_norm": 0.7092869877815247,
+      "learning_rate": 0.0002,
+      "loss": 1.6353,
+      "step": 280
+    },
+    {
+      "epoch": 0.20991675714802752,
+      "grad_norm": 0.7256413698196411,
+      "learning_rate": 0.0002,
+      "loss": 1.5996,
+      "step": 290
+    },
+    {
+      "epoch": 0.21715526601520088,
+      "grad_norm": 0.6398681402206421,
+      "learning_rate": 0.0002,
+      "loss": 1.6754,
+      "step": 300
+    },
+    {
+      "epoch": 0.22439377488237422,
+      "grad_norm": 0.6273287534713745,
+      "learning_rate": 0.0002,
+      "loss": 1.397,
+      "step": 310
+    },
+    {
+      "epoch": 0.23163228374954759,
+      "grad_norm": 0.511648416519165,
+      "learning_rate": 0.0002,
+      "loss": 1.5115,
+      "step": 320
+    },
+    {
+      "epoch": 0.23887079261672095,
+      "grad_norm": 0.8677352070808411,
+      "learning_rate": 0.0002,
+      "loss": 1.5424,
+      "step": 330
+    },
+    {
+      "epoch": 0.24610930148389432,
+      "grad_norm": 0.6270743012428284,
+      "learning_rate": 0.0002,
+      "loss": 1.6779,
+      "step": 340
+    },
+    {
+      "epoch": 0.2533478103510677,
+      "grad_norm": 0.7980281114578247,
+      "learning_rate": 0.0002,
+      "loss": 1.626,
+      "step": 350
+    },
+    {
+      "epoch": 0.26058631921824105,
+      "grad_norm": 0.632486879825592,
+      "learning_rate": 0.0002,
+      "loss": 1.5238,
+      "step": 360
+    },
+    {
+      "epoch": 0.2678248280854144,
+      "grad_norm": 0.6527034640312195,
+      "learning_rate": 0.0002,
+      "loss": 1.5175,
+      "step": 370
+    },
+    {
+      "epoch": 0.2750633369525878,
+      "grad_norm": 0.7672118544578552,
+      "learning_rate": 0.0002,
+      "loss": 1.627,
+      "step": 380
+    },
+    {
+      "epoch": 0.28230184581976114,
+      "grad_norm": 0.6035117506980896,
+      "learning_rate": 0.0002,
+      "loss": 1.5605,
+      "step": 390
+    },
+    {
+      "epoch": 0.2895403546869345,
+      "grad_norm": 0.5955103039741516,
+      "learning_rate": 0.0002,
+      "loss": 1.4603,
+      "step": 400
+    },
+    {
+      "epoch": 0.2967788635541079,
+      "grad_norm": 0.6015191674232483,
+      "learning_rate": 0.0002,
+      "loss": 1.558,
+      "step": 410
+    },
+    {
+      "epoch": 0.30401737242128124,
+      "grad_norm": 0.6380982398986816,
+      "learning_rate": 0.0002,
+      "loss": 1.6091,
+      "step": 420
+    },
+    {
+      "epoch": 0.3112558812884546,
+      "grad_norm": 0.6707863211631775,
+      "learning_rate": 0.0002,
+      "loss": 1.5292,
+      "step": 430
+    },
+    {
+      "epoch": 0.3184943901556279,
+      "grad_norm": 0.7010176777839661,
+      "learning_rate": 0.0002,
+      "loss": 1.4426,
+      "step": 440
+    },
+    {
+      "epoch": 0.3257328990228013,
+      "grad_norm": 0.8263739943504333,
+      "learning_rate": 0.0002,
+      "loss": 1.5572,
+      "step": 450
+    },
+    {
+      "epoch": 0.33297140788997465,
+      "grad_norm": 0.7253276109695435,
+      "learning_rate": 0.0002,
+      "loss": 1.5188,
+      "step": 460
+    },
+    {
+      "epoch": 0.340209916757148,
+      "grad_norm": 0.5238934755325317,
+      "learning_rate": 0.0002,
+      "loss": 1.584,
+      "step": 470
+    },
+    {
+      "epoch": 0.3474484256243214,
+      "grad_norm": 0.7869495749473572,
+      "learning_rate": 0.0002,
+      "loss": 1.7035,
+      "step": 480
+    },
+    {
+      "epoch": 0.35468693449149474,
+      "grad_norm": 0.7485215663909912,
+      "learning_rate": 0.0002,
+      "loss": 1.5776,
+      "step": 490
+    },
+    {
+      "epoch": 0.3619254433586681,
+      "grad_norm": 0.5413193106651306,
+      "learning_rate": 0.0002,
+      "loss": 1.6274,
+      "step": 500
+    },
+    {
+      "epoch": 0.3691639522258415,
+      "grad_norm": 0.7615048885345459,
+      "learning_rate": 0.0002,
+      "loss": 1.7323,
+      "step": 510
+    },
+    {
+      "epoch": 0.37640246109301484,
+      "grad_norm": 0.7685340046882629,
+      "learning_rate": 0.0002,
+      "loss": 1.532,
+      "step": 520
+    },
+    {
+      "epoch": 0.3836409699601882,
+      "grad_norm": 0.6379081010818481,
+      "learning_rate": 0.0002,
+      "loss": 1.6312,
+      "step": 530
+    },
+    {
+      "epoch": 0.39087947882736157,
+      "grad_norm": 0.7946939468383789,
+      "learning_rate": 0.0002,
+      "loss": 1.5645,
+      "step": 540
+    },
+    {
+      "epoch": 0.39811798769453494,
+      "grad_norm": 0.6287278532981873,
+      "learning_rate": 0.0002,
+      "loss": 1.4001,
+      "step": 550
+    },
+    {
+      "epoch": 0.4053564965617083,
+      "grad_norm": 0.6811642646789551,
+      "learning_rate": 0.0002,
+      "loss": 1.5982,
+      "step": 560
+    },
+    {
+      "epoch": 0.41259500542888167,
+      "grad_norm": 0.671073317527771,
+      "learning_rate": 0.0002,
+      "loss": 1.4953,
+      "step": 570
+    },
+    {
+      "epoch": 0.41983351429605503,
+      "grad_norm": 0.6313900351524353,
+      "learning_rate": 0.0002,
+      "loss": 1.6753,
+      "step": 580
+    },
+    {
+      "epoch": 0.4270720231632284,
+      "grad_norm": 0.5291772484779358,
+      "learning_rate": 0.0002,
+      "loss": 1.546,
+      "step": 590
+    },
+    {
+      "epoch": 0.43431053203040176,
+      "grad_norm": 0.62503582239151,
+      "learning_rate": 0.0002,
+      "loss": 1.5441,
+      "step": 600
+    },
+    {
+      "epoch": 0.4415490408975751,
+      "grad_norm": 0.5777305364608765,
+      "learning_rate": 0.0002,
+      "loss": 1.6276,
+      "step": 610
+    },
+    {
+      "epoch": 0.44878754976474844,
+      "grad_norm": 0.7013497352600098,
+      "learning_rate": 0.0002,
+      "loss": 1.4758,
+      "step": 620
+    },
+    {
+      "epoch": 0.4560260586319218,
+      "grad_norm": 0.8044822216033936,
+      "learning_rate": 0.0002,
+      "loss": 1.4029,
+      "step": 630
+    },
+    {
+      "epoch": 0.46326456749909517,
+      "grad_norm": 0.672531247138977,
+      "learning_rate": 0.0002,
+      "loss": 1.7195,
+      "step": 640
+    },
+    {
+      "epoch": 0.47050307636626854,
+      "grad_norm": 0.6233910322189331,
+      "learning_rate": 0.0002,
+      "loss": 1.614,
+      "step": 650
+    },
+    {
+      "epoch": 0.4777415852334419,
+      "grad_norm": 0.651524543762207,
+      "learning_rate": 0.0002,
+      "loss": 1.6041,
+      "step": 660
+    },
+    {
+      "epoch": 0.48498009410061527,
+      "grad_norm": 0.7213939428329468,
+      "learning_rate": 0.0002,
+      "loss": 1.5842,
+      "step": 670
+    },
+    {
+      "epoch": 0.49221860296778863,
+      "grad_norm": 0.6541454792022705,
+      "learning_rate": 0.0002,
+      "loss": 1.5453,
+      "step": 680
+    },
+    {
+      "epoch": 0.499457111834962,
+      "grad_norm": 0.6568936109542847,
+      "learning_rate": 0.0002,
+      "loss": 1.662,
+      "step": 690
+    },
+    {
+      "epoch": 0.5066956207021354,
+      "grad_norm": 0.7176415324211121,
+      "learning_rate": 0.0002,
+      "loss": 1.624,
+      "step": 700
+    },
+    {
+      "epoch": 0.5139341295693087,
+      "grad_norm": 0.6553855538368225,
+      "learning_rate": 0.0002,
+      "loss": 1.6099,
+      "step": 710
+    },
+    {
+      "epoch": 0.5211726384364821,
+      "grad_norm": 0.5654335618019104,
+      "learning_rate": 0.0002,
+      "loss": 1.5508,
+      "step": 720
+    },
+    {
+      "epoch": 0.5284111473036555,
+      "grad_norm": 0.5671001672744751,
+      "learning_rate": 0.0002,
+      "loss": 1.392,
+      "step": 730
+    },
+    {
+      "epoch": 0.5356496561708288,
+      "grad_norm": 0.7914412021636963,
+      "learning_rate": 0.0002,
+      "loss": 1.388,
+      "step": 740
+    },
+    {
+      "epoch": 0.5428881650380022,
+      "grad_norm": 0.6172138452529907,
+      "learning_rate": 0.0002,
+      "loss": 1.5931,
+      "step": 750
+    },
+    {
+      "epoch": 0.5501266739051756,
+      "grad_norm": 0.6132623553276062,
+      "learning_rate": 0.0002,
+      "loss": 1.4018,
+      "step": 760
+    },
+    {
+      "epoch": 0.5573651827723489,
+      "grad_norm": 0.654000461101532,
+      "learning_rate": 0.0002,
+      "loss": 1.513,
+      "step": 770
+    },
+    {
+      "epoch": 0.5646036916395223,
+      "grad_norm": 0.5691370964050293,
+      "learning_rate": 0.0002,
+      "loss": 1.5035,
+      "step": 780
+    },
+    {
+      "epoch": 0.5718422005066957,
+      "grad_norm": 0.7922580242156982,
+      "learning_rate": 0.0002,
+      "loss": 1.65,
+      "step": 790
+    },
+    {
+      "epoch": 0.579080709373869,
+      "grad_norm": 0.6831880211830139,
+      "learning_rate": 0.0002,
+      "loss": 1.4521,
+      "step": 800
+    },
+    {
+      "epoch": 0.5863192182410424,
+      "grad_norm": 0.6740124821662903,
+      "learning_rate": 0.0002,
+      "loss": 1.4734,
+      "step": 810
+    },
+    {
+      "epoch": 0.5935577271082157,
+      "grad_norm": 1.380016803741455,
+      "learning_rate": 0.0002,
+      "loss": 1.6498,
+      "step": 820
+    },
+    {
+      "epoch": 0.6007962359753891,
+      "grad_norm": 0.6552878022193909,
+      "learning_rate": 0.0002,
+      "loss": 1.4642,
+      "step": 830
+    },
+    {
+      "epoch": 0.6080347448425625,
+      "grad_norm": 0.6649535298347473,
+      "learning_rate": 0.0002,
+      "loss": 1.6271,
+      "step": 840
+    },
+    {
+      "epoch": 0.6152732537097358,
+      "grad_norm": 0.561738133430481,
+      "learning_rate": 0.0002,
+      "loss": 1.5886,
+      "step": 850
+    },
+    {
+      "epoch": 0.6225117625769092,
+      "grad_norm": 0.6133047938346863,
+      "learning_rate": 0.0002,
+      "loss": 1.5364,
+      "step": 860
+    },
+    {
+      "epoch": 0.6297502714440825,
+      "grad_norm": 0.559843122959137,
+      "learning_rate": 0.0002,
+      "loss": 1.3489,
+      "step": 870
+    },
+    {
+      "epoch": 0.6369887803112558,
+      "grad_norm": 0.6117811799049377,
+      "learning_rate": 0.0002,
+      "loss": 1.4878,
+      "step": 880
+    },
+    {
+      "epoch": 0.6442272891784292,
+      "grad_norm": 0.6209776401519775,
+      "learning_rate": 0.0002,
+      "loss": 1.56,
+      "step": 890
+    },
+    {
+      "epoch": 0.6514657980456026,
+      "grad_norm": 0.6234082579612732,
+      "learning_rate": 0.0002,
+      "loss": 1.6747,
+      "step": 900
+    },
+    {
+      "epoch": 0.6587043069127759,
+      "grad_norm": 0.7623258233070374,
+      "learning_rate": 0.0002,
+      "loss": 1.6963,
+      "step": 910
+    },
+    {
+      "epoch": 0.6659428157799493,
+      "grad_norm": 0.6148061752319336,
+      "learning_rate": 0.0002,
+      "loss": 1.2424,
+      "step": 920
+    },
+    {
+      "epoch": 0.6731813246471227,
+      "grad_norm": 0.6682973504066467,
+      "learning_rate": 0.0002,
+      "loss": 1.4319,
+      "step": 930
+    },
+    {
+      "epoch": 0.680419833514296,
+      "grad_norm": 0.5513041615486145,
+      "learning_rate": 0.0002,
+      "loss": 1.5377,
+      "step": 940
+    },
+    {
+      "epoch": 0.6876583423814694,
+      "grad_norm": 0.5197525024414062,
+      "learning_rate": 0.0002,
+      "loss": 1.3991,
+      "step": 950
+    },
+    {
+      "epoch": 0.6948968512486428,
+      "grad_norm": 0.6490758061408997,
+      "learning_rate": 0.0002,
+      "loss": 1.4398,
+      "step": 960
+    },
+    {
+      "epoch": 0.7021353601158161,
+      "grad_norm": 0.6450682878494263,
+      "learning_rate": 0.0002,
+      "loss": 1.5251,
+      "step": 970
+    },
+    {
+      "epoch": 0.7093738689829895,
+      "grad_norm": 0.6203766465187073,
+      "learning_rate": 0.0002,
+      "loss": 1.5417,
+      "step": 980
+    },
+    {
+      "epoch": 0.7166123778501629,
+      "grad_norm": 0.6023609638214111,
+      "learning_rate": 0.0002,
+      "loss": 1.4575,
+      "step": 990
+    },
+    {
+      "epoch": 0.7238508867173362,
+      "grad_norm": 0.5765255093574524,
+      "learning_rate": 0.0002,
+      "loss": 1.4973,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7310893955845096,
+      "grad_norm": 0.6650075316429138,
+      "learning_rate": 0.0002,
+      "loss": 1.483,
+      "step": 1010
+    },
+    {
+      "epoch": 0.738327904451683,
+      "grad_norm": 0.5610854029655457,
+      "learning_rate": 0.0002,
+      "loss": 1.5959,
+      "step": 1020
+    },
+    {
+      "epoch": 0.7455664133188563,
+      "grad_norm": 0.7072813510894775,
+      "learning_rate": 0.0002,
+      "loss": 1.5248,
+      "step": 1030
+    },
+    {
+      "epoch": 0.7528049221860297,
+      "grad_norm": 0.6815407872200012,
+      "learning_rate": 0.0002,
+      "loss": 1.5776,
+      "step": 1040
+    },
+    {
+      "epoch": 0.760043431053203,
+      "grad_norm": 0.7932390570640564,
+      "learning_rate": 0.0002,
+      "loss": 1.4577,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7672819399203764,
+      "grad_norm": 0.5798183083534241,
+      "learning_rate": 0.0002,
+      "loss": 1.4515,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7745204487875498,
+      "grad_norm": 0.7898504137992859,
+      "learning_rate": 0.0002,
+      "loss": 1.5053,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7817589576547231,
+      "grad_norm": 0.4983280301094055,
+      "learning_rate": 0.0002,
+      "loss": 1.4776,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7889974665218965,
+      "grad_norm": 0.691403329372406,
+      "learning_rate": 0.0002,
+      "loss": 1.5007,
+      "step": 1090
+    },
+    {
+      "epoch": 0.7962359753890699,
+      "grad_norm": 0.5394481420516968,
+      "learning_rate": 0.0002,
+      "loss": 1.5153,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8034744842562432,
+      "grad_norm": 0.5136822462081909,
+      "learning_rate": 0.0002,
+      "loss": 1.6892,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8107129931234166,
+      "grad_norm": 0.6828126907348633,
+      "learning_rate": 0.0002,
+      "loss": 1.4902,
+      "step": 1120
+    },
+    {
+      "epoch": 0.81795150199059,
+      "grad_norm": 0.6799656748771667,
+      "learning_rate": 0.0002,
+      "loss": 1.4346,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8251900108577633,
+      "grad_norm": 0.5428406000137329,
+      "learning_rate": 0.0002,
+      "loss": 1.2678,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8324285197249367,
+      "grad_norm": 0.4811290502548218,
+      "learning_rate": 0.0002,
+      "loss": 1.4072,
+      "step": 1150
+    },
+    {
+      "epoch": 0.8396670285921101,
+      "grad_norm": 0.5519434809684753,
+      "learning_rate": 0.0002,
+      "loss": 1.4512,
+      "step": 1160
+    },
+    {
+      "epoch": 0.8469055374592834,
+      "grad_norm": 0.9748060703277588,
+      "learning_rate": 0.0002,
+      "loss": 1.4072,
+      "step": 1170
+    },
+    {
+      "epoch": 0.8541440463264568,
+      "grad_norm": 0.712609589099884,
+      "learning_rate": 0.0002,
+      "loss": 1.4309,
+      "step": 1180
+    },
+    {
+      "epoch": 0.8613825551936302,
+      "grad_norm": 0.6866157054901123,
+      "learning_rate": 0.0002,
+      "loss": 1.434,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8686210640608035,
+      "grad_norm": 0.5068854093551636,
+      "learning_rate": 0.0002,
+      "loss": 1.3704,
+      "step": 1200
+    },
+    {
+      "epoch": 0.8758595729279768,
+      "grad_norm": 0.6333245038986206,
+      "learning_rate": 0.0002,
+      "loss": 1.5601,
+      "step": 1210
+    },
+    {
+      "epoch": 0.8830980817951501,
+      "grad_norm": 0.6424421072006226,
+      "learning_rate": 0.0002,
+      "loss": 1.4636,
+      "step": 1220
+    },
+    {
+      "epoch": 0.8903365906623235,
+      "grad_norm": 0.4771921932697296,
+      "learning_rate": 0.0002,
+      "loss": 1.4186,
+      "step": 1230
+    },
+    {
+      "epoch": 0.8975750995294969,
+      "grad_norm": 0.5191764235496521,
+      "learning_rate": 0.0002,
+      "loss": 1.6323,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9048136083966702,
+      "grad_norm": 0.756222128868103,
+      "learning_rate": 0.0002,
+      "loss": 1.6105,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9120521172638436,
+      "grad_norm": 0.623823881149292,
+      "learning_rate": 0.0002,
+      "loss": 1.4396,
+      "step": 1260
+    },
+    {
+      "epoch": 0.919290626131017,
+      "grad_norm": 0.8166571259498596,
+      "learning_rate": 0.0002,
+      "loss": 1.3097,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9265291349981903,
+      "grad_norm": 0.6059346795082092,
+      "learning_rate": 0.0002,
+      "loss": 1.4625,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9337676438653637,
+      "grad_norm": 0.5842690467834473,
+      "learning_rate": 0.0002,
+      "loss": 1.3555,
+      "step": 1290
+    },
+    {
+      "epoch": 0.9410061527325371,
+      "grad_norm": 0.7649800777435303,
+      "learning_rate": 0.0002,
+      "loss": 1.5859,
+      "step": 1300
+    },
+    {
+      "epoch": 0.9482446615997104,
+      "grad_norm": 0.6420919895172119,
+      "learning_rate": 0.0002,
+      "loss": 1.5915,
+      "step": 1310
+    },
+    {
+      "epoch": 0.9554831704668838,
+      "grad_norm": 0.7011452913284302,
+      "learning_rate": 0.0002,
+      "loss": 1.453,
+      "step": 1320
+    },
+    {
+      "epoch": 0.9627216793340572,
+      "grad_norm": 0.5783746242523193,
+      "learning_rate": 0.0002,
+      "loss": 1.6766,
+      "step": 1330
+    },
+    {
+      "epoch": 0.9699601882012305,
+      "grad_norm": 0.5973192453384399,
+      "learning_rate": 0.0002,
+      "loss": 1.6308,
+      "step": 1340
+    },
+    {
+      "epoch": 0.9771986970684039,
+      "grad_norm": 0.6181833744049072,
+      "learning_rate": 0.0002,
+      "loss": 1.5901,
+      "step": 1350
+    },
+    {
+      "epoch": 0.9844372059355773,
+      "grad_norm": 0.5563396215438843,
+      "learning_rate": 0.0002,
+      "loss": 1.5258,
+      "step": 1360
+    },
+    {
+      "epoch": 0.9916757148027506,
+      "grad_norm": 0.45723360776901245,
+      "learning_rate": 0.0002,
+      "loss": 1.4508,
+      "step": 1370
+    },
+    {
+      "epoch": 0.998914223669924,
+      "grad_norm": 0.5947498679161072,
+      "learning_rate": 0.0002,
+      "loss": 1.3291,
+      "step": 1380
+    },
+    {
+      "epoch": 0.9996380745566413,
+      "eval_loss": 1.480796456336975,
+      "eval_runtime": 27.3103,
+      "eval_samples_per_second": 15.965,
+      "eval_steps_per_second": 2.014,
+      "step": 1381
+    },
+    {
+      "epoch": 1.0061527325370974,
+      "grad_norm": 0.5599952936172485,
+      "learning_rate": 0.0002,
+      "loss": 1.3057,
+      "step": 1390
+    },
+    {
+      "epoch": 1.0133912414042707,
+      "grad_norm": 0.5932008028030396,
+      "learning_rate": 0.0002,
+      "loss": 1.4991,
+      "step": 1400
+    },
+    {
+      "epoch": 1.020629750271444,
+      "grad_norm": 0.6194121837615967,
+      "learning_rate": 0.0002,
+      "loss": 1.4506,
+      "step": 1410
+    },
+    {
+      "epoch": 1.0278682591386175,
+      "grad_norm": 0.6995621919631958,
+      "learning_rate": 0.0002,
+      "loss": 1.5966,
+      "step": 1420
+    },
+    {
+      "epoch": 1.0351067680057908,
+      "grad_norm": 0.7905810475349426,
+      "learning_rate": 0.0002,
+      "loss": 1.4153,
+      "step": 1430
+    },
+    {
+      "epoch": 1.0423452768729642,
+      "grad_norm": 0.7221615314483643,
+      "learning_rate": 0.0002,
+      "loss": 1.4414,
+      "step": 1440
+    },
+    {
+      "epoch": 1.0495837857401376,
+      "grad_norm": 0.6170642375946045,
+      "learning_rate": 0.0002,
+      "loss": 1.3859,
+      "step": 1450
+    },
+    {
+      "epoch": 1.056822294607311,
+      "grad_norm": 0.5844094753265381,
+      "learning_rate": 0.0002,
+      "loss": 1.3806,
+      "step": 1460
+    },
+    {
+      "epoch": 1.0640608034744843,
+      "grad_norm": 0.7731822729110718,
+      "learning_rate": 0.0002,
+      "loss": 1.4871,
+      "step": 1470
+    },
+    {
+      "epoch": 1.0712993123416577,
+      "grad_norm": 0.4554748237133026,
+      "learning_rate": 0.0002,
+      "loss": 1.4286,
+      "step": 1480
+    },
+    {
+      "epoch": 1.078537821208831,
+      "grad_norm": 0.6923259496688843,
+      "learning_rate": 0.0002,
+      "loss": 1.3977,
+      "step": 1490
+    },
+    {
+      "epoch": 1.0857763300760044,
+      "grad_norm": 0.6008219122886658,
+      "learning_rate": 0.0002,
+      "loss": 1.3936,
+      "step": 1500
+    },
+    {
+      "epoch": 1.0930148389431777,
+      "grad_norm": 0.6450045704841614,
+      "learning_rate": 0.0002,
+      "loss": 1.4821,
+      "step": 1510
+    },
+    {
+      "epoch": 1.1002533478103511,
+      "grad_norm": 0.7833753824234009,
+      "learning_rate": 0.0002,
+      "loss": 1.3295,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1074918566775245,
+      "grad_norm": 0.5076758861541748,
+      "learning_rate": 0.0002,
+      "loss": 1.3424,
+      "step": 1530
+    },
+    {
+      "epoch": 1.1147303655446978,
+      "grad_norm": 0.5661332011222839,
+      "learning_rate": 0.0002,
+      "loss": 1.4043,
+      "step": 1540
+    },
+    {
+      "epoch": 1.1219688744118712,
+      "grad_norm": 0.6526919603347778,
+      "learning_rate": 0.0002,
+      "loss": 1.4963,
+      "step": 1550
+    },
+    {
+      "epoch": 1.1292073832790446,
+      "grad_norm": 0.5613082647323608,
+      "learning_rate": 0.0002,
+      "loss": 1.3671,
+      "step": 1560
+    },
+    {
+      "epoch": 1.136445892146218,
+      "grad_norm": 0.6113885641098022,
+      "learning_rate": 0.0002,
+      "loss": 1.4458,
+      "step": 1570
+    },
+    {
+      "epoch": 1.1436844010133913,
+      "grad_norm": 0.6732510328292847,
+      "learning_rate": 0.0002,
+      "loss": 1.3552,
+      "step": 1580
+    },
+    {
+      "epoch": 1.1509229098805647,
+      "grad_norm": 0.6146392226219177,
+      "learning_rate": 0.0002,
+      "loss": 1.3114,
+      "step": 1590
+    },
+    {
+      "epoch": 1.158161418747738,
+      "grad_norm": 0.6766974329948425,
+      "learning_rate": 0.0002,
+      "loss": 1.411,
+      "step": 1600
+    },
+    {
+      "epoch": 1.1653999276149114,
+      "grad_norm": 0.7621957659721375,
+      "learning_rate": 0.0002,
+      "loss": 1.2401,
+      "step": 1610
+    },
+    {
+      "epoch": 1.1726384364820848,
+      "grad_norm": 0.6959581971168518,
+      "learning_rate": 0.0002,
+      "loss": 1.3758,
+      "step": 1620
+    },
+    {
+      "epoch": 1.1798769453492581,
+      "grad_norm": 0.6691278219223022,
+      "learning_rate": 0.0002,
+      "loss": 1.382,
+      "step": 1630
+    },
+    {
+      "epoch": 1.1871154542164315,
+      "grad_norm": 0.4927774965763092,
+      "learning_rate": 0.0002,
+      "loss": 1.4147,
+      "step": 1640
+    },
+    {
+      "epoch": 1.1943539630836049,
+      "grad_norm": 0.7724234461784363,
+      "learning_rate": 0.0002,
+      "loss": 1.449,
+      "step": 1650
+    },
+    {
+      "epoch": 1.2015924719507782,
+      "grad_norm": 0.6817787885665894,
+      "learning_rate": 0.0002,
+      "loss": 1.4778,
+      "step": 1660
+    },
+    {
+      "epoch": 1.2088309808179516,
+      "grad_norm": 0.6500699520111084,
+      "learning_rate": 0.0002,
+      "loss": 1.3776,
+      "step": 1670
+    },
+    {
+      "epoch": 1.216069489685125,
+      "grad_norm": 0.5703568458557129,
+      "learning_rate": 0.0002,
+      "loss": 1.3875,
+      "step": 1680
+    },
+    {
+      "epoch": 1.2233079985522983,
+      "grad_norm": 0.6261579990386963,
+      "learning_rate": 0.0002,
+      "loss": 1.4735,
+      "step": 1690
+    },
+    {
+      "epoch": 1.2305465074194717,
+      "grad_norm": 0.651713490486145,
+      "learning_rate": 0.0002,
+      "loss": 1.3898,
+      "step": 1700
+    },
+    {
+      "epoch": 1.237785016286645,
+      "grad_norm": 0.684399425983429,
+      "learning_rate": 0.0002,
+      "loss": 1.4002,
+      "step": 1710
+    },
+    {
+      "epoch": 1.2450235251538184,
+      "grad_norm": 0.6996857523918152,
+      "learning_rate": 0.0002,
+      "loss": 1.5027,
+      "step": 1720
+    },
+    {
+      "epoch": 1.2522620340209918,
+      "grad_norm": 0.7102537751197815,
+      "learning_rate": 0.0002,
+      "loss": 1.3326,
+      "step": 1730
+    },
+    {
+      "epoch": 1.2595005428881652,
+      "grad_norm": 0.45809897780418396,
+      "learning_rate": 0.0002,
+      "loss": 1.3675,
+      "step": 1740
+    },
+    {
+      "epoch": 1.2667390517553385,
+      "grad_norm": 0.6377046704292297,
+      "learning_rate": 0.0002,
+      "loss": 1.4175,
+      "step": 1750
+    },
+    {
+      "epoch": 1.2739775606225119,
+      "grad_norm": 0.6965704560279846,
+      "learning_rate": 0.0002,
+      "loss": 1.3479,
+      "step": 1760
+    },
+    {
+      "epoch": 1.2812160694896852,
+      "grad_norm": 0.5688214302062988,
+      "learning_rate": 0.0002,
+      "loss": 1.5647,
+      "step": 1770
+    },
+    {
+      "epoch": 1.2884545783568586,
+      "grad_norm": 0.6384190320968628,
+      "learning_rate": 0.0002,
+      "loss": 1.3967,
+      "step": 1780
+    },
+    {
+      "epoch": 1.295693087224032,
+      "grad_norm": 0.5629363656044006,
+      "learning_rate": 0.0002,
+      "loss": 1.3671,
+      "step": 1790
+    },
+    {
+      "epoch": 1.3029315960912053,
+      "grad_norm": 0.6148255467414856,
+      "learning_rate": 0.0002,
+      "loss": 1.2292,
+      "step": 1800
+    },
+    {
+      "epoch": 1.3101701049583787,
+      "grad_norm": 0.655580997467041,
+      "learning_rate": 0.0002,
+      "loss": 1.5806,
+      "step": 1810
+    },
+    {
+      "epoch": 1.3174086138255519,
+      "grad_norm": 0.5642657279968262,
+      "learning_rate": 0.0002,
+      "loss": 1.2398,
+      "step": 1820
+    },
+    {
+      "epoch": 1.3246471226927252,
+      "grad_norm": 0.59607994556427,
+      "learning_rate": 0.0002,
+      "loss": 1.3246,
+      "step": 1830
+    },
+    {
+      "epoch": 1.3318856315598986,
+      "grad_norm": 0.5564199090003967,
+      "learning_rate": 0.0002,
+      "loss": 1.3274,
+      "step": 1840
+    },
+    {
+      "epoch": 1.339124140427072,
+      "grad_norm": 0.6949955821037292,
+      "learning_rate": 0.0002,
+      "loss": 1.5834,
+      "step": 1850
+    },
+    {
+      "epoch": 1.3463626492942453,
+      "grad_norm": 0.7036856412887573,
+      "learning_rate": 0.0002,
+      "loss": 1.4722,
+      "step": 1860
+    },
+    {
+      "epoch": 1.3536011581614187,
+      "grad_norm": 0.722062885761261,
+      "learning_rate": 0.0002,
+      "loss": 1.333,
+      "step": 1870
+    },
+    {
+      "epoch": 1.360839667028592,
+      "grad_norm": 0.6098677515983582,
+      "learning_rate": 0.0002,
+      "loss": 1.4044,
+      "step": 1880
+    },
+    {
+      "epoch": 1.3680781758957654,
+      "grad_norm": 0.5376402735710144,
+      "learning_rate": 0.0002,
+      "loss": 1.6217,
+      "step": 1890
+    },
+    {
+      "epoch": 1.3753166847629388,
+      "grad_norm": 0.6974610090255737,
+      "learning_rate": 0.0002,
+      "loss": 1.5071,
+      "step": 1900
+    },
+    {
+      "epoch": 1.3825551936301121,
+      "grad_norm": 0.6520763635635376,
+      "learning_rate": 0.0002,
+      "loss": 1.5854,
+      "step": 1910
+    },
+    {
+      "epoch": 1.3897937024972855,
+      "grad_norm": 0.6604374647140503,
+      "learning_rate": 0.0002,
+      "loss": 1.4271,
+      "step": 1920
+    },
+    {
+      "epoch": 1.3970322113644589,
+      "grad_norm": 0.7364398241043091,
+      "learning_rate": 0.0002,
+      "loss": 1.419,
+      "step": 1930
+    },
+    {
+      "epoch": 1.4042707202316322,
+      "grad_norm": 0.6849475502967834,
+      "learning_rate": 0.0002,
+      "loss": 1.4585,
+      "step": 1940
+    },
+    {
+      "epoch": 1.4115092290988056,
+      "grad_norm": 0.6562670469284058,
+      "learning_rate": 0.0002,
+      "loss": 1.5577,
+      "step": 1950
+    },
+    {
+      "epoch": 1.418747737965979,
+      "grad_norm": 0.5695616006851196,
+      "learning_rate": 0.0002,
+      "loss": 1.4725,
+      "step": 1960
+    },
+    {
+      "epoch": 1.4259862468331523,
+      "grad_norm": 0.5244464874267578,
+      "learning_rate": 0.0002,
+      "loss": 1.3088,
+      "step": 1970
+    },
+    {
+      "epoch": 1.4332247557003257,
+      "grad_norm": 0.6347293257713318,
+      "learning_rate": 0.0002,
+      "loss": 1.5069,
+      "step": 1980
+    },
+    {
+      "epoch": 1.440463264567499,
+      "grad_norm": 0.5528361201286316,
+      "learning_rate": 0.0002,
+      "loss": 1.3502,
+      "step": 1990
+    },
+    {
+      "epoch": 1.4477017734346724,
+      "grad_norm": 0.6987585425376892,
+      "learning_rate": 0.0002,
+      "loss": 1.3978,
+      "step": 2000
+    },
+    {
+      "epoch": 1.4549402823018458,
+      "grad_norm": 0.6568987369537354,
+      "learning_rate": 0.0002,
+      "loss": 1.4262,
+      "step": 2010
+    },
+    {
+      "epoch": 1.4621787911690192,
+      "grad_norm": 0.7665994763374329,
+      "learning_rate": 0.0002,
+      "loss": 1.4175,
+      "step": 2020
+    },
+    {
+      "epoch": 1.4694173000361925,
+      "grad_norm": 0.5127707123756409,
+      "learning_rate": 0.0002,
+      "loss": 1.244,
+      "step": 2030
+    },
+    {
+      "epoch": 1.476655808903366,
+      "grad_norm": 0.5406824946403503,
+      "learning_rate": 0.0002,
+      "loss": 1.3699,
+      "step": 2040
+    },
+    {
+      "epoch": 1.4838943177705393,
+      "grad_norm": 0.5990166664123535,
+      "learning_rate": 0.0002,
+      "loss": 1.3353,
+      "step": 2050
+    },
+    {
+      "epoch": 1.4911328266377126,
+      "grad_norm": 0.6186193823814392,
+      "learning_rate": 0.0002,
+      "loss": 1.2454,
+      "step": 2060
+    },
+    {
+      "epoch": 1.498371335504886,
+      "grad_norm": 0.6154307126998901,
+      "learning_rate": 0.0002,
+      "loss": 1.428,
+      "step": 2070
+    },
+    {
+      "epoch": 1.5056098443720594,
+      "grad_norm": 0.5606056451797485,
+      "learning_rate": 0.0002,
+      "loss": 1.4528,
+      "step": 2080
+    },
+    {
+      "epoch": 1.5128483532392327,
+      "grad_norm": 0.5006417036056519,
+      "learning_rate": 0.0002,
+      "loss": 1.2405,
+      "step": 2090
+    },
+    {
+      "epoch": 1.520086862106406,
+      "grad_norm": 0.5968486070632935,
+      "learning_rate": 0.0002,
+      "loss": 1.4258,
+      "step": 2100
+    },
+    {
+      "epoch": 1.5273253709735795,
+      "grad_norm": 0.5835496187210083,
+      "learning_rate": 0.0002,
+      "loss": 1.2752,
+      "step": 2110
+    },
+    {
+      "epoch": 1.5345638798407528,
+      "grad_norm": 0.6753535270690918,
+      "learning_rate": 0.0002,
+      "loss": 1.5443,
+      "step": 2120
+    },
+    {
+      "epoch": 1.5418023887079262,
+      "grad_norm": 0.7299720644950867,
+      "learning_rate": 0.0002,
+      "loss": 1.2139,
+      "step": 2130
+    },
+    {
+      "epoch": 1.5490408975750996,
+      "grad_norm": 0.5105988383293152,
+      "learning_rate": 0.0002,
+      "loss": 1.2364,
+      "step": 2140
+    },
+    {
+      "epoch": 1.556279406442273,
+      "grad_norm": 0.5675431489944458,
+      "learning_rate": 0.0002,
+      "loss": 1.4528,
+      "step": 2150
+    },
+    {
+      "epoch": 1.5635179153094463,
+      "grad_norm": 0.6246723532676697,
+      "learning_rate": 0.0002,
+      "loss": 1.4563,
+      "step": 2160
+    },
+    {
+      "epoch": 1.5707564241766196,
+      "grad_norm": 0.7291720509529114,
+      "learning_rate": 0.0002,
+      "loss": 1.5255,
+      "step": 2170
+    },
+    {
+      "epoch": 1.577994933043793,
+      "grad_norm": 0.678114116191864,
+      "learning_rate": 0.0002,
+      "loss": 1.5432,
+      "step": 2180
+    },
+    {
+      "epoch": 1.5852334419109664,
+      "grad_norm": 0.5136260986328125,
+      "learning_rate": 0.0002,
+      "loss": 1.5212,
+      "step": 2190
+    },
+    {
+      "epoch": 1.5924719507781397,
+      "grad_norm": 0.6359935998916626,
+      "learning_rate": 0.0002,
+      "loss": 1.3271,
+      "step": 2200
+    },
+    {
+      "epoch": 1.599710459645313,
+      "grad_norm": 0.7650278806686401,
+      "learning_rate": 0.0002,
+      "loss": 1.4038,
+      "step": 2210
+    },
+    {
+      "epoch": 1.6069489685124865,
+      "grad_norm": 0.7256110906600952,
+      "learning_rate": 0.0002,
+      "loss": 1.5478,
+      "step": 2220
+    },
+    {
+      "epoch": 1.6141874773796598,
+      "grad_norm": 0.688689649105072,
+      "learning_rate": 0.0002,
+      "loss": 1.4387,
+      "step": 2230
+    },
+    {
+      "epoch": 1.6214259862468332,
+      "grad_norm": 0.6045311093330383,
+      "learning_rate": 0.0002,
+      "loss": 1.4096,
+      "step": 2240
+    },
+    {
+      "epoch": 1.6286644951140063,
+      "grad_norm": 0.7064604163169861,
+      "learning_rate": 0.0002,
+      "loss": 1.4097,
+      "step": 2250
+    },
+    {
+      "epoch": 1.6359030039811797,
+      "grad_norm": 0.5309562087059021,
+      "learning_rate": 0.0002,
+      "loss": 1.3477,
+      "step": 2260
+    },
+    {
+      "epoch": 1.643141512848353,
+      "grad_norm": 0.5687053203582764,
+      "learning_rate": 0.0002,
+      "loss": 1.4022,
+      "step": 2270
+    },
+    {
+      "epoch": 1.6503800217155264,
+      "grad_norm": 0.535872757434845,
+      "learning_rate": 0.0002,
+      "loss": 1.2977,
+      "step": 2280
+    },
+    {
+      "epoch": 1.6576185305826998,
+      "grad_norm": 0.5502381920814514,
+      "learning_rate": 0.0002,
+      "loss": 1.3844,
+      "step": 2290
+    },
+    {
+      "epoch": 1.6648570394498732,
+      "grad_norm": 0.6158602237701416,
+      "learning_rate": 0.0002,
+      "loss": 1.3764,
+      "step": 2300
+    },
+    {
+      "epoch": 1.6720955483170465,
+      "grad_norm": 0.5804675817489624,
+      "learning_rate": 0.0002,
+      "loss": 1.3515,
+      "step": 2310
+    },
+    {
+      "epoch": 1.67933405718422,
+      "grad_norm": 0.600742757320404,
+      "learning_rate": 0.0002,
+      "loss": 1.2532,
+      "step": 2320
+    },
+    {
+      "epoch": 1.6865725660513933,
+      "grad_norm": 0.7101941108703613,
+      "learning_rate": 0.0002,
+      "loss": 1.477,
+      "step": 2330
+    },
+    {
+      "epoch": 1.6938110749185666,
+      "grad_norm": 0.7507809996604919,
+      "learning_rate": 0.0002,
+      "loss": 1.4849,
+      "step": 2340
+    },
+    {
+      "epoch": 1.70104958378574,
+      "grad_norm": 0.768502414226532,
+      "learning_rate": 0.0002,
+      "loss": 1.2703,
+      "step": 2350
+    },
+    {
+      "epoch": 1.7082880926529134,
+      "grad_norm": 0.4801851212978363,
+      "learning_rate": 0.0002,
+      "loss": 1.3332,
+      "step": 2360
+    },
+    {
+      "epoch": 1.7155266015200867,
+      "grad_norm": 0.5322122573852539,
+      "learning_rate": 0.0002,
+      "loss": 1.4158,
+      "step": 2370
+    },
+    {
+      "epoch": 1.72276511038726,
+      "grad_norm": 0.587661862373352,
+      "learning_rate": 0.0002,
+      "loss": 1.4136,
+      "step": 2380
+    },
+    {
+      "epoch": 1.7300036192544335,
+      "grad_norm": 0.6073525547981262,
+      "learning_rate": 0.0002,
+      "loss": 1.3771,
+      "step": 2390
+    },
+    {
+      "epoch": 1.7372421281216068,
+      "grad_norm": 0.6950460076332092,
+      "learning_rate": 0.0002,
+      "loss": 1.2754,
+      "step": 2400
+    },
+    {
+      "epoch": 1.7444806369887802,
+      "grad_norm": 0.5981102585792542,
+      "learning_rate": 0.0002,
+      "loss": 1.3858,
+      "step": 2410
+    },
+    {
+      "epoch": 1.7517191458559536,
+      "grad_norm": 0.544570803642273,
+      "learning_rate": 0.0002,
+      "loss": 1.4075,
+      "step": 2420
+    },
+    {
+      "epoch": 1.758957654723127,
+      "grad_norm": 0.5304399728775024,
+      "learning_rate": 0.0002,
+      "loss": 1.3861,
+      "step": 2430
+    },
+    {
+      "epoch": 1.7661961635903003,
+      "grad_norm": 0.7921594977378845,
+      "learning_rate": 0.0002,
+      "loss": 1.4244,
+      "step": 2440
+    },
+    {
+      "epoch": 1.7734346724574737,
+      "grad_norm": 0.6084808707237244,
+      "learning_rate": 0.0002,
+      "loss": 1.3053,
+      "step": 2450
+    },
+    {
+      "epoch": 1.780673181324647,
+      "grad_norm": 0.8844701051712036,
+      "learning_rate": 0.0002,
+      "loss": 1.3781,
+      "step": 2460
+    },
+    {
+      "epoch": 1.7879116901918204,
+      "grad_norm": 0.5729258060455322,
+      "learning_rate": 0.0002,
+      "loss": 1.3227,
+      "step": 2470
+    },
+    {
+      "epoch": 1.7951501990589938,
+      "grad_norm": 0.6303611993789673,
+      "learning_rate": 0.0002,
+      "loss": 1.3422,
+      "step": 2480
+    },
+    {
+      "epoch": 1.8023887079261671,
+      "grad_norm": 0.5627942085266113,
+      "learning_rate": 0.0002,
+      "loss": 1.3926,
+      "step": 2490
+    },
+    {
+      "epoch": 1.8096272167933405,
+      "grad_norm": 0.6724274158477783,
+      "learning_rate": 0.0002,
+      "loss": 1.3816,
+      "step": 2500
+    },
+    {
+      "epoch": 1.8168657256605139,
+      "grad_norm": 0.5030826330184937,
+      "learning_rate": 0.0002,
+      "loss": 1.2951,
+      "step": 2510
+    },
+    {
+      "epoch": 1.8241042345276872,
+      "grad_norm": 0.5504099130630493,
+      "learning_rate": 0.0002,
+      "loss": 1.2839,
+      "step": 2520
+    },
+    {
+      "epoch": 1.8313427433948606,
+      "grad_norm": 0.6338945627212524,
+      "learning_rate": 0.0002,
+      "loss": 1.4264,
+      "step": 2530
+    },
+    {
+      "epoch": 1.838581252262034,
+      "grad_norm": 0.5902037620544434,
+      "learning_rate": 0.0002,
+      "loss": 1.563,
+      "step": 2540
+    },
+    {
+      "epoch": 1.8458197611292073,
+      "grad_norm": 0.48814457654953003,
+      "learning_rate": 0.0002,
+      "loss": 1.2961,
+      "step": 2550
+    },
+    {
+      "epoch": 1.8530582699963807,
+      "grad_norm": 0.6216312646865845,
+      "learning_rate": 0.0002,
+      "loss": 1.466,
+      "step": 2560
+    },
+    {
+      "epoch": 1.860296778863554,
+      "grad_norm": 0.635603666305542,
+      "learning_rate": 0.0002,
+      "loss": 1.5123,
+      "step": 2570
+    },
+    {
+      "epoch": 1.8675352877307274,
+      "grad_norm": 0.6938216090202332,
+      "learning_rate": 0.0002,
+      "loss": 1.372,
+      "step": 2580
+    },
+    {
+      "epoch": 1.8747737965979008,
+      "grad_norm": 0.599557638168335,
+      "learning_rate": 0.0002,
+      "loss": 1.5011,
+      "step": 2590
+    },
+    {
+      "epoch": 1.8820123054650741,
+      "grad_norm": 0.564424455165863,
+      "learning_rate": 0.0002,
+      "loss": 1.2714,
+      "step": 2600
+    },
+    {
+      "epoch": 1.8892508143322475,
+      "grad_norm": 0.5430700182914734,
+      "learning_rate": 0.0002,
+      "loss": 1.3403,
+      "step": 2610
+    },
+    {
+      "epoch": 1.8964893231994209,
+      "grad_norm": 0.6150169372558594,
+      "learning_rate": 0.0002,
+      "loss": 1.4347,
+      "step": 2620
+    },
+    {
+      "epoch": 1.9037278320665942,
+      "grad_norm": 0.48159119486808777,
+      "learning_rate": 0.0002,
+      "loss": 1.2474,
+      "step": 2630
+    },
+    {
+      "epoch": 1.9109663409337676,
+      "grad_norm": 0.5608997941017151,
+      "learning_rate": 0.0002,
+      "loss": 1.3716,
+      "step": 2640
+    },
+    {
+      "epoch": 1.918204849800941,
+      "grad_norm": 0.6454501748085022,
+      "learning_rate": 0.0002,
+      "loss": 1.5787,
+      "step": 2650
+    },
+    {
+      "epoch": 1.9254433586681143,
+      "grad_norm": 0.5458073616027832,
+      "learning_rate": 0.0002,
+      "loss": 1.3238,
+      "step": 2660
+    },
+    {
+      "epoch": 1.9326818675352877,
+      "grad_norm": 0.5328490734100342,
+      "learning_rate": 0.0002,
+      "loss": 1.3208,
+      "step": 2670
+    },
+    {
+      "epoch": 1.939920376402461,
+      "grad_norm": 0.6444696187973022,
+      "learning_rate": 0.0002,
+      "loss": 1.4971,
+      "step": 2680
+    },
+    {
+      "epoch": 1.9471588852696344,
+      "grad_norm": 0.7126023769378662,
+      "learning_rate": 0.0002,
+      "loss": 1.5387,
+      "step": 2690
+    },
+    {
+      "epoch": 1.9543973941368078,
+      "grad_norm": 0.5164045095443726,
+      "learning_rate": 0.0002,
+      "loss": 1.3637,
+      "step": 2700
+    },
+    {
+      "epoch": 1.9616359030039812,
+      "grad_norm": 0.5347061157226562,
+      "learning_rate": 0.0002,
+      "loss": 1.5303,
+      "step": 2710
+    },
+    {
+      "epoch": 1.9688744118711545,
+      "grad_norm": 0.5297950506210327,
+      "learning_rate": 0.0002,
+      "loss": 1.2815,
+      "step": 2720
+    },
+    {
+      "epoch": 1.976112920738328,
+      "grad_norm": 0.6537790298461914,
+      "learning_rate": 0.0002,
+      "loss": 1.3566,
+      "step": 2730
+    },
+    {
+      "epoch": 1.9833514296055013,
+      "grad_norm": 0.5536222457885742,
+      "learning_rate": 0.0002,
+      "loss": 1.332,
+      "step": 2740
+    },
+    {
+      "epoch": 1.9905899384726746,
+      "grad_norm": 0.4856105446815491,
+      "learning_rate": 0.0002,
+      "loss": 1.3333,
+      "step": 2750
+    },
+    {
+      "epoch": 1.997828447339848,
+      "grad_norm": 0.6642730832099915,
+      "learning_rate": 0.0002,
+      "loss": 1.3521,
+      "step": 2760
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.4366681575775146,
+      "eval_runtime": 27.3729,
+      "eval_samples_per_second": 15.928,
+      "eval_steps_per_second": 2.009,
+      "step": 2763
+    },
+    {
+      "epoch": 2.0050669562070214,
+      "grad_norm": 0.740253210067749,
+      "learning_rate": 0.0002,
+      "loss": 1.4322,
+      "step": 2770
+    },
+    {
+      "epoch": 2.0123054650741947,
+      "grad_norm": 0.5826276540756226,
+      "learning_rate": 0.0002,
+      "loss": 1.277,
+      "step": 2780
+    },
+    {
+      "epoch": 2.019543973941368,
+      "grad_norm": 0.607356071472168,
+      "learning_rate": 0.0002,
+      "loss": 1.2424,
+      "step": 2790
+    },
+    {
+      "epoch": 2.0267824828085415,
+      "grad_norm": 0.5918063521385193,
+      "learning_rate": 0.0002,
+      "loss": 1.2601,
+      "step": 2800
+    },
+    {
+      "epoch": 2.034020991675715,
+      "grad_norm": 0.5610089898109436,
+      "learning_rate": 0.0002,
+      "loss": 1.3715,
+      "step": 2810
+    },
+    {
+      "epoch": 2.041259500542888,
+      "grad_norm": 0.5869926810264587,
+      "learning_rate": 0.0002,
+      "loss": 1.2092,
+      "step": 2820
+    },
+    {
+      "epoch": 2.0484980094100615,
+      "grad_norm": 0.5753467679023743,
+      "learning_rate": 0.0002,
+      "loss": 1.1929,
+      "step": 2830
+    },
+    {
+      "epoch": 2.055736518277235,
+      "grad_norm": 0.7096508145332336,
+      "learning_rate": 0.0002,
+      "loss": 1.333,
+      "step": 2840
+    },
+    {
+      "epoch": 2.0629750271444083,
+      "grad_norm": 0.7653635144233704,
+      "learning_rate": 0.0002,
+      "loss": 1.1766,
+      "step": 2850
+    },
+    {
+      "epoch": 2.0702135360115816,
+      "grad_norm": 0.6202841997146606,
+      "learning_rate": 0.0002,
+      "loss": 1.2331,
+      "step": 2860
+    },
+    {
+      "epoch": 2.077452044878755,
+      "grad_norm": 0.6810227632522583,
+      "learning_rate": 0.0002,
+      "loss": 1.3298,
+      "step": 2870
+    },
+    {
+      "epoch": 2.0846905537459284,
+      "grad_norm": 0.7481493353843689,
+      "learning_rate": 0.0002,
+      "loss": 1.2505,
+      "step": 2880
+    },
+    {
+      "epoch": 2.0919290626131017,
+      "grad_norm": 0.7089637517929077,
+      "learning_rate": 0.0002,
+      "loss": 1.2484,
+      "step": 2890
+    },
+    {
+      "epoch": 2.099167571480275,
+      "grad_norm": 0.7472923398017883,
+      "learning_rate": 0.0002,
+      "loss": 1.3095,
+      "step": 2900
+    },
+    {
+      "epoch": 2.1064060803474485,
+      "grad_norm": 0.8135465979576111,
+      "learning_rate": 0.0002,
+      "loss": 1.304,
+      "step": 2910
+    },
+    {
+      "epoch": 2.113644589214622,
+      "grad_norm": 0.6097133159637451,
+      "learning_rate": 0.0002,
+      "loss": 1.273,
+      "step": 2920
+    },
+    {
+      "epoch": 2.120883098081795,
+      "grad_norm": 0.5970117449760437,
+      "learning_rate": 0.0002,
+      "loss": 1.3384,
+      "step": 2930
+    },
+    {
+      "epoch": 2.1281216069489686,
+      "grad_norm": 0.6169309616088867,
+      "learning_rate": 0.0002,
+      "loss": 1.3233,
+      "step": 2940
+    },
+    {
+      "epoch": 2.135360115816142,
+      "grad_norm": 0.9428738355636597,
+      "learning_rate": 0.0002,
+      "loss": 1.4246,
+      "step": 2950
+    },
+    {
+      "epoch": 2.1425986246833153,
+      "grad_norm": 0.5671679973602295,
+      "learning_rate": 0.0002,
+      "loss": 1.3527,
+      "step": 2960
+    },
+    {
+      "epoch": 2.1498371335504887,
+      "grad_norm": 0.7007262110710144,
+      "learning_rate": 0.0002,
+      "loss": 1.1375,
+      "step": 2970
+    },
+    {
+      "epoch": 2.157075642417662,
+      "grad_norm": 0.6294044256210327,
+      "learning_rate": 0.0002,
+      "loss": 1.2015,
+      "step": 2980
+    },
+    {
+      "epoch": 2.1643141512848354,
+      "grad_norm": 0.6105241775512695,
+      "learning_rate": 0.0002,
+      "loss": 1.2167,
+      "step": 2990
+    },
+    {
+      "epoch": 2.1715526601520088,
+      "grad_norm": 0.557124137878418,
+      "learning_rate": 0.0002,
+      "loss": 1.2065,
+      "step": 3000
+    },
+    {
+      "epoch": 2.178791169019182,
+      "grad_norm": 0.6250392198562622,
+      "learning_rate": 0.0002,
+      "loss": 1.2515,
+      "step": 3010
+    },
+    {
+      "epoch": 2.1860296778863555,
+      "grad_norm": 0.645218551158905,
+      "learning_rate": 0.0002,
+      "loss": 1.385,
+      "step": 3020
+    },
+    {
+      "epoch": 2.193268186753529,
+      "grad_norm": 0.9033605456352234,
+      "learning_rate": 0.0002,
+      "loss": 1.3928,
+      "step": 3030
+    },
+    {
+      "epoch": 2.2005066956207022,
+      "grad_norm": 0.5325747132301331,
+      "learning_rate": 0.0002,
+      "loss": 1.2458,
+      "step": 3040
+    },
+    {
+      "epoch": 2.2077452044878756,
+      "grad_norm": 0.6334700584411621,
+      "learning_rate": 0.0002,
+      "loss": 1.261,
+      "step": 3050
+    },
+    {
+      "epoch": 2.214983713355049,
+      "grad_norm": 0.5206325054168701,
+      "learning_rate": 0.0002,
+      "loss": 1.2385,
+      "step": 3060
+    },
+    {
+      "epoch": 2.2222222222222223,
+      "grad_norm": 0.5987200140953064,
+      "learning_rate": 0.0002,
+      "loss": 1.3103,
+      "step": 3070
+    },
+    {
+      "epoch": 2.2294607310893957,
+      "grad_norm": 0.5893264412879944,
+      "learning_rate": 0.0002,
+      "loss": 1.1756,
+      "step": 3080
+    },
+    {
+      "epoch": 2.236699239956569,
+      "grad_norm": 0.6869237422943115,
+      "learning_rate": 0.0002,
+      "loss": 1.235,
+      "step": 3090
+    },
+    {
+      "epoch": 2.2439377488237424,
+      "grad_norm": 0.5040048360824585,
+      "learning_rate": 0.0002,
+      "loss": 1.3285,
+      "step": 3100
+    },
+    {
+      "epoch": 2.251176257690916,
+      "grad_norm": 0.6660613417625427,
+      "learning_rate": 0.0002,
+      "loss": 1.3316,
+      "step": 3110
+    },
+    {
+      "epoch": 2.258414766558089,
+      "grad_norm": 0.5890918970108032,
+      "learning_rate": 0.0002,
+      "loss": 1.3108,
+      "step": 3120
+    },
+    {
+      "epoch": 2.2656532754252625,
+      "grad_norm": 0.6458896994590759,
+      "learning_rate": 0.0002,
+      "loss": 1.248,
+      "step": 3130
+    },
+    {
+      "epoch": 2.272891784292436,
+      "grad_norm": 0.6832690834999084,
+      "learning_rate": 0.0002,
+      "loss": 1.4151,
+      "step": 3140
+    },
+    {
+      "epoch": 2.2801302931596092,
+      "grad_norm": 0.833908200263977,
+      "learning_rate": 0.0002,
+      "loss": 1.4458,
+      "step": 3150
+    },
+    {
+      "epoch": 2.2873688020267826,
+      "grad_norm": 0.4596034586429596,
+      "learning_rate": 0.0002,
+      "loss": 1.2931,
+      "step": 3160
+    },
+    {
+      "epoch": 2.294607310893956,
+      "grad_norm": 0.9130966067314148,
+      "learning_rate": 0.0002,
+      "loss": 1.449,
+      "step": 3170
+    },
+    {
+      "epoch": 2.3018458197611293,
+      "grad_norm": 0.7143292427062988,
+      "learning_rate": 0.0002,
+      "loss": 1.3806,
+      "step": 3180
+    },
+    {
+      "epoch": 2.3090843286283027,
+      "grad_norm": 0.5388900637626648,
+      "learning_rate": 0.0002,
+      "loss": 1.2692,
+      "step": 3190
+    },
+    {
+      "epoch": 2.316322837495476,
+      "grad_norm": 0.5607513189315796,
+      "learning_rate": 0.0002,
+      "loss": 1.2402,
+      "step": 3200
+    },
+    {
+      "epoch": 2.3235613463626494,
+      "grad_norm": 0.6795142292976379,
+      "learning_rate": 0.0002,
+      "loss": 1.3874,
+      "step": 3210
+    },
+    {
+      "epoch": 2.330799855229823,
+      "grad_norm": 0.6561070680618286,
+      "learning_rate": 0.0002,
+      "loss": 1.3042,
+      "step": 3220
+    },
+    {
+      "epoch": 2.338038364096996,
+      "grad_norm": 0.8858118057250977,
+      "learning_rate": 0.0002,
+      "loss": 1.4636,
+      "step": 3230
+    },
+    {
+      "epoch": 2.3452768729641695,
+      "grad_norm": 0.6604151725769043,
+      "learning_rate": 0.0002,
+      "loss": 1.3214,
+      "step": 3240
+    },
+    {
+      "epoch": 2.352515381831343,
+      "grad_norm": 0.6755785346031189,
+      "learning_rate": 0.0002,
+      "loss": 1.4004,
+      "step": 3250
+    },
+    {
+      "epoch": 2.3597538906985163,
+      "grad_norm": 0.6981677412986755,
+      "learning_rate": 0.0002,
+      "loss": 1.2503,
+      "step": 3260
+    },
+    {
+      "epoch": 2.3669923995656896,
+      "grad_norm": 0.6338568329811096,
+      "learning_rate": 0.0002,
+      "loss": 1.3078,
+      "step": 3270
+    },
+    {
+      "epoch": 2.374230908432863,
+      "grad_norm": 0.5754265785217285,
+      "learning_rate": 0.0002,
+      "loss": 1.285,
+      "step": 3280
+    },
+    {
+      "epoch": 2.3814694173000364,
+      "grad_norm": 0.7533153295516968,
+      "learning_rate": 0.0002,
+      "loss": 1.2924,
+      "step": 3290
+    },
+    {
+      "epoch": 2.3887079261672097,
+      "grad_norm": 0.675065279006958,
+      "learning_rate": 0.0002,
+      "loss": 1.3711,
+      "step": 3300
+    },
+    {
+      "epoch": 2.395946435034383,
+      "grad_norm": 0.5686452984809875,
+      "learning_rate": 0.0002,
+      "loss": 1.3548,
+      "step": 3310
+    },
+    {
+      "epoch": 2.4031849439015565,
+      "grad_norm": 0.8129481673240662,
+      "learning_rate": 0.0002,
+      "loss": 1.1998,
+      "step": 3320
+    },
+    {
+      "epoch": 2.41042345276873,
+      "grad_norm": 0.6615934371948242,
+      "learning_rate": 0.0002,
+      "loss": 1.2584,
+      "step": 3330
+    },
+    {
+      "epoch": 2.417661961635903,
+      "grad_norm": 0.6678834557533264,
+      "learning_rate": 0.0002,
+      "loss": 1.3691,
+      "step": 3340
+    },
+    {
+      "epoch": 2.4249004705030766,
+      "grad_norm": 0.5581308007240295,
+      "learning_rate": 0.0002,
+      "loss": 1.2381,
+      "step": 3350
+    },
+    {
+      "epoch": 2.43213897937025,
+      "grad_norm": 0.6098920106887817,
+      "learning_rate": 0.0002,
+      "loss": 1.3853,
+      "step": 3360
+    },
+    {
+      "epoch": 2.4393774882374233,
+      "grad_norm": 0.8101736903190613,
+      "learning_rate": 0.0002,
+      "loss": 1.3692,
+      "step": 3370
+    },
+    {
+      "epoch": 2.4466159971045967,
+      "grad_norm": 0.6621488928794861,
+      "learning_rate": 0.0002,
+      "loss": 1.4418,
+      "step": 3380
+    },
+    {
+      "epoch": 2.45385450597177,
+      "grad_norm": 0.8693289160728455,
+      "learning_rate": 0.0002,
+      "loss": 1.4579,
+      "step": 3390
+    },
+    {
+      "epoch": 2.4610930148389434,
+      "grad_norm": 0.6724580526351929,
+      "learning_rate": 0.0002,
+      "loss": 1.3644,
+      "step": 3400
+    },
+    {
+      "epoch": 2.4683315237061167,
+      "grad_norm": 0.6776891946792603,
+      "learning_rate": 0.0002,
+      "loss": 1.2006,
+      "step": 3410
+    },
+    {
+      "epoch": 2.47557003257329,
+      "grad_norm": 0.7214453816413879,
+      "learning_rate": 0.0002,
+      "loss": 1.2937,
+      "step": 3420
+    },
+    {
+      "epoch": 2.4828085414404635,
+      "grad_norm": 0.8390451073646545,
+      "learning_rate": 0.0002,
+      "loss": 1.4051,
+      "step": 3430
+    },
+    {
+      "epoch": 2.490047050307637,
+      "grad_norm": 0.7130982279777527,
+      "learning_rate": 0.0002,
+      "loss": 1.25,
+      "step": 3440
+    },
+    {
+      "epoch": 2.49728555917481,
+      "grad_norm": 0.8873937129974365,
+      "learning_rate": 0.0002,
+      "loss": 1.2231,
+      "step": 3450
+    },
+    {
+      "epoch": 2.5045240680419836,
+      "grad_norm": 0.725185751914978,
+      "learning_rate": 0.0002,
+      "loss": 1.1429,
+      "step": 3460
+    },
+    {
+      "epoch": 2.511762576909157,
+      "grad_norm": 0.6120352149009705,
+      "learning_rate": 0.0002,
+      "loss": 1.2699,
+      "step": 3470
+    },
+    {
+      "epoch": 2.5190010857763303,
+      "grad_norm": 0.7713613510131836,
+      "learning_rate": 0.0002,
+      "loss": 1.2552,
+      "step": 3480
+    },
+    {
+      "epoch": 2.5262395946435037,
+      "grad_norm": 0.895309567451477,
+      "learning_rate": 0.0002,
+      "loss": 1.4648,
+      "step": 3490
+    },
+    {
+      "epoch": 2.533478103510677,
+      "grad_norm": 0.9631021022796631,
+      "learning_rate": 0.0002,
+      "loss": 1.3043,
+      "step": 3500
+    },
+    {
+      "epoch": 2.5407166123778504,
+      "grad_norm": 0.7475683093070984,
+      "learning_rate": 0.0002,
+      "loss": 1.3492,
+      "step": 3510
+    },
+    {
+      "epoch": 2.5479551212450238,
+      "grad_norm": 0.7271341681480408,
+      "learning_rate": 0.0002,
+      "loss": 1.3637,
+      "step": 3520
+    },
+    {
+      "epoch": 2.555193630112197,
+      "grad_norm": 0.6979510188102722,
+      "learning_rate": 0.0002,
+      "loss": 1.304,
+      "step": 3530
+    },
+    {
+      "epoch": 2.5624321389793705,
+      "grad_norm": 0.6504196524620056,
+      "learning_rate": 0.0002,
+      "loss": 1.2353,
+      "step": 3540
+    },
+    {
+      "epoch": 2.569670647846544,
+      "grad_norm": 0.7226675748825073,
+      "learning_rate": 0.0002,
+      "loss": 1.2699,
+      "step": 3550
+    },
+    {
+      "epoch": 2.5769091567137172,
+      "grad_norm": 0.6143222451210022,
+      "learning_rate": 0.0002,
+      "loss": 1.3002,
+      "step": 3560
+    },
+    {
+      "epoch": 2.5841476655808906,
+      "grad_norm": 0.7245154976844788,
+      "learning_rate": 0.0002,
+      "loss": 1.1585,
+      "step": 3570
+    },
+    {
+      "epoch": 2.591386174448064,
+      "grad_norm": 0.943540632724762,
+      "learning_rate": 0.0002,
+      "loss": 1.3651,
+      "step": 3580
+    },
+    {
+      "epoch": 2.5986246833152373,
+      "grad_norm": 0.7707241773605347,
+      "learning_rate": 0.0002,
+      "loss": 1.3034,
+      "step": 3590
+    },
+    {
+      "epoch": 2.6058631921824107,
+      "grad_norm": 0.6705001592636108,
+      "learning_rate": 0.0002,
+      "loss": 1.3063,
+      "step": 3600
+    },
+    {
+      "epoch": 2.613101701049584,
+      "grad_norm": 0.6360933780670166,
+      "learning_rate": 0.0002,
+      "loss": 1.2437,
+      "step": 3610
+    },
+    {
+      "epoch": 2.6203402099167574,
+      "grad_norm": 0.5846424698829651,
+      "learning_rate": 0.0002,
+      "loss": 1.1844,
+      "step": 3620
+    },
+    {
+      "epoch": 2.6275787187839303,
+      "grad_norm": 0.5958625674247742,
+      "learning_rate": 0.0002,
+      "loss": 1.3674,
+      "step": 3630
+    },
+    {
+      "epoch": 2.6348172276511037,
+      "grad_norm": 0.6819243431091309,
+      "learning_rate": 0.0002,
+      "loss": 1.3599,
+      "step": 3640
+    },
+    {
+      "epoch": 2.642055736518277,
+      "grad_norm": 0.7033445835113525,
+      "learning_rate": 0.0002,
+      "loss": 1.3884,
+      "step": 3650
+    },
+    {
+      "epoch": 2.6492942453854504,
+      "grad_norm": 0.6134849786758423,
+      "learning_rate": 0.0002,
+      "loss": 1.3392,
+      "step": 3660
+    },
+    {
+      "epoch": 2.656532754252624,
+      "grad_norm": 0.658009946346283,
+      "learning_rate": 0.0002,
+      "loss": 1.2661,
+      "step": 3670
+    },
+    {
+      "epoch": 2.663771263119797,
+      "grad_norm": 0.6280999779701233,
+      "learning_rate": 0.0002,
+      "loss": 1.3987,
+      "step": 3680
+    },
+    {
+      "epoch": 2.6710097719869705,
+      "grad_norm": 0.5536085963249207,
+      "learning_rate": 0.0002,
+      "loss": 1.2995,
+      "step": 3690
+    },
+    {
+      "epoch": 2.678248280854144,
+      "grad_norm": 0.8603981733322144,
+      "learning_rate": 0.0002,
+      "loss": 1.2044,
+      "step": 3700
+    },
+    {
+      "epoch": 2.6854867897213173,
+      "grad_norm": 0.5509994626045227,
+      "learning_rate": 0.0002,
+      "loss": 1.3879,
+      "step": 3710
+    },
+    {
+      "epoch": 2.6927252985884906,
+      "grad_norm": 0.9093621969223022,
+      "learning_rate": 0.0002,
+      "loss": 1.3253,
+      "step": 3720
+    },
+    {
+      "epoch": 2.699963807455664,
+      "grad_norm": 0.7525952458381653,
+      "learning_rate": 0.0002,
+      "loss": 1.2668,
+      "step": 3730
+    },
+    {
+      "epoch": 2.7072023163228374,
+      "grad_norm": 0.6737023591995239,
+      "learning_rate": 0.0002,
+      "loss": 1.248,
+      "step": 3740
+    },
+    {
+      "epoch": 2.7144408251900107,
+      "grad_norm": 0.8656924962997437,
+      "learning_rate": 0.0002,
+      "loss": 1.2981,
+      "step": 3750
+    },
+    {
+      "epoch": 2.721679334057184,
+      "grad_norm": 0.7494133114814758,
+      "learning_rate": 0.0002,
+      "loss": 1.2342,
+      "step": 3760
+    },
+    {
+      "epoch": 2.7289178429243575,
+      "grad_norm": 0.5725520849227905,
+      "learning_rate": 0.0002,
+      "loss": 1.2417,
+      "step": 3770
+    },
+    {
+      "epoch": 2.736156351791531,
+      "grad_norm": 0.836412787437439,
+      "learning_rate": 0.0002,
+      "loss": 1.28,
+      "step": 3780
+    },
+    {
+      "epoch": 2.743394860658704,
+      "grad_norm": 0.6893242597579956,
+      "learning_rate": 0.0002,
+      "loss": 1.3784,
+      "step": 3790
+    },
+    {
+      "epoch": 2.7506333695258776,
+      "grad_norm": 0.6696223020553589,
+      "learning_rate": 0.0002,
+      "loss": 1.2929,
+      "step": 3800
+    },
+    {
+      "epoch": 2.757871878393051,
+      "grad_norm": 0.6483015418052673,
+      "learning_rate": 0.0002,
+      "loss": 1.2449,
+      "step": 3810
+    },
+    {
+      "epoch": 2.7651103872602243,
+      "grad_norm": 0.8084456920623779,
+      "learning_rate": 0.0002,
+      "loss": 1.3282,
+      "step": 3820
+    },
+    {
+      "epoch": 2.7723488961273977,
+      "grad_norm": 0.6601949334144592,
+      "learning_rate": 0.0002,
+      "loss": 1.3694,
+      "step": 3830
+    },
+    {
+      "epoch": 2.779587404994571,
+      "grad_norm": 0.6905533671379089,
+      "learning_rate": 0.0002,
+      "loss": 1.3568,
+      "step": 3840
+    },
+    {
+      "epoch": 2.7868259138617444,
+      "grad_norm": 0.619318425655365,
+      "learning_rate": 0.0002,
+      "loss": 1.3854,
+      "step": 3850
+    },
+    {
+      "epoch": 2.7940644227289178,
+      "grad_norm": 0.5994023084640503,
+      "learning_rate": 0.0002,
+      "loss": 1.2551,
+      "step": 3860
+    },
+    {
+      "epoch": 2.801302931596091,
+      "grad_norm": 0.5627168416976929,
+      "learning_rate": 0.0002,
+      "loss": 1.2022,
+      "step": 3870
+    },
+    {
+      "epoch": 2.8085414404632645,
+      "grad_norm": 0.6001605987548828,
+      "learning_rate": 0.0002,
+      "loss": 1.3921,
+      "step": 3880
+    },
+    {
+      "epoch": 2.815779949330438,
+      "grad_norm": 0.6022412776947021,
+      "learning_rate": 0.0002,
+      "loss": 1.3026,
+      "step": 3890
+    },
+    {
+      "epoch": 2.823018458197611,
+      "grad_norm": 0.6832426190376282,
+      "learning_rate": 0.0002,
+      "loss": 1.2765,
+      "step": 3900
+    },
+    {
+      "epoch": 2.8302569670647846,
+      "grad_norm": 0.5936811566352844,
+      "learning_rate": 0.0002,
+      "loss": 1.1363,
+      "step": 3910
+    },
+    {
+      "epoch": 2.837495475931958,
+      "grad_norm": 0.6960572600364685,
+      "learning_rate": 0.0002,
+      "loss": 1.1707,
+      "step": 3920
+    },
+    {
+      "epoch": 2.8447339847991313,
+      "grad_norm": 0.5913406610488892,
+      "learning_rate": 0.0002,
+      "loss": 1.4063,
+      "step": 3930
+    },
+    {
+      "epoch": 2.8519724936663047,
+      "grad_norm": 0.678154706954956,
+      "learning_rate": 0.0002,
+      "loss": 1.3245,
+      "step": 3940
+    },
+    {
+      "epoch": 2.859211002533478,
+      "grad_norm": 0.7898936867713928,
+      "learning_rate": 0.0002,
+      "loss": 1.366,
+      "step": 3950
+    },
+    {
+      "epoch": 2.8664495114006514,
+      "grad_norm": 0.9234195351600647,
+      "learning_rate": 0.0002,
+      "loss": 1.3948,
+      "step": 3960
+    },
+    {
+      "epoch": 2.8736880202678248,
+      "grad_norm": 0.5960825085639954,
+      "learning_rate": 0.0002,
+      "loss": 1.2773,
+      "step": 3970
+    },
+    {
+      "epoch": 2.880926529134998,
+      "grad_norm": 0.677118182182312,
+      "learning_rate": 0.0002,
+      "loss": 1.3127,
+      "step": 3980
+    },
+    {
+      "epoch": 2.8881650380021715,
+      "grad_norm": 0.6505142450332642,
+      "learning_rate": 0.0002,
+      "loss": 1.2652,
+      "step": 3990
+    },
+    {
+      "epoch": 2.895403546869345,
+      "grad_norm": 0.550826907157898,
+      "learning_rate": 0.0002,
+      "loss": 1.2078,
+      "step": 4000
+    },
+    {
+      "epoch": 2.9026420557365182,
+      "grad_norm": 0.6209215521812439,
+      "learning_rate": 0.0002,
+      "loss": 1.1811,
+      "step": 4010
+    },
+    {
+      "epoch": 2.9098805646036916,
+      "grad_norm": 0.6549018025398254,
+      "learning_rate": 0.0002,
+      "loss": 1.4001,
+      "step": 4020
+    },
+    {
+      "epoch": 2.917119073470865,
+      "grad_norm": 0.570682168006897,
+      "learning_rate": 0.0002,
+      "loss": 1.2285,
+      "step": 4030
+    },
+    {
+      "epoch": 2.9243575823380383,
+      "grad_norm": 1.1807632446289062,
+      "learning_rate": 0.0002,
+      "loss": 1.0832,
+      "step": 4040
+    },
+    {
+      "epoch": 2.9315960912052117,
+      "grad_norm": 0.7058857679367065,
+      "learning_rate": 0.0002,
+      "loss": 1.2693,
+      "step": 4050
+    },
+    {
+      "epoch": 2.938834600072385,
+      "grad_norm": 0.5542812943458557,
+      "learning_rate": 0.0002,
+      "loss": 1.2905,
+      "step": 4060
+    },
+    {
+      "epoch": 2.9460731089395584,
+      "grad_norm": 0.63167804479599,
+      "learning_rate": 0.0002,
+      "loss": 1.33,
+      "step": 4070
+    },
+    {
+      "epoch": 2.953311617806732,
+      "grad_norm": 0.5702962279319763,
+      "learning_rate": 0.0002,
+      "loss": 1.3075,
+      "step": 4080
+    },
+    {
+      "epoch": 2.960550126673905,
+      "grad_norm": 0.620944082736969,
+      "learning_rate": 0.0002,
+      "loss": 1.2007,
+      "step": 4090
+    },
+    {
+      "epoch": 2.9677886355410785,
+      "grad_norm": 0.5866289734840393,
+      "learning_rate": 0.0002,
+      "loss": 1.2864,
+      "step": 4100
+    },
+    {
+      "epoch": 2.975027144408252,
+      "grad_norm": 0.560170590877533,
+      "learning_rate": 0.0002,
+      "loss": 1.3293,
+      "step": 4110
+    },
+    {
+      "epoch": 2.9822656532754253,
+      "grad_norm": 0.675082802772522,
+      "learning_rate": 0.0002,
+      "loss": 1.2071,
+      "step": 4120
+    },
+    {
+      "epoch": 2.9895041621425986,
+      "grad_norm": 0.62708580493927,
+      "learning_rate": 0.0002,
+      "loss": 1.2981,
+      "step": 4130
+    },
+    {
+      "epoch": 2.996742671009772,
+      "grad_norm": 0.7893929481506348,
+      "learning_rate": 0.0002,
+      "loss": 1.2758,
+      "step": 4140
+    },
+    {
+      "epoch": 2.9996380745566413,
+      "eval_loss": 1.4217946529388428,
+      "eval_runtime": 27.1596,
+      "eval_samples_per_second": 16.053,
+      "eval_steps_per_second": 2.025,
+      "step": 4144
+    },
+    {
+      "epoch": 3.0039811798769454,
+      "grad_norm": 0.7043836116790771,
+      "learning_rate": 0.0002,
+      "loss": 1.2152,
+      "step": 4150
+    },
+    {
+      "epoch": 3.0112196887441187,
+      "grad_norm": 0.6806283593177795,
+      "learning_rate": 0.0002,
+      "loss": 1.1664,
+      "step": 4160
+    },
+    {
+      "epoch": 3.018458197611292,
+      "grad_norm": 0.7684550285339355,
+      "learning_rate": 0.0002,
+      "loss": 1.292,
+      "step": 4170
+    },
+    {
+      "epoch": 3.0256967064784654,
+      "grad_norm": 0.7895237803459167,
+      "learning_rate": 0.0002,
+      "loss": 1.3467,
+      "step": 4180
+    },
+    {
+      "epoch": 3.032935215345639,
+      "grad_norm": 0.7464531064033508,
+      "learning_rate": 0.0002,
+      "loss": 1.1324,
+      "step": 4190
+    },
+    {
+      "epoch": 3.040173724212812,
+      "grad_norm": 0.9358500838279724,
+      "learning_rate": 0.0002,
+      "loss": 1.1614,
+      "step": 4200
+    },
+    {
+      "epoch": 3.0474122330799855,
+      "grad_norm": 1.1066628694534302,
+      "learning_rate": 0.0002,
+      "loss": 1.1834,
+      "step": 4210
+    },
+    {
+      "epoch": 3.054650741947159,
+      "grad_norm": 0.6663267612457275,
+      "learning_rate": 0.0002,
+      "loss": 1.1557,
+      "step": 4220
+    },
+    {
+      "epoch": 3.0618892508143323,
+      "grad_norm": 0.6669464707374573,
+      "learning_rate": 0.0002,
+      "loss": 1.1707,
+      "step": 4230
+    },
+    {
+      "epoch": 3.0691277596815056,
+      "grad_norm": 0.7052164077758789,
+      "learning_rate": 0.0002,
+      "loss": 1.1841,
+      "step": 4240
+    },
+    {
+      "epoch": 3.076366268548679,
+      "grad_norm": 0.6118432879447937,
+      "learning_rate": 0.0002,
+      "loss": 1.2913,
+      "step": 4250
+    },
+    {
+      "epoch": 3.0836047774158524,
+      "grad_norm": 0.6915903687477112,
+      "learning_rate": 0.0002,
+      "loss": 1.1526,
+      "step": 4260
+    },
+    {
+      "epoch": 3.0908432862830257,
+      "grad_norm": 0.7441644668579102,
+      "learning_rate": 0.0002,
+      "loss": 1.1348,
+      "step": 4270
+    },
+    {
+      "epoch": 3.098081795150199,
+      "grad_norm": 0.823850691318512,
+      "learning_rate": 0.0002,
+      "loss": 1.1672,
+      "step": 4280
+    },
+    {
+      "epoch": 3.1053203040173725,
+      "grad_norm": 0.9677883386611938,
+      "learning_rate": 0.0002,
+      "loss": 1.2655,
+      "step": 4290
+    },
+    {
+      "epoch": 3.112558812884546,
+      "grad_norm": 0.7002579569816589,
+      "learning_rate": 0.0002,
+      "loss": 1.1794,
+      "step": 4300
+    },
+    {
+      "epoch": 3.119797321751719,
+      "grad_norm": 0.778789758682251,
+      "learning_rate": 0.0002,
+      "loss": 1.135,
+      "step": 4310
+    },
+    {
+      "epoch": 3.1270358306188926,
+      "grad_norm": 0.7236007452011108,
+      "learning_rate": 0.0002,
+      "loss": 1.0818,
+      "step": 4320
+    },
+    {
+      "epoch": 3.134274339486066,
+      "grad_norm": 0.8809133768081665,
+      "learning_rate": 0.0002,
+      "loss": 1.1803,
+      "step": 4330
+    },
+    {
+      "epoch": 3.1415128483532393,
+      "grad_norm": 0.7924913167953491,
+      "learning_rate": 0.0002,
+      "loss": 1.2571,
+      "step": 4340
+    },
+    {
+      "epoch": 3.1487513572204127,
+      "grad_norm": 0.7437422275543213,
+      "learning_rate": 0.0002,
+      "loss": 1.1413,
+      "step": 4350
+    },
+    {
+      "epoch": 3.155989866087586,
+      "grad_norm": 0.6428450345993042,
+      "learning_rate": 0.0002,
+      "loss": 1.2088,
+      "step": 4360
+    },
+    {
+      "epoch": 3.1632283749547594,
+      "grad_norm": 0.7922873497009277,
+      "learning_rate": 0.0002,
+      "loss": 1.3032,
+      "step": 4370
+    },
+    {
+      "epoch": 3.1704668838219328,
+      "grad_norm": 0.5252506732940674,
+      "learning_rate": 0.0002,
+      "loss": 1.216,
+      "step": 4380
+    },
+    {
+      "epoch": 3.177705392689106,
+      "grad_norm": 0.8570457696914673,
+      "learning_rate": 0.0002,
+      "loss": 1.1297,
+      "step": 4390
+    },
+    {
+      "epoch": 3.1849439015562795,
+      "grad_norm": 0.7218987345695496,
+      "learning_rate": 0.0002,
+      "loss": 1.0994,
+      "step": 4400
+    },
+    {
+      "epoch": 3.192182410423453,
+      "grad_norm": 0.6921393275260925,
+      "learning_rate": 0.0002,
+      "loss": 1.2891,
+      "step": 4410
+    },
+    {
+      "epoch": 3.199420919290626,
+      "grad_norm": 0.7386137843132019,
+      "learning_rate": 0.0002,
+      "loss": 1.2668,
+      "step": 4420
+    },
+    {
+      "epoch": 3.2066594281577996,
+      "grad_norm": 0.6227759122848511,
+      "learning_rate": 0.0002,
+      "loss": 1.1654,
+      "step": 4430
+    },
+    {
+      "epoch": 3.213897937024973,
+      "grad_norm": 0.7180278897285461,
+      "learning_rate": 0.0002,
+      "loss": 1.1752,
+      "step": 4440
+    },
+    {
+      "epoch": 3.2211364458921463,
+      "grad_norm": 0.745830774307251,
+      "learning_rate": 0.0002,
+      "loss": 1.1757,
+      "step": 4450
+    },
+    {
+      "epoch": 3.2283749547593197,
+      "grad_norm": 0.6766072511672974,
+      "learning_rate": 0.0002,
+      "loss": 1.234,
+      "step": 4460
+    },
+    {
+      "epoch": 3.235613463626493,
+      "grad_norm": 0.8325067162513733,
+      "learning_rate": 0.0002,
+      "loss": 1.1999,
+      "step": 4470
+    },
+    {
+      "epoch": 3.2428519724936664,
+      "grad_norm": 0.7148305177688599,
+      "learning_rate": 0.0002,
+      "loss": 1.1606,
+      "step": 4480
+    },
+    {
+      "epoch": 3.25009048136084,
+      "grad_norm": 0.7752676010131836,
+      "learning_rate": 0.0002,
+      "loss": 1.1383,
+      "step": 4490
+    },
+    {
+      "epoch": 3.257328990228013,
+      "grad_norm": 0.6776860952377319,
+      "learning_rate": 0.0002,
+      "loss": 1.3006,
+      "step": 4500
+    },
+    {
+      "epoch": 3.2645674990951865,
+      "grad_norm": 0.704359769821167,
+      "learning_rate": 0.0002,
+      "loss": 1.0796,
+      "step": 4510
+    },
+    {
+      "epoch": 3.27180600796236,
+      "grad_norm": 0.6880282163619995,
+      "learning_rate": 0.0002,
+      "loss": 1.2496,
+      "step": 4520
+    },
+    {
+      "epoch": 3.2790445168295332,
+      "grad_norm": 0.8179270029067993,
+      "learning_rate": 0.0002,
+      "loss": 1.0947,
+      "step": 4530
+    },
+    {
+      "epoch": 3.2862830256967066,
+      "grad_norm": 0.6718448996543884,
+      "learning_rate": 0.0002,
+      "loss": 1.1909,
+      "step": 4540
+    },
+    {
+      "epoch": 3.29352153456388,
+      "grad_norm": 0.8300657868385315,
+      "learning_rate": 0.0002,
+      "loss": 1.2708,
+      "step": 4550
+    },
+    {
+      "epoch": 3.3007600434310533,
+      "grad_norm": 0.6433690786361694,
+      "learning_rate": 0.0002,
+      "loss": 1.2594,
+      "step": 4560
+    },
+    {
+      "epoch": 3.3079985522982267,
+      "grad_norm": 0.690262496471405,
+      "learning_rate": 0.0002,
+      "loss": 1.2479,
+      "step": 4570
+    },
+    {
+      "epoch": 3.3152370611654,
+      "grad_norm": 0.7022852301597595,
+      "learning_rate": 0.0002,
+      "loss": 1.1342,
+      "step": 4580
+    },
+    {
+      "epoch": 3.3224755700325734,
+      "grad_norm": 0.6438387632369995,
+      "learning_rate": 0.0002,
+      "loss": 1.0844,
+      "step": 4590
+    },
+    {
+      "epoch": 3.329714078899747,
+      "grad_norm": 0.6866899132728577,
+      "learning_rate": 0.0002,
+      "loss": 1.17,
+      "step": 4600
+    },
+    {
+      "epoch": 3.33695258776692,
+      "grad_norm": 0.8233968019485474,
+      "learning_rate": 0.0002,
+      "loss": 1.1289,
+      "step": 4610
+    },
+    {
+      "epoch": 3.3441910966340935,
+      "grad_norm": 0.7251574993133545,
+      "learning_rate": 0.0002,
+      "loss": 1.1855,
+      "step": 4620
+    },
+    {
+      "epoch": 3.351429605501267,
+      "grad_norm": 0.7855110168457031,
+      "learning_rate": 0.0002,
+      "loss": 1.3403,
+      "step": 4630
+    },
+    {
+      "epoch": 3.3586681143684403,
+      "grad_norm": 0.8487356305122375,
+      "learning_rate": 0.0002,
+      "loss": 1.2922,
+      "step": 4640
+    },
+    {
+      "epoch": 3.3659066232356136,
+      "grad_norm": 0.6429011225700378,
+      "learning_rate": 0.0002,
+      "loss": 1.2462,
+      "step": 4650
+    },
+    {
+      "epoch": 3.373145132102787,
+      "grad_norm": 0.7095270156860352,
+      "learning_rate": 0.0002,
+      "loss": 1.129,
+      "step": 4660
+    },
+    {
+      "epoch": 3.3803836409699604,
+      "grad_norm": 0.6792303323745728,
+      "learning_rate": 0.0002,
+      "loss": 1.262,
+      "step": 4670
+    },
+    {
+      "epoch": 3.3876221498371337,
+      "grad_norm": 0.6784825921058655,
+      "learning_rate": 0.0002,
+      "loss": 1.256,
+      "step": 4680
+    },
+    {
+      "epoch": 3.394860658704307,
+      "grad_norm": 0.6362888216972351,
+      "learning_rate": 0.0002,
+      "loss": 1.0838,
+      "step": 4690
+    },
+    {
+      "epoch": 3.4020991675714805,
+      "grad_norm": 0.7794778943061829,
+      "learning_rate": 0.0002,
+      "loss": 1.2165,
+      "step": 4700
+    },
+    {
+      "epoch": 3.409337676438654,
+      "grad_norm": 0.7287485003471375,
+      "learning_rate": 0.0002,
+      "loss": 1.0644,
+      "step": 4710
+    },
+    {
+      "epoch": 3.416576185305827,
+      "grad_norm": 0.6481451392173767,
+      "learning_rate": 0.0002,
+      "loss": 1.2925,
+      "step": 4720
+    },
+    {
+      "epoch": 3.4238146941730006,
+      "grad_norm": 0.9200371503829956,
+      "learning_rate": 0.0002,
+      "loss": 1.2121,
+      "step": 4730
+    },
+    {
+      "epoch": 3.431053203040174,
+      "grad_norm": 1.074180245399475,
+      "learning_rate": 0.0002,
+      "loss": 1.072,
+      "step": 4740
+    },
+    {
+      "epoch": 3.438291711907347,
+      "grad_norm": 0.6722986698150635,
+      "learning_rate": 0.0002,
+      "loss": 1.0421,
+      "step": 4750
+    },
+    {
+      "epoch": 3.44553022077452,
+      "grad_norm": 0.7945933938026428,
+      "learning_rate": 0.0002,
+      "loss": 1.2258,
+      "step": 4760
+    },
+    {
+      "epoch": 3.4527687296416936,
+      "grad_norm": 0.7624640464782715,
+      "learning_rate": 0.0002,
+      "loss": 1.0927,
+      "step": 4770
+    },
+    {
+      "epoch": 3.460007238508867,
+      "grad_norm": 0.7763656377792358,
+      "learning_rate": 0.0002,
+      "loss": 1.2428,
+      "step": 4780
+    },
+    {
+      "epoch": 3.4672457473760403,
+      "grad_norm": 0.7736947536468506,
+      "learning_rate": 0.0002,
+      "loss": 1.2584,
+      "step": 4790
+    },
+    {
+      "epoch": 3.4744842562432137,
+      "grad_norm": 0.8450354933738708,
+      "learning_rate": 0.0002,
+      "loss": 1.1953,
+      "step": 4800
+    },
+    {
+      "epoch": 3.481722765110387,
+      "grad_norm": 0.6480133533477783,
+      "learning_rate": 0.0002,
+      "loss": 1.1362,
+      "step": 4810
+    },
+    {
+      "epoch": 3.4889612739775604,
+      "grad_norm": 0.8437445759773254,
+      "learning_rate": 0.0002,
+      "loss": 1.1882,
+      "step": 4820
+    },
+    {
+      "epoch": 3.4961997828447338,
+      "grad_norm": 0.7781730890274048,
+      "learning_rate": 0.0002,
+      "loss": 1.1519,
+      "step": 4830
+    },
+    {
+      "epoch": 3.503438291711907,
+      "grad_norm": 0.8523228168487549,
+      "learning_rate": 0.0002,
+      "loss": 1.1836,
+      "step": 4840
+    },
+    {
+      "epoch": 3.5106768005790805,
+      "grad_norm": 0.6236732006072998,
+      "learning_rate": 0.0002,
+      "loss": 1.1672,
+      "step": 4850
+    },
+    {
+      "epoch": 3.517915309446254,
+      "grad_norm": 0.7500787377357483,
+      "learning_rate": 0.0002,
+      "loss": 1.1926,
+      "step": 4860
+    },
+    {
+      "epoch": 3.5251538183134272,
+      "grad_norm": 0.7665374875068665,
+      "learning_rate": 0.0002,
+      "loss": 1.1998,
+      "step": 4870
+    },
+    {
+      "epoch": 3.5323923271806006,
+      "grad_norm": 0.787857711315155,
+      "learning_rate": 0.0002,
+      "loss": 1.1551,
+      "step": 4880
+    },
+    {
+      "epoch": 3.539630836047774,
+      "grad_norm": 0.970595121383667,
+      "learning_rate": 0.0002,
+      "loss": 1.2758,
+      "step": 4890
+    },
+    {
+      "epoch": 3.5468693449149473,
+      "grad_norm": 0.6409347057342529,
+      "learning_rate": 0.0002,
+      "loss": 1.1274,
+      "step": 4900
+    },
+    {
+      "epoch": 3.5541078537821207,
+      "grad_norm": 0.888551652431488,
+      "learning_rate": 0.0002,
+      "loss": 1.1596,
+      "step": 4910
+    },
+    {
+      "epoch": 3.561346362649294,
+      "grad_norm": 1.0808377265930176,
+      "learning_rate": 0.0002,
+      "loss": 1.1644,
+      "step": 4920
+    },
+    {
+      "epoch": 3.5685848715164674,
+      "grad_norm": 0.7501053214073181,
+      "learning_rate": 0.0002,
+      "loss": 1.2564,
+      "step": 4930
+    },
+    {
+      "epoch": 3.575823380383641,
+      "grad_norm": 0.7375240325927734,
+      "learning_rate": 0.0002,
+      "loss": 1.2351,
+      "step": 4940
+    },
+    {
+      "epoch": 3.583061889250814,
+      "grad_norm": 0.7075039744377136,
+      "learning_rate": 0.0002,
+      "loss": 1.3568,
+      "step": 4950
+    },
+    {
+      "epoch": 3.5903003981179875,
+      "grad_norm": 0.939337432384491,
+      "learning_rate": 0.0002,
+      "loss": 1.3355,
+      "step": 4960
+    },
+    {
+      "epoch": 3.597538906985161,
+      "grad_norm": 0.6717396974563599,
+      "learning_rate": 0.0002,
+      "loss": 1.1722,
+      "step": 4970
+    },
+    {
+      "epoch": 3.6047774158523342,
+      "grad_norm": 0.7141643762588501,
+      "learning_rate": 0.0002,
+      "loss": 1.1186,
+      "step": 4980
+    },
+    {
+      "epoch": 3.6120159247195076,
+      "grad_norm": 0.7109216451644897,
+      "learning_rate": 0.0002,
+      "loss": 1.1011,
+      "step": 4990
+    },
+    {
+      "epoch": 3.619254433586681,
+      "grad_norm": 0.7020776867866516,
+      "learning_rate": 0.0002,
+      "loss": 1.2178,
+      "step": 5000
+    },
+    {
+      "epoch": 3.6264929424538543,
+      "grad_norm": 0.7158873677253723,
+      "learning_rate": 0.0002,
+      "loss": 1.1939,
+      "step": 5010
+    },
+    {
+      "epoch": 3.6337314513210277,
+      "grad_norm": 0.7062035202980042,
+      "learning_rate": 0.0002,
+      "loss": 1.2624,
+      "step": 5020
+    },
+    {
+      "epoch": 3.640969960188201,
+      "grad_norm": 0.7081155776977539,
+      "learning_rate": 0.0002,
+      "loss": 1.0224,
+      "step": 5030
+    },
+    {
+      "epoch": 3.6482084690553744,
+      "grad_norm": 1.2210607528686523,
+      "learning_rate": 0.0002,
+      "loss": 1.2195,
+      "step": 5040
+    },
+    {
+      "epoch": 3.655446977922548,
+      "grad_norm": 0.6650236248970032,
+      "learning_rate": 0.0002,
+      "loss": 1.2596,
+      "step": 5050
+    },
+    {
+      "epoch": 3.662685486789721,
+      "grad_norm": 0.6884829998016357,
+      "learning_rate": 0.0002,
+      "loss": 1.1072,
+      "step": 5060
+    },
+    {
+      "epoch": 3.6699239956568945,
+      "grad_norm": 0.7317819595336914,
+      "learning_rate": 0.0002,
+      "loss": 1.2292,
+      "step": 5070
+    },
+    {
+      "epoch": 3.677162504524068,
+      "grad_norm": 0.7406691908836365,
+      "learning_rate": 0.0002,
+      "loss": 1.1917,
+      "step": 5080
+    },
+    {
+      "epoch": 3.6844010133912413,
+      "grad_norm": 0.9009454250335693,
+      "learning_rate": 0.0002,
+      "loss": 1.2949,
+      "step": 5090
+    },
+    {
+      "epoch": 3.6916395222584146,
+      "grad_norm": 0.8189385533332825,
+      "learning_rate": 0.0002,
+      "loss": 1.1528,
+      "step": 5100
+    },
+    {
+      "epoch": 3.698878031125588,
+      "grad_norm": 1.0793628692626953,
+      "learning_rate": 0.0002,
+      "loss": 1.3408,
+      "step": 5110
+    },
+    {
+      "epoch": 3.7061165399927614,
+      "grad_norm": 0.8593027591705322,
+      "learning_rate": 0.0002,
+      "loss": 1.2417,
+      "step": 5120
+    },
+    {
+      "epoch": 3.7133550488599347,
+      "grad_norm": 0.8481812477111816,
+      "learning_rate": 0.0002,
+      "loss": 1.2141,
+      "step": 5130
+    },
+    {
+      "epoch": 3.720593557727108,
+      "grad_norm": 0.6527451276779175,
+      "learning_rate": 0.0002,
+      "loss": 1.125,
+      "step": 5140
+    },
+    {
+      "epoch": 3.7278320665942815,
+      "grad_norm": 0.9220114350318909,
+      "learning_rate": 0.0002,
+      "loss": 1.1584,
+      "step": 5150
+    },
+    {
+      "epoch": 3.735070575461455,
+      "grad_norm": 1.0842019319534302,
+      "learning_rate": 0.0002,
+      "loss": 1.2267,
+      "step": 5160
+    },
+    {
+      "epoch": 3.742309084328628,
+      "grad_norm": 0.965453565120697,
+      "learning_rate": 0.0002,
+      "loss": 1.3083,
+      "step": 5170
+    },
+    {
+      "epoch": 3.7495475931958016,
+      "grad_norm": 0.9903319478034973,
+      "learning_rate": 0.0002,
+      "loss": 1.1772,
+      "step": 5180
+    },
+    {
+      "epoch": 3.756786102062975,
+      "grad_norm": 0.7434818148612976,
+      "learning_rate": 0.0002,
+      "loss": 1.2515,
+      "step": 5190
+    },
+    {
+      "epoch": 3.7640246109301483,
+      "grad_norm": 0.6717280745506287,
+      "learning_rate": 0.0002,
+      "loss": 1.2631,
+      "step": 5200
+    },
+    {
+      "epoch": 3.7712631197973217,
+      "grad_norm": 0.7754665613174438,
+      "learning_rate": 0.0002,
+      "loss": 1.2012,
+      "step": 5210
+    },
+    {
+      "epoch": 3.778501628664495,
+      "grad_norm": 1.028374433517456,
+      "learning_rate": 0.0002,
+      "loss": 1.305,
+      "step": 5220
+    },
+    {
+      "epoch": 3.7857401375316684,
+      "grad_norm": 0.6026996374130249,
+      "learning_rate": 0.0002,
+      "loss": 1.1866,
+      "step": 5230
+    },
+    {
+      "epoch": 3.7929786463988417,
+      "grad_norm": 0.6978490948677063,
+      "learning_rate": 0.0002,
+      "loss": 1.1901,
+      "step": 5240
+    },
+    {
+      "epoch": 3.800217155266015,
+      "grad_norm": 0.7303446531295776,
+      "learning_rate": 0.0002,
+      "loss": 1.2576,
+      "step": 5250
+    },
+    {
+      "epoch": 3.8074556641331885,
+      "grad_norm": 1.0734210014343262,
+      "learning_rate": 0.0002,
+      "loss": 1.3173,
+      "step": 5260
+    },
+    {
+      "epoch": 3.814694173000362,
+      "grad_norm": 0.6383201479911804,
+      "learning_rate": 0.0002,
+      "loss": 1.1137,
+      "step": 5270
+    },
+    {
+      "epoch": 3.821932681867535,
+      "grad_norm": 0.7742630243301392,
+      "learning_rate": 0.0002,
+      "loss": 1.0904,
+      "step": 5280
+    },
+    {
+      "epoch": 3.8291711907347086,
+      "grad_norm": 0.8477074503898621,
+      "learning_rate": 0.0002,
+      "loss": 1.2232,
+      "step": 5290
+    },
+    {
+      "epoch": 3.836409699601882,
+      "grad_norm": 0.6675317883491516,
+      "learning_rate": 0.0002,
+      "loss": 1.2047,
+      "step": 5300
+    },
+    {
+      "epoch": 3.8436482084690553,
+      "grad_norm": 0.7515445351600647,
+      "learning_rate": 0.0002,
+      "loss": 1.2275,
+      "step": 5310
+    },
+    {
+      "epoch": 3.8508867173362287,
+      "grad_norm": 1.1441220045089722,
+      "learning_rate": 0.0002,
+      "loss": 1.2569,
+      "step": 5320
+    },
+    {
+      "epoch": 3.858125226203402,
+      "grad_norm": 0.7968795895576477,
+      "learning_rate": 0.0002,
+      "loss": 1.1512,
+      "step": 5330
+    },
+    {
+      "epoch": 3.8653637350705754,
+      "grad_norm": 0.7842824459075928,
+      "learning_rate": 0.0002,
+      "loss": 1.232,
+      "step": 5340
+    },
+    {
+      "epoch": 3.8726022439377488,
+      "grad_norm": 0.8272225260734558,
+      "learning_rate": 0.0002,
+      "loss": 1.1847,
+      "step": 5350
+    },
+    {
+      "epoch": 3.879840752804922,
+      "grad_norm": 0.8413397669792175,
+      "learning_rate": 0.0002,
+      "loss": 1.1381,
+      "step": 5360
+    },
+    {
+      "epoch": 3.8870792616720955,
+      "grad_norm": 1.141764760017395,
+      "learning_rate": 0.0002,
+      "loss": 1.2349,
+      "step": 5370
+    },
+    {
+      "epoch": 3.894317770539269,
+      "grad_norm": 0.9826975464820862,
+      "learning_rate": 0.0002,
+      "loss": 1.212,
+      "step": 5380
+    },
+    {
+      "epoch": 3.9015562794064422,
+      "grad_norm": 0.8598255515098572,
+      "learning_rate": 0.0002,
+      "loss": 1.1833,
+      "step": 5390
+    },
+    {
+      "epoch": 3.9087947882736156,
+      "grad_norm": 0.6271058320999146,
+      "learning_rate": 0.0002,
+      "loss": 1.1247,
+      "step": 5400
+    },
+    {
+      "epoch": 3.916033297140789,
+      "grad_norm": 0.6379870772361755,
+      "learning_rate": 0.0002,
+      "loss": 1.2212,
+      "step": 5410
+    },
+    {
+      "epoch": 3.9232718060079623,
+      "grad_norm": 1.0313376188278198,
+      "learning_rate": 0.0002,
+      "loss": 1.2481,
+      "step": 5420
+    },
+    {
+      "epoch": 3.9305103148751357,
+      "grad_norm": 0.8220619559288025,
+      "learning_rate": 0.0002,
+      "loss": 1.1872,
+      "step": 5430
+    },
+    {
+      "epoch": 3.937748823742309,
+      "grad_norm": 0.7576116919517517,
+      "learning_rate": 0.0002,
+      "loss": 1.2006,
+      "step": 5440
+    },
+    {
+      "epoch": 3.9449873326094824,
+      "grad_norm": 1.226235032081604,
+      "learning_rate": 0.0002,
+      "loss": 1.1969,
+      "step": 5450
+    },
+    {
+      "epoch": 3.952225841476656,
+      "grad_norm": 0.7979229688644409,
+      "learning_rate": 0.0002,
+      "loss": 1.2945,
+      "step": 5460
+    },
+    {
+      "epoch": 3.959464350343829,
+      "grad_norm": 0.9911929965019226,
+      "learning_rate": 0.0002,
+      "loss": 1.1922,
+      "step": 5470
+    },
+    {
+      "epoch": 3.9667028592110025,
+      "grad_norm": 0.643738865852356,
+      "learning_rate": 0.0002,
+      "loss": 1.0924,
+      "step": 5480
+    },
+    {
+      "epoch": 3.973941368078176,
+      "grad_norm": 0.682305634021759,
+      "learning_rate": 0.0002,
+      "loss": 1.0607,
+      "step": 5490
+    },
+    {
+      "epoch": 3.9811798769453492,
+      "grad_norm": 1.18373441696167,
+      "learning_rate": 0.0002,
+      "loss": 1.2908,
+      "step": 5500
+    },
+    {
+      "epoch": 3.9884183858125226,
+      "grad_norm": 0.7190203070640564,
+      "learning_rate": 0.0002,
+      "loss": 1.0889,
+      "step": 5510
+    },
+    {
+      "epoch": 3.995656894679696,
+      "grad_norm": 0.7516948580741882,
+      "learning_rate": 0.0002,
+      "loss": 1.2745,
+      "step": 5520
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.4252897500991821,
+      "eval_runtime": 27.235,
+      "eval_samples_per_second": 16.009,
+      "eval_steps_per_second": 2.019,
+      "step": 5526
+    },
+    {
+      "epoch": 4.002895403546869,
+      "grad_norm": 0.6353074312210083,
+      "learning_rate": 0.0002,
+      "loss": 1.0088,
+      "step": 5530
+    },
+    {
+      "epoch": 4.010133912414043,
+      "grad_norm": 0.7424906492233276,
+      "learning_rate": 0.0002,
+      "loss": 1.0326,
+      "step": 5540
+    },
+    {
+      "epoch": 4.017372421281216,
+      "grad_norm": 0.8856638073921204,
+      "learning_rate": 0.0002,
+      "loss": 1.0667,
+      "step": 5550
+    },
+    {
+      "epoch": 4.024610930148389,
+      "grad_norm": 0.9627974033355713,
+      "learning_rate": 0.0002,
+      "loss": 1.0905,
+      "step": 5560
+    },
+    {
+      "epoch": 4.031849439015563,
+      "grad_norm": 0.9048978686332703,
+      "learning_rate": 0.0002,
+      "loss": 1.0965,
+      "step": 5570
+    },
+    {
+      "epoch": 4.039087947882736,
+      "grad_norm": 0.921119213104248,
+      "learning_rate": 0.0002,
+      "loss": 1.1108,
+      "step": 5580
+    },
+    {
+      "epoch": 4.0463264567499095,
+      "grad_norm": 0.8654361963272095,
+      "learning_rate": 0.0002,
+      "loss": 1.1235,
+      "step": 5590
+    },
+    {
+      "epoch": 4.053564965617083,
+      "grad_norm": 0.7947945594787598,
+      "learning_rate": 0.0002,
+      "loss": 1.0794,
+      "step": 5600
+    },
+    {
+      "epoch": 4.060803474484256,
+      "grad_norm": 0.8307326436042786,
+      "learning_rate": 0.0002,
+      "loss": 1.0674,
+      "step": 5610
+    },
+    {
+      "epoch": 4.06804198335143,
+      "grad_norm": 0.793273389339447,
+      "learning_rate": 0.0002,
+      "loss": 1.0076,
+      "step": 5620
+    },
+    {
+      "epoch": 4.075280492218603,
+      "grad_norm": 0.8748673796653748,
+      "learning_rate": 0.0002,
+      "loss": 1.0651,
+      "step": 5630
+    },
+    {
+      "epoch": 4.082519001085776,
+      "grad_norm": 0.7926856279373169,
+      "learning_rate": 0.0002,
+      "loss": 1.111,
+      "step": 5640
+    },
+    {
+      "epoch": 4.08975750995295,
+      "grad_norm": 0.922645092010498,
+      "learning_rate": 0.0002,
+      "loss": 1.044,
+      "step": 5650
+    },
+    {
+      "epoch": 4.096996018820123,
+      "grad_norm": 0.9539641737937927,
+      "learning_rate": 0.0002,
+      "loss": 1.109,
+      "step": 5660
+    },
+    {
+      "epoch": 4.1042345276872965,
+      "grad_norm": 0.8674443364143372,
+      "learning_rate": 0.0002,
+      "loss": 1.0788,
+      "step": 5670
+    },
+    {
+      "epoch": 4.11147303655447,
+      "grad_norm": 0.7097609043121338,
+      "learning_rate": 0.0002,
+      "loss": 0.9867,
+      "step": 5680
+    },
+    {
+      "epoch": 4.118711545421643,
+      "grad_norm": 0.8875522613525391,
+      "learning_rate": 0.0002,
+      "loss": 1.1154,
+      "step": 5690
+    },
+    {
+      "epoch": 4.125950054288817,
+      "grad_norm": 0.8583634495735168,
+      "learning_rate": 0.0002,
+      "loss": 1.1217,
+      "step": 5700
+    },
+    {
+      "epoch": 4.13318856315599,
+      "grad_norm": 0.6736377477645874,
+      "learning_rate": 0.0002,
+      "loss": 1.0973,
+      "step": 5710
+    },
+    {
+      "epoch": 4.140427072023163,
+      "grad_norm": 0.9349062442779541,
+      "learning_rate": 0.0002,
+      "loss": 1.1199,
+      "step": 5720
+    },
+    {
+      "epoch": 4.147665580890337,
+      "grad_norm": 1.0610365867614746,
+      "learning_rate": 0.0002,
+      "loss": 1.0508,
+      "step": 5730
+    },
+    {
+      "epoch": 4.15490408975751,
+      "grad_norm": 1.5838189125061035,
+      "learning_rate": 0.0002,
+      "loss": 1.1146,
+      "step": 5740
+    },
+    {
+      "epoch": 4.162142598624683,
+      "grad_norm": 0.747522234916687,
+      "learning_rate": 0.0002,
+      "loss": 1.0222,
+      "step": 5750
+    },
+    {
+      "epoch": 4.169381107491857,
+      "grad_norm": 1.3247915506362915,
+      "learning_rate": 0.0002,
+      "loss": 1.1328,
+      "step": 5760
+    },
+    {
+      "epoch": 4.17661961635903,
+      "grad_norm": 0.8750247955322266,
+      "learning_rate": 0.0002,
+      "loss": 1.1655,
+      "step": 5770
+    },
+    {
+      "epoch": 4.1838581252262035,
+      "grad_norm": 0.7914144992828369,
+      "learning_rate": 0.0002,
+      "loss": 1.199,
+      "step": 5780
+    },
+    {
+      "epoch": 4.191096634093377,
+      "grad_norm": 0.9493299126625061,
+      "learning_rate": 0.0002,
+      "loss": 1.1213,
+      "step": 5790
+    },
+    {
+      "epoch": 4.19833514296055,
+      "grad_norm": 0.7802295088768005,
+      "learning_rate": 0.0002,
+      "loss": 1.1515,
+      "step": 5800
+    },
+    {
+      "epoch": 4.205573651827724,
+      "grad_norm": 0.6987314820289612,
+      "learning_rate": 0.0002,
+      "loss": 1.0704,
+      "step": 5810
+    },
+    {
+      "epoch": 4.212812160694897,
+      "grad_norm": 0.9220341444015503,
+      "learning_rate": 0.0002,
+      "loss": 1.1699,
+      "step": 5820
+    },
+    {
+      "epoch": 4.22005066956207,
+      "grad_norm": 0.8932939767837524,
+      "learning_rate": 0.0002,
+      "loss": 1.1394,
+      "step": 5830
+    },
+    {
+      "epoch": 4.227289178429244,
+      "grad_norm": 0.920002818107605,
+      "learning_rate": 0.0002,
+      "loss": 1.0048,
+      "step": 5840
+    },
+    {
+      "epoch": 4.234527687296417,
+      "grad_norm": 0.6662752032279968,
+      "learning_rate": 0.0002,
+      "loss": 0.964,
+      "step": 5850
+    },
+    {
+      "epoch": 4.24176619616359,
+      "grad_norm": 0.8679718971252441,
+      "learning_rate": 0.0002,
+      "loss": 0.986,
+      "step": 5860
+    },
+    {
+      "epoch": 4.249004705030764,
+      "grad_norm": 0.7020887732505798,
+      "learning_rate": 0.0002,
+      "loss": 0.8991,
+      "step": 5870
+    },
+    {
+      "epoch": 4.256243213897937,
+      "grad_norm": 0.869611382484436,
+      "learning_rate": 0.0002,
+      "loss": 1.1132,
+      "step": 5880
+    },
+    {
+      "epoch": 4.2634817227651105,
+      "grad_norm": 0.7796585559844971,
+      "learning_rate": 0.0002,
+      "loss": 1.1026,
+      "step": 5890
+    },
+    {
+      "epoch": 4.270720231632284,
+      "grad_norm": 0.8978819251060486,
+      "learning_rate": 0.0002,
+      "loss": 1.0957,
+      "step": 5900
+    },
+    {
+      "epoch": 4.277958740499457,
+      "grad_norm": 1.0837205648422241,
+      "learning_rate": 0.0002,
+      "loss": 1.1325,
+      "step": 5910
+    },
+    {
+      "epoch": 4.285197249366631,
+      "grad_norm": 0.7584353089332581,
+      "learning_rate": 0.0002,
+      "loss": 1.1279,
+      "step": 5920
+    },
+    {
+      "epoch": 4.292435758233804,
+      "grad_norm": 0.7313185334205627,
+      "learning_rate": 0.0002,
+      "loss": 1.0513,
+      "step": 5930
+    },
+    {
+      "epoch": 4.299674267100977,
+      "grad_norm": 0.8004671335220337,
+      "learning_rate": 0.0002,
+      "loss": 1.1101,
+      "step": 5940
+    },
+    {
+      "epoch": 4.306912775968151,
+      "grad_norm": 2.154958724975586,
+      "learning_rate": 0.0002,
+      "loss": 1.14,
+      "step": 5950
+    },
+    {
+      "epoch": 4.314151284835324,
+      "grad_norm": 0.9163479804992676,
+      "learning_rate": 0.0002,
+      "loss": 1.1206,
+      "step": 5960
+    },
+    {
+      "epoch": 4.321389793702497,
+      "grad_norm": 0.9151589274406433,
+      "learning_rate": 0.0002,
+      "loss": 0.9941,
+      "step": 5970
+    },
+    {
+      "epoch": 4.328628302569671,
+      "grad_norm": 0.8624112010002136,
+      "learning_rate": 0.0002,
+      "loss": 1.0606,
+      "step": 5980
+    },
+    {
+      "epoch": 4.335866811436844,
+      "grad_norm": 0.9357741475105286,
+      "learning_rate": 0.0002,
+      "loss": 1.1625,
+      "step": 5990
+    },
+    {
+      "epoch": 4.3431053203040175,
+      "grad_norm": 1.3482335805892944,
+      "learning_rate": 0.0002,
+      "loss": 1.0712,
+      "step": 6000
+    },
+    {
+      "epoch": 4.350343829171191,
+      "grad_norm": 0.7156149744987488,
+      "learning_rate": 0.0002,
+      "loss": 1.1224,
+      "step": 6010
+    },
+    {
+      "epoch": 4.357582338038364,
+      "grad_norm": 0.8480049967765808,
+      "learning_rate": 0.0002,
+      "loss": 1.0753,
+      "step": 6020
+    },
+    {
+      "epoch": 4.364820846905538,
+      "grad_norm": 0.8262244462966919,
+      "learning_rate": 0.0002,
+      "loss": 1.051,
+      "step": 6030
+    },
+    {
+      "epoch": 4.372059355772711,
+      "grad_norm": 0.7733905911445618,
+      "learning_rate": 0.0002,
+      "loss": 0.9966,
+      "step": 6040
+    },
+    {
+      "epoch": 4.379297864639884,
+      "grad_norm": 0.8553919792175293,
+      "learning_rate": 0.0002,
+      "loss": 1.1008,
+      "step": 6050
+    },
+    {
+      "epoch": 4.386536373507058,
+      "grad_norm": 0.8666832447052002,
+      "learning_rate": 0.0002,
+      "loss": 1.1777,
+      "step": 6060
+    },
+    {
+      "epoch": 4.393774882374231,
+      "grad_norm": 0.9168295860290527,
+      "learning_rate": 0.0002,
+      "loss": 1.1934,
+      "step": 6070
+    },
+    {
+      "epoch": 4.4010133912414044,
+      "grad_norm": 0.7315238118171692,
+      "learning_rate": 0.0002,
+      "loss": 1.0988,
+      "step": 6080
+    },
+    {
+      "epoch": 4.408251900108578,
+      "grad_norm": 1.020263433456421,
+      "learning_rate": 0.0002,
+      "loss": 1.1599,
+      "step": 6090
+    },
+    {
+      "epoch": 4.415490408975751,
+      "grad_norm": 0.9978243708610535,
+      "learning_rate": 0.0002,
+      "loss": 1.133,
+      "step": 6100
+    },
+    {
+      "epoch": 4.4227289178429245,
+      "grad_norm": 0.995453953742981,
+      "learning_rate": 0.0002,
+      "loss": 1.1324,
+      "step": 6110
+    },
+    {
+      "epoch": 4.429967426710098,
+      "grad_norm": 0.9360884428024292,
+      "learning_rate": 0.0002,
+      "loss": 1.0957,
+      "step": 6120
+    },
+    {
+      "epoch": 4.437205935577271,
+      "grad_norm": 0.8099448084831238,
+      "learning_rate": 0.0002,
+      "loss": 0.9506,
+      "step": 6130
+    },
+    {
+      "epoch": 4.444444444444445,
+      "grad_norm": 0.8173841238021851,
+      "learning_rate": 0.0002,
+      "loss": 1.0887,
+      "step": 6140
+    },
+    {
+      "epoch": 4.451682953311618,
+      "grad_norm": 0.7972666025161743,
+      "learning_rate": 0.0002,
+      "loss": 1.1219,
+      "step": 6150
+    },
+    {
+      "epoch": 4.458921462178791,
+      "grad_norm": 0.7685779333114624,
+      "learning_rate": 0.0002,
+      "loss": 1.0226,
+      "step": 6160
+    },
+    {
+      "epoch": 4.466159971045965,
+      "grad_norm": 0.7872623801231384,
+      "learning_rate": 0.0002,
+      "loss": 1.0732,
+      "step": 6170
+    },
+    {
+      "epoch": 4.473398479913138,
+      "grad_norm": 0.7677070498466492,
+      "learning_rate": 0.0002,
+      "loss": 0.9911,
+      "step": 6180
+    },
+    {
+      "epoch": 4.4806369887803115,
+      "grad_norm": 0.7878316044807434,
+      "learning_rate": 0.0002,
+      "loss": 1.0919,
+      "step": 6190
+    },
+    {
+      "epoch": 4.487875497647485,
+      "grad_norm": 0.8178079724311829,
+      "learning_rate": 0.0002,
+      "loss": 1.018,
+      "step": 6200
+    },
+    {
+      "epoch": 4.495114006514658,
+      "grad_norm": 1.2820082902908325,
+      "learning_rate": 0.0002,
+      "loss": 1.0517,
+      "step": 6210
+    },
+    {
+      "epoch": 4.502352515381832,
+      "grad_norm": 0.9380832314491272,
+      "learning_rate": 0.0002,
+      "loss": 1.3101,
+      "step": 6220
+    },
+    {
+      "epoch": 4.509591024249005,
+      "grad_norm": 0.7810422778129578,
+      "learning_rate": 0.0002,
+      "loss": 0.9818,
+      "step": 6230
+    },
+    {
+      "epoch": 4.516829533116178,
+      "grad_norm": 1.1022917032241821,
+      "learning_rate": 0.0002,
+      "loss": 1.1677,
+      "step": 6240
+    },
+    {
+      "epoch": 4.524068041983352,
+      "grad_norm": 1.4275553226470947,
+      "learning_rate": 0.0002,
+      "loss": 1.1579,
+      "step": 6250
+    },
+    {
+      "epoch": 4.531306550850525,
+      "grad_norm": 0.7597777247428894,
+      "learning_rate": 0.0002,
+      "loss": 1.3237,
+      "step": 6260
+    },
+    {
+      "epoch": 4.538545059717698,
+      "grad_norm": 1.10992431640625,
+      "learning_rate": 0.0002,
+      "loss": 1.1529,
+      "step": 6270
+    },
+    {
+      "epoch": 4.545783568584872,
+      "grad_norm": 0.8981178998947144,
+      "learning_rate": 0.0002,
+      "loss": 1.0732,
+      "step": 6280
+    },
+    {
+      "epoch": 4.553022077452045,
+      "grad_norm": 0.7863979339599609,
+      "learning_rate": 0.0002,
+      "loss": 1.086,
+      "step": 6290
+    },
+    {
+      "epoch": 4.5602605863192185,
+      "grad_norm": 0.9071474671363831,
+      "learning_rate": 0.0002,
+      "loss": 1.2008,
+      "step": 6300
+    },
+    {
+      "epoch": 4.567499095186392,
+      "grad_norm": 0.7429424524307251,
+      "learning_rate": 0.0002,
+      "loss": 1.0916,
+      "step": 6310
+    },
+    {
+      "epoch": 4.574737604053565,
+      "grad_norm": 1.0767850875854492,
+      "learning_rate": 0.0002,
+      "loss": 1.095,
+      "step": 6320
+    },
+    {
+      "epoch": 4.581976112920739,
+      "grad_norm": 0.7885915637016296,
+      "learning_rate": 0.0002,
+      "loss": 1.1023,
+      "step": 6330
+    },
+    {
+      "epoch": 4.589214621787912,
+      "grad_norm": 0.8350457549095154,
+      "learning_rate": 0.0002,
+      "loss": 1.1131,
+      "step": 6340
+    },
+    {
+      "epoch": 4.596453130655085,
+      "grad_norm": 0.7853530645370483,
+      "learning_rate": 0.0002,
+      "loss": 1.0743,
+      "step": 6350
+    },
+    {
+      "epoch": 4.603691639522259,
+      "grad_norm": 1.1220661401748657,
+      "learning_rate": 0.0002,
+      "loss": 1.1912,
+      "step": 6360
+    },
+    {
+      "epoch": 4.610930148389432,
+      "grad_norm": 0.7959423065185547,
+      "learning_rate": 0.0002,
+      "loss": 1.0927,
+      "step": 6370
+    },
+    {
+      "epoch": 4.618168657256605,
+      "grad_norm": 0.7782652378082275,
+      "learning_rate": 0.0002,
+      "loss": 1.1542,
+      "step": 6380
+    },
+    {
+      "epoch": 4.625407166123779,
+      "grad_norm": 0.7882203459739685,
+      "learning_rate": 0.0002,
+      "loss": 1.0753,
+      "step": 6390
+    },
+    {
+      "epoch": 4.632645674990952,
+      "grad_norm": 0.8841899037361145,
+      "learning_rate": 0.0002,
+      "loss": 1.0676,
+      "step": 6400
+    },
+    {
+      "epoch": 4.6398841838581255,
+      "grad_norm": 0.7936127781867981,
+      "learning_rate": 0.0002,
+      "loss": 1.0815,
+      "step": 6410
+    },
+    {
+      "epoch": 4.647122692725299,
+      "grad_norm": 0.9213966131210327,
+      "learning_rate": 0.0002,
+      "loss": 1.0198,
+      "step": 6420
+    },
+    {
+      "epoch": 4.654361201592472,
+      "grad_norm": 0.9246473908424377,
+      "learning_rate": 0.0002,
+      "loss": 0.9872,
+      "step": 6430
+    },
+    {
+      "epoch": 4.661599710459646,
+      "grad_norm": 0.766572892665863,
+      "learning_rate": 0.0002,
+      "loss": 1.1309,
+      "step": 6440
+    },
+    {
+      "epoch": 4.668838219326819,
+      "grad_norm": 0.8596171736717224,
+      "learning_rate": 0.0002,
+      "loss": 1.1095,
+      "step": 6450
+    },
+    {
+      "epoch": 4.676076728193992,
+      "grad_norm": 0.8482751846313477,
+      "learning_rate": 0.0002,
+      "loss": 1.1869,
+      "step": 6460
+    },
+    {
+      "epoch": 4.683315237061166,
+      "grad_norm": 1.0826905965805054,
+      "learning_rate": 0.0002,
+      "loss": 1.0622,
+      "step": 6470
+    },
+    {
+      "epoch": 4.690553745928339,
+      "grad_norm": 1.1048457622528076,
+      "learning_rate": 0.0002,
+      "loss": 1.0256,
+      "step": 6480
+    },
+    {
+      "epoch": 4.697792254795512,
+      "grad_norm": 0.9429134726524353,
+      "learning_rate": 0.0002,
+      "loss": 1.0514,
+      "step": 6490
+    },
+    {
+      "epoch": 4.705030763662686,
+      "grad_norm": 0.8587502837181091,
+      "learning_rate": 0.0002,
+      "loss": 1.1351,
+      "step": 6500
+    },
+    {
+      "epoch": 4.712269272529859,
+      "grad_norm": 1.0387083292007446,
+      "learning_rate": 0.0002,
+      "loss": 1.0969,
+      "step": 6510
+    },
+    {
+      "epoch": 4.7195077813970325,
+      "grad_norm": 0.7471951842308044,
+      "learning_rate": 0.0002,
+      "loss": 1.0493,
+      "step": 6520
+    },
+    {
+      "epoch": 4.726746290264206,
+      "grad_norm": 0.8800424933433533,
+      "learning_rate": 0.0002,
+      "loss": 1.2632,
+      "step": 6530
+    },
+    {
+      "epoch": 4.733984799131379,
+      "grad_norm": 0.8136811852455139,
+      "learning_rate": 0.0002,
+      "loss": 1.2126,
+      "step": 6540
+    },
+    {
+      "epoch": 4.741223307998553,
+      "grad_norm": 0.9910339713096619,
+      "learning_rate": 0.0002,
+      "loss": 1.195,
+      "step": 6550
+    },
+    {
+      "epoch": 4.748461816865726,
+      "grad_norm": 1.0679163932800293,
+      "learning_rate": 0.0002,
+      "loss": 1.1201,
+      "step": 6560
+    },
+    {
+      "epoch": 4.755700325732899,
+      "grad_norm": 0.8468248248100281,
+      "learning_rate": 0.0002,
+      "loss": 1.0297,
+      "step": 6570
+    },
+    {
+      "epoch": 4.762938834600073,
+      "grad_norm": 0.8771235942840576,
+      "learning_rate": 0.0002,
+      "loss": 1.0858,
+      "step": 6580
+    },
+    {
+      "epoch": 4.770177343467246,
+      "grad_norm": 0.7024846076965332,
+      "learning_rate": 0.0002,
+      "loss": 1.077,
+      "step": 6590
+    },
+    {
+      "epoch": 4.7774158523344195,
+      "grad_norm": 0.7836683392524719,
+      "learning_rate": 0.0002,
+      "loss": 1.0876,
+      "step": 6600
+    },
+    {
+      "epoch": 4.784654361201593,
+      "grad_norm": 0.7717288136482239,
+      "learning_rate": 0.0002,
+      "loss": 1.1006,
+      "step": 6610
+    },
+    {
+      "epoch": 4.791892870068766,
+      "grad_norm": 0.884183943271637,
+      "learning_rate": 0.0002,
+      "loss": 1.0376,
+      "step": 6620
+    },
+    {
+      "epoch": 4.7991313789359396,
+      "grad_norm": 1.383867621421814,
+      "learning_rate": 0.0002,
+      "loss": 1.1757,
+      "step": 6630
+    },
+    {
+      "epoch": 4.806369887803113,
+      "grad_norm": 0.9741523861885071,
+      "learning_rate": 0.0002,
+      "loss": 1.0861,
+      "step": 6640
+    },
+    {
+      "epoch": 4.813608396670286,
+      "grad_norm": 0.9723693132400513,
+      "learning_rate": 0.0002,
+      "loss": 1.0884,
+      "step": 6650
+    },
+    {
+      "epoch": 4.82084690553746,
+      "grad_norm": 1.8324809074401855,
+      "learning_rate": 0.0002,
+      "loss": 1.2203,
+      "step": 6660
+    },
+    {
+      "epoch": 4.828085414404633,
+      "grad_norm": 0.904909074306488,
+      "learning_rate": 0.0002,
+      "loss": 1.0292,
+      "step": 6670
+    },
+    {
+      "epoch": 4.835323923271806,
+      "grad_norm": 0.7355411648750305,
+      "learning_rate": 0.0002,
+      "loss": 1.0349,
+      "step": 6680
+    },
+    {
+      "epoch": 4.84256243213898,
+      "grad_norm": 0.8934960961341858,
+      "learning_rate": 0.0002,
+      "loss": 1.0793,
+      "step": 6690
+    },
+    {
+      "epoch": 4.849800941006153,
+      "grad_norm": 1.4596954584121704,
+      "learning_rate": 0.0002,
+      "loss": 1.0375,
+      "step": 6700
+    },
+    {
+      "epoch": 4.8570394498733265,
+      "grad_norm": 0.8310341238975525,
+      "learning_rate": 0.0002,
+      "loss": 1.1065,
+      "step": 6710
+    },
+    {
+      "epoch": 4.8642779587405,
+      "grad_norm": 0.9709894061088562,
+      "learning_rate": 0.0002,
+      "loss": 1.1089,
+      "step": 6720
+    },
+    {
+      "epoch": 4.871516467607673,
+      "grad_norm": 0.852142333984375,
+      "learning_rate": 0.0002,
+      "loss": 1.0069,
+      "step": 6730
+    },
+    {
+      "epoch": 4.878754976474847,
+      "grad_norm": 1.0643625259399414,
+      "learning_rate": 0.0002,
+      "loss": 1.0507,
+      "step": 6740
+    },
+    {
+      "epoch": 4.88599348534202,
+      "grad_norm": 0.9419508576393127,
+      "learning_rate": 0.0002,
+      "loss": 1.056,
+      "step": 6750
+    },
+    {
+      "epoch": 4.893231994209193,
+      "grad_norm": 1.1818498373031616,
+      "learning_rate": 0.0002,
+      "loss": 1.1995,
+      "step": 6760
+    },
+    {
+      "epoch": 4.900470503076367,
+      "grad_norm": 0.9369569420814514,
+      "learning_rate": 0.0002,
+      "loss": 1.0925,
+      "step": 6770
+    },
+    {
+      "epoch": 4.90770901194354,
+      "grad_norm": 0.7012579441070557,
+      "learning_rate": 0.0002,
+      "loss": 1.1648,
+      "step": 6780
+    },
+    {
+      "epoch": 4.914947520810713,
+      "grad_norm": 0.9109319448471069,
+      "learning_rate": 0.0002,
+      "loss": 1.0926,
+      "step": 6790
+    },
+    {
+      "epoch": 4.922186029677887,
+      "grad_norm": 0.8077534437179565,
+      "learning_rate": 0.0002,
+      "loss": 1.0358,
+      "step": 6800
+    },
+    {
+      "epoch": 4.92942453854506,
+      "grad_norm": 0.7571148872375488,
+      "learning_rate": 0.0002,
+      "loss": 1.2549,
+      "step": 6810
+    },
+    {
+      "epoch": 4.9366630474122335,
+      "grad_norm": 0.7325633764266968,
+      "learning_rate": 0.0002,
+      "loss": 0.9638,
+      "step": 6820
+    },
+    {
+      "epoch": 4.943901556279407,
+      "grad_norm": 0.8465084433555603,
+      "learning_rate": 0.0002,
+      "loss": 1.0128,
+      "step": 6830
+    },
+    {
+      "epoch": 4.95114006514658,
+      "grad_norm": 0.8753737807273865,
+      "learning_rate": 0.0002,
+      "loss": 1.153,
+      "step": 6840
+    },
+    {
+      "epoch": 4.958378574013754,
+      "grad_norm": 0.9421748518943787,
+      "learning_rate": 0.0002,
+      "loss": 1.0247,
+      "step": 6850
+    },
+    {
+      "epoch": 4.965617082880927,
+      "grad_norm": 0.8245896697044373,
+      "learning_rate": 0.0002,
+      "loss": 1.1483,
+      "step": 6860
+    },
+    {
+      "epoch": 4.9728555917481,
+      "grad_norm": 0.8823089599609375,
+      "learning_rate": 0.0002,
+      "loss": 0.9905,
+      "step": 6870
+    },
+    {
+      "epoch": 4.980094100615274,
+      "grad_norm": 0.8406389355659485,
+      "learning_rate": 0.0002,
+      "loss": 1.1664,
+      "step": 6880
+    },
+    {
+      "epoch": 4.987332609482447,
+      "grad_norm": 0.9732868075370789,
+      "learning_rate": 0.0002,
+      "loss": 1.0944,
+      "step": 6890
+    },
+    {
+      "epoch": 4.99457111834962,
+      "grad_norm": 2.125141143798828,
+      "learning_rate": 0.0002,
+      "loss": 1.1776,
+      "step": 6900
+    },
+    {
+      "epoch": 4.999638074556641,
+      "eval_loss": 1.445176601409912,
+      "eval_runtime": 27.2351,
+      "eval_samples_per_second": 16.009,
+      "eval_steps_per_second": 2.019,
+      "step": 6907
+    },
+    {
+      "epoch": 5.001809627216793,
+      "grad_norm": 0.9465792775154114,
+      "learning_rate": 0.0002,
+      "loss": 1.1362,
+      "step": 6910
+    },
+    {
+      "epoch": 5.009048136083966,
+      "grad_norm": 1.2834891080856323,
+      "learning_rate": 0.0002,
+      "loss": 0.982,
+      "step": 6920
+    },
+    {
+      "epoch": 5.01628664495114,
+      "grad_norm": 1.0297378301620483,
+      "learning_rate": 0.0002,
+      "loss": 0.9803,
+      "step": 6930
+    },
+    {
+      "epoch": 5.023525153818313,
+      "grad_norm": 1.1705161333084106,
+      "learning_rate": 0.0002,
+      "loss": 1.0447,
+      "step": 6940
+    },
+    {
+      "epoch": 5.030763662685486,
+      "grad_norm": 0.8293961882591248,
+      "learning_rate": 0.0002,
+      "loss": 1.0113,
+      "step": 6950
+    },
+    {
+      "epoch": 5.03800217155266,
+      "grad_norm": 1.0422210693359375,
+      "learning_rate": 0.0002,
+      "loss": 0.9203,
+      "step": 6960
+    },
+    {
+      "epoch": 5.045240680419833,
+      "grad_norm": 1.116104245185852,
+      "learning_rate": 0.0002,
+      "loss": 1.0553,
+      "step": 6970
+    },
+    {
+      "epoch": 5.0524791892870065,
+      "grad_norm": 1.5118416547775269,
+      "learning_rate": 0.0002,
+      "loss": 0.9011,
+      "step": 6980
+    },
+    {
+      "epoch": 5.05971769815418,
+      "grad_norm": 0.8383979797363281,
+      "learning_rate": 0.0002,
+      "loss": 0.9969,
+      "step": 6990
+    },
+    {
+      "epoch": 5.066956207021353,
+      "grad_norm": 1.3378649950027466,
+      "learning_rate": 0.0002,
+      "loss": 0.9659,
+      "step": 7000
+    },
+    {
+      "epoch": 5.0741947158885266,
+      "grad_norm": 1.1840510368347168,
+      "learning_rate": 0.0002,
+      "loss": 1.0212,
+      "step": 7010
+    },
+    {
+      "epoch": 5.0814332247557,
+      "grad_norm": 1.2354751825332642,
+      "learning_rate": 0.0002,
+      "loss": 0.9939,
+      "step": 7020
+    },
+    {
+      "epoch": 5.088671733622873,
+      "grad_norm": 1.3830451965332031,
+      "learning_rate": 0.0002,
+      "loss": 0.9831,
+      "step": 7030
+    },
+    {
+      "epoch": 5.095910242490047,
+      "grad_norm": 0.8101674318313599,
+      "learning_rate": 0.0002,
+      "loss": 1.1827,
+      "step": 7040
+    },
+    {
+      "epoch": 5.10314875135722,
+      "grad_norm": 0.897982656955719,
+      "learning_rate": 0.0002,
+      "loss": 0.9255,
+      "step": 7050
+    },
+    {
+      "epoch": 5.110387260224393,
+      "grad_norm": 1.2049678564071655,
+      "learning_rate": 0.0002,
+      "loss": 0.8784,
+      "step": 7060
+    },
+    {
+      "epoch": 5.117625769091567,
+      "grad_norm": 1.5912116765975952,
+      "learning_rate": 0.0002,
+      "loss": 1.0182,
+      "step": 7070
+    },
+    {
+      "epoch": 5.12486427795874,
+      "grad_norm": 0.9261530041694641,
+      "learning_rate": 0.0002,
+      "loss": 1.0909,
+      "step": 7080
+    },
+    {
+      "epoch": 5.1321027868259135,
+      "grad_norm": 1.1454812288284302,
+      "learning_rate": 0.0002,
+      "loss": 0.9603,
+      "step": 7090
+    },
+    {
+      "epoch": 5.139341295693087,
+      "grad_norm": 1.0049978494644165,
+      "learning_rate": 0.0002,
+      "loss": 0.9149,
+      "step": 7100
+    },
+    {
+      "epoch": 5.14657980456026,
+      "grad_norm": 1.4513251781463623,
+      "learning_rate": 0.0002,
+      "loss": 0.9463,
+      "step": 7110
+    },
+    {
+      "epoch": 5.153818313427434,
+      "grad_norm": 0.9800849556922913,
+      "learning_rate": 0.0002,
+      "loss": 0.8995,
+      "step": 7120
+    },
+    {
+      "epoch": 5.161056822294607,
+      "grad_norm": 0.9698708653450012,
+      "learning_rate": 0.0002,
+      "loss": 0.9835,
+      "step": 7130
+    },
+    {
+      "epoch": 5.16829533116178,
+      "grad_norm": 1.1126646995544434,
+      "learning_rate": 0.0002,
+      "loss": 0.9672,
+      "step": 7140
+    },
+    {
+      "epoch": 5.175533840028954,
+      "grad_norm": 0.9248330593109131,
+      "learning_rate": 0.0002,
+      "loss": 0.9384,
+      "step": 7150
+    },
+    {
+      "epoch": 5.182772348896127,
+      "grad_norm": 0.7967255711555481,
+      "learning_rate": 0.0002,
+      "loss": 0.826,
+      "step": 7160
+    },
+    {
+      "epoch": 5.1900108577633,
+      "grad_norm": 0.9933333992958069,
+      "learning_rate": 0.0002,
+      "loss": 1.0078,
+      "step": 7170
+    },
+    {
+      "epoch": 5.197249366630474,
+      "grad_norm": 1.0080649852752686,
+      "learning_rate": 0.0002,
+      "loss": 1.0276,
+      "step": 7180
+    },
+    {
+      "epoch": 5.204487875497647,
+      "grad_norm": 1.3954921960830688,
+      "learning_rate": 0.0002,
+      "loss": 1.0201,
+      "step": 7190
+    },
+    {
+      "epoch": 5.2117263843648205,
+      "grad_norm": 1.2386271953582764,
+      "learning_rate": 0.0002,
+      "loss": 1.0863,
+      "step": 7200
+    },
+    {
+      "epoch": 5.218964893231994,
+      "grad_norm": 1.2379488945007324,
+      "learning_rate": 0.0002,
+      "loss": 0.8863,
+      "step": 7210
+    },
+    {
+      "epoch": 5.226203402099167,
+      "grad_norm": 0.9882503747940063,
+      "learning_rate": 0.0002,
+      "loss": 1.0518,
+      "step": 7220
+    },
+    {
+      "epoch": 5.233441910966341,
+      "grad_norm": 1.1728729009628296,
+      "learning_rate": 0.0002,
+      "loss": 0.9834,
+      "step": 7230
+    },
+    {
+      "epoch": 5.240680419833514,
+      "grad_norm": 0.9849673509597778,
+      "learning_rate": 0.0002,
+      "loss": 0.9269,
+      "step": 7240
+    },
+    {
+      "epoch": 5.247918928700687,
+      "grad_norm": 1.177639365196228,
+      "learning_rate": 0.0002,
+      "loss": 0.9935,
+      "step": 7250
+    },
+    {
+      "epoch": 5.255157437567861,
+      "grad_norm": 1.2395055294036865,
+      "learning_rate": 0.0002,
+      "loss": 1.0639,
+      "step": 7260
+    },
+    {
+      "epoch": 5.262395946435034,
+      "grad_norm": 1.3999171257019043,
+      "learning_rate": 0.0002,
+      "loss": 1.0138,
+      "step": 7270
+    },
+    {
+      "epoch": 5.269634455302207,
+      "grad_norm": 0.7698732018470764,
+      "learning_rate": 0.0002,
+      "loss": 0.9745,
+      "step": 7280
+    },
+    {
+      "epoch": 5.276872964169381,
+      "grad_norm": 0.9167453646659851,
+      "learning_rate": 0.0002,
+      "loss": 1.0389,
+      "step": 7290
+    },
+    {
+      "epoch": 5.284111473036554,
+      "grad_norm": 1.113830804824829,
+      "learning_rate": 0.0002,
+      "loss": 0.9858,
+      "step": 7300
+    },
+    {
+      "epoch": 5.2913499819037275,
+      "grad_norm": 0.9644396901130676,
+      "learning_rate": 0.0002,
+      "loss": 0.9577,
+      "step": 7310
+    },
+    {
+      "epoch": 5.298588490770901,
+      "grad_norm": 1.462435007095337,
+      "learning_rate": 0.0002,
+      "loss": 1.0556,
+      "step": 7320
+    },
+    {
+      "epoch": 5.305826999638074,
+      "grad_norm": 0.9406287670135498,
+      "learning_rate": 0.0002,
+      "loss": 0.871,
+      "step": 7330
+    },
+    {
+      "epoch": 5.313065508505248,
+      "grad_norm": 0.9698247909545898,
+      "learning_rate": 0.0002,
+      "loss": 1.0022,
+      "step": 7340
+    },
+    {
+      "epoch": 5.320304017372421,
+      "grad_norm": 1.12003755569458,
+      "learning_rate": 0.0002,
+      "loss": 0.915,
+      "step": 7350
+    },
+    {
+      "epoch": 5.327542526239594,
+      "grad_norm": 1.598681926727295,
+      "learning_rate": 0.0002,
+      "loss": 0.9838,
+      "step": 7360
+    },
+    {
+      "epoch": 5.334781035106768,
+      "grad_norm": 1.0450010299682617,
+      "learning_rate": 0.0002,
+      "loss": 1.0,
+      "step": 7370
+    },
+    {
+      "epoch": 5.342019543973941,
+      "grad_norm": 0.8680008053779602,
+      "learning_rate": 0.0002,
+      "loss": 0.9983,
+      "step": 7380
+    },
+    {
+      "epoch": 5.349258052841114,
+      "grad_norm": 1.0115476846694946,
+      "learning_rate": 0.0002,
+      "loss": 0.9851,
+      "step": 7390
+    },
+    {
+      "epoch": 5.356496561708288,
+      "grad_norm": 0.9589748382568359,
+      "learning_rate": 0.0002,
+      "loss": 1.0702,
+      "step": 7400
+    },
+    {
+      "epoch": 5.363735070575461,
+      "grad_norm": 0.6729998588562012,
+      "learning_rate": 0.0002,
+      "loss": 0.9366,
+      "step": 7410
+    },
+    {
+      "epoch": 5.3709735794426345,
+      "grad_norm": 0.9246699213981628,
+      "learning_rate": 0.0002,
+      "loss": 1.0126,
+      "step": 7420
+    },
+    {
+      "epoch": 5.378212088309808,
+      "grad_norm": 1.1266791820526123,
+      "learning_rate": 0.0002,
+      "loss": 0.9815,
+      "step": 7430
+    },
+    {
+      "epoch": 5.385450597176981,
+      "grad_norm": 1.8056942224502563,
+      "learning_rate": 0.0002,
+      "loss": 1.1166,
+      "step": 7440
+    },
+    {
+      "epoch": 5.392689106044155,
+      "grad_norm": 0.9802932739257812,
+      "learning_rate": 0.0002,
+      "loss": 0.9604,
+      "step": 7450
+    },
+    {
+      "epoch": 5.399927614911328,
+      "grad_norm": 1.0504707098007202,
+      "learning_rate": 0.0002,
+      "loss": 0.9656,
+      "step": 7460
+    },
+    {
+      "epoch": 5.407166123778501,
+      "grad_norm": 1.1915022134780884,
+      "learning_rate": 0.0002,
+      "loss": 1.0132,
+      "step": 7470
+    },
+    {
+      "epoch": 5.414404632645675,
+      "grad_norm": 1.1856611967086792,
+      "learning_rate": 0.0002,
+      "loss": 1.0041,
+      "step": 7480
+    },
+    {
+      "epoch": 5.421643141512848,
+      "grad_norm": 1.292152762413025,
+      "learning_rate": 0.0002,
+      "loss": 0.9747,
+      "step": 7490
+    },
+    {
+      "epoch": 5.4288816503800215,
+      "grad_norm": 1.2675740718841553,
+      "learning_rate": 0.0002,
+      "loss": 0.9659,
+      "step": 7500
+    },
+    {
+      "epoch": 5.436120159247195,
+      "grad_norm": 1.4034695625305176,
+      "learning_rate": 0.0002,
+      "loss": 1.0271,
+      "step": 7510
+    },
+    {
+      "epoch": 5.443358668114368,
+      "grad_norm": 0.984588623046875,
+      "learning_rate": 0.0002,
+      "loss": 1.0318,
+      "step": 7520
+    },
+    {
+      "epoch": 5.450597176981542,
+      "grad_norm": 0.8419108390808105,
+      "learning_rate": 0.0002,
+      "loss": 1.0726,
+      "step": 7530
+    },
+    {
+      "epoch": 5.457835685848715,
+      "grad_norm": 1.0270143747329712,
+      "learning_rate": 0.0002,
+      "loss": 1.0499,
+      "step": 7540
+    },
+    {
+      "epoch": 5.465074194715888,
+      "grad_norm": 2.2158689498901367,
+      "learning_rate": 0.0002,
+      "loss": 0.9804,
+      "step": 7550
+    },
+    {
+      "epoch": 5.472312703583062,
+      "grad_norm": 1.0740524530410767,
+      "learning_rate": 0.0002,
+      "loss": 0.9856,
+      "step": 7560
+    },
+    {
+      "epoch": 5.479551212450235,
+      "grad_norm": 1.3804482221603394,
+      "learning_rate": 0.0002,
+      "loss": 1.0522,
+      "step": 7570
+    },
+    {
+      "epoch": 5.486789721317408,
+      "grad_norm": 0.9428979754447937,
+      "learning_rate": 0.0002,
+      "loss": 1.0297,
+      "step": 7580
+    },
+    {
+      "epoch": 5.494028230184582,
+      "grad_norm": 0.9548295736312866,
+      "learning_rate": 0.0002,
+      "loss": 1.0906,
+      "step": 7590
+    },
+    {
+      "epoch": 5.501266739051755,
+      "grad_norm": 1.0691065788269043,
+      "learning_rate": 0.0002,
+      "loss": 0.8853,
+      "step": 7600
+    },
+    {
+      "epoch": 5.5085052479189285,
+      "grad_norm": 1.0987380743026733,
+      "learning_rate": 0.0002,
+      "loss": 1.0375,
+      "step": 7610
+    },
+    {
+      "epoch": 5.515743756786102,
+      "grad_norm": 0.9483979344367981,
+      "learning_rate": 0.0002,
+      "loss": 1.0162,
+      "step": 7620
+    },
+    {
+      "epoch": 5.522982265653275,
+      "grad_norm": 1.16624915599823,
+      "learning_rate": 0.0002,
+      "loss": 1.105,
+      "step": 7630
+    },
+    {
+      "epoch": 5.530220774520449,
+      "grad_norm": 0.8563777208328247,
+      "learning_rate": 0.0002,
+      "loss": 0.8695,
+      "step": 7640
+    },
+    {
+      "epoch": 5.537459283387622,
+      "grad_norm": 1.268186092376709,
+      "learning_rate": 0.0002,
+      "loss": 0.9297,
+      "step": 7650
+    },
+    {
+      "epoch": 5.544697792254795,
+      "grad_norm": 1.0752092599868774,
+      "learning_rate": 0.0002,
+      "loss": 1.1152,
+      "step": 7660
+    },
+    {
+      "epoch": 5.551936301121969,
+      "grad_norm": 1.210389256477356,
+      "learning_rate": 0.0002,
+      "loss": 0.9344,
+      "step": 7670
+    },
+    {
+      "epoch": 5.559174809989142,
+      "grad_norm": 1.669063925743103,
+      "learning_rate": 0.0002,
+      "loss": 1.0349,
+      "step": 7680
+    },
+    {
+      "epoch": 5.566413318856315,
+      "grad_norm": 1.038020133972168,
+      "learning_rate": 0.0002,
+      "loss": 0.9833,
+      "step": 7690
+    },
+    {
+      "epoch": 5.573651827723489,
+      "grad_norm": 1.316673994064331,
+      "learning_rate": 0.0002,
+      "loss": 0.8907,
+      "step": 7700
+    },
+    {
+      "epoch": 5.580890336590662,
+      "grad_norm": 1.029935359954834,
+      "learning_rate": 0.0002,
+      "loss": 0.9614,
+      "step": 7710
+    },
+    {
+      "epoch": 5.5881288454578355,
+      "grad_norm": 0.9401940703392029,
+      "learning_rate": 0.0002,
+      "loss": 1.0409,
+      "step": 7720
+    },
+    {
+      "epoch": 5.595367354325009,
+      "grad_norm": 2.4811816215515137,
+      "learning_rate": 0.0002,
+      "loss": 0.9272,
+      "step": 7730
+    },
+    {
+      "epoch": 5.602605863192182,
+      "grad_norm": 1.0329105854034424,
+      "learning_rate": 0.0002,
+      "loss": 0.992,
+      "step": 7740
+    },
+    {
+      "epoch": 5.609844372059356,
+      "grad_norm": 1.479629635810852,
+      "learning_rate": 0.0002,
+      "loss": 0.9493,
+      "step": 7750
+    },
+    {
+      "epoch": 5.617082880926529,
+      "grad_norm": 1.9232319593429565,
+      "learning_rate": 0.0002,
+      "loss": 1.0727,
+      "step": 7760
+    },
+    {
+      "epoch": 5.624321389793702,
+      "grad_norm": 1.0055509805679321,
+      "learning_rate": 0.0002,
+      "loss": 1.0741,
+      "step": 7770
+    },
+    {
+      "epoch": 5.631559898660876,
+      "grad_norm": 1.0037437677383423,
+      "learning_rate": 0.0002,
+      "loss": 1.0731,
+      "step": 7780
+    },
+    {
+      "epoch": 5.638798407528049,
+      "grad_norm": 1.4245030879974365,
+      "learning_rate": 0.0002,
+      "loss": 1.0913,
+      "step": 7790
+    },
+    {
+      "epoch": 5.646036916395222,
+      "grad_norm": 1.080687403678894,
+      "learning_rate": 0.0002,
+      "loss": 0.9711,
+      "step": 7800
+    },
+    {
+      "epoch": 5.653275425262396,
+      "grad_norm": 1.354953408241272,
+      "learning_rate": 0.0002,
+      "loss": 1.0276,
+      "step": 7810
+    },
+    {
+      "epoch": 5.660513934129569,
+      "grad_norm": 0.8966761231422424,
+      "learning_rate": 0.0002,
+      "loss": 1.0534,
+      "step": 7820
+    },
+    {
+      "epoch": 5.6677524429967425,
+      "grad_norm": 1.0675480365753174,
+      "learning_rate": 0.0002,
+      "loss": 1.0662,
+      "step": 7830
+    },
+    {
+      "epoch": 5.674990951863916,
+      "grad_norm": 1.2104216814041138,
+      "learning_rate": 0.0002,
+      "loss": 1.1077,
+      "step": 7840
+    },
+    {
+      "epoch": 5.682229460731089,
+      "grad_norm": 1.105790376663208,
+      "learning_rate": 0.0002,
+      "loss": 0.9627,
+      "step": 7850
+    },
+    {
+      "epoch": 5.689467969598263,
+      "grad_norm": 1.0915391445159912,
+      "learning_rate": 0.0002,
+      "loss": 1.0483,
+      "step": 7860
+    },
+    {
+      "epoch": 5.696706478465436,
+      "grad_norm": 0.8957812786102295,
+      "learning_rate": 0.0002,
+      "loss": 1.0291,
+      "step": 7870
+    },
+    {
+      "epoch": 5.703944987332609,
+      "grad_norm": 1.9189311265945435,
+      "learning_rate": 0.0002,
+      "loss": 0.9785,
+      "step": 7880
+    },
+    {
+      "epoch": 5.711183496199783,
+      "grad_norm": 1.0867321491241455,
+      "learning_rate": 0.0002,
+      "loss": 1.0076,
+      "step": 7890
+    },
+    {
+      "epoch": 5.718422005066956,
+      "grad_norm": 1.0233147144317627,
+      "learning_rate": 0.0002,
+      "loss": 1.0236,
+      "step": 7900
+    },
+    {
+      "epoch": 5.7256605139341294,
+      "grad_norm": 1.16460382938385,
+      "learning_rate": 0.0002,
+      "loss": 0.9872,
+      "step": 7910
+    },
+    {
+      "epoch": 5.732899022801303,
+      "grad_norm": 1.1098358631134033,
+      "learning_rate": 0.0002,
+      "loss": 1.0762,
+      "step": 7920
+    },
+    {
+      "epoch": 5.740137531668476,
+      "grad_norm": 0.8555701375007629,
+      "learning_rate": 0.0002,
+      "loss": 0.9937,
+      "step": 7930
+    },
+    {
+      "epoch": 5.7473760405356495,
+      "grad_norm": 0.9885705709457397,
+      "learning_rate": 0.0002,
+      "loss": 1.0081,
+      "step": 7940
+    },
+    {
+      "epoch": 5.754614549402823,
+      "grad_norm": 0.9184203147888184,
+      "learning_rate": 0.0002,
+      "loss": 0.9909,
+      "step": 7950
+    },
+    {
+      "epoch": 5.761853058269996,
+      "grad_norm": 0.9653698205947876,
+      "learning_rate": 0.0002,
+      "loss": 1.0767,
+      "step": 7960
+    },
+    {
+      "epoch": 5.76909156713717,
+      "grad_norm": 1.0014251470565796,
+      "learning_rate": 0.0002,
+      "loss": 0.9317,
+      "step": 7970
+    },
+    {
+      "epoch": 5.776330076004343,
+      "grad_norm": 1.004701018333435,
+      "learning_rate": 0.0002,
+      "loss": 1.0271,
+      "step": 7980
+    },
+    {
+      "epoch": 5.783568584871516,
+      "grad_norm": 0.950577974319458,
+      "learning_rate": 0.0002,
+      "loss": 1.0397,
+      "step": 7990
+    },
+    {
+      "epoch": 5.79080709373869,
+      "grad_norm": 1.2986834049224854,
+      "learning_rate": 0.0002,
+      "loss": 0.9725,
+      "step": 8000
+    },
+    {
+      "epoch": 5.798045602605863,
+      "grad_norm": 1.3353424072265625,
+      "learning_rate": 0.0002,
+      "loss": 1.039,
+      "step": 8010
+    },
+    {
+      "epoch": 5.8052841114730365,
+      "grad_norm": 0.7650562524795532,
+      "learning_rate": 0.0002,
+      "loss": 1.0626,
+      "step": 8020
+    },
+    {
+      "epoch": 5.81252262034021,
+      "grad_norm": 1.0156235694885254,
+      "learning_rate": 0.0002,
+      "loss": 1.0802,
+      "step": 8030
+    },
+    {
+      "epoch": 5.819761129207383,
+      "grad_norm": 1.3092900514602661,
+      "learning_rate": 0.0002,
+      "loss": 1.0185,
+      "step": 8040
+    },
+    {
+      "epoch": 5.826999638074557,
+      "grad_norm": 1.184428095817566,
+      "learning_rate": 0.0002,
+      "loss": 0.9905,
+      "step": 8050
+    },
+    {
+      "epoch": 5.83423814694173,
+      "grad_norm": 0.979401707649231,
+      "learning_rate": 0.0002,
+      "loss": 1.0548,
+      "step": 8060
+    },
+    {
+      "epoch": 5.841476655808903,
+      "grad_norm": 1.3557400703430176,
+      "learning_rate": 0.0002,
+      "loss": 0.9721,
+      "step": 8070
+    },
+    {
+      "epoch": 5.848715164676077,
+      "grad_norm": 0.8429333567619324,
+      "learning_rate": 0.0002,
+      "loss": 1.0235,
+      "step": 8080
+    },
+    {
+      "epoch": 5.85595367354325,
+      "grad_norm": 1.3167692422866821,
+      "learning_rate": 0.0002,
+      "loss": 0.952,
+      "step": 8090
+    },
+    {
+      "epoch": 5.863192182410423,
+      "grad_norm": 0.9750998020172119,
+      "learning_rate": 0.0002,
+      "loss": 0.9609,
+      "step": 8100
+    },
+    {
+      "epoch": 5.870430691277597,
+      "grad_norm": 1.1869813203811646,
+      "learning_rate": 0.0002,
+      "loss": 1.0789,
+      "step": 8110
+    },
+    {
+      "epoch": 5.87766920014477,
+      "grad_norm": 1.508615255355835,
+      "learning_rate": 0.0002,
+      "loss": 1.0331,
+      "step": 8120
+    },
+    {
+      "epoch": 5.8849077090119435,
+      "grad_norm": 0.9439908266067505,
+      "learning_rate": 0.0002,
+      "loss": 1.0171,
+      "step": 8130
+    },
+    {
+      "epoch": 5.892146217879117,
+      "grad_norm": 0.910508930683136,
+      "learning_rate": 0.0002,
+      "loss": 0.9682,
+      "step": 8140
+    },
+    {
+      "epoch": 5.89938472674629,
+      "grad_norm": 1.111501932144165,
+      "learning_rate": 0.0002,
+      "loss": 1.0032,
+      "step": 8150
+    },
+    {
+      "epoch": 5.906623235613464,
+      "grad_norm": 0.726554274559021,
+      "learning_rate": 0.0002,
+      "loss": 1.0266,
+      "step": 8160
+    },
+    {
+      "epoch": 5.913861744480637,
+      "grad_norm": 1.1084556579589844,
+      "learning_rate": 0.0002,
+      "loss": 1.0681,
+      "step": 8170
+    },
+    {
+      "epoch": 5.92110025334781,
+      "grad_norm": 0.9695167541503906,
+      "learning_rate": 0.0002,
+      "loss": 0.969,
+      "step": 8180
+    },
+    {
+      "epoch": 5.928338762214984,
+      "grad_norm": 1.1169592142105103,
+      "learning_rate": 0.0002,
+      "loss": 0.9858,
+      "step": 8190
+    },
+    {
+      "epoch": 5.935577271082157,
+      "grad_norm": 1.5116780996322632,
+      "learning_rate": 0.0002,
+      "loss": 1.0924,
+      "step": 8200
+    },
+    {
+      "epoch": 5.94281577994933,
+      "grad_norm": 1.0073388814926147,
+      "learning_rate": 0.0002,
+      "loss": 0.878,
+      "step": 8210
+    },
+    {
+      "epoch": 5.950054288816504,
+      "grad_norm": 0.9323263168334961,
+      "learning_rate": 0.0002,
+      "loss": 1.0462,
+      "step": 8220
+    },
+    {
+      "epoch": 5.957292797683677,
+      "grad_norm": 0.9422887563705444,
+      "learning_rate": 0.0002,
+      "loss": 1.0291,
+      "step": 8230
+    },
+    {
+      "epoch": 5.9645313065508505,
+      "grad_norm": 0.9691047668457031,
+      "learning_rate": 0.0002,
+      "loss": 0.953,
+      "step": 8240
+    },
+    {
+      "epoch": 5.971769815418024,
+      "grad_norm": 0.9650622606277466,
+      "learning_rate": 0.0002,
+      "loss": 0.9842,
+      "step": 8250
+    },
+    {
+      "epoch": 5.979008324285197,
+      "grad_norm": 1.077958345413208,
+      "learning_rate": 0.0002,
+      "loss": 0.907,
+      "step": 8260
+    },
+    {
+      "epoch": 5.986246833152371,
+      "grad_norm": 0.8946306109428406,
+      "learning_rate": 0.0002,
+      "loss": 0.9162,
+      "step": 8270
+    },
+    {
+      "epoch": 5.993485342019544,
+      "grad_norm": 1.34098219871521,
+      "learning_rate": 0.0002,
+      "loss": 1.0439,
+      "step": 8280
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 1.4714229106903076,
+      "eval_runtime": 26.301,
+      "eval_samples_per_second": 16.577,
+      "eval_steps_per_second": 2.091,
+      "step": 8289
+    },
+    {
+      "epoch": 6.000723850886717,
+      "grad_norm": 0.9737564325332642,
+      "learning_rate": 0.0002,
+      "loss": 1.1403,
+      "step": 8290
+    },
+    {
+      "epoch": 6.007962359753891,
+      "grad_norm": 1.2205945253372192,
+      "learning_rate": 0.0002,
+      "loss": 0.8875,
+      "step": 8300
+    },
+    {
+      "epoch": 6.015200868621064,
+      "grad_norm": 1.3529434204101562,
+      "learning_rate": 0.0002,
+      "loss": 0.8623,
+      "step": 8310
+    },
+    {
+      "epoch": 6.022439377488237,
+      "grad_norm": 1.2300174236297607,
+      "learning_rate": 0.0002,
+      "loss": 0.9427,
+      "step": 8320
+    },
+    {
+      "epoch": 6.029677886355411,
+      "grad_norm": 0.9248194098472595,
+      "learning_rate": 0.0002,
+      "loss": 0.9322,
+      "step": 8330
+    },
+    {
+      "epoch": 6.036916395222584,
+      "grad_norm": 1.1140035390853882,
+      "learning_rate": 0.0002,
+      "loss": 0.9302,
+      "step": 8340
+    },
+    {
+      "epoch": 6.0441549040897575,
+      "grad_norm": 1.2097352743148804,
+      "learning_rate": 0.0002,
+      "loss": 0.8255,
+      "step": 8350
+    },
+    {
+      "epoch": 6.051393412956931,
+      "grad_norm": 0.9472483396530151,
+      "learning_rate": 0.0002,
+      "loss": 0.8792,
+      "step": 8360
+    },
+    {
+      "epoch": 6.058631921824104,
+      "grad_norm": 1.0195368528366089,
+      "learning_rate": 0.0002,
+      "loss": 0.8865,
+      "step": 8370
+    },
+    {
+      "epoch": 6.065870430691278,
+      "grad_norm": 1.182735562324524,
+      "learning_rate": 0.0002,
+      "loss": 0.8858,
+      "step": 8380
+    },
+    {
+      "epoch": 6.073108939558451,
+      "grad_norm": 1.1042858362197876,
+      "learning_rate": 0.0002,
+      "loss": 0.9455,
+      "step": 8390
+    },
+    {
+      "epoch": 6.080347448425624,
+      "grad_norm": 0.8606401085853577,
+      "learning_rate": 0.0002,
+      "loss": 0.9723,
+      "step": 8400
+    },
+    {
+      "epoch": 6.087585957292798,
+      "grad_norm": 1.1015676259994507,
+      "learning_rate": 0.0002,
+      "loss": 0.8436,
+      "step": 8410
+    },
+    {
+      "epoch": 6.094824466159971,
+      "grad_norm": 1.690224289894104,
+      "learning_rate": 0.0002,
+      "loss": 0.8845,
+      "step": 8420
+    },
+    {
+      "epoch": 6.1020629750271445,
+      "grad_norm": 1.1928749084472656,
+      "learning_rate": 0.0002,
+      "loss": 0.8484,
+      "step": 8430
+    },
+    {
+      "epoch": 6.109301483894318,
+      "grad_norm": 1.0816864967346191,
+      "learning_rate": 0.0002,
+      "loss": 0.9546,
+      "step": 8440
+    },
+    {
+      "epoch": 6.116539992761491,
+      "grad_norm": 1.1638226509094238,
+      "learning_rate": 0.0002,
+      "loss": 0.8286,
+      "step": 8450
+    },
+    {
+      "epoch": 6.1237785016286646,
+      "grad_norm": 1.3782968521118164,
+      "learning_rate": 0.0002,
+      "loss": 0.8749,
+      "step": 8460
+    },
+    {
+      "epoch": 6.131017010495838,
+      "grad_norm": 1.2030094861984253,
+      "learning_rate": 0.0002,
+      "loss": 0.7956,
+      "step": 8470
+    },
+    {
+      "epoch": 6.138255519363011,
+      "grad_norm": 1.3227659463882446,
+      "learning_rate": 0.0002,
+      "loss": 0.8393,
+      "step": 8480
+    },
+    {
+      "epoch": 6.145494028230185,
+      "grad_norm": 1.104384422302246,
+      "learning_rate": 0.0002,
+      "loss": 0.9175,
+      "step": 8490
+    },
+    {
+      "epoch": 6.152732537097358,
+      "grad_norm": 1.518805980682373,
+      "learning_rate": 0.0002,
+      "loss": 0.861,
+      "step": 8500
+    },
+    {
+      "epoch": 6.159971045964531,
+      "grad_norm": 1.2029093503952026,
+      "learning_rate": 0.0002,
+      "loss": 0.9169,
+      "step": 8510
+    },
+    {
+      "epoch": 6.167209554831705,
+      "grad_norm": 1.2991217374801636,
+      "learning_rate": 0.0002,
+      "loss": 0.8701,
+      "step": 8520
+    },
+    {
+      "epoch": 6.174448063698878,
+      "grad_norm": 1.7002956867218018,
+      "learning_rate": 0.0002,
+      "loss": 0.9748,
+      "step": 8530
+    },
+    {
+      "epoch": 6.1816865725660515,
+      "grad_norm": 1.6653581857681274,
+      "learning_rate": 0.0002,
+      "loss": 0.8881,
+      "step": 8540
+    },
+    {
+      "epoch": 6.188925081433225,
+      "grad_norm": 1.0493303537368774,
+      "learning_rate": 0.0002,
+      "loss": 0.817,
+      "step": 8550
+    },
+    {
+      "epoch": 6.196163590300398,
+      "grad_norm": 1.539345622062683,
+      "learning_rate": 0.0002,
+      "loss": 0.8726,
+      "step": 8560
+    },
+    {
+      "epoch": 6.203402099167572,
+      "grad_norm": 1.2757070064544678,
+      "learning_rate": 0.0002,
+      "loss": 0.9452,
+      "step": 8570
+    },
+    {
+      "epoch": 6.210640608034745,
+      "grad_norm": 1.2416890859603882,
+      "learning_rate": 0.0002,
+      "loss": 0.8773,
+      "step": 8580
+    },
+    {
+      "epoch": 6.217879116901918,
+      "grad_norm": 1.617621898651123,
+      "learning_rate": 0.0002,
+      "loss": 0.815,
+      "step": 8590
+    },
+    {
+      "epoch": 6.225117625769092,
+      "grad_norm": 1.058962106704712,
+      "learning_rate": 0.0002,
+      "loss": 0.9137,
+      "step": 8600
+    },
+    {
+      "epoch": 6.232356134636265,
+      "grad_norm": 1.1489088535308838,
+      "learning_rate": 0.0002,
+      "loss": 0.8164,
+      "step": 8610
+    },
+    {
+      "epoch": 6.239594643503438,
+      "grad_norm": 0.9391577243804932,
+      "learning_rate": 0.0002,
+      "loss": 0.9476,
+      "step": 8620
+    },
+    {
+      "epoch": 6.246833152370612,
+      "grad_norm": 1.363706111907959,
+      "learning_rate": 0.0002,
+      "loss": 0.932,
+      "step": 8630
+    },
+    {
+      "epoch": 6.254071661237785,
+      "grad_norm": 0.779502809047699,
+      "learning_rate": 0.0002,
+      "loss": 0.8917,
+      "step": 8640
+    },
+    {
+      "epoch": 6.2613101701049585,
+      "grad_norm": 2.000821590423584,
+      "learning_rate": 0.0002,
+      "loss": 0.9196,
+      "step": 8650
+    },
+    {
+      "epoch": 6.268548678972132,
+      "grad_norm": 1.1521023511886597,
+      "learning_rate": 0.0002,
+      "loss": 0.9794,
+      "step": 8660
+    },
+    {
+      "epoch": 6.275787187839305,
+      "grad_norm": 1.3734570741653442,
+      "learning_rate": 0.0002,
+      "loss": 0.9147,
+      "step": 8670
+    },
+    {
+      "epoch": 6.283025696706479,
+      "grad_norm": 0.9550670385360718,
+      "learning_rate": 0.0002,
+      "loss": 0.795,
+      "step": 8680
+    },
+    {
+      "epoch": 6.290264205573652,
+      "grad_norm": 0.8937032222747803,
+      "learning_rate": 0.0002,
+      "loss": 0.9049,
+      "step": 8690
+    },
+    {
+      "epoch": 6.297502714440825,
+      "grad_norm": 1.3352779150009155,
+      "learning_rate": 0.0002,
+      "loss": 0.8526,
+      "step": 8700
+    },
+    {
+      "epoch": 6.304741223307999,
+      "grad_norm": 1.3057222366333008,
+      "learning_rate": 0.0002,
+      "loss": 0.8572,
+      "step": 8710
+    },
+    {
+      "epoch": 6.311979732175172,
+      "grad_norm": 0.9078314304351807,
+      "learning_rate": 0.0002,
+      "loss": 0.8825,
+      "step": 8720
+    },
+    {
+      "epoch": 6.319218241042345,
+      "grad_norm": 1.6663457155227661,
+      "learning_rate": 0.0002,
+      "loss": 0.8666,
+      "step": 8730
+    },
+    {
+      "epoch": 6.326456749909519,
+      "grad_norm": 1.2043739557266235,
+      "learning_rate": 0.0002,
+      "loss": 0.927,
+      "step": 8740
+    },
+    {
+      "epoch": 6.333695258776692,
+      "grad_norm": 0.9165967702865601,
+      "learning_rate": 0.0002,
+      "loss": 0.8014,
+      "step": 8750
+    },
+    {
+      "epoch": 6.3409337676438655,
+      "grad_norm": 1.016452670097351,
+      "learning_rate": 0.0002,
+      "loss": 0.9761,
+      "step": 8760
+    },
+    {
+      "epoch": 6.348172276511039,
+      "grad_norm": 1.2209261655807495,
+      "learning_rate": 0.0002,
+      "loss": 1.022,
+      "step": 8770
+    },
+    {
+      "epoch": 6.355410785378212,
+      "grad_norm": 1.3380663394927979,
+      "learning_rate": 0.0002,
+      "loss": 0.8012,
+      "step": 8780
+    },
+    {
+      "epoch": 6.362649294245386,
+      "grad_norm": 2.3311562538146973,
+      "learning_rate": 0.0002,
+      "loss": 0.9553,
+      "step": 8790
+    },
+    {
+      "epoch": 6.369887803112559,
+      "grad_norm": 1.0330604314804077,
+      "learning_rate": 0.0002,
+      "loss": 0.8676,
+      "step": 8800
+    },
+    {
+      "epoch": 6.377126311979732,
+      "grad_norm": 0.9655511975288391,
+      "learning_rate": 0.0002,
+      "loss": 0.98,
+      "step": 8810
+    },
+    {
+      "epoch": 6.384364820846906,
+      "grad_norm": 1.1065765619277954,
+      "learning_rate": 0.0002,
+      "loss": 1.0324,
+      "step": 8820
+    },
+    {
+      "epoch": 6.391603329714079,
+      "grad_norm": 1.2631285190582275,
+      "learning_rate": 0.0002,
+      "loss": 1.0078,
+      "step": 8830
+    },
+    {
+      "epoch": 6.398841838581252,
+      "grad_norm": 0.92459636926651,
+      "learning_rate": 0.0002,
+      "loss": 0.8989,
+      "step": 8840
+    },
+    {
+      "epoch": 6.406080347448426,
+      "grad_norm": 0.9982633590698242,
+      "learning_rate": 0.0002,
+      "loss": 0.8536,
+      "step": 8850
+    },
+    {
+      "epoch": 6.413318856315599,
+      "grad_norm": 1.0746768712997437,
+      "learning_rate": 0.0002,
+      "loss": 0.8949,
+      "step": 8860
+    },
+    {
+      "epoch": 6.4205573651827725,
+      "grad_norm": 1.3024073839187622,
+      "learning_rate": 0.0002,
+      "loss": 0.8547,
+      "step": 8870
+    },
+    {
+      "epoch": 6.427795874049946,
+      "grad_norm": 1.2764527797698975,
+      "learning_rate": 0.0002,
+      "loss": 0.9618,
+      "step": 8880
+    },
+    {
+      "epoch": 6.435034382917119,
+      "grad_norm": 0.8318809270858765,
+      "learning_rate": 0.0002,
+      "loss": 0.8905,
+      "step": 8890
+    },
+    {
+      "epoch": 6.442272891784293,
+      "grad_norm": 1.7350783348083496,
+      "learning_rate": 0.0002,
+      "loss": 0.917,
+      "step": 8900
+    },
+    {
+      "epoch": 6.449511400651466,
+      "grad_norm": 1.3430488109588623,
+      "learning_rate": 0.0002,
+      "loss": 1.0229,
+      "step": 8910
+    },
+    {
+      "epoch": 6.456749909518639,
+      "grad_norm": 1.5907495021820068,
+      "learning_rate": 0.0002,
+      "loss": 0.9678,
+      "step": 8920
+    },
+    {
+      "epoch": 6.463988418385813,
+      "grad_norm": 1.8579202890396118,
+      "learning_rate": 0.0002,
+      "loss": 0.9639,
+      "step": 8930
+    },
+    {
+      "epoch": 6.471226927252986,
+      "grad_norm": 1.2233413457870483,
+      "learning_rate": 0.0002,
+      "loss": 0.9302,
+      "step": 8940
+    },
+    {
+      "epoch": 6.4784654361201595,
+      "grad_norm": 1.009103775024414,
+      "learning_rate": 0.0002,
+      "loss": 0.9169,
+      "step": 8950
+    },
+    {
+      "epoch": 6.485703944987333,
+      "grad_norm": 1.1265181303024292,
+      "learning_rate": 0.0002,
+      "loss": 0.8969,
+      "step": 8960
+    },
+    {
+      "epoch": 6.492942453854506,
+      "grad_norm": 1.1733338832855225,
+      "learning_rate": 0.0002,
+      "loss": 0.8374,
+      "step": 8970
+    },
+    {
+      "epoch": 6.50018096272168,
+      "grad_norm": 1.0444518327713013,
+      "learning_rate": 0.0002,
+      "loss": 0.8764,
+      "step": 8980
+    },
+    {
+      "epoch": 6.507419471588853,
+      "grad_norm": 1.2296479940414429,
+      "learning_rate": 0.0002,
+      "loss": 0.9582,
+      "step": 8990
+    },
+    {
+      "epoch": 6.514657980456026,
+      "grad_norm": 1.370417833328247,
+      "learning_rate": 0.0002,
+      "loss": 0.8557,
+      "step": 9000
+    },
+    {
+      "epoch": 6.5218964893232,
+      "grad_norm": 1.4787620306015015,
+      "learning_rate": 0.0002,
+      "loss": 0.9787,
+      "step": 9010
+    },
+    {
+      "epoch": 6.529134998190373,
+      "grad_norm": 0.8550514578819275,
+      "learning_rate": 0.0002,
+      "loss": 0.967,
+      "step": 9020
+    },
+    {
+      "epoch": 6.536373507057546,
+      "grad_norm": 1.2327991724014282,
+      "learning_rate": 0.0002,
+      "loss": 0.9755,
+      "step": 9030
+    },
+    {
+      "epoch": 6.54361201592472,
+      "grad_norm": 1.0915621519088745,
+      "learning_rate": 0.0002,
+      "loss": 0.9248,
+      "step": 9040
+    },
+    {
+      "epoch": 6.550850524791893,
+      "grad_norm": 1.7243309020996094,
+      "learning_rate": 0.0002,
+      "loss": 1.0024,
+      "step": 9050
+    },
+    {
+      "epoch": 6.5580890336590665,
+      "grad_norm": 0.954359769821167,
+      "learning_rate": 0.0002,
+      "loss": 1.0123,
+      "step": 9060
+    },
+    {
+      "epoch": 6.56532754252624,
+      "grad_norm": 1.066051959991455,
+      "learning_rate": 0.0002,
+      "loss": 0.8261,
+      "step": 9070
+    },
+    {
+      "epoch": 6.572566051393413,
+      "grad_norm": 1.200271487236023,
+      "learning_rate": 0.0002,
+      "loss": 0.944,
+      "step": 9080
+    },
+    {
+      "epoch": 6.579804560260587,
+      "grad_norm": 1.4331457614898682,
+      "learning_rate": 0.0002,
+      "loss": 0.9788,
+      "step": 9090
+    },
+    {
+      "epoch": 6.58704306912776,
+      "grad_norm": 1.0892444849014282,
+      "learning_rate": 0.0002,
+      "loss": 1.0216,
+      "step": 9100
+    },
+    {
+      "epoch": 6.594281577994933,
+      "grad_norm": 1.849726915359497,
+      "learning_rate": 0.0002,
+      "loss": 0.8557,
+      "step": 9110
+    },
+    {
+      "epoch": 6.601520086862107,
+      "grad_norm": 1.1228708028793335,
+      "learning_rate": 0.0002,
+      "loss": 0.9495,
+      "step": 9120
+    },
+    {
+      "epoch": 6.60875859572928,
+      "grad_norm": 1.0928595066070557,
+      "learning_rate": 0.0002,
+      "loss": 1.0169,
+      "step": 9130
+    },
+    {
+      "epoch": 6.615997104596453,
+      "grad_norm": 1.2138155698776245,
+      "learning_rate": 0.0002,
+      "loss": 0.9342,
+      "step": 9140
+    },
+    {
+      "epoch": 6.623235613463627,
+      "grad_norm": 1.5155235528945923,
+      "learning_rate": 0.0002,
+      "loss": 0.8715,
+      "step": 9150
+    },
+    {
+      "epoch": 6.6304741223308,
+      "grad_norm": 1.3194212913513184,
+      "learning_rate": 0.0002,
+      "loss": 0.9806,
+      "step": 9160
+    },
+    {
+      "epoch": 6.6377126311979735,
+      "grad_norm": 1.045623779296875,
+      "learning_rate": 0.0002,
+      "loss": 0.8958,
+      "step": 9170
+    },
+    {
+      "epoch": 6.644951140065147,
+      "grad_norm": 0.9647570252418518,
+      "learning_rate": 0.0002,
+      "loss": 0.8698,
+      "step": 9180
+    },
+    {
+      "epoch": 6.65218964893232,
+      "grad_norm": 1.0818220376968384,
+      "learning_rate": 0.0002,
+      "loss": 0.8829,
+      "step": 9190
+    },
+    {
+      "epoch": 6.659428157799494,
+      "grad_norm": 1.2792822122573853,
+      "learning_rate": 0.0002,
+      "loss": 0.9745,
+      "step": 9200
+    },
+    {
+      "epoch": 6.666666666666667,
+      "grad_norm": 1.2764191627502441,
+      "learning_rate": 0.0002,
+      "loss": 0.8854,
+      "step": 9210
+    },
+    {
+      "epoch": 6.67390517553384,
+      "grad_norm": 1.0552066564559937,
+      "learning_rate": 0.0002,
+      "loss": 0.9709,
+      "step": 9220
+    },
+    {
+      "epoch": 6.681143684401014,
+      "grad_norm": 1.082476019859314,
+      "learning_rate": 0.0002,
+      "loss": 0.8855,
+      "step": 9230
+    },
+    {
+      "epoch": 6.688382193268187,
+      "grad_norm": 1.3313323259353638,
+      "learning_rate": 0.0002,
+      "loss": 0.9779,
+      "step": 9240
+    },
+    {
+      "epoch": 6.69562070213536,
+      "grad_norm": 1.130048394203186,
+      "learning_rate": 0.0002,
+      "loss": 1.005,
+      "step": 9250
+    },
+    {
+      "epoch": 6.702859211002534,
+      "grad_norm": 1.1997296810150146,
+      "learning_rate": 0.0002,
+      "loss": 0.9969,
+      "step": 9260
+    },
+    {
+      "epoch": 6.710097719869707,
+      "grad_norm": 1.0591834783554077,
+      "learning_rate": 0.0002,
+      "loss": 0.8691,
+      "step": 9270
+    },
+    {
+      "epoch": 6.7173362287368805,
+      "grad_norm": 1.2722901105880737,
+      "learning_rate": 0.0002,
+      "loss": 0.9603,
+      "step": 9280
+    },
+    {
+      "epoch": 6.724574737604054,
+      "grad_norm": 1.1150950193405151,
+      "learning_rate": 0.0002,
+      "loss": 0.9227,
+      "step": 9290
+    },
+    {
+      "epoch": 6.731813246471227,
+      "grad_norm": 1.1575992107391357,
+      "learning_rate": 0.0002,
+      "loss": 0.95,
+      "step": 9300
+    },
+    {
+      "epoch": 6.739051755338401,
+      "grad_norm": 0.9371691346168518,
+      "learning_rate": 0.0002,
+      "loss": 0.9822,
+      "step": 9310
+    },
+    {
+      "epoch": 6.746290264205574,
+      "grad_norm": 1.4924226999282837,
+      "learning_rate": 0.0002,
+      "loss": 0.9773,
+      "step": 9320
+    },
+    {
+      "epoch": 6.753528773072747,
+      "grad_norm": 1.1524218320846558,
+      "learning_rate": 0.0002,
+      "loss": 0.969,
+      "step": 9330
+    },
+    {
+      "epoch": 6.760767281939921,
+      "grad_norm": 0.9500471949577332,
+      "learning_rate": 0.0002,
+      "loss": 0.9271,
+      "step": 9340
+    },
+    {
+      "epoch": 6.768005790807094,
+      "grad_norm": 1.2062290906906128,
+      "learning_rate": 0.0002,
+      "loss": 0.9029,
+      "step": 9350
+    },
+    {
+      "epoch": 6.7752442996742674,
+      "grad_norm": 1.212631106376648,
+      "learning_rate": 0.0002,
+      "loss": 0.9121,
+      "step": 9360
+    },
+    {
+      "epoch": 6.782482808541441,
+      "grad_norm": 1.9135472774505615,
+      "learning_rate": 0.0002,
+      "loss": 0.8486,
+      "step": 9370
+    },
+    {
+      "epoch": 6.789721317408614,
+      "grad_norm": 0.9682775139808655,
+      "learning_rate": 0.0002,
+      "loss": 0.9332,
+      "step": 9380
+    },
+    {
+      "epoch": 6.7969598262757875,
+      "grad_norm": 1.1405237913131714,
+      "learning_rate": 0.0002,
+      "loss": 0.8548,
+      "step": 9390
+    },
+    {
+      "epoch": 6.804198335142961,
+      "grad_norm": 1.6855751276016235,
+      "learning_rate": 0.0002,
+      "loss": 0.8922,
+      "step": 9400
+    },
+    {
+      "epoch": 6.811436844010134,
+      "grad_norm": 1.6590169668197632,
+      "learning_rate": 0.0002,
+      "loss": 0.9417,
+      "step": 9410
+    },
+    {
+      "epoch": 6.818675352877308,
+      "grad_norm": 1.8795170783996582,
+      "learning_rate": 0.0002,
+      "loss": 0.868,
+      "step": 9420
+    },
+    {
+      "epoch": 6.825913861744481,
+      "grad_norm": 1.1087183952331543,
+      "learning_rate": 0.0002,
+      "loss": 0.9142,
+      "step": 9430
+    },
+    {
+      "epoch": 6.833152370611654,
+      "grad_norm": 1.4178446531295776,
+      "learning_rate": 0.0002,
+      "loss": 1.1427,
+      "step": 9440
+    },
+    {
+      "epoch": 6.840390879478828,
+      "grad_norm": 1.0792350769042969,
+      "learning_rate": 0.0002,
+      "loss": 0.8325,
+      "step": 9450
+    },
+    {
+      "epoch": 6.847629388346001,
+      "grad_norm": 1.2159196138381958,
+      "learning_rate": 0.0002,
+      "loss": 1.0078,
+      "step": 9460
+    },
+    {
+      "epoch": 6.8548678972131745,
+      "grad_norm": 0.9998821020126343,
+      "learning_rate": 0.0002,
+      "loss": 0.9536,
+      "step": 9470
+    },
+    {
+      "epoch": 6.862106406080348,
+      "grad_norm": 0.7940687537193298,
+      "learning_rate": 0.0002,
+      "loss": 0.9277,
+      "step": 9480
+    },
+    {
+      "epoch": 6.869344914947521,
+      "grad_norm": 0.9572826027870178,
+      "learning_rate": 0.0002,
+      "loss": 0.8612,
+      "step": 9490
+    },
+    {
+      "epoch": 6.876583423814694,
+      "grad_norm": 1.1086537837982178,
+      "learning_rate": 0.0002,
+      "loss": 0.9611,
+      "step": 9500
+    },
+    {
+      "epoch": 6.883821932681867,
+      "grad_norm": 1.1934887170791626,
+      "learning_rate": 0.0002,
+      "loss": 0.9276,
+      "step": 9510
+    },
+    {
+      "epoch": 6.89106044154904,
+      "grad_norm": 1.207324504852295,
+      "learning_rate": 0.0002,
+      "loss": 0.8416,
+      "step": 9520
+    },
+    {
+      "epoch": 6.898298950416214,
+      "grad_norm": 1.1303677558898926,
+      "learning_rate": 0.0002,
+      "loss": 0.9378,
+      "step": 9530
+    },
+    {
+      "epoch": 6.905537459283387,
+      "grad_norm": 1.4958926439285278,
+      "learning_rate": 0.0002,
+      "loss": 0.9599,
+      "step": 9540
+    },
+    {
+      "epoch": 6.9127759681505605,
+      "grad_norm": 1.2141553163528442,
+      "learning_rate": 0.0002,
+      "loss": 0.9365,
+      "step": 9550
+    },
+    {
+      "epoch": 6.920014477017734,
+      "grad_norm": 1.6544346809387207,
+      "learning_rate": 0.0002,
+      "loss": 1.0291,
+      "step": 9560
+    },
+    {
+      "epoch": 6.927252985884907,
+      "grad_norm": 1.0540320873260498,
+      "learning_rate": 0.0002,
+      "loss": 0.8439,
+      "step": 9570
+    },
+    {
+      "epoch": 6.934491494752081,
+      "grad_norm": 1.3095581531524658,
+      "learning_rate": 0.0002,
+      "loss": 0.9831,
+      "step": 9580
+    },
+    {
+      "epoch": 6.941730003619254,
+      "grad_norm": 1.4509341716766357,
+      "learning_rate": 0.0002,
+      "loss": 0.8694,
+      "step": 9590
+    },
+    {
+      "epoch": 6.948968512486427,
+      "grad_norm": 1.1091740131378174,
+      "learning_rate": 0.0002,
+      "loss": 0.983,
+      "step": 9600
+    },
+    {
+      "epoch": 6.956207021353601,
+      "grad_norm": 1.102929949760437,
+      "learning_rate": 0.0002,
+      "loss": 0.9126,
+      "step": 9610
+    },
+    {
+      "epoch": 6.963445530220774,
+      "grad_norm": 1.1377743482589722,
+      "learning_rate": 0.0002,
+      "loss": 0.9622,
+      "step": 9620
+    },
+    {
+      "epoch": 6.970684039087947,
+      "grad_norm": 1.2070361375808716,
+      "learning_rate": 0.0002,
+      "loss": 0.9045,
+      "step": 9630
+    },
+    {
+      "epoch": 6.977922547955121,
+      "grad_norm": 1.30153489112854,
+      "learning_rate": 0.0002,
+      "loss": 0.9714,
+      "step": 9640
+    },
+    {
+      "epoch": 6.985161056822294,
+      "grad_norm": 1.4641543626785278,
+      "learning_rate": 0.0002,
+      "loss": 0.9555,
+      "step": 9650
+    },
+    {
+      "epoch": 6.9923995656894675,
+      "grad_norm": 1.0497819185256958,
+      "learning_rate": 0.0002,
+      "loss": 0.9177,
+      "step": 9660
+    },
+    {
+      "epoch": 6.999638074556641,
+      "grad_norm": 1.2500354051589966,
+      "learning_rate": 0.0002,
+      "loss": 0.8369,
+      "step": 9670
+    },
+    {
+      "epoch": 6.999638074556641,
+      "eval_loss": 1.518465518951416,
+      "eval_runtime": 26.4525,
+      "eval_samples_per_second": 16.482,
+      "eval_steps_per_second": 2.079,
+      "step": 9670
+    },
+    {
+      "epoch": 7.006876583423814,
+      "grad_norm": 1.0240943431854248,
+      "learning_rate": 0.0002,
+      "loss": 0.7792,
+      "step": 9680
+    },
+    {
+      "epoch": 7.014115092290988,
+      "grad_norm": 1.2250308990478516,
+      "learning_rate": 0.0002,
+      "loss": 0.7812,
+      "step": 9690
+    },
+    {
+      "epoch": 7.021353601158161,
+      "grad_norm": 1.397510290145874,
+      "learning_rate": 0.0002,
+      "loss": 0.8463,
+      "step": 9700
+    },
+    {
+      "epoch": 7.028592110025334,
+      "grad_norm": 1.9754822254180908,
+      "learning_rate": 0.0002,
+      "loss": 0.738,
+      "step": 9710
+    },
+    {
+      "epoch": 7.035830618892508,
+      "grad_norm": 1.7932360172271729,
+      "learning_rate": 0.0002,
+      "loss": 0.8017,
+      "step": 9720
+    },
+    {
+      "epoch": 7.043069127759681,
+      "grad_norm": 0.8552590608596802,
+      "learning_rate": 0.0002,
+      "loss": 0.7427,
+      "step": 9730
+    },
+    {
+      "epoch": 7.0503076366268544,
+      "grad_norm": 1.758886694908142,
+      "learning_rate": 0.0002,
+      "loss": 0.8204,
+      "step": 9740
+    },
+    {
+      "epoch": 7.057546145494028,
+      "grad_norm": 1.5239284038543701,
+      "learning_rate": 0.0002,
+      "loss": 0.8444,
+      "step": 9750
+    },
+    {
+      "epoch": 7.064784654361201,
+      "grad_norm": 1.2506821155548096,
+      "learning_rate": 0.0002,
+      "loss": 0.9078,
+      "step": 9760
+    },
+    {
+      "epoch": 7.0720231632283745,
+      "grad_norm": 1.274202823638916,
+      "learning_rate": 0.0002,
+      "loss": 0.7512,
+      "step": 9770
+    },
+    {
+      "epoch": 7.079261672095548,
+      "grad_norm": 1.296419620513916,
+      "learning_rate": 0.0002,
+      "loss": 0.7693,
+      "step": 9780
+    },
+    {
+      "epoch": 7.086500180962721,
+      "grad_norm": 1.3418468236923218,
+      "learning_rate": 0.0002,
+      "loss": 0.8311,
+      "step": 9790
+    },
+    {
+      "epoch": 7.093738689829895,
+      "grad_norm": 1.1935685873031616,
+      "learning_rate": 0.0002,
+      "loss": 0.8302,
+      "step": 9800
+    },
+    {
+      "epoch": 7.100977198697068,
+      "grad_norm": 1.2341830730438232,
+      "learning_rate": 0.0002,
+      "loss": 0.8037,
+      "step": 9810
+    },
+    {
+      "epoch": 7.108215707564241,
+      "grad_norm": 1.3398581743240356,
+      "learning_rate": 0.0002,
+      "loss": 0.7887,
+      "step": 9820
+    },
+    {
+      "epoch": 7.115454216431415,
+      "grad_norm": 1.1919665336608887,
+      "learning_rate": 0.0002,
+      "loss": 0.824,
+      "step": 9830
+    },
+    {
+      "epoch": 7.122692725298588,
+      "grad_norm": 0.9331274032592773,
+      "learning_rate": 0.0002,
+      "loss": 0.7296,
+      "step": 9840
+    },
+    {
+      "epoch": 7.1299312341657615,
+      "grad_norm": 1.0933221578598022,
+      "learning_rate": 0.0002,
+      "loss": 0.8208,
+      "step": 9850
+    },
+    {
+      "epoch": 7.137169743032935,
+      "grad_norm": 1.0350896120071411,
+      "learning_rate": 0.0002,
+      "loss": 0.7717,
+      "step": 9860
+    },
+    {
+      "epoch": 7.144408251900108,
+      "grad_norm": 1.334342360496521,
+      "learning_rate": 0.0002,
+      "loss": 0.903,
+      "step": 9870
+    },
+    {
+      "epoch": 7.151646760767282,
+      "grad_norm": 1.5271754264831543,
+      "learning_rate": 0.0002,
+      "loss": 0.7759,
+      "step": 9880
+    },
+    {
+      "epoch": 7.158885269634455,
+      "grad_norm": 1.0453001260757446,
+      "learning_rate": 0.0002,
+      "loss": 0.8223,
+      "step": 9890
+    },
+    {
+      "epoch": 7.166123778501628,
+      "grad_norm": 1.204174518585205,
+      "learning_rate": 0.0002,
+      "loss": 0.8334,
+      "step": 9900
+    },
+    {
+      "epoch": 7.173362287368802,
+      "grad_norm": 1.0774344205856323,
+      "learning_rate": 0.0002,
+      "loss": 0.82,
+      "step": 9910
+    },
+    {
+      "epoch": 7.180600796235975,
+      "grad_norm": 1.282188892364502,
+      "learning_rate": 0.0002,
+      "loss": 0.7772,
+      "step": 9920
+    },
+    {
+      "epoch": 7.187839305103148,
+      "grad_norm": 1.1413695812225342,
+      "learning_rate": 0.0002,
+      "loss": 0.7619,
+      "step": 9930
+    },
+    {
+      "epoch": 7.195077813970322,
+      "grad_norm": 1.2970763444900513,
+      "learning_rate": 0.0002,
+      "loss": 0.7194,
+      "step": 9940
+    },
+    {
+      "epoch": 7.202316322837495,
+      "grad_norm": 1.2535417079925537,
+      "learning_rate": 0.0002,
+      "loss": 0.857,
+      "step": 9950
+    },
+    {
+      "epoch": 7.2095548317046685,
+      "grad_norm": 1.3520581722259521,
+      "learning_rate": 0.0002,
+      "loss": 0.839,
+      "step": 9960
+    },
+    {
+      "epoch": 7.216793340571842,
+      "grad_norm": 1.288572072982788,
+      "learning_rate": 0.0002,
+      "loss": 0.8308,
+      "step": 9970
+    },
+    {
+      "epoch": 7.224031849439015,
+      "grad_norm": 1.4298021793365479,
+      "learning_rate": 0.0002,
+      "loss": 0.8102,
+      "step": 9980
+    },
+    {
+      "epoch": 7.231270358306189,
+      "grad_norm": 1.4797194004058838,
+      "learning_rate": 0.0002,
+      "loss": 0.9437,
+      "step": 9990
+    },
+    {
+      "epoch": 7.238508867173362,
+      "grad_norm": 1.5020978450775146,
+      "learning_rate": 0.0002,
+      "loss": 0.8486,
+      "step": 10000
+    },
+    {
+      "epoch": 7.245747376040535,
+      "grad_norm": 1.1417840719223022,
+      "learning_rate": 0.0002,
+      "loss": 0.8584,
+      "step": 10010
+    },
+    {
+      "epoch": 7.252985884907709,
+      "grad_norm": 1.746782898902893,
+      "learning_rate": 0.0002,
+      "loss": 0.8547,
+      "step": 10020
+    },
+    {
+      "epoch": 7.260224393774882,
+      "grad_norm": 1.019222617149353,
+      "learning_rate": 0.0002,
+      "loss": 0.8721,
+      "step": 10030
+    },
+    {
+      "epoch": 7.267462902642055,
+      "grad_norm": 1.3712849617004395,
+      "learning_rate": 0.0002,
+      "loss": 0.836,
+      "step": 10040
+    },
+    {
+      "epoch": 7.274701411509229,
+      "grad_norm": 1.5264288187026978,
+      "learning_rate": 0.0002,
+      "loss": 0.9228,
+      "step": 10050
+    },
+    {
+      "epoch": 7.281939920376402,
+      "grad_norm": 1.2784953117370605,
+      "learning_rate": 0.0002,
+      "loss": 0.8803,
+      "step": 10060
+    },
+    {
+      "epoch": 7.2891784292435755,
+      "grad_norm": 1.0246731042861938,
+      "learning_rate": 0.0002,
+      "loss": 0.9144,
+      "step": 10070
+    },
+    {
+      "epoch": 7.296416938110749,
+      "grad_norm": 1.2060108184814453,
+      "learning_rate": 0.0002,
+      "loss": 0.7846,
+      "step": 10080
+    },
+    {
+      "epoch": 7.303655446977922,
+      "grad_norm": 1.0908410549163818,
+      "learning_rate": 0.0002,
+      "loss": 0.8715,
+      "step": 10090
+    },
+    {
+      "epoch": 7.310893955845096,
+      "grad_norm": 1.2308661937713623,
+      "learning_rate": 0.0002,
+      "loss": 0.7516,
+      "step": 10100
+    },
+    {
+      "epoch": 7.318132464712269,
+      "grad_norm": 1.185610055923462,
+      "learning_rate": 0.0002,
+      "loss": 0.7477,
+      "step": 10110
+    },
+    {
+      "epoch": 7.325370973579442,
+      "grad_norm": 1.0026527643203735,
+      "learning_rate": 0.0002,
+      "loss": 0.8495,
+      "step": 10120
+    },
+    {
+      "epoch": 7.332609482446616,
+      "grad_norm": 1.3346470594406128,
+      "learning_rate": 0.0002,
+      "loss": 0.9155,
+      "step": 10130
+    },
+    {
+      "epoch": 7.339847991313789,
+      "grad_norm": 1.5946321487426758,
+      "learning_rate": 0.0002,
+      "loss": 0.975,
+      "step": 10140
+    },
+    {
+      "epoch": 7.347086500180962,
+      "grad_norm": 1.3622175455093384,
+      "learning_rate": 0.0002,
+      "loss": 0.7774,
+      "step": 10150
+    },
+    {
+      "epoch": 7.354325009048136,
+      "grad_norm": 1.0937085151672363,
+      "learning_rate": 0.0002,
+      "loss": 0.8455,
+      "step": 10160
+    },
+    {
+      "epoch": 7.361563517915309,
+      "grad_norm": 1.6057474613189697,
+      "learning_rate": 0.0002,
+      "loss": 0.8548,
+      "step": 10170
+    },
+    {
+      "epoch": 7.3688020267824825,
+      "grad_norm": 1.234887719154358,
+      "learning_rate": 0.0002,
+      "loss": 0.8345,
+      "step": 10180
+    },
+    {
+      "epoch": 7.376040535649656,
+      "grad_norm": 1.2238616943359375,
+      "learning_rate": 0.0002,
+      "loss": 0.9041,
+      "step": 10190
+    },
+    {
+      "epoch": 7.383279044516829,
+      "grad_norm": 1.2640055418014526,
+      "learning_rate": 0.0002,
+      "loss": 0.8568,
+      "step": 10200
+    },
+    {
+      "epoch": 7.390517553384003,
+      "grad_norm": 1.2917805910110474,
+      "learning_rate": 0.0002,
+      "loss": 0.8233,
+      "step": 10210
+    },
+    {
+      "epoch": 7.397756062251176,
+      "grad_norm": 1.096583366394043,
+      "learning_rate": 0.0002,
+      "loss": 0.8748,
+      "step": 10220
+    },
+    {
+      "epoch": 7.404994571118349,
+      "grad_norm": 1.1854201555252075,
+      "learning_rate": 0.0002,
+      "loss": 0.7882,
+      "step": 10230
+    },
+    {
+      "epoch": 7.412233079985523,
+      "grad_norm": 1.2318766117095947,
+      "learning_rate": 0.0002,
+      "loss": 0.7746,
+      "step": 10240
+    },
+    {
+      "epoch": 7.419471588852696,
+      "grad_norm": 1.395302414894104,
+      "learning_rate": 0.0002,
+      "loss": 0.9144,
+      "step": 10250
+    },
+    {
+      "epoch": 7.4267100977198695,
+      "grad_norm": 1.118148922920227,
+      "learning_rate": 0.0002,
+      "loss": 0.7673,
+      "step": 10260
+    },
+    {
+      "epoch": 7.433948606587043,
+      "grad_norm": 1.1969468593597412,
+      "learning_rate": 0.0002,
+      "loss": 0.8743,
+      "step": 10270
+    },
+    {
+      "epoch": 7.441187115454216,
+      "grad_norm": 1.434050440788269,
+      "learning_rate": 0.0002,
+      "loss": 0.8955,
+      "step": 10280
+    },
+    {
+      "epoch": 7.4484256243213895,
+      "grad_norm": 1.2344770431518555,
+      "learning_rate": 0.0002,
+      "loss": 0.7716,
+      "step": 10290
+    },
+    {
+      "epoch": 7.455664133188563,
+      "grad_norm": 1.2186434268951416,
+      "learning_rate": 0.0002,
+      "loss": 0.801,
+      "step": 10300
+    },
+    {
+      "epoch": 7.462902642055736,
+      "grad_norm": 1.482475757598877,
+      "learning_rate": 0.0002,
+      "loss": 0.8937,
+      "step": 10310
+    },
+    {
+      "epoch": 7.47014115092291,
+      "grad_norm": 1.8391777276992798,
+      "learning_rate": 0.0002,
+      "loss": 0.7954,
+      "step": 10320
+    },
+    {
+      "epoch": 7.477379659790083,
+      "grad_norm": 1.9489128589630127,
+      "learning_rate": 0.0002,
+      "loss": 0.8711,
+      "step": 10330
+    },
+    {
+      "epoch": 7.484618168657256,
+      "grad_norm": 1.369743824005127,
+      "learning_rate": 0.0002,
+      "loss": 0.8455,
+      "step": 10340
+    },
+    {
+      "epoch": 7.49185667752443,
+      "grad_norm": 1.3188602924346924,
+      "learning_rate": 0.0002,
+      "loss": 0.9501,
+      "step": 10350
+    },
+    {
+      "epoch": 7.499095186391603,
+      "grad_norm": 1.1885292530059814,
+      "learning_rate": 0.0002,
+      "loss": 0.8293,
+      "step": 10360
+    },
+    {
+      "epoch": 7.5063336952587765,
+      "grad_norm": 1.4873403310775757,
+      "learning_rate": 0.0002,
+      "loss": 0.8832,
+      "step": 10370
+    },
+    {
+      "epoch": 7.51357220412595,
+      "grad_norm": 1.8681598901748657,
+      "learning_rate": 0.0002,
+      "loss": 0.8531,
+      "step": 10380
+    },
+    {
+      "epoch": 7.520810712993123,
+      "grad_norm": 1.398186445236206,
+      "learning_rate": 0.0002,
+      "loss": 0.8172,
+      "step": 10390
+    },
+    {
+      "epoch": 7.528049221860297,
+      "grad_norm": 1.272192358970642,
+      "learning_rate": 0.0002,
+      "loss": 0.9239,
+      "step": 10400
+    },
+    {
+      "epoch": 7.53528773072747,
+      "grad_norm": 0.9671797752380371,
+      "learning_rate": 0.0002,
+      "loss": 0.818,
+      "step": 10410
+    },
+    {
+      "epoch": 7.542526239594643,
+      "grad_norm": 0.9752382040023804,
+      "learning_rate": 0.0002,
+      "loss": 0.8393,
+      "step": 10420
+    },
+    {
+      "epoch": 7.549764748461817,
+      "grad_norm": 1.2241966724395752,
+      "learning_rate": 0.0002,
+      "loss": 0.8011,
+      "step": 10430
+    },
+    {
+      "epoch": 7.55700325732899,
+      "grad_norm": 1.4615166187286377,
+      "learning_rate": 0.0002,
+      "loss": 0.8497,
+      "step": 10440
+    },
+    {
+      "epoch": 7.564241766196163,
+      "grad_norm": 1.123205542564392,
+      "learning_rate": 0.0002,
+      "loss": 0.8475,
+      "step": 10450
+    },
+    {
+      "epoch": 7.571480275063337,
+      "grad_norm": 1.3798918724060059,
+      "learning_rate": 0.0002,
+      "loss": 0.893,
+      "step": 10460
+    },
+    {
+      "epoch": 7.57871878393051,
+      "grad_norm": 1.3772553205490112,
+      "learning_rate": 0.0002,
+      "loss": 0.8109,
+      "step": 10470
+    },
+    {
+      "epoch": 7.5859572927976835,
+      "grad_norm": 1.4591912031173706,
+      "learning_rate": 0.0002,
+      "loss": 0.7907,
+      "step": 10480
+    },
+    {
+      "epoch": 7.593195801664857,
+      "grad_norm": 1.4248491525650024,
+      "learning_rate": 0.0002,
+      "loss": 0.878,
+      "step": 10490
+    },
+    {
+      "epoch": 7.60043431053203,
+      "grad_norm": 1.2663065195083618,
+      "learning_rate": 0.0002,
+      "loss": 0.7818,
+      "step": 10500
+    },
+    {
+      "epoch": 7.607672819399204,
+      "grad_norm": 1.1095938682556152,
+      "learning_rate": 0.0002,
+      "loss": 0.8349,
+      "step": 10510
+    },
+    {
+      "epoch": 7.614911328266377,
+      "grad_norm": 1.8462796211242676,
+      "learning_rate": 0.0002,
+      "loss": 0.7793,
+      "step": 10520
+    },
+    {
+      "epoch": 7.62214983713355,
+      "grad_norm": 1.1936118602752686,
+      "learning_rate": 0.0002,
+      "loss": 0.7695,
+      "step": 10530
+    },
+    {
+      "epoch": 7.629388346000724,
+      "grad_norm": 1.3520885705947876,
+      "learning_rate": 0.0002,
+      "loss": 0.8911,
+      "step": 10540
+    },
+    {
+      "epoch": 7.636626854867897,
+      "grad_norm": 1.2915338277816772,
+      "learning_rate": 0.0002,
+      "loss": 0.8589,
+      "step": 10550
+    },
+    {
+      "epoch": 7.64386536373507,
+      "grad_norm": 1.125656008720398,
+      "learning_rate": 0.0002,
+      "loss": 0.8932,
+      "step": 10560
+    },
+    {
+      "epoch": 7.651103872602244,
+      "grad_norm": 1.419791579246521,
+      "learning_rate": 0.0002,
+      "loss": 0.8689,
+      "step": 10570
+    },
+    {
+      "epoch": 7.658342381469417,
+      "grad_norm": 1.2106866836547852,
+      "learning_rate": 0.0002,
+      "loss": 0.8113,
+      "step": 10580
+    },
+    {
+      "epoch": 7.6655808903365905,
+      "grad_norm": 1.359818458557129,
+      "learning_rate": 0.0002,
+      "loss": 0.7383,
+      "step": 10590
+    },
+    {
+      "epoch": 7.672819399203764,
+      "grad_norm": 1.3971713781356812,
+      "learning_rate": 0.0002,
+      "loss": 0.8986,
+      "step": 10600
+    },
+    {
+      "epoch": 7.680057908070937,
+      "grad_norm": 1.287888765335083,
+      "learning_rate": 0.0002,
+      "loss": 0.942,
+      "step": 10610
+    },
+    {
+      "epoch": 7.687296416938111,
+      "grad_norm": 0.9856569766998291,
+      "learning_rate": 0.0002,
+      "loss": 0.8835,
+      "step": 10620
+    },
+    {
+      "epoch": 7.694534925805284,
+      "grad_norm": 1.5403797626495361,
+      "learning_rate": 0.0002,
+      "loss": 0.8463,
+      "step": 10630
+    },
+    {
+      "epoch": 7.701773434672457,
+      "grad_norm": 1.204551339149475,
+      "learning_rate": 0.0002,
+      "loss": 0.7888,
+      "step": 10640
+    },
+    {
+      "epoch": 7.709011943539631,
+      "grad_norm": 1.3801014423370361,
+      "learning_rate": 0.0002,
+      "loss": 0.7582,
+      "step": 10650
+    },
+    {
+      "epoch": 7.716250452406804,
+      "grad_norm": 1.3335949182510376,
+      "learning_rate": 0.0002,
+      "loss": 0.8715,
+      "step": 10660
+    },
+    {
+      "epoch": 7.723488961273977,
+      "grad_norm": 1.1740102767944336,
+      "learning_rate": 0.0002,
+      "loss": 0.9412,
+      "step": 10670
+    },
+    {
+      "epoch": 7.730727470141151,
+      "grad_norm": 1.1663082838058472,
+      "learning_rate": 0.0002,
+      "loss": 0.832,
+      "step": 10680
+    },
+    {
+      "epoch": 7.737965979008324,
+      "grad_norm": 1.3149393796920776,
+      "learning_rate": 0.0002,
+      "loss": 0.9191,
+      "step": 10690
+    },
+    {
+      "epoch": 7.7452044878754975,
+      "grad_norm": 1.3169108629226685,
+      "learning_rate": 0.0002,
+      "loss": 0.8697,
+      "step": 10700
+    },
+    {
+      "epoch": 7.752442996742671,
+      "grad_norm": 1.4583594799041748,
+      "learning_rate": 0.0002,
+      "loss": 0.8623,
+      "step": 10710
+    },
+    {
+      "epoch": 7.759681505609844,
+      "grad_norm": 1.1077126264572144,
+      "learning_rate": 0.0002,
+      "loss": 0.75,
+      "step": 10720
+    },
+    {
+      "epoch": 7.766920014477018,
+      "grad_norm": 1.5475820302963257,
+      "learning_rate": 0.0002,
+      "loss": 0.9109,
+      "step": 10730
+    },
+    {
+      "epoch": 7.774158523344191,
+      "grad_norm": 1.2319282293319702,
+      "learning_rate": 0.0002,
+      "loss": 0.8153,
+      "step": 10740
+    },
+    {
+      "epoch": 7.781397032211364,
+      "grad_norm": 0.9938047528266907,
+      "learning_rate": 0.0002,
+      "loss": 0.8324,
+      "step": 10750
+    },
+    {
+      "epoch": 7.788635541078538,
+      "grad_norm": 1.2498962879180908,
+      "learning_rate": 0.0002,
+      "loss": 0.9356,
+      "step": 10760
+    },
+    {
+      "epoch": 7.795874049945711,
+      "grad_norm": 2.192695379257202,
+      "learning_rate": 0.0002,
+      "loss": 0.8744,
+      "step": 10770
+    },
+    {
+      "epoch": 7.8031125588128845,
+      "grad_norm": 1.1851826906204224,
+      "learning_rate": 0.0002,
+      "loss": 0.9072,
+      "step": 10780
+    },
+    {
+      "epoch": 7.810351067680058,
+      "grad_norm": 1.0591034889221191,
+      "learning_rate": 0.0002,
+      "loss": 0.8933,
+      "step": 10790
+    },
+    {
+      "epoch": 7.817589576547231,
+      "grad_norm": 0.9350354671478271,
+      "learning_rate": 0.0002,
+      "loss": 0.8298,
+      "step": 10800
+    },
+    {
+      "epoch": 7.8248280854144046,
+      "grad_norm": 1.5080015659332275,
+      "learning_rate": 0.0002,
+      "loss": 0.8701,
+      "step": 10810
+    },
+    {
+      "epoch": 7.832066594281578,
+      "grad_norm": 2.136425495147705,
+      "learning_rate": 0.0002,
+      "loss": 0.9315,
+      "step": 10820
+    },
+    {
+      "epoch": 7.839305103148751,
+      "grad_norm": 1.5646673440933228,
+      "learning_rate": 0.0002,
+      "loss": 0.8204,
+      "step": 10830
+    },
+    {
+      "epoch": 7.846543612015925,
+      "grad_norm": 1.381301999092102,
+      "learning_rate": 0.0002,
+      "loss": 0.7794,
+      "step": 10840
+    },
+    {
+      "epoch": 7.853782120883098,
+      "grad_norm": 1.9323210716247559,
+      "learning_rate": 0.0002,
+      "loss": 0.9088,
+      "step": 10850
+    },
+    {
+      "epoch": 7.861020629750271,
+      "grad_norm": 1.020809531211853,
+      "learning_rate": 0.0002,
+      "loss": 0.8854,
+      "step": 10860
+    },
+    {
+      "epoch": 7.868259138617445,
+      "grad_norm": 1.1488909721374512,
+      "learning_rate": 0.0002,
+      "loss": 0.8144,
+      "step": 10870
+    },
+    {
+      "epoch": 7.875497647484618,
+      "grad_norm": 1.4068763256072998,
+      "learning_rate": 0.0002,
+      "loss": 0.883,
+      "step": 10880
+    },
+    {
+      "epoch": 7.8827361563517915,
+      "grad_norm": 0.9201020002365112,
+      "learning_rate": 0.0002,
+      "loss": 0.7431,
+      "step": 10890
+    },
+    {
+      "epoch": 7.889974665218965,
+      "grad_norm": 1.3163132667541504,
+      "learning_rate": 0.0002,
+      "loss": 0.8358,
+      "step": 10900
+    },
+    {
+      "epoch": 7.897213174086138,
+      "grad_norm": 1.65055513381958,
+      "learning_rate": 0.0002,
+      "loss": 0.9908,
+      "step": 10910
+    },
+    {
+      "epoch": 7.904451682953312,
+      "grad_norm": 1.1068748235702515,
+      "learning_rate": 0.0002,
+      "loss": 0.9883,
+      "step": 10920
+    },
+    {
+      "epoch": 7.911690191820485,
+      "grad_norm": 1.8744254112243652,
+      "learning_rate": 0.0002,
+      "loss": 0.8553,
+      "step": 10930
+    },
+    {
+      "epoch": 7.918928700687658,
+      "grad_norm": 1.3279157876968384,
+      "learning_rate": 0.0002,
+      "loss": 0.8969,
+      "step": 10940
+    },
+    {
+      "epoch": 7.926167209554832,
+      "grad_norm": 1.0890769958496094,
+      "learning_rate": 0.0002,
+      "loss": 0.8642,
+      "step": 10950
+    },
+    {
+      "epoch": 7.933405718422005,
+      "grad_norm": 1.3951836824417114,
+      "learning_rate": 0.0002,
+      "loss": 0.8742,
+      "step": 10960
+    },
+    {
+      "epoch": 7.940644227289178,
+      "grad_norm": 1.2761356830596924,
+      "learning_rate": 0.0002,
+      "loss": 0.895,
+      "step": 10970
+    },
+    {
+      "epoch": 7.947882736156352,
+      "grad_norm": 1.2073882818222046,
+      "learning_rate": 0.0002,
+      "loss": 0.8303,
+      "step": 10980
+    },
+    {
+      "epoch": 7.955121245023525,
+      "grad_norm": 1.1899374723434448,
+      "learning_rate": 0.0002,
+      "loss": 0.8675,
+      "step": 10990
+    },
+    {
+      "epoch": 7.9623597538906985,
+      "grad_norm": 1.3041194677352905,
+      "learning_rate": 0.0002,
+      "loss": 0.8373,
+      "step": 11000
+    },
+    {
+      "epoch": 7.969598262757872,
+      "grad_norm": 1.3564491271972656,
+      "learning_rate": 0.0002,
+      "loss": 0.8207,
+      "step": 11010
+    },
+    {
+      "epoch": 7.976836771625045,
+      "grad_norm": 1.1411082744598389,
+      "learning_rate": 0.0002,
+      "loss": 0.895,
+      "step": 11020
+    },
+    {
+      "epoch": 7.984075280492219,
+      "grad_norm": 1.1378493309020996,
+      "learning_rate": 0.0002,
+      "loss": 0.811,
+      "step": 11030
+    },
+    {
+      "epoch": 7.991313789359392,
+      "grad_norm": 1.5072855949401855,
+      "learning_rate": 0.0002,
+      "loss": 0.8257,
+      "step": 11040
+    },
+    {
+      "epoch": 7.997104596453131,
+      "eval_loss": 1.5476553440093994,
+      "eval_runtime": 26.3225,
+      "eval_samples_per_second": 16.564,
+      "eval_steps_per_second": 2.089,
+      "step": 11048
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 11048,
+  "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": 1.35028615510229e+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-0.9-num-8828-sd-1/checkpoint-11048/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-0.9-num-8828-sd-1/checkpoint-11048/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..4d7b7431bbbe8c9bf29b925bca391a558af5ff8c
--- /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-0.9-num-8828-sd-1/checkpoint-11048/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ad613885e4f267fc04125f1a836d42cfa796bbe12e536f9ee60c955de02cdb5a
+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-0.9-num-8828-sd-1/checkpoint-1381/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-0.9-num-8828-sd-1/checkpoint-1381/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-0.9-num-8828-sd-1/checkpoint-1381/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-0.9-num-8828-sd-1/checkpoint-1381/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-0.9-num-8828-sd-1/checkpoint-1381/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e99bbcd43df1c19d98706c7e3be95c93844c5349
--- /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-0.9-num-8828-sd-1/checkpoint-1381/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-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-8828-sd-1/checkpoint-1381/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-0.9-num-8828-sd-1/checkpoint-1381/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..3892c93516f4e1f5762f7c87679d013348b5051f
--- /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-0.9-num-8828-sd-1/checkpoint-1381/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c71d459969d8bed507ce9131e6643b65ab1d862f073a373123b5b5b6fb91eb26
+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-0.9-num-8828-sd-1/checkpoint-1381/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-0.9-num-8828-sd-1/checkpoint-1381/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..8205864177320ea73a81c1d8515f6139843e2e0d
--- /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-0.9-num-8828-sd-1/checkpoint-1381/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7f78e3a9f40d3b81ab4f3ca6497abb80b966ab129bf8b0e5327f7142e8c0d421
+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-0.9-num-8828-sd-1/checkpoint-1381/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-0.9-num-8828-sd-1/checkpoint-1381/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..81321b6bce511583b7000f2a0b7a954ce14c7d68
--- /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-0.9-num-8828-sd-1/checkpoint-1381/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2e317f41062559ca5eb1af54e847c5d7cf85da67608a9312cf4265eb0c1c2611
+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-0.9-num-8828-sd-1/checkpoint-1381/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-0.9-num-8828-sd-1/checkpoint-1381/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..93d93154c7de3957b2ca32bd9583ae698db6750f
--- /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-0.9-num-8828-sd-1/checkpoint-1381/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:568cc02dab4cab04c964cedf1df761956570e2e74c904d82b8bb30340ddb6861
+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-0.9-num-8828-sd-1/checkpoint-1381/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-0.9-num-8828-sd-1/checkpoint-1381/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-0.9-num-8828-sd-1/checkpoint-1381/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-0.9-num-8828-sd-1/checkpoint-1381/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-0.9-num-8828-sd-1/checkpoint-1381/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-0.9-num-8828-sd-1/checkpoint-1381/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-0.9-num-8828-sd-1/checkpoint-1381/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-0.9-num-8828-sd-1/checkpoint-1381/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-0.9-num-8828-sd-1/checkpoint-1381/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-0.9-num-8828-sd-1/checkpoint-1381/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-0.9-num-8828-sd-1/checkpoint-1381/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-0.9-num-8828-sd-1/checkpoint-1381/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-0.9-num-8828-sd-1/checkpoint-1381/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-0.9-num-8828-sd-1/checkpoint-1381/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..96436c688664ec881c52140e90adcc41b536d363
--- /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-0.9-num-8828-sd-1/checkpoint-1381/trainer_state.json
@@ -0,0 +1,1007 @@
+{
+  "best_metric": 1.480796456336975,
+  "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-0.9-num-8828-sd-1/checkpoint-1381",
+  "epoch": 0.9996380745566413,
+  "eval_steps": 10,
+  "global_step": 1381,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007238508867173362,
+      "grad_norm": 1.2523442506790161,
+      "learning_rate": 0.0002,
+      "loss": 4.7061,
+      "step": 10
+    },
+    {
+      "epoch": 0.014477017734346724,
+      "grad_norm": 1.8887330293655396,
+      "learning_rate": 0.0002,
+      "loss": 3.3493,
+      "step": 20
+    },
+    {
+      "epoch": 0.021715526601520086,
+      "grad_norm": 0.9668035507202148,
+      "learning_rate": 0.0002,
+      "loss": 2.7585,
+      "step": 30
+    },
+    {
+      "epoch": 0.028954035468693448,
+      "grad_norm": 2.9167306423187256,
+      "learning_rate": 0.0002,
+      "loss": 2.3699,
+      "step": 40
+    },
+    {
+      "epoch": 0.036192544335866814,
+      "grad_norm": 2.649867296218872,
+      "learning_rate": 0.0002,
+      "loss": 2.2679,
+      "step": 50
+    },
+    {
+      "epoch": 0.04343105320304017,
+      "grad_norm": 1.5120655298233032,
+      "learning_rate": 0.0002,
+      "loss": 2.2202,
+      "step": 60
+    },
+    {
+      "epoch": 0.05066956207021354,
+      "grad_norm": 0.7879868149757385,
+      "learning_rate": 0.0002,
+      "loss": 2.2026,
+      "step": 70
+    },
+    {
+      "epoch": 0.057908070937386896,
+      "grad_norm": 0.7616953253746033,
+      "learning_rate": 0.0002,
+      "loss": 1.9447,
+      "step": 80
+    },
+    {
+      "epoch": 0.06514657980456026,
+      "grad_norm": 1.8809149265289307,
+      "learning_rate": 0.0002,
+      "loss": 2.0112,
+      "step": 90
+    },
+    {
+      "epoch": 0.07238508867173363,
+      "grad_norm": 0.9294016361236572,
+      "learning_rate": 0.0002,
+      "loss": 1.8337,
+      "step": 100
+    },
+    {
+      "epoch": 0.07962359753890698,
+      "grad_norm": 0.7145281434059143,
+      "learning_rate": 0.0002,
+      "loss": 1.8419,
+      "step": 110
+    },
+    {
+      "epoch": 0.08686210640608034,
+      "grad_norm": 0.7564446330070496,
+      "learning_rate": 0.0002,
+      "loss": 2.0036,
+      "step": 120
+    },
+    {
+      "epoch": 0.09410061527325371,
+      "grad_norm": 1.1681925058364868,
+      "learning_rate": 0.0002,
+      "loss": 1.9306,
+      "step": 130
+    },
+    {
+      "epoch": 0.10133912414042708,
+      "grad_norm": 0.6708641648292542,
+      "learning_rate": 0.0002,
+      "loss": 1.7875,
+      "step": 140
+    },
+    {
+      "epoch": 0.10857763300760044,
+      "grad_norm": 0.7625647783279419,
+      "learning_rate": 0.0002,
+      "loss": 1.786,
+      "step": 150
+    },
+    {
+      "epoch": 0.11581614187477379,
+      "grad_norm": 0.8463464975357056,
+      "learning_rate": 0.0002,
+      "loss": 1.6687,
+      "step": 160
+    },
+    {
+      "epoch": 0.12305465074194716,
+      "grad_norm": 0.7502335906028748,
+      "learning_rate": 0.0002,
+      "loss": 1.6214,
+      "step": 170
+    },
+    {
+      "epoch": 0.13029315960912052,
+      "grad_norm": 0.6929958462715149,
+      "learning_rate": 0.0002,
+      "loss": 1.7433,
+      "step": 180
+    },
+    {
+      "epoch": 0.1375316684762939,
+      "grad_norm": 0.6798707842826843,
+      "learning_rate": 0.0002,
+      "loss": 1.6009,
+      "step": 190
+    },
+    {
+      "epoch": 0.14477017734346725,
+      "grad_norm": 0.7566508650779724,
+      "learning_rate": 0.0002,
+      "loss": 1.6208,
+      "step": 200
+    },
+    {
+      "epoch": 0.15200868621064062,
+      "grad_norm": 0.7196869850158691,
+      "learning_rate": 0.0002,
+      "loss": 1.5823,
+      "step": 210
+    },
+    {
+      "epoch": 0.15924719507781396,
+      "grad_norm": 0.8401045799255371,
+      "learning_rate": 0.0002,
+      "loss": 1.738,
+      "step": 220
+    },
+    {
+      "epoch": 0.16648570394498732,
+      "grad_norm": 0.8503773212432861,
+      "learning_rate": 0.0002,
+      "loss": 1.7574,
+      "step": 230
+    },
+    {
+      "epoch": 0.1737242128121607,
+      "grad_norm": 0.7183733582496643,
+      "learning_rate": 0.0002,
+      "loss": 1.7861,
+      "step": 240
+    },
+    {
+      "epoch": 0.18096272167933405,
+      "grad_norm": 0.7082605957984924,
+      "learning_rate": 0.0002,
+      "loss": 1.6693,
+      "step": 250
+    },
+    {
+      "epoch": 0.18820123054650742,
+      "grad_norm": 0.9386326670646667,
+      "learning_rate": 0.0002,
+      "loss": 1.619,
+      "step": 260
+    },
+    {
+      "epoch": 0.19543973941368079,
+      "grad_norm": 0.7332451939582825,
+      "learning_rate": 0.0002,
+      "loss": 1.6511,
+      "step": 270
+    },
+    {
+      "epoch": 0.20267824828085415,
+      "grad_norm": 0.7092869877815247,
+      "learning_rate": 0.0002,
+      "loss": 1.6353,
+      "step": 280
+    },
+    {
+      "epoch": 0.20991675714802752,
+      "grad_norm": 0.7256413698196411,
+      "learning_rate": 0.0002,
+      "loss": 1.5996,
+      "step": 290
+    },
+    {
+      "epoch": 0.21715526601520088,
+      "grad_norm": 0.6398681402206421,
+      "learning_rate": 0.0002,
+      "loss": 1.6754,
+      "step": 300
+    },
+    {
+      "epoch": 0.22439377488237422,
+      "grad_norm": 0.6273287534713745,
+      "learning_rate": 0.0002,
+      "loss": 1.397,
+      "step": 310
+    },
+    {
+      "epoch": 0.23163228374954759,
+      "grad_norm": 0.511648416519165,
+      "learning_rate": 0.0002,
+      "loss": 1.5115,
+      "step": 320
+    },
+    {
+      "epoch": 0.23887079261672095,
+      "grad_norm": 0.8677352070808411,
+      "learning_rate": 0.0002,
+      "loss": 1.5424,
+      "step": 330
+    },
+    {
+      "epoch": 0.24610930148389432,
+      "grad_norm": 0.6270743012428284,
+      "learning_rate": 0.0002,
+      "loss": 1.6779,
+      "step": 340
+    },
+    {
+      "epoch": 0.2533478103510677,
+      "grad_norm": 0.7980281114578247,
+      "learning_rate": 0.0002,
+      "loss": 1.626,
+      "step": 350
+    },
+    {
+      "epoch": 0.26058631921824105,
+      "grad_norm": 0.632486879825592,
+      "learning_rate": 0.0002,
+      "loss": 1.5238,
+      "step": 360
+    },
+    {
+      "epoch": 0.2678248280854144,
+      "grad_norm": 0.6527034640312195,
+      "learning_rate": 0.0002,
+      "loss": 1.5175,
+      "step": 370
+    },
+    {
+      "epoch": 0.2750633369525878,
+      "grad_norm": 0.7672118544578552,
+      "learning_rate": 0.0002,
+      "loss": 1.627,
+      "step": 380
+    },
+    {
+      "epoch": 0.28230184581976114,
+      "grad_norm": 0.6035117506980896,
+      "learning_rate": 0.0002,
+      "loss": 1.5605,
+      "step": 390
+    },
+    {
+      "epoch": 0.2895403546869345,
+      "grad_norm": 0.5955103039741516,
+      "learning_rate": 0.0002,
+      "loss": 1.4603,
+      "step": 400
+    },
+    {
+      "epoch": 0.2967788635541079,
+      "grad_norm": 0.6015191674232483,
+      "learning_rate": 0.0002,
+      "loss": 1.558,
+      "step": 410
+    },
+    {
+      "epoch": 0.30401737242128124,
+      "grad_norm": 0.6380982398986816,
+      "learning_rate": 0.0002,
+      "loss": 1.6091,
+      "step": 420
+    },
+    {
+      "epoch": 0.3112558812884546,
+      "grad_norm": 0.6707863211631775,
+      "learning_rate": 0.0002,
+      "loss": 1.5292,
+      "step": 430
+    },
+    {
+      "epoch": 0.3184943901556279,
+      "grad_norm": 0.7010176777839661,
+      "learning_rate": 0.0002,
+      "loss": 1.4426,
+      "step": 440
+    },
+    {
+      "epoch": 0.3257328990228013,
+      "grad_norm": 0.8263739943504333,
+      "learning_rate": 0.0002,
+      "loss": 1.5572,
+      "step": 450
+    },
+    {
+      "epoch": 0.33297140788997465,
+      "grad_norm": 0.7253276109695435,
+      "learning_rate": 0.0002,
+      "loss": 1.5188,
+      "step": 460
+    },
+    {
+      "epoch": 0.340209916757148,
+      "grad_norm": 0.5238934755325317,
+      "learning_rate": 0.0002,
+      "loss": 1.584,
+      "step": 470
+    },
+    {
+      "epoch": 0.3474484256243214,
+      "grad_norm": 0.7869495749473572,
+      "learning_rate": 0.0002,
+      "loss": 1.7035,
+      "step": 480
+    },
+    {
+      "epoch": 0.35468693449149474,
+      "grad_norm": 0.7485215663909912,
+      "learning_rate": 0.0002,
+      "loss": 1.5776,
+      "step": 490
+    },
+    {
+      "epoch": 0.3619254433586681,
+      "grad_norm": 0.5413193106651306,
+      "learning_rate": 0.0002,
+      "loss": 1.6274,
+      "step": 500
+    },
+    {
+      "epoch": 0.3691639522258415,
+      "grad_norm": 0.7615048885345459,
+      "learning_rate": 0.0002,
+      "loss": 1.7323,
+      "step": 510
+    },
+    {
+      "epoch": 0.37640246109301484,
+      "grad_norm": 0.7685340046882629,
+      "learning_rate": 0.0002,
+      "loss": 1.532,
+      "step": 520
+    },
+    {
+      "epoch": 0.3836409699601882,
+      "grad_norm": 0.6379081010818481,
+      "learning_rate": 0.0002,
+      "loss": 1.6312,
+      "step": 530
+    },
+    {
+      "epoch": 0.39087947882736157,
+      "grad_norm": 0.7946939468383789,
+      "learning_rate": 0.0002,
+      "loss": 1.5645,
+      "step": 540
+    },
+    {
+      "epoch": 0.39811798769453494,
+      "grad_norm": 0.6287278532981873,
+      "learning_rate": 0.0002,
+      "loss": 1.4001,
+      "step": 550
+    },
+    {
+      "epoch": 0.4053564965617083,
+      "grad_norm": 0.6811642646789551,
+      "learning_rate": 0.0002,
+      "loss": 1.5982,
+      "step": 560
+    },
+    {
+      "epoch": 0.41259500542888167,
+      "grad_norm": 0.671073317527771,
+      "learning_rate": 0.0002,
+      "loss": 1.4953,
+      "step": 570
+    },
+    {
+      "epoch": 0.41983351429605503,
+      "grad_norm": 0.6313900351524353,
+      "learning_rate": 0.0002,
+      "loss": 1.6753,
+      "step": 580
+    },
+    {
+      "epoch": 0.4270720231632284,
+      "grad_norm": 0.5291772484779358,
+      "learning_rate": 0.0002,
+      "loss": 1.546,
+      "step": 590
+    },
+    {
+      "epoch": 0.43431053203040176,
+      "grad_norm": 0.62503582239151,
+      "learning_rate": 0.0002,
+      "loss": 1.5441,
+      "step": 600
+    },
+    {
+      "epoch": 0.4415490408975751,
+      "grad_norm": 0.5777305364608765,
+      "learning_rate": 0.0002,
+      "loss": 1.6276,
+      "step": 610
+    },
+    {
+      "epoch": 0.44878754976474844,
+      "grad_norm": 0.7013497352600098,
+      "learning_rate": 0.0002,
+      "loss": 1.4758,
+      "step": 620
+    },
+    {
+      "epoch": 0.4560260586319218,
+      "grad_norm": 0.8044822216033936,
+      "learning_rate": 0.0002,
+      "loss": 1.4029,
+      "step": 630
+    },
+    {
+      "epoch": 0.46326456749909517,
+      "grad_norm": 0.672531247138977,
+      "learning_rate": 0.0002,
+      "loss": 1.7195,
+      "step": 640
+    },
+    {
+      "epoch": 0.47050307636626854,
+      "grad_norm": 0.6233910322189331,
+      "learning_rate": 0.0002,
+      "loss": 1.614,
+      "step": 650
+    },
+    {
+      "epoch": 0.4777415852334419,
+      "grad_norm": 0.651524543762207,
+      "learning_rate": 0.0002,
+      "loss": 1.6041,
+      "step": 660
+    },
+    {
+      "epoch": 0.48498009410061527,
+      "grad_norm": 0.7213939428329468,
+      "learning_rate": 0.0002,
+      "loss": 1.5842,
+      "step": 670
+    },
+    {
+      "epoch": 0.49221860296778863,
+      "grad_norm": 0.6541454792022705,
+      "learning_rate": 0.0002,
+      "loss": 1.5453,
+      "step": 680
+    },
+    {
+      "epoch": 0.499457111834962,
+      "grad_norm": 0.6568936109542847,
+      "learning_rate": 0.0002,
+      "loss": 1.662,
+      "step": 690
+    },
+    {
+      "epoch": 0.5066956207021354,
+      "grad_norm": 0.7176415324211121,
+      "learning_rate": 0.0002,
+      "loss": 1.624,
+      "step": 700
+    },
+    {
+      "epoch": 0.5139341295693087,
+      "grad_norm": 0.6553855538368225,
+      "learning_rate": 0.0002,
+      "loss": 1.6099,
+      "step": 710
+    },
+    {
+      "epoch": 0.5211726384364821,
+      "grad_norm": 0.5654335618019104,
+      "learning_rate": 0.0002,
+      "loss": 1.5508,
+      "step": 720
+    },
+    {
+      "epoch": 0.5284111473036555,
+      "grad_norm": 0.5671001672744751,
+      "learning_rate": 0.0002,
+      "loss": 1.392,
+      "step": 730
+    },
+    {
+      "epoch": 0.5356496561708288,
+      "grad_norm": 0.7914412021636963,
+      "learning_rate": 0.0002,
+      "loss": 1.388,
+      "step": 740
+    },
+    {
+      "epoch": 0.5428881650380022,
+      "grad_norm": 0.6172138452529907,
+      "learning_rate": 0.0002,
+      "loss": 1.5931,
+      "step": 750
+    },
+    {
+      "epoch": 0.5501266739051756,
+      "grad_norm": 0.6132623553276062,
+      "learning_rate": 0.0002,
+      "loss": 1.4018,
+      "step": 760
+    },
+    {
+      "epoch": 0.5573651827723489,
+      "grad_norm": 0.654000461101532,
+      "learning_rate": 0.0002,
+      "loss": 1.513,
+      "step": 770
+    },
+    {
+      "epoch": 0.5646036916395223,
+      "grad_norm": 0.5691370964050293,
+      "learning_rate": 0.0002,
+      "loss": 1.5035,
+      "step": 780
+    },
+    {
+      "epoch": 0.5718422005066957,
+      "grad_norm": 0.7922580242156982,
+      "learning_rate": 0.0002,
+      "loss": 1.65,
+      "step": 790
+    },
+    {
+      "epoch": 0.579080709373869,
+      "grad_norm": 0.6831880211830139,
+      "learning_rate": 0.0002,
+      "loss": 1.4521,
+      "step": 800
+    },
+    {
+      "epoch": 0.5863192182410424,
+      "grad_norm": 0.6740124821662903,
+      "learning_rate": 0.0002,
+      "loss": 1.4734,
+      "step": 810
+    },
+    {
+      "epoch": 0.5935577271082157,
+      "grad_norm": 1.380016803741455,
+      "learning_rate": 0.0002,
+      "loss": 1.6498,
+      "step": 820
+    },
+    {
+      "epoch": 0.6007962359753891,
+      "grad_norm": 0.6552878022193909,
+      "learning_rate": 0.0002,
+      "loss": 1.4642,
+      "step": 830
+    },
+    {
+      "epoch": 0.6080347448425625,
+      "grad_norm": 0.6649535298347473,
+      "learning_rate": 0.0002,
+      "loss": 1.6271,
+      "step": 840
+    },
+    {
+      "epoch": 0.6152732537097358,
+      "grad_norm": 0.561738133430481,
+      "learning_rate": 0.0002,
+      "loss": 1.5886,
+      "step": 850
+    },
+    {
+      "epoch": 0.6225117625769092,
+      "grad_norm": 0.6133047938346863,
+      "learning_rate": 0.0002,
+      "loss": 1.5364,
+      "step": 860
+    },
+    {
+      "epoch": 0.6297502714440825,
+      "grad_norm": 0.559843122959137,
+      "learning_rate": 0.0002,
+      "loss": 1.3489,
+      "step": 870
+    },
+    {
+      "epoch": 0.6369887803112558,
+      "grad_norm": 0.6117811799049377,
+      "learning_rate": 0.0002,
+      "loss": 1.4878,
+      "step": 880
+    },
+    {
+      "epoch": 0.6442272891784292,
+      "grad_norm": 0.6209776401519775,
+      "learning_rate": 0.0002,
+      "loss": 1.56,
+      "step": 890
+    },
+    {
+      "epoch": 0.6514657980456026,
+      "grad_norm": 0.6234082579612732,
+      "learning_rate": 0.0002,
+      "loss": 1.6747,
+      "step": 900
+    },
+    {
+      "epoch": 0.6587043069127759,
+      "grad_norm": 0.7623258233070374,
+      "learning_rate": 0.0002,
+      "loss": 1.6963,
+      "step": 910
+    },
+    {
+      "epoch": 0.6659428157799493,
+      "grad_norm": 0.6148061752319336,
+      "learning_rate": 0.0002,
+      "loss": 1.2424,
+      "step": 920
+    },
+    {
+      "epoch": 0.6731813246471227,
+      "grad_norm": 0.6682973504066467,
+      "learning_rate": 0.0002,
+      "loss": 1.4319,
+      "step": 930
+    },
+    {
+      "epoch": 0.680419833514296,
+      "grad_norm": 0.5513041615486145,
+      "learning_rate": 0.0002,
+      "loss": 1.5377,
+      "step": 940
+    },
+    {
+      "epoch": 0.6876583423814694,
+      "grad_norm": 0.5197525024414062,
+      "learning_rate": 0.0002,
+      "loss": 1.3991,
+      "step": 950
+    },
+    {
+      "epoch": 0.6948968512486428,
+      "grad_norm": 0.6490758061408997,
+      "learning_rate": 0.0002,
+      "loss": 1.4398,
+      "step": 960
+    },
+    {
+      "epoch": 0.7021353601158161,
+      "grad_norm": 0.6450682878494263,
+      "learning_rate": 0.0002,
+      "loss": 1.5251,
+      "step": 970
+    },
+    {
+      "epoch": 0.7093738689829895,
+      "grad_norm": 0.6203766465187073,
+      "learning_rate": 0.0002,
+      "loss": 1.5417,
+      "step": 980
+    },
+    {
+      "epoch": 0.7166123778501629,
+      "grad_norm": 0.6023609638214111,
+      "learning_rate": 0.0002,
+      "loss": 1.4575,
+      "step": 990
+    },
+    {
+      "epoch": 0.7238508867173362,
+      "grad_norm": 0.5765255093574524,
+      "learning_rate": 0.0002,
+      "loss": 1.4973,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7310893955845096,
+      "grad_norm": 0.6650075316429138,
+      "learning_rate": 0.0002,
+      "loss": 1.483,
+      "step": 1010
+    },
+    {
+      "epoch": 0.738327904451683,
+      "grad_norm": 0.5610854029655457,
+      "learning_rate": 0.0002,
+      "loss": 1.5959,
+      "step": 1020
+    },
+    {
+      "epoch": 0.7455664133188563,
+      "grad_norm": 0.7072813510894775,
+      "learning_rate": 0.0002,
+      "loss": 1.5248,
+      "step": 1030
+    },
+    {
+      "epoch": 0.7528049221860297,
+      "grad_norm": 0.6815407872200012,
+      "learning_rate": 0.0002,
+      "loss": 1.5776,
+      "step": 1040
+    },
+    {
+      "epoch": 0.760043431053203,
+      "grad_norm": 0.7932390570640564,
+      "learning_rate": 0.0002,
+      "loss": 1.4577,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7672819399203764,
+      "grad_norm": 0.5798183083534241,
+      "learning_rate": 0.0002,
+      "loss": 1.4515,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7745204487875498,
+      "grad_norm": 0.7898504137992859,
+      "learning_rate": 0.0002,
+      "loss": 1.5053,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7817589576547231,
+      "grad_norm": 0.4983280301094055,
+      "learning_rate": 0.0002,
+      "loss": 1.4776,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7889974665218965,
+      "grad_norm": 0.691403329372406,
+      "learning_rate": 0.0002,
+      "loss": 1.5007,
+      "step": 1090
+    },
+    {
+      "epoch": 0.7962359753890699,
+      "grad_norm": 0.5394481420516968,
+      "learning_rate": 0.0002,
+      "loss": 1.5153,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8034744842562432,
+      "grad_norm": 0.5136822462081909,
+      "learning_rate": 0.0002,
+      "loss": 1.6892,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8107129931234166,
+      "grad_norm": 0.6828126907348633,
+      "learning_rate": 0.0002,
+      "loss": 1.4902,
+      "step": 1120
+    },
+    {
+      "epoch": 0.81795150199059,
+      "grad_norm": 0.6799656748771667,
+      "learning_rate": 0.0002,
+      "loss": 1.4346,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8251900108577633,
+      "grad_norm": 0.5428406000137329,
+      "learning_rate": 0.0002,
+      "loss": 1.2678,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8324285197249367,
+      "grad_norm": 0.4811290502548218,
+      "learning_rate": 0.0002,
+      "loss": 1.4072,
+      "step": 1150
+    },
+    {
+      "epoch": 0.8396670285921101,
+      "grad_norm": 0.5519434809684753,
+      "learning_rate": 0.0002,
+      "loss": 1.4512,
+      "step": 1160
+    },
+    {
+      "epoch": 0.8469055374592834,
+      "grad_norm": 0.9748060703277588,
+      "learning_rate": 0.0002,
+      "loss": 1.4072,
+      "step": 1170
+    },
+    {
+      "epoch": 0.8541440463264568,
+      "grad_norm": 0.712609589099884,
+      "learning_rate": 0.0002,
+      "loss": 1.4309,
+      "step": 1180
+    },
+    {
+      "epoch": 0.8613825551936302,
+      "grad_norm": 0.6866157054901123,
+      "learning_rate": 0.0002,
+      "loss": 1.434,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8686210640608035,
+      "grad_norm": 0.5068854093551636,
+      "learning_rate": 0.0002,
+      "loss": 1.3704,
+      "step": 1200
+    },
+    {
+      "epoch": 0.8758595729279768,
+      "grad_norm": 0.6333245038986206,
+      "learning_rate": 0.0002,
+      "loss": 1.5601,
+      "step": 1210
+    },
+    {
+      "epoch": 0.8830980817951501,
+      "grad_norm": 0.6424421072006226,
+      "learning_rate": 0.0002,
+      "loss": 1.4636,
+      "step": 1220
+    },
+    {
+      "epoch": 0.8903365906623235,
+      "grad_norm": 0.4771921932697296,
+      "learning_rate": 0.0002,
+      "loss": 1.4186,
+      "step": 1230
+    },
+    {
+      "epoch": 0.8975750995294969,
+      "grad_norm": 0.5191764235496521,
+      "learning_rate": 0.0002,
+      "loss": 1.6323,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9048136083966702,
+      "grad_norm": 0.756222128868103,
+      "learning_rate": 0.0002,
+      "loss": 1.6105,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9120521172638436,
+      "grad_norm": 0.623823881149292,
+      "learning_rate": 0.0002,
+      "loss": 1.4396,
+      "step": 1260
+    },
+    {
+      "epoch": 0.919290626131017,
+      "grad_norm": 0.8166571259498596,
+      "learning_rate": 0.0002,
+      "loss": 1.3097,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9265291349981903,
+      "grad_norm": 0.6059346795082092,
+      "learning_rate": 0.0002,
+      "loss": 1.4625,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9337676438653637,
+      "grad_norm": 0.5842690467834473,
+      "learning_rate": 0.0002,
+      "loss": 1.3555,
+      "step": 1290
+    },
+    {
+      "epoch": 0.9410061527325371,
+      "grad_norm": 0.7649800777435303,
+      "learning_rate": 0.0002,
+      "loss": 1.5859,
+      "step": 1300
+    },
+    {
+      "epoch": 0.9482446615997104,
+      "grad_norm": 0.6420919895172119,
+      "learning_rate": 0.0002,
+      "loss": 1.5915,
+      "step": 1310
+    },
+    {
+      "epoch": 0.9554831704668838,
+      "grad_norm": 0.7011452913284302,
+      "learning_rate": 0.0002,
+      "loss": 1.453,
+      "step": 1320
+    },
+    {
+      "epoch": 0.9627216793340572,
+      "grad_norm": 0.5783746242523193,
+      "learning_rate": 0.0002,
+      "loss": 1.6766,
+      "step": 1330
+    },
+    {
+      "epoch": 0.9699601882012305,
+      "grad_norm": 0.5973192453384399,
+      "learning_rate": 0.0002,
+      "loss": 1.6308,
+      "step": 1340
+    },
+    {
+      "epoch": 0.9771986970684039,
+      "grad_norm": 0.6181833744049072,
+      "learning_rate": 0.0002,
+      "loss": 1.5901,
+      "step": 1350
+    },
+    {
+      "epoch": 0.9844372059355773,
+      "grad_norm": 0.5563396215438843,
+      "learning_rate": 0.0002,
+      "loss": 1.5258,
+      "step": 1360
+    },
+    {
+      "epoch": 0.9916757148027506,
+      "grad_norm": 0.45723360776901245,
+      "learning_rate": 0.0002,
+      "loss": 1.4508,
+      "step": 1370
+    },
+    {
+      "epoch": 0.998914223669924,
+      "grad_norm": 0.5947498679161072,
+      "learning_rate": 0.0002,
+      "loss": 1.3291,
+      "step": 1380
+    },
+    {
+      "epoch": 0.9996380745566413,
+      "eval_loss": 1.480796456336975,
+      "eval_runtime": 27.3103,
+      "eval_samples_per_second": 15.965,
+      "eval_steps_per_second": 2.014,
+      "step": 1381
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 11048,
+  "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": 1.6884687936946176e+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-0.9-num-8828-sd-1/checkpoint-1381/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-0.9-num-8828-sd-1/checkpoint-1381/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..4d7b7431bbbe8c9bf29b925bca391a558af5ff8c
--- /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-0.9-num-8828-sd-1/checkpoint-1381/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ad613885e4f267fc04125f1a836d42cfa796bbe12e536f9ee60c955de02cdb5a
+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-0.9-num-8828-sd-1/checkpoint-2763/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-0.9-num-8828-sd-1/checkpoint-2763/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-0.9-num-8828-sd-1/checkpoint-2763/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-0.9-num-8828-sd-1/checkpoint-2763/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-0.9-num-8828-sd-1/checkpoint-2763/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e99bbcd43df1c19d98706c7e3be95c93844c5349
--- /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-0.9-num-8828-sd-1/checkpoint-2763/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-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-8828-sd-1/checkpoint-2763/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-0.9-num-8828-sd-1/checkpoint-2763/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..8e0a88f4ddbdcab4a213de1d48b25947c90523fe
--- /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-0.9-num-8828-sd-1/checkpoint-2763/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b9da8a367d56151653f5ff9143fe9d10f3ec83a5a6610e31d53748ea75673f4b
+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-0.9-num-8828-sd-1/checkpoint-2763/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-0.9-num-8828-sd-1/checkpoint-2763/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..32b375f4a1edfa10ae8c3dc8083d8bc0a68b984c
--- /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-0.9-num-8828-sd-1/checkpoint-2763/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e73a5b960c502949a09578886e280eade3328143ddecbd6edc6542b05162277f
+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-0.9-num-8828-sd-1/checkpoint-2763/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-0.9-num-8828-sd-1/checkpoint-2763/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..2dd55fc461ae4f6d9ef68bed1742e771b59bb42b
--- /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-0.9-num-8828-sd-1/checkpoint-2763/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6acfc6453c61add2025a81b7aeb4cca24d4c821e87be1d086631674d7f9c4a33
+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-0.9-num-8828-sd-1/checkpoint-2763/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-0.9-num-8828-sd-1/checkpoint-2763/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..7709aa944d4094e1f170ebf504e64d0bf6ce6fd9
--- /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-0.9-num-8828-sd-1/checkpoint-2763/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dfbb5e3cc39c3fb3bb05a3dbc06549de20a41e5f1432694f538f56fe06ec0d46
+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-0.9-num-8828-sd-1/checkpoint-2763/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-0.9-num-8828-sd-1/checkpoint-2763/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-0.9-num-8828-sd-1/checkpoint-2763/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-0.9-num-8828-sd-1/checkpoint-2763/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-0.9-num-8828-sd-1/checkpoint-2763/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-0.9-num-8828-sd-1/checkpoint-2763/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-0.9-num-8828-sd-1/checkpoint-2763/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-0.9-num-8828-sd-1/checkpoint-2763/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-0.9-num-8828-sd-1/checkpoint-2763/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-0.9-num-8828-sd-1/checkpoint-2763/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-0.9-num-8828-sd-1/checkpoint-2763/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-0.9-num-8828-sd-1/checkpoint-2763/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-0.9-num-8828-sd-1/checkpoint-2763/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-0.9-num-8828-sd-1/checkpoint-2763/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..8aa433345b63380da5893753c4fc4b9a2f7dead2
--- /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-0.9-num-8828-sd-1/checkpoint-2763/trainer_state.json
@@ -0,0 +1,1981 @@
+{
+  "best_metric": 1.4366681575775146,
+  "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-0.9-num-8828-sd-1/checkpoint-2763",
+  "epoch": 2.0,
+  "eval_steps": 10,
+  "global_step": 2763,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007238508867173362,
+      "grad_norm": 1.2523442506790161,
+      "learning_rate": 0.0002,
+      "loss": 4.7061,
+      "step": 10
+    },
+    {
+      "epoch": 0.014477017734346724,
+      "grad_norm": 1.8887330293655396,
+      "learning_rate": 0.0002,
+      "loss": 3.3493,
+      "step": 20
+    },
+    {
+      "epoch": 0.021715526601520086,
+      "grad_norm": 0.9668035507202148,
+      "learning_rate": 0.0002,
+      "loss": 2.7585,
+      "step": 30
+    },
+    {
+      "epoch": 0.028954035468693448,
+      "grad_norm": 2.9167306423187256,
+      "learning_rate": 0.0002,
+      "loss": 2.3699,
+      "step": 40
+    },
+    {
+      "epoch": 0.036192544335866814,
+      "grad_norm": 2.649867296218872,
+      "learning_rate": 0.0002,
+      "loss": 2.2679,
+      "step": 50
+    },
+    {
+      "epoch": 0.04343105320304017,
+      "grad_norm": 1.5120655298233032,
+      "learning_rate": 0.0002,
+      "loss": 2.2202,
+      "step": 60
+    },
+    {
+      "epoch": 0.05066956207021354,
+      "grad_norm": 0.7879868149757385,
+      "learning_rate": 0.0002,
+      "loss": 2.2026,
+      "step": 70
+    },
+    {
+      "epoch": 0.057908070937386896,
+      "grad_norm": 0.7616953253746033,
+      "learning_rate": 0.0002,
+      "loss": 1.9447,
+      "step": 80
+    },
+    {
+      "epoch": 0.06514657980456026,
+      "grad_norm": 1.8809149265289307,
+      "learning_rate": 0.0002,
+      "loss": 2.0112,
+      "step": 90
+    },
+    {
+      "epoch": 0.07238508867173363,
+      "grad_norm": 0.9294016361236572,
+      "learning_rate": 0.0002,
+      "loss": 1.8337,
+      "step": 100
+    },
+    {
+      "epoch": 0.07962359753890698,
+      "grad_norm": 0.7145281434059143,
+      "learning_rate": 0.0002,
+      "loss": 1.8419,
+      "step": 110
+    },
+    {
+      "epoch": 0.08686210640608034,
+      "grad_norm": 0.7564446330070496,
+      "learning_rate": 0.0002,
+      "loss": 2.0036,
+      "step": 120
+    },
+    {
+      "epoch": 0.09410061527325371,
+      "grad_norm": 1.1681925058364868,
+      "learning_rate": 0.0002,
+      "loss": 1.9306,
+      "step": 130
+    },
+    {
+      "epoch": 0.10133912414042708,
+      "grad_norm": 0.6708641648292542,
+      "learning_rate": 0.0002,
+      "loss": 1.7875,
+      "step": 140
+    },
+    {
+      "epoch": 0.10857763300760044,
+      "grad_norm": 0.7625647783279419,
+      "learning_rate": 0.0002,
+      "loss": 1.786,
+      "step": 150
+    },
+    {
+      "epoch": 0.11581614187477379,
+      "grad_norm": 0.8463464975357056,
+      "learning_rate": 0.0002,
+      "loss": 1.6687,
+      "step": 160
+    },
+    {
+      "epoch": 0.12305465074194716,
+      "grad_norm": 0.7502335906028748,
+      "learning_rate": 0.0002,
+      "loss": 1.6214,
+      "step": 170
+    },
+    {
+      "epoch": 0.13029315960912052,
+      "grad_norm": 0.6929958462715149,
+      "learning_rate": 0.0002,
+      "loss": 1.7433,
+      "step": 180
+    },
+    {
+      "epoch": 0.1375316684762939,
+      "grad_norm": 0.6798707842826843,
+      "learning_rate": 0.0002,
+      "loss": 1.6009,
+      "step": 190
+    },
+    {
+      "epoch": 0.14477017734346725,
+      "grad_norm": 0.7566508650779724,
+      "learning_rate": 0.0002,
+      "loss": 1.6208,
+      "step": 200
+    },
+    {
+      "epoch": 0.15200868621064062,
+      "grad_norm": 0.7196869850158691,
+      "learning_rate": 0.0002,
+      "loss": 1.5823,
+      "step": 210
+    },
+    {
+      "epoch": 0.15924719507781396,
+      "grad_norm": 0.8401045799255371,
+      "learning_rate": 0.0002,
+      "loss": 1.738,
+      "step": 220
+    },
+    {
+      "epoch": 0.16648570394498732,
+      "grad_norm": 0.8503773212432861,
+      "learning_rate": 0.0002,
+      "loss": 1.7574,
+      "step": 230
+    },
+    {
+      "epoch": 0.1737242128121607,
+      "grad_norm": 0.7183733582496643,
+      "learning_rate": 0.0002,
+      "loss": 1.7861,
+      "step": 240
+    },
+    {
+      "epoch": 0.18096272167933405,
+      "grad_norm": 0.7082605957984924,
+      "learning_rate": 0.0002,
+      "loss": 1.6693,
+      "step": 250
+    },
+    {
+      "epoch": 0.18820123054650742,
+      "grad_norm": 0.9386326670646667,
+      "learning_rate": 0.0002,
+      "loss": 1.619,
+      "step": 260
+    },
+    {
+      "epoch": 0.19543973941368079,
+      "grad_norm": 0.7332451939582825,
+      "learning_rate": 0.0002,
+      "loss": 1.6511,
+      "step": 270
+    },
+    {
+      "epoch": 0.20267824828085415,
+      "grad_norm": 0.7092869877815247,
+      "learning_rate": 0.0002,
+      "loss": 1.6353,
+      "step": 280
+    },
+    {
+      "epoch": 0.20991675714802752,
+      "grad_norm": 0.7256413698196411,
+      "learning_rate": 0.0002,
+      "loss": 1.5996,
+      "step": 290
+    },
+    {
+      "epoch": 0.21715526601520088,
+      "grad_norm": 0.6398681402206421,
+      "learning_rate": 0.0002,
+      "loss": 1.6754,
+      "step": 300
+    },
+    {
+      "epoch": 0.22439377488237422,
+      "grad_norm": 0.6273287534713745,
+      "learning_rate": 0.0002,
+      "loss": 1.397,
+      "step": 310
+    },
+    {
+      "epoch": 0.23163228374954759,
+      "grad_norm": 0.511648416519165,
+      "learning_rate": 0.0002,
+      "loss": 1.5115,
+      "step": 320
+    },
+    {
+      "epoch": 0.23887079261672095,
+      "grad_norm": 0.8677352070808411,
+      "learning_rate": 0.0002,
+      "loss": 1.5424,
+      "step": 330
+    },
+    {
+      "epoch": 0.24610930148389432,
+      "grad_norm": 0.6270743012428284,
+      "learning_rate": 0.0002,
+      "loss": 1.6779,
+      "step": 340
+    },
+    {
+      "epoch": 0.2533478103510677,
+      "grad_norm": 0.7980281114578247,
+      "learning_rate": 0.0002,
+      "loss": 1.626,
+      "step": 350
+    },
+    {
+      "epoch": 0.26058631921824105,
+      "grad_norm": 0.632486879825592,
+      "learning_rate": 0.0002,
+      "loss": 1.5238,
+      "step": 360
+    },
+    {
+      "epoch": 0.2678248280854144,
+      "grad_norm": 0.6527034640312195,
+      "learning_rate": 0.0002,
+      "loss": 1.5175,
+      "step": 370
+    },
+    {
+      "epoch": 0.2750633369525878,
+      "grad_norm": 0.7672118544578552,
+      "learning_rate": 0.0002,
+      "loss": 1.627,
+      "step": 380
+    },
+    {
+      "epoch": 0.28230184581976114,
+      "grad_norm": 0.6035117506980896,
+      "learning_rate": 0.0002,
+      "loss": 1.5605,
+      "step": 390
+    },
+    {
+      "epoch": 0.2895403546869345,
+      "grad_norm": 0.5955103039741516,
+      "learning_rate": 0.0002,
+      "loss": 1.4603,
+      "step": 400
+    },
+    {
+      "epoch": 0.2967788635541079,
+      "grad_norm": 0.6015191674232483,
+      "learning_rate": 0.0002,
+      "loss": 1.558,
+      "step": 410
+    },
+    {
+      "epoch": 0.30401737242128124,
+      "grad_norm": 0.6380982398986816,
+      "learning_rate": 0.0002,
+      "loss": 1.6091,
+      "step": 420
+    },
+    {
+      "epoch": 0.3112558812884546,
+      "grad_norm": 0.6707863211631775,
+      "learning_rate": 0.0002,
+      "loss": 1.5292,
+      "step": 430
+    },
+    {
+      "epoch": 0.3184943901556279,
+      "grad_norm": 0.7010176777839661,
+      "learning_rate": 0.0002,
+      "loss": 1.4426,
+      "step": 440
+    },
+    {
+      "epoch": 0.3257328990228013,
+      "grad_norm": 0.8263739943504333,
+      "learning_rate": 0.0002,
+      "loss": 1.5572,
+      "step": 450
+    },
+    {
+      "epoch": 0.33297140788997465,
+      "grad_norm": 0.7253276109695435,
+      "learning_rate": 0.0002,
+      "loss": 1.5188,
+      "step": 460
+    },
+    {
+      "epoch": 0.340209916757148,
+      "grad_norm": 0.5238934755325317,
+      "learning_rate": 0.0002,
+      "loss": 1.584,
+      "step": 470
+    },
+    {
+      "epoch": 0.3474484256243214,
+      "grad_norm": 0.7869495749473572,
+      "learning_rate": 0.0002,
+      "loss": 1.7035,
+      "step": 480
+    },
+    {
+      "epoch": 0.35468693449149474,
+      "grad_norm": 0.7485215663909912,
+      "learning_rate": 0.0002,
+      "loss": 1.5776,
+      "step": 490
+    },
+    {
+      "epoch": 0.3619254433586681,
+      "grad_norm": 0.5413193106651306,
+      "learning_rate": 0.0002,
+      "loss": 1.6274,
+      "step": 500
+    },
+    {
+      "epoch": 0.3691639522258415,
+      "grad_norm": 0.7615048885345459,
+      "learning_rate": 0.0002,
+      "loss": 1.7323,
+      "step": 510
+    },
+    {
+      "epoch": 0.37640246109301484,
+      "grad_norm": 0.7685340046882629,
+      "learning_rate": 0.0002,
+      "loss": 1.532,
+      "step": 520
+    },
+    {
+      "epoch": 0.3836409699601882,
+      "grad_norm": 0.6379081010818481,
+      "learning_rate": 0.0002,
+      "loss": 1.6312,
+      "step": 530
+    },
+    {
+      "epoch": 0.39087947882736157,
+      "grad_norm": 0.7946939468383789,
+      "learning_rate": 0.0002,
+      "loss": 1.5645,
+      "step": 540
+    },
+    {
+      "epoch": 0.39811798769453494,
+      "grad_norm": 0.6287278532981873,
+      "learning_rate": 0.0002,
+      "loss": 1.4001,
+      "step": 550
+    },
+    {
+      "epoch": 0.4053564965617083,
+      "grad_norm": 0.6811642646789551,
+      "learning_rate": 0.0002,
+      "loss": 1.5982,
+      "step": 560
+    },
+    {
+      "epoch": 0.41259500542888167,
+      "grad_norm": 0.671073317527771,
+      "learning_rate": 0.0002,
+      "loss": 1.4953,
+      "step": 570
+    },
+    {
+      "epoch": 0.41983351429605503,
+      "grad_norm": 0.6313900351524353,
+      "learning_rate": 0.0002,
+      "loss": 1.6753,
+      "step": 580
+    },
+    {
+      "epoch": 0.4270720231632284,
+      "grad_norm": 0.5291772484779358,
+      "learning_rate": 0.0002,
+      "loss": 1.546,
+      "step": 590
+    },
+    {
+      "epoch": 0.43431053203040176,
+      "grad_norm": 0.62503582239151,
+      "learning_rate": 0.0002,
+      "loss": 1.5441,
+      "step": 600
+    },
+    {
+      "epoch": 0.4415490408975751,
+      "grad_norm": 0.5777305364608765,
+      "learning_rate": 0.0002,
+      "loss": 1.6276,
+      "step": 610
+    },
+    {
+      "epoch": 0.44878754976474844,
+      "grad_norm": 0.7013497352600098,
+      "learning_rate": 0.0002,
+      "loss": 1.4758,
+      "step": 620
+    },
+    {
+      "epoch": 0.4560260586319218,
+      "grad_norm": 0.8044822216033936,
+      "learning_rate": 0.0002,
+      "loss": 1.4029,
+      "step": 630
+    },
+    {
+      "epoch": 0.46326456749909517,
+      "grad_norm": 0.672531247138977,
+      "learning_rate": 0.0002,
+      "loss": 1.7195,
+      "step": 640
+    },
+    {
+      "epoch": 0.47050307636626854,
+      "grad_norm": 0.6233910322189331,
+      "learning_rate": 0.0002,
+      "loss": 1.614,
+      "step": 650
+    },
+    {
+      "epoch": 0.4777415852334419,
+      "grad_norm": 0.651524543762207,
+      "learning_rate": 0.0002,
+      "loss": 1.6041,
+      "step": 660
+    },
+    {
+      "epoch": 0.48498009410061527,
+      "grad_norm": 0.7213939428329468,
+      "learning_rate": 0.0002,
+      "loss": 1.5842,
+      "step": 670
+    },
+    {
+      "epoch": 0.49221860296778863,
+      "grad_norm": 0.6541454792022705,
+      "learning_rate": 0.0002,
+      "loss": 1.5453,
+      "step": 680
+    },
+    {
+      "epoch": 0.499457111834962,
+      "grad_norm": 0.6568936109542847,
+      "learning_rate": 0.0002,
+      "loss": 1.662,
+      "step": 690
+    },
+    {
+      "epoch": 0.5066956207021354,
+      "grad_norm": 0.7176415324211121,
+      "learning_rate": 0.0002,
+      "loss": 1.624,
+      "step": 700
+    },
+    {
+      "epoch": 0.5139341295693087,
+      "grad_norm": 0.6553855538368225,
+      "learning_rate": 0.0002,
+      "loss": 1.6099,
+      "step": 710
+    },
+    {
+      "epoch": 0.5211726384364821,
+      "grad_norm": 0.5654335618019104,
+      "learning_rate": 0.0002,
+      "loss": 1.5508,
+      "step": 720
+    },
+    {
+      "epoch": 0.5284111473036555,
+      "grad_norm": 0.5671001672744751,
+      "learning_rate": 0.0002,
+      "loss": 1.392,
+      "step": 730
+    },
+    {
+      "epoch": 0.5356496561708288,
+      "grad_norm": 0.7914412021636963,
+      "learning_rate": 0.0002,
+      "loss": 1.388,
+      "step": 740
+    },
+    {
+      "epoch": 0.5428881650380022,
+      "grad_norm": 0.6172138452529907,
+      "learning_rate": 0.0002,
+      "loss": 1.5931,
+      "step": 750
+    },
+    {
+      "epoch": 0.5501266739051756,
+      "grad_norm": 0.6132623553276062,
+      "learning_rate": 0.0002,
+      "loss": 1.4018,
+      "step": 760
+    },
+    {
+      "epoch": 0.5573651827723489,
+      "grad_norm": 0.654000461101532,
+      "learning_rate": 0.0002,
+      "loss": 1.513,
+      "step": 770
+    },
+    {
+      "epoch": 0.5646036916395223,
+      "grad_norm": 0.5691370964050293,
+      "learning_rate": 0.0002,
+      "loss": 1.5035,
+      "step": 780
+    },
+    {
+      "epoch": 0.5718422005066957,
+      "grad_norm": 0.7922580242156982,
+      "learning_rate": 0.0002,
+      "loss": 1.65,
+      "step": 790
+    },
+    {
+      "epoch": 0.579080709373869,
+      "grad_norm": 0.6831880211830139,
+      "learning_rate": 0.0002,
+      "loss": 1.4521,
+      "step": 800
+    },
+    {
+      "epoch": 0.5863192182410424,
+      "grad_norm": 0.6740124821662903,
+      "learning_rate": 0.0002,
+      "loss": 1.4734,
+      "step": 810
+    },
+    {
+      "epoch": 0.5935577271082157,
+      "grad_norm": 1.380016803741455,
+      "learning_rate": 0.0002,
+      "loss": 1.6498,
+      "step": 820
+    },
+    {
+      "epoch": 0.6007962359753891,
+      "grad_norm": 0.6552878022193909,
+      "learning_rate": 0.0002,
+      "loss": 1.4642,
+      "step": 830
+    },
+    {
+      "epoch": 0.6080347448425625,
+      "grad_norm": 0.6649535298347473,
+      "learning_rate": 0.0002,
+      "loss": 1.6271,
+      "step": 840
+    },
+    {
+      "epoch": 0.6152732537097358,
+      "grad_norm": 0.561738133430481,
+      "learning_rate": 0.0002,
+      "loss": 1.5886,
+      "step": 850
+    },
+    {
+      "epoch": 0.6225117625769092,
+      "grad_norm": 0.6133047938346863,
+      "learning_rate": 0.0002,
+      "loss": 1.5364,
+      "step": 860
+    },
+    {
+      "epoch": 0.6297502714440825,
+      "grad_norm": 0.559843122959137,
+      "learning_rate": 0.0002,
+      "loss": 1.3489,
+      "step": 870
+    },
+    {
+      "epoch": 0.6369887803112558,
+      "grad_norm": 0.6117811799049377,
+      "learning_rate": 0.0002,
+      "loss": 1.4878,
+      "step": 880
+    },
+    {
+      "epoch": 0.6442272891784292,
+      "grad_norm": 0.6209776401519775,
+      "learning_rate": 0.0002,
+      "loss": 1.56,
+      "step": 890
+    },
+    {
+      "epoch": 0.6514657980456026,
+      "grad_norm": 0.6234082579612732,
+      "learning_rate": 0.0002,
+      "loss": 1.6747,
+      "step": 900
+    },
+    {
+      "epoch": 0.6587043069127759,
+      "grad_norm": 0.7623258233070374,
+      "learning_rate": 0.0002,
+      "loss": 1.6963,
+      "step": 910
+    },
+    {
+      "epoch": 0.6659428157799493,
+      "grad_norm": 0.6148061752319336,
+      "learning_rate": 0.0002,
+      "loss": 1.2424,
+      "step": 920
+    },
+    {
+      "epoch": 0.6731813246471227,
+      "grad_norm": 0.6682973504066467,
+      "learning_rate": 0.0002,
+      "loss": 1.4319,
+      "step": 930
+    },
+    {
+      "epoch": 0.680419833514296,
+      "grad_norm": 0.5513041615486145,
+      "learning_rate": 0.0002,
+      "loss": 1.5377,
+      "step": 940
+    },
+    {
+      "epoch": 0.6876583423814694,
+      "grad_norm": 0.5197525024414062,
+      "learning_rate": 0.0002,
+      "loss": 1.3991,
+      "step": 950
+    },
+    {
+      "epoch": 0.6948968512486428,
+      "grad_norm": 0.6490758061408997,
+      "learning_rate": 0.0002,
+      "loss": 1.4398,
+      "step": 960
+    },
+    {
+      "epoch": 0.7021353601158161,
+      "grad_norm": 0.6450682878494263,
+      "learning_rate": 0.0002,
+      "loss": 1.5251,
+      "step": 970
+    },
+    {
+      "epoch": 0.7093738689829895,
+      "grad_norm": 0.6203766465187073,
+      "learning_rate": 0.0002,
+      "loss": 1.5417,
+      "step": 980
+    },
+    {
+      "epoch": 0.7166123778501629,
+      "grad_norm": 0.6023609638214111,
+      "learning_rate": 0.0002,
+      "loss": 1.4575,
+      "step": 990
+    },
+    {
+      "epoch": 0.7238508867173362,
+      "grad_norm": 0.5765255093574524,
+      "learning_rate": 0.0002,
+      "loss": 1.4973,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7310893955845096,
+      "grad_norm": 0.6650075316429138,
+      "learning_rate": 0.0002,
+      "loss": 1.483,
+      "step": 1010
+    },
+    {
+      "epoch": 0.738327904451683,
+      "grad_norm": 0.5610854029655457,
+      "learning_rate": 0.0002,
+      "loss": 1.5959,
+      "step": 1020
+    },
+    {
+      "epoch": 0.7455664133188563,
+      "grad_norm": 0.7072813510894775,
+      "learning_rate": 0.0002,
+      "loss": 1.5248,
+      "step": 1030
+    },
+    {
+      "epoch": 0.7528049221860297,
+      "grad_norm": 0.6815407872200012,
+      "learning_rate": 0.0002,
+      "loss": 1.5776,
+      "step": 1040
+    },
+    {
+      "epoch": 0.760043431053203,
+      "grad_norm": 0.7932390570640564,
+      "learning_rate": 0.0002,
+      "loss": 1.4577,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7672819399203764,
+      "grad_norm": 0.5798183083534241,
+      "learning_rate": 0.0002,
+      "loss": 1.4515,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7745204487875498,
+      "grad_norm": 0.7898504137992859,
+      "learning_rate": 0.0002,
+      "loss": 1.5053,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7817589576547231,
+      "grad_norm": 0.4983280301094055,
+      "learning_rate": 0.0002,
+      "loss": 1.4776,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7889974665218965,
+      "grad_norm": 0.691403329372406,
+      "learning_rate": 0.0002,
+      "loss": 1.5007,
+      "step": 1090
+    },
+    {
+      "epoch": 0.7962359753890699,
+      "grad_norm": 0.5394481420516968,
+      "learning_rate": 0.0002,
+      "loss": 1.5153,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8034744842562432,
+      "grad_norm": 0.5136822462081909,
+      "learning_rate": 0.0002,
+      "loss": 1.6892,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8107129931234166,
+      "grad_norm": 0.6828126907348633,
+      "learning_rate": 0.0002,
+      "loss": 1.4902,
+      "step": 1120
+    },
+    {
+      "epoch": 0.81795150199059,
+      "grad_norm": 0.6799656748771667,
+      "learning_rate": 0.0002,
+      "loss": 1.4346,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8251900108577633,
+      "grad_norm": 0.5428406000137329,
+      "learning_rate": 0.0002,
+      "loss": 1.2678,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8324285197249367,
+      "grad_norm": 0.4811290502548218,
+      "learning_rate": 0.0002,
+      "loss": 1.4072,
+      "step": 1150
+    },
+    {
+      "epoch": 0.8396670285921101,
+      "grad_norm": 0.5519434809684753,
+      "learning_rate": 0.0002,
+      "loss": 1.4512,
+      "step": 1160
+    },
+    {
+      "epoch": 0.8469055374592834,
+      "grad_norm": 0.9748060703277588,
+      "learning_rate": 0.0002,
+      "loss": 1.4072,
+      "step": 1170
+    },
+    {
+      "epoch": 0.8541440463264568,
+      "grad_norm": 0.712609589099884,
+      "learning_rate": 0.0002,
+      "loss": 1.4309,
+      "step": 1180
+    },
+    {
+      "epoch": 0.8613825551936302,
+      "grad_norm": 0.6866157054901123,
+      "learning_rate": 0.0002,
+      "loss": 1.434,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8686210640608035,
+      "grad_norm": 0.5068854093551636,
+      "learning_rate": 0.0002,
+      "loss": 1.3704,
+      "step": 1200
+    },
+    {
+      "epoch": 0.8758595729279768,
+      "grad_norm": 0.6333245038986206,
+      "learning_rate": 0.0002,
+      "loss": 1.5601,
+      "step": 1210
+    },
+    {
+      "epoch": 0.8830980817951501,
+      "grad_norm": 0.6424421072006226,
+      "learning_rate": 0.0002,
+      "loss": 1.4636,
+      "step": 1220
+    },
+    {
+      "epoch": 0.8903365906623235,
+      "grad_norm": 0.4771921932697296,
+      "learning_rate": 0.0002,
+      "loss": 1.4186,
+      "step": 1230
+    },
+    {
+      "epoch": 0.8975750995294969,
+      "grad_norm": 0.5191764235496521,
+      "learning_rate": 0.0002,
+      "loss": 1.6323,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9048136083966702,
+      "grad_norm": 0.756222128868103,
+      "learning_rate": 0.0002,
+      "loss": 1.6105,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9120521172638436,
+      "grad_norm": 0.623823881149292,
+      "learning_rate": 0.0002,
+      "loss": 1.4396,
+      "step": 1260
+    },
+    {
+      "epoch": 0.919290626131017,
+      "grad_norm": 0.8166571259498596,
+      "learning_rate": 0.0002,
+      "loss": 1.3097,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9265291349981903,
+      "grad_norm": 0.6059346795082092,
+      "learning_rate": 0.0002,
+      "loss": 1.4625,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9337676438653637,
+      "grad_norm": 0.5842690467834473,
+      "learning_rate": 0.0002,
+      "loss": 1.3555,
+      "step": 1290
+    },
+    {
+      "epoch": 0.9410061527325371,
+      "grad_norm": 0.7649800777435303,
+      "learning_rate": 0.0002,
+      "loss": 1.5859,
+      "step": 1300
+    },
+    {
+      "epoch": 0.9482446615997104,
+      "grad_norm": 0.6420919895172119,
+      "learning_rate": 0.0002,
+      "loss": 1.5915,
+      "step": 1310
+    },
+    {
+      "epoch": 0.9554831704668838,
+      "grad_norm": 0.7011452913284302,
+      "learning_rate": 0.0002,
+      "loss": 1.453,
+      "step": 1320
+    },
+    {
+      "epoch": 0.9627216793340572,
+      "grad_norm": 0.5783746242523193,
+      "learning_rate": 0.0002,
+      "loss": 1.6766,
+      "step": 1330
+    },
+    {
+      "epoch": 0.9699601882012305,
+      "grad_norm": 0.5973192453384399,
+      "learning_rate": 0.0002,
+      "loss": 1.6308,
+      "step": 1340
+    },
+    {
+      "epoch": 0.9771986970684039,
+      "grad_norm": 0.6181833744049072,
+      "learning_rate": 0.0002,
+      "loss": 1.5901,
+      "step": 1350
+    },
+    {
+      "epoch": 0.9844372059355773,
+      "grad_norm": 0.5563396215438843,
+      "learning_rate": 0.0002,
+      "loss": 1.5258,
+      "step": 1360
+    },
+    {
+      "epoch": 0.9916757148027506,
+      "grad_norm": 0.45723360776901245,
+      "learning_rate": 0.0002,
+      "loss": 1.4508,
+      "step": 1370
+    },
+    {
+      "epoch": 0.998914223669924,
+      "grad_norm": 0.5947498679161072,
+      "learning_rate": 0.0002,
+      "loss": 1.3291,
+      "step": 1380
+    },
+    {
+      "epoch": 0.9996380745566413,
+      "eval_loss": 1.480796456336975,
+      "eval_runtime": 27.3103,
+      "eval_samples_per_second": 15.965,
+      "eval_steps_per_second": 2.014,
+      "step": 1381
+    },
+    {
+      "epoch": 1.0061527325370974,
+      "grad_norm": 0.5599952936172485,
+      "learning_rate": 0.0002,
+      "loss": 1.3057,
+      "step": 1390
+    },
+    {
+      "epoch": 1.0133912414042707,
+      "grad_norm": 0.5932008028030396,
+      "learning_rate": 0.0002,
+      "loss": 1.4991,
+      "step": 1400
+    },
+    {
+      "epoch": 1.020629750271444,
+      "grad_norm": 0.6194121837615967,
+      "learning_rate": 0.0002,
+      "loss": 1.4506,
+      "step": 1410
+    },
+    {
+      "epoch": 1.0278682591386175,
+      "grad_norm": 0.6995621919631958,
+      "learning_rate": 0.0002,
+      "loss": 1.5966,
+      "step": 1420
+    },
+    {
+      "epoch": 1.0351067680057908,
+      "grad_norm": 0.7905810475349426,
+      "learning_rate": 0.0002,
+      "loss": 1.4153,
+      "step": 1430
+    },
+    {
+      "epoch": 1.0423452768729642,
+      "grad_norm": 0.7221615314483643,
+      "learning_rate": 0.0002,
+      "loss": 1.4414,
+      "step": 1440
+    },
+    {
+      "epoch": 1.0495837857401376,
+      "grad_norm": 0.6170642375946045,
+      "learning_rate": 0.0002,
+      "loss": 1.3859,
+      "step": 1450
+    },
+    {
+      "epoch": 1.056822294607311,
+      "grad_norm": 0.5844094753265381,
+      "learning_rate": 0.0002,
+      "loss": 1.3806,
+      "step": 1460
+    },
+    {
+      "epoch": 1.0640608034744843,
+      "grad_norm": 0.7731822729110718,
+      "learning_rate": 0.0002,
+      "loss": 1.4871,
+      "step": 1470
+    },
+    {
+      "epoch": 1.0712993123416577,
+      "grad_norm": 0.4554748237133026,
+      "learning_rate": 0.0002,
+      "loss": 1.4286,
+      "step": 1480
+    },
+    {
+      "epoch": 1.078537821208831,
+      "grad_norm": 0.6923259496688843,
+      "learning_rate": 0.0002,
+      "loss": 1.3977,
+      "step": 1490
+    },
+    {
+      "epoch": 1.0857763300760044,
+      "grad_norm": 0.6008219122886658,
+      "learning_rate": 0.0002,
+      "loss": 1.3936,
+      "step": 1500
+    },
+    {
+      "epoch": 1.0930148389431777,
+      "grad_norm": 0.6450045704841614,
+      "learning_rate": 0.0002,
+      "loss": 1.4821,
+      "step": 1510
+    },
+    {
+      "epoch": 1.1002533478103511,
+      "grad_norm": 0.7833753824234009,
+      "learning_rate": 0.0002,
+      "loss": 1.3295,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1074918566775245,
+      "grad_norm": 0.5076758861541748,
+      "learning_rate": 0.0002,
+      "loss": 1.3424,
+      "step": 1530
+    },
+    {
+      "epoch": 1.1147303655446978,
+      "grad_norm": 0.5661332011222839,
+      "learning_rate": 0.0002,
+      "loss": 1.4043,
+      "step": 1540
+    },
+    {
+      "epoch": 1.1219688744118712,
+      "grad_norm": 0.6526919603347778,
+      "learning_rate": 0.0002,
+      "loss": 1.4963,
+      "step": 1550
+    },
+    {
+      "epoch": 1.1292073832790446,
+      "grad_norm": 0.5613082647323608,
+      "learning_rate": 0.0002,
+      "loss": 1.3671,
+      "step": 1560
+    },
+    {
+      "epoch": 1.136445892146218,
+      "grad_norm": 0.6113885641098022,
+      "learning_rate": 0.0002,
+      "loss": 1.4458,
+      "step": 1570
+    },
+    {
+      "epoch": 1.1436844010133913,
+      "grad_norm": 0.6732510328292847,
+      "learning_rate": 0.0002,
+      "loss": 1.3552,
+      "step": 1580
+    },
+    {
+      "epoch": 1.1509229098805647,
+      "grad_norm": 0.6146392226219177,
+      "learning_rate": 0.0002,
+      "loss": 1.3114,
+      "step": 1590
+    },
+    {
+      "epoch": 1.158161418747738,
+      "grad_norm": 0.6766974329948425,
+      "learning_rate": 0.0002,
+      "loss": 1.411,
+      "step": 1600
+    },
+    {
+      "epoch": 1.1653999276149114,
+      "grad_norm": 0.7621957659721375,
+      "learning_rate": 0.0002,
+      "loss": 1.2401,
+      "step": 1610
+    },
+    {
+      "epoch": 1.1726384364820848,
+      "grad_norm": 0.6959581971168518,
+      "learning_rate": 0.0002,
+      "loss": 1.3758,
+      "step": 1620
+    },
+    {
+      "epoch": 1.1798769453492581,
+      "grad_norm": 0.6691278219223022,
+      "learning_rate": 0.0002,
+      "loss": 1.382,
+      "step": 1630
+    },
+    {
+      "epoch": 1.1871154542164315,
+      "grad_norm": 0.4927774965763092,
+      "learning_rate": 0.0002,
+      "loss": 1.4147,
+      "step": 1640
+    },
+    {
+      "epoch": 1.1943539630836049,
+      "grad_norm": 0.7724234461784363,
+      "learning_rate": 0.0002,
+      "loss": 1.449,
+      "step": 1650
+    },
+    {
+      "epoch": 1.2015924719507782,
+      "grad_norm": 0.6817787885665894,
+      "learning_rate": 0.0002,
+      "loss": 1.4778,
+      "step": 1660
+    },
+    {
+      "epoch": 1.2088309808179516,
+      "grad_norm": 0.6500699520111084,
+      "learning_rate": 0.0002,
+      "loss": 1.3776,
+      "step": 1670
+    },
+    {
+      "epoch": 1.216069489685125,
+      "grad_norm": 0.5703568458557129,
+      "learning_rate": 0.0002,
+      "loss": 1.3875,
+      "step": 1680
+    },
+    {
+      "epoch": 1.2233079985522983,
+      "grad_norm": 0.6261579990386963,
+      "learning_rate": 0.0002,
+      "loss": 1.4735,
+      "step": 1690
+    },
+    {
+      "epoch": 1.2305465074194717,
+      "grad_norm": 0.651713490486145,
+      "learning_rate": 0.0002,
+      "loss": 1.3898,
+      "step": 1700
+    },
+    {
+      "epoch": 1.237785016286645,
+      "grad_norm": 0.684399425983429,
+      "learning_rate": 0.0002,
+      "loss": 1.4002,
+      "step": 1710
+    },
+    {
+      "epoch": 1.2450235251538184,
+      "grad_norm": 0.6996857523918152,
+      "learning_rate": 0.0002,
+      "loss": 1.5027,
+      "step": 1720
+    },
+    {
+      "epoch": 1.2522620340209918,
+      "grad_norm": 0.7102537751197815,
+      "learning_rate": 0.0002,
+      "loss": 1.3326,
+      "step": 1730
+    },
+    {
+      "epoch": 1.2595005428881652,
+      "grad_norm": 0.45809897780418396,
+      "learning_rate": 0.0002,
+      "loss": 1.3675,
+      "step": 1740
+    },
+    {
+      "epoch": 1.2667390517553385,
+      "grad_norm": 0.6377046704292297,
+      "learning_rate": 0.0002,
+      "loss": 1.4175,
+      "step": 1750
+    },
+    {
+      "epoch": 1.2739775606225119,
+      "grad_norm": 0.6965704560279846,
+      "learning_rate": 0.0002,
+      "loss": 1.3479,
+      "step": 1760
+    },
+    {
+      "epoch": 1.2812160694896852,
+      "grad_norm": 0.5688214302062988,
+      "learning_rate": 0.0002,
+      "loss": 1.5647,
+      "step": 1770
+    },
+    {
+      "epoch": 1.2884545783568586,
+      "grad_norm": 0.6384190320968628,
+      "learning_rate": 0.0002,
+      "loss": 1.3967,
+      "step": 1780
+    },
+    {
+      "epoch": 1.295693087224032,
+      "grad_norm": 0.5629363656044006,
+      "learning_rate": 0.0002,
+      "loss": 1.3671,
+      "step": 1790
+    },
+    {
+      "epoch": 1.3029315960912053,
+      "grad_norm": 0.6148255467414856,
+      "learning_rate": 0.0002,
+      "loss": 1.2292,
+      "step": 1800
+    },
+    {
+      "epoch": 1.3101701049583787,
+      "grad_norm": 0.655580997467041,
+      "learning_rate": 0.0002,
+      "loss": 1.5806,
+      "step": 1810
+    },
+    {
+      "epoch": 1.3174086138255519,
+      "grad_norm": 0.5642657279968262,
+      "learning_rate": 0.0002,
+      "loss": 1.2398,
+      "step": 1820
+    },
+    {
+      "epoch": 1.3246471226927252,
+      "grad_norm": 0.59607994556427,
+      "learning_rate": 0.0002,
+      "loss": 1.3246,
+      "step": 1830
+    },
+    {
+      "epoch": 1.3318856315598986,
+      "grad_norm": 0.5564199090003967,
+      "learning_rate": 0.0002,
+      "loss": 1.3274,
+      "step": 1840
+    },
+    {
+      "epoch": 1.339124140427072,
+      "grad_norm": 0.6949955821037292,
+      "learning_rate": 0.0002,
+      "loss": 1.5834,
+      "step": 1850
+    },
+    {
+      "epoch": 1.3463626492942453,
+      "grad_norm": 0.7036856412887573,
+      "learning_rate": 0.0002,
+      "loss": 1.4722,
+      "step": 1860
+    },
+    {
+      "epoch": 1.3536011581614187,
+      "grad_norm": 0.722062885761261,
+      "learning_rate": 0.0002,
+      "loss": 1.333,
+      "step": 1870
+    },
+    {
+      "epoch": 1.360839667028592,
+      "grad_norm": 0.6098677515983582,
+      "learning_rate": 0.0002,
+      "loss": 1.4044,
+      "step": 1880
+    },
+    {
+      "epoch": 1.3680781758957654,
+      "grad_norm": 0.5376402735710144,
+      "learning_rate": 0.0002,
+      "loss": 1.6217,
+      "step": 1890
+    },
+    {
+      "epoch": 1.3753166847629388,
+      "grad_norm": 0.6974610090255737,
+      "learning_rate": 0.0002,
+      "loss": 1.5071,
+      "step": 1900
+    },
+    {
+      "epoch": 1.3825551936301121,
+      "grad_norm": 0.6520763635635376,
+      "learning_rate": 0.0002,
+      "loss": 1.5854,
+      "step": 1910
+    },
+    {
+      "epoch": 1.3897937024972855,
+      "grad_norm": 0.6604374647140503,
+      "learning_rate": 0.0002,
+      "loss": 1.4271,
+      "step": 1920
+    },
+    {
+      "epoch": 1.3970322113644589,
+      "grad_norm": 0.7364398241043091,
+      "learning_rate": 0.0002,
+      "loss": 1.419,
+      "step": 1930
+    },
+    {
+      "epoch": 1.4042707202316322,
+      "grad_norm": 0.6849475502967834,
+      "learning_rate": 0.0002,
+      "loss": 1.4585,
+      "step": 1940
+    },
+    {
+      "epoch": 1.4115092290988056,
+      "grad_norm": 0.6562670469284058,
+      "learning_rate": 0.0002,
+      "loss": 1.5577,
+      "step": 1950
+    },
+    {
+      "epoch": 1.418747737965979,
+      "grad_norm": 0.5695616006851196,
+      "learning_rate": 0.0002,
+      "loss": 1.4725,
+      "step": 1960
+    },
+    {
+      "epoch": 1.4259862468331523,
+      "grad_norm": 0.5244464874267578,
+      "learning_rate": 0.0002,
+      "loss": 1.3088,
+      "step": 1970
+    },
+    {
+      "epoch": 1.4332247557003257,
+      "grad_norm": 0.6347293257713318,
+      "learning_rate": 0.0002,
+      "loss": 1.5069,
+      "step": 1980
+    },
+    {
+      "epoch": 1.440463264567499,
+      "grad_norm": 0.5528361201286316,
+      "learning_rate": 0.0002,
+      "loss": 1.3502,
+      "step": 1990
+    },
+    {
+      "epoch": 1.4477017734346724,
+      "grad_norm": 0.6987585425376892,
+      "learning_rate": 0.0002,
+      "loss": 1.3978,
+      "step": 2000
+    },
+    {
+      "epoch": 1.4549402823018458,
+      "grad_norm": 0.6568987369537354,
+      "learning_rate": 0.0002,
+      "loss": 1.4262,
+      "step": 2010
+    },
+    {
+      "epoch": 1.4621787911690192,
+      "grad_norm": 0.7665994763374329,
+      "learning_rate": 0.0002,
+      "loss": 1.4175,
+      "step": 2020
+    },
+    {
+      "epoch": 1.4694173000361925,
+      "grad_norm": 0.5127707123756409,
+      "learning_rate": 0.0002,
+      "loss": 1.244,
+      "step": 2030
+    },
+    {
+      "epoch": 1.476655808903366,
+      "grad_norm": 0.5406824946403503,
+      "learning_rate": 0.0002,
+      "loss": 1.3699,
+      "step": 2040
+    },
+    {
+      "epoch": 1.4838943177705393,
+      "grad_norm": 0.5990166664123535,
+      "learning_rate": 0.0002,
+      "loss": 1.3353,
+      "step": 2050
+    },
+    {
+      "epoch": 1.4911328266377126,
+      "grad_norm": 0.6186193823814392,
+      "learning_rate": 0.0002,
+      "loss": 1.2454,
+      "step": 2060
+    },
+    {
+      "epoch": 1.498371335504886,
+      "grad_norm": 0.6154307126998901,
+      "learning_rate": 0.0002,
+      "loss": 1.428,
+      "step": 2070
+    },
+    {
+      "epoch": 1.5056098443720594,
+      "grad_norm": 0.5606056451797485,
+      "learning_rate": 0.0002,
+      "loss": 1.4528,
+      "step": 2080
+    },
+    {
+      "epoch": 1.5128483532392327,
+      "grad_norm": 0.5006417036056519,
+      "learning_rate": 0.0002,
+      "loss": 1.2405,
+      "step": 2090
+    },
+    {
+      "epoch": 1.520086862106406,
+      "grad_norm": 0.5968486070632935,
+      "learning_rate": 0.0002,
+      "loss": 1.4258,
+      "step": 2100
+    },
+    {
+      "epoch": 1.5273253709735795,
+      "grad_norm": 0.5835496187210083,
+      "learning_rate": 0.0002,
+      "loss": 1.2752,
+      "step": 2110
+    },
+    {
+      "epoch": 1.5345638798407528,
+      "grad_norm": 0.6753535270690918,
+      "learning_rate": 0.0002,
+      "loss": 1.5443,
+      "step": 2120
+    },
+    {
+      "epoch": 1.5418023887079262,
+      "grad_norm": 0.7299720644950867,
+      "learning_rate": 0.0002,
+      "loss": 1.2139,
+      "step": 2130
+    },
+    {
+      "epoch": 1.5490408975750996,
+      "grad_norm": 0.5105988383293152,
+      "learning_rate": 0.0002,
+      "loss": 1.2364,
+      "step": 2140
+    },
+    {
+      "epoch": 1.556279406442273,
+      "grad_norm": 0.5675431489944458,
+      "learning_rate": 0.0002,
+      "loss": 1.4528,
+      "step": 2150
+    },
+    {
+      "epoch": 1.5635179153094463,
+      "grad_norm": 0.6246723532676697,
+      "learning_rate": 0.0002,
+      "loss": 1.4563,
+      "step": 2160
+    },
+    {
+      "epoch": 1.5707564241766196,
+      "grad_norm": 0.7291720509529114,
+      "learning_rate": 0.0002,
+      "loss": 1.5255,
+      "step": 2170
+    },
+    {
+      "epoch": 1.577994933043793,
+      "grad_norm": 0.678114116191864,
+      "learning_rate": 0.0002,
+      "loss": 1.5432,
+      "step": 2180
+    },
+    {
+      "epoch": 1.5852334419109664,
+      "grad_norm": 0.5136260986328125,
+      "learning_rate": 0.0002,
+      "loss": 1.5212,
+      "step": 2190
+    },
+    {
+      "epoch": 1.5924719507781397,
+      "grad_norm": 0.6359935998916626,
+      "learning_rate": 0.0002,
+      "loss": 1.3271,
+      "step": 2200
+    },
+    {
+      "epoch": 1.599710459645313,
+      "grad_norm": 0.7650278806686401,
+      "learning_rate": 0.0002,
+      "loss": 1.4038,
+      "step": 2210
+    },
+    {
+      "epoch": 1.6069489685124865,
+      "grad_norm": 0.7256110906600952,
+      "learning_rate": 0.0002,
+      "loss": 1.5478,
+      "step": 2220
+    },
+    {
+      "epoch": 1.6141874773796598,
+      "grad_norm": 0.688689649105072,
+      "learning_rate": 0.0002,
+      "loss": 1.4387,
+      "step": 2230
+    },
+    {
+      "epoch": 1.6214259862468332,
+      "grad_norm": 0.6045311093330383,
+      "learning_rate": 0.0002,
+      "loss": 1.4096,
+      "step": 2240
+    },
+    {
+      "epoch": 1.6286644951140063,
+      "grad_norm": 0.7064604163169861,
+      "learning_rate": 0.0002,
+      "loss": 1.4097,
+      "step": 2250
+    },
+    {
+      "epoch": 1.6359030039811797,
+      "grad_norm": 0.5309562087059021,
+      "learning_rate": 0.0002,
+      "loss": 1.3477,
+      "step": 2260
+    },
+    {
+      "epoch": 1.643141512848353,
+      "grad_norm": 0.5687053203582764,
+      "learning_rate": 0.0002,
+      "loss": 1.4022,
+      "step": 2270
+    },
+    {
+      "epoch": 1.6503800217155264,
+      "grad_norm": 0.535872757434845,
+      "learning_rate": 0.0002,
+      "loss": 1.2977,
+      "step": 2280
+    },
+    {
+      "epoch": 1.6576185305826998,
+      "grad_norm": 0.5502381920814514,
+      "learning_rate": 0.0002,
+      "loss": 1.3844,
+      "step": 2290
+    },
+    {
+      "epoch": 1.6648570394498732,
+      "grad_norm": 0.6158602237701416,
+      "learning_rate": 0.0002,
+      "loss": 1.3764,
+      "step": 2300
+    },
+    {
+      "epoch": 1.6720955483170465,
+      "grad_norm": 0.5804675817489624,
+      "learning_rate": 0.0002,
+      "loss": 1.3515,
+      "step": 2310
+    },
+    {
+      "epoch": 1.67933405718422,
+      "grad_norm": 0.600742757320404,
+      "learning_rate": 0.0002,
+      "loss": 1.2532,
+      "step": 2320
+    },
+    {
+      "epoch": 1.6865725660513933,
+      "grad_norm": 0.7101941108703613,
+      "learning_rate": 0.0002,
+      "loss": 1.477,
+      "step": 2330
+    },
+    {
+      "epoch": 1.6938110749185666,
+      "grad_norm": 0.7507809996604919,
+      "learning_rate": 0.0002,
+      "loss": 1.4849,
+      "step": 2340
+    },
+    {
+      "epoch": 1.70104958378574,
+      "grad_norm": 0.768502414226532,
+      "learning_rate": 0.0002,
+      "loss": 1.2703,
+      "step": 2350
+    },
+    {
+      "epoch": 1.7082880926529134,
+      "grad_norm": 0.4801851212978363,
+      "learning_rate": 0.0002,
+      "loss": 1.3332,
+      "step": 2360
+    },
+    {
+      "epoch": 1.7155266015200867,
+      "grad_norm": 0.5322122573852539,
+      "learning_rate": 0.0002,
+      "loss": 1.4158,
+      "step": 2370
+    },
+    {
+      "epoch": 1.72276511038726,
+      "grad_norm": 0.587661862373352,
+      "learning_rate": 0.0002,
+      "loss": 1.4136,
+      "step": 2380
+    },
+    {
+      "epoch": 1.7300036192544335,
+      "grad_norm": 0.6073525547981262,
+      "learning_rate": 0.0002,
+      "loss": 1.3771,
+      "step": 2390
+    },
+    {
+      "epoch": 1.7372421281216068,
+      "grad_norm": 0.6950460076332092,
+      "learning_rate": 0.0002,
+      "loss": 1.2754,
+      "step": 2400
+    },
+    {
+      "epoch": 1.7444806369887802,
+      "grad_norm": 0.5981102585792542,
+      "learning_rate": 0.0002,
+      "loss": 1.3858,
+      "step": 2410
+    },
+    {
+      "epoch": 1.7517191458559536,
+      "grad_norm": 0.544570803642273,
+      "learning_rate": 0.0002,
+      "loss": 1.4075,
+      "step": 2420
+    },
+    {
+      "epoch": 1.758957654723127,
+      "grad_norm": 0.5304399728775024,
+      "learning_rate": 0.0002,
+      "loss": 1.3861,
+      "step": 2430
+    },
+    {
+      "epoch": 1.7661961635903003,
+      "grad_norm": 0.7921594977378845,
+      "learning_rate": 0.0002,
+      "loss": 1.4244,
+      "step": 2440
+    },
+    {
+      "epoch": 1.7734346724574737,
+      "grad_norm": 0.6084808707237244,
+      "learning_rate": 0.0002,
+      "loss": 1.3053,
+      "step": 2450
+    },
+    {
+      "epoch": 1.780673181324647,
+      "grad_norm": 0.8844701051712036,
+      "learning_rate": 0.0002,
+      "loss": 1.3781,
+      "step": 2460
+    },
+    {
+      "epoch": 1.7879116901918204,
+      "grad_norm": 0.5729258060455322,
+      "learning_rate": 0.0002,
+      "loss": 1.3227,
+      "step": 2470
+    },
+    {
+      "epoch": 1.7951501990589938,
+      "grad_norm": 0.6303611993789673,
+      "learning_rate": 0.0002,
+      "loss": 1.3422,
+      "step": 2480
+    },
+    {
+      "epoch": 1.8023887079261671,
+      "grad_norm": 0.5627942085266113,
+      "learning_rate": 0.0002,
+      "loss": 1.3926,
+      "step": 2490
+    },
+    {
+      "epoch": 1.8096272167933405,
+      "grad_norm": 0.6724274158477783,
+      "learning_rate": 0.0002,
+      "loss": 1.3816,
+      "step": 2500
+    },
+    {
+      "epoch": 1.8168657256605139,
+      "grad_norm": 0.5030826330184937,
+      "learning_rate": 0.0002,
+      "loss": 1.2951,
+      "step": 2510
+    },
+    {
+      "epoch": 1.8241042345276872,
+      "grad_norm": 0.5504099130630493,
+      "learning_rate": 0.0002,
+      "loss": 1.2839,
+      "step": 2520
+    },
+    {
+      "epoch": 1.8313427433948606,
+      "grad_norm": 0.6338945627212524,
+      "learning_rate": 0.0002,
+      "loss": 1.4264,
+      "step": 2530
+    },
+    {
+      "epoch": 1.838581252262034,
+      "grad_norm": 0.5902037620544434,
+      "learning_rate": 0.0002,
+      "loss": 1.563,
+      "step": 2540
+    },
+    {
+      "epoch": 1.8458197611292073,
+      "grad_norm": 0.48814457654953003,
+      "learning_rate": 0.0002,
+      "loss": 1.2961,
+      "step": 2550
+    },
+    {
+      "epoch": 1.8530582699963807,
+      "grad_norm": 0.6216312646865845,
+      "learning_rate": 0.0002,
+      "loss": 1.466,
+      "step": 2560
+    },
+    {
+      "epoch": 1.860296778863554,
+      "grad_norm": 0.635603666305542,
+      "learning_rate": 0.0002,
+      "loss": 1.5123,
+      "step": 2570
+    },
+    {
+      "epoch": 1.8675352877307274,
+      "grad_norm": 0.6938216090202332,
+      "learning_rate": 0.0002,
+      "loss": 1.372,
+      "step": 2580
+    },
+    {
+      "epoch": 1.8747737965979008,
+      "grad_norm": 0.599557638168335,
+      "learning_rate": 0.0002,
+      "loss": 1.5011,
+      "step": 2590
+    },
+    {
+      "epoch": 1.8820123054650741,
+      "grad_norm": 0.564424455165863,
+      "learning_rate": 0.0002,
+      "loss": 1.2714,
+      "step": 2600
+    },
+    {
+      "epoch": 1.8892508143322475,
+      "grad_norm": 0.5430700182914734,
+      "learning_rate": 0.0002,
+      "loss": 1.3403,
+      "step": 2610
+    },
+    {
+      "epoch": 1.8964893231994209,
+      "grad_norm": 0.6150169372558594,
+      "learning_rate": 0.0002,
+      "loss": 1.4347,
+      "step": 2620
+    },
+    {
+      "epoch": 1.9037278320665942,
+      "grad_norm": 0.48159119486808777,
+      "learning_rate": 0.0002,
+      "loss": 1.2474,
+      "step": 2630
+    },
+    {
+      "epoch": 1.9109663409337676,
+      "grad_norm": 0.5608997941017151,
+      "learning_rate": 0.0002,
+      "loss": 1.3716,
+      "step": 2640
+    },
+    {
+      "epoch": 1.918204849800941,
+      "grad_norm": 0.6454501748085022,
+      "learning_rate": 0.0002,
+      "loss": 1.5787,
+      "step": 2650
+    },
+    {
+      "epoch": 1.9254433586681143,
+      "grad_norm": 0.5458073616027832,
+      "learning_rate": 0.0002,
+      "loss": 1.3238,
+      "step": 2660
+    },
+    {
+      "epoch": 1.9326818675352877,
+      "grad_norm": 0.5328490734100342,
+      "learning_rate": 0.0002,
+      "loss": 1.3208,
+      "step": 2670
+    },
+    {
+      "epoch": 1.939920376402461,
+      "grad_norm": 0.6444696187973022,
+      "learning_rate": 0.0002,
+      "loss": 1.4971,
+      "step": 2680
+    },
+    {
+      "epoch": 1.9471588852696344,
+      "grad_norm": 0.7126023769378662,
+      "learning_rate": 0.0002,
+      "loss": 1.5387,
+      "step": 2690
+    },
+    {
+      "epoch": 1.9543973941368078,
+      "grad_norm": 0.5164045095443726,
+      "learning_rate": 0.0002,
+      "loss": 1.3637,
+      "step": 2700
+    },
+    {
+      "epoch": 1.9616359030039812,
+      "grad_norm": 0.5347061157226562,
+      "learning_rate": 0.0002,
+      "loss": 1.5303,
+      "step": 2710
+    },
+    {
+      "epoch": 1.9688744118711545,
+      "grad_norm": 0.5297950506210327,
+      "learning_rate": 0.0002,
+      "loss": 1.2815,
+      "step": 2720
+    },
+    {
+      "epoch": 1.976112920738328,
+      "grad_norm": 0.6537790298461914,
+      "learning_rate": 0.0002,
+      "loss": 1.3566,
+      "step": 2730
+    },
+    {
+      "epoch": 1.9833514296055013,
+      "grad_norm": 0.5536222457885742,
+      "learning_rate": 0.0002,
+      "loss": 1.332,
+      "step": 2740
+    },
+    {
+      "epoch": 1.9905899384726746,
+      "grad_norm": 0.4856105446815491,
+      "learning_rate": 0.0002,
+      "loss": 1.3333,
+      "step": 2750
+    },
+    {
+      "epoch": 1.997828447339848,
+      "grad_norm": 0.6642730832099915,
+      "learning_rate": 0.0002,
+      "loss": 1.3521,
+      "step": 2760
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.4366681575775146,
+      "eval_runtime": 27.3729,
+      "eval_samples_per_second": 15.928,
+      "eval_steps_per_second": 2.009,
+      "step": 2763
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 11048,
+  "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": 3.376937587389235e+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-0.9-num-8828-sd-1/checkpoint-2763/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-0.9-num-8828-sd-1/checkpoint-2763/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..4d7b7431bbbe8c9bf29b925bca391a558af5ff8c
--- /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-0.9-num-8828-sd-1/checkpoint-2763/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ad613885e4f267fc04125f1a836d42cfa796bbe12e536f9ee60c955de02cdb5a
+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-0.9-num-8828-sd-1/checkpoint-4144/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-0.9-num-8828-sd-1/checkpoint-4144/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-0.9-num-8828-sd-1/checkpoint-4144/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-0.9-num-8828-sd-1/checkpoint-4144/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-0.9-num-8828-sd-1/checkpoint-4144/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e99bbcd43df1c19d98706c7e3be95c93844c5349
--- /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-0.9-num-8828-sd-1/checkpoint-4144/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-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-8828-sd-1/checkpoint-4144/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-0.9-num-8828-sd-1/checkpoint-4144/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..c79d70858ce055c6f1bb883cb35180f5113e2f20
--- /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-0.9-num-8828-sd-1/checkpoint-4144/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bdd71ca55a2af1b8facc7c698205b6599fe0e892ac0b224b8ae93a917b4ed891
+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-0.9-num-8828-sd-1/checkpoint-4144/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-0.9-num-8828-sd-1/checkpoint-4144/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..fefe026d77588a2fb04504506ae4b453fbff41e5
--- /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-0.9-num-8828-sd-1/checkpoint-4144/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2e6774e4d3087b746b73c316efd2376e80be183d49d96bbe52e95af6194cfc6b
+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-0.9-num-8828-sd-1/checkpoint-4144/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-0.9-num-8828-sd-1/checkpoint-4144/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..cf374c934593037876e73e7d373f33b62283978a
--- /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-0.9-num-8828-sd-1/checkpoint-4144/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4fc5b5b8871dcac290402ff2806fc4f44c48ede96564614f00630c0c35f35799
+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-0.9-num-8828-sd-1/checkpoint-4144/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-0.9-num-8828-sd-1/checkpoint-4144/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..3f4c2699ed751970f04911bcaaac08b3a60cf292
--- /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-0.9-num-8828-sd-1/checkpoint-4144/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0bee20d46fc8ef8a7ae4eaca3f0c175688567933abd0069997d944e0d19b0cff
+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-0.9-num-8828-sd-1/checkpoint-4144/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-0.9-num-8828-sd-1/checkpoint-4144/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-0.9-num-8828-sd-1/checkpoint-4144/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-0.9-num-8828-sd-1/checkpoint-4144/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-0.9-num-8828-sd-1/checkpoint-4144/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-0.9-num-8828-sd-1/checkpoint-4144/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-0.9-num-8828-sd-1/checkpoint-4144/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-0.9-num-8828-sd-1/checkpoint-4144/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-0.9-num-8828-sd-1/checkpoint-4144/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-0.9-num-8828-sd-1/checkpoint-4144/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-0.9-num-8828-sd-1/checkpoint-4144/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-0.9-num-8828-sd-1/checkpoint-4144/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-0.9-num-8828-sd-1/checkpoint-4144/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-0.9-num-8828-sd-1/checkpoint-4144/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..3ab605fc65ec684264e9e3fdf98ee1ec92b78a8a
--- /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-0.9-num-8828-sd-1/checkpoint-4144/trainer_state.json
@@ -0,0 +1,2955 @@
+{
+  "best_metric": 1.4217946529388428,
+  "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-0.9-num-8828-sd-1/checkpoint-4144",
+  "epoch": 2.9996380745566413,
+  "eval_steps": 10,
+  "global_step": 4144,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007238508867173362,
+      "grad_norm": 1.2523442506790161,
+      "learning_rate": 0.0002,
+      "loss": 4.7061,
+      "step": 10
+    },
+    {
+      "epoch": 0.014477017734346724,
+      "grad_norm": 1.8887330293655396,
+      "learning_rate": 0.0002,
+      "loss": 3.3493,
+      "step": 20
+    },
+    {
+      "epoch": 0.021715526601520086,
+      "grad_norm": 0.9668035507202148,
+      "learning_rate": 0.0002,
+      "loss": 2.7585,
+      "step": 30
+    },
+    {
+      "epoch": 0.028954035468693448,
+      "grad_norm": 2.9167306423187256,
+      "learning_rate": 0.0002,
+      "loss": 2.3699,
+      "step": 40
+    },
+    {
+      "epoch": 0.036192544335866814,
+      "grad_norm": 2.649867296218872,
+      "learning_rate": 0.0002,
+      "loss": 2.2679,
+      "step": 50
+    },
+    {
+      "epoch": 0.04343105320304017,
+      "grad_norm": 1.5120655298233032,
+      "learning_rate": 0.0002,
+      "loss": 2.2202,
+      "step": 60
+    },
+    {
+      "epoch": 0.05066956207021354,
+      "grad_norm": 0.7879868149757385,
+      "learning_rate": 0.0002,
+      "loss": 2.2026,
+      "step": 70
+    },
+    {
+      "epoch": 0.057908070937386896,
+      "grad_norm": 0.7616953253746033,
+      "learning_rate": 0.0002,
+      "loss": 1.9447,
+      "step": 80
+    },
+    {
+      "epoch": 0.06514657980456026,
+      "grad_norm": 1.8809149265289307,
+      "learning_rate": 0.0002,
+      "loss": 2.0112,
+      "step": 90
+    },
+    {
+      "epoch": 0.07238508867173363,
+      "grad_norm": 0.9294016361236572,
+      "learning_rate": 0.0002,
+      "loss": 1.8337,
+      "step": 100
+    },
+    {
+      "epoch": 0.07962359753890698,
+      "grad_norm": 0.7145281434059143,
+      "learning_rate": 0.0002,
+      "loss": 1.8419,
+      "step": 110
+    },
+    {
+      "epoch": 0.08686210640608034,
+      "grad_norm": 0.7564446330070496,
+      "learning_rate": 0.0002,
+      "loss": 2.0036,
+      "step": 120
+    },
+    {
+      "epoch": 0.09410061527325371,
+      "grad_norm": 1.1681925058364868,
+      "learning_rate": 0.0002,
+      "loss": 1.9306,
+      "step": 130
+    },
+    {
+      "epoch": 0.10133912414042708,
+      "grad_norm": 0.6708641648292542,
+      "learning_rate": 0.0002,
+      "loss": 1.7875,
+      "step": 140
+    },
+    {
+      "epoch": 0.10857763300760044,
+      "grad_norm": 0.7625647783279419,
+      "learning_rate": 0.0002,
+      "loss": 1.786,
+      "step": 150
+    },
+    {
+      "epoch": 0.11581614187477379,
+      "grad_norm": 0.8463464975357056,
+      "learning_rate": 0.0002,
+      "loss": 1.6687,
+      "step": 160
+    },
+    {
+      "epoch": 0.12305465074194716,
+      "grad_norm": 0.7502335906028748,
+      "learning_rate": 0.0002,
+      "loss": 1.6214,
+      "step": 170
+    },
+    {
+      "epoch": 0.13029315960912052,
+      "grad_norm": 0.6929958462715149,
+      "learning_rate": 0.0002,
+      "loss": 1.7433,
+      "step": 180
+    },
+    {
+      "epoch": 0.1375316684762939,
+      "grad_norm": 0.6798707842826843,
+      "learning_rate": 0.0002,
+      "loss": 1.6009,
+      "step": 190
+    },
+    {
+      "epoch": 0.14477017734346725,
+      "grad_norm": 0.7566508650779724,
+      "learning_rate": 0.0002,
+      "loss": 1.6208,
+      "step": 200
+    },
+    {
+      "epoch": 0.15200868621064062,
+      "grad_norm": 0.7196869850158691,
+      "learning_rate": 0.0002,
+      "loss": 1.5823,
+      "step": 210
+    },
+    {
+      "epoch": 0.15924719507781396,
+      "grad_norm": 0.8401045799255371,
+      "learning_rate": 0.0002,
+      "loss": 1.738,
+      "step": 220
+    },
+    {
+      "epoch": 0.16648570394498732,
+      "grad_norm": 0.8503773212432861,
+      "learning_rate": 0.0002,
+      "loss": 1.7574,
+      "step": 230
+    },
+    {
+      "epoch": 0.1737242128121607,
+      "grad_norm": 0.7183733582496643,
+      "learning_rate": 0.0002,
+      "loss": 1.7861,
+      "step": 240
+    },
+    {
+      "epoch": 0.18096272167933405,
+      "grad_norm": 0.7082605957984924,
+      "learning_rate": 0.0002,
+      "loss": 1.6693,
+      "step": 250
+    },
+    {
+      "epoch": 0.18820123054650742,
+      "grad_norm": 0.9386326670646667,
+      "learning_rate": 0.0002,
+      "loss": 1.619,
+      "step": 260
+    },
+    {
+      "epoch": 0.19543973941368079,
+      "grad_norm": 0.7332451939582825,
+      "learning_rate": 0.0002,
+      "loss": 1.6511,
+      "step": 270
+    },
+    {
+      "epoch": 0.20267824828085415,
+      "grad_norm": 0.7092869877815247,
+      "learning_rate": 0.0002,
+      "loss": 1.6353,
+      "step": 280
+    },
+    {
+      "epoch": 0.20991675714802752,
+      "grad_norm": 0.7256413698196411,
+      "learning_rate": 0.0002,
+      "loss": 1.5996,
+      "step": 290
+    },
+    {
+      "epoch": 0.21715526601520088,
+      "grad_norm": 0.6398681402206421,
+      "learning_rate": 0.0002,
+      "loss": 1.6754,
+      "step": 300
+    },
+    {
+      "epoch": 0.22439377488237422,
+      "grad_norm": 0.6273287534713745,
+      "learning_rate": 0.0002,
+      "loss": 1.397,
+      "step": 310
+    },
+    {
+      "epoch": 0.23163228374954759,
+      "grad_norm": 0.511648416519165,
+      "learning_rate": 0.0002,
+      "loss": 1.5115,
+      "step": 320
+    },
+    {
+      "epoch": 0.23887079261672095,
+      "grad_norm": 0.8677352070808411,
+      "learning_rate": 0.0002,
+      "loss": 1.5424,
+      "step": 330
+    },
+    {
+      "epoch": 0.24610930148389432,
+      "grad_norm": 0.6270743012428284,
+      "learning_rate": 0.0002,
+      "loss": 1.6779,
+      "step": 340
+    },
+    {
+      "epoch": 0.2533478103510677,
+      "grad_norm": 0.7980281114578247,
+      "learning_rate": 0.0002,
+      "loss": 1.626,
+      "step": 350
+    },
+    {
+      "epoch": 0.26058631921824105,
+      "grad_norm": 0.632486879825592,
+      "learning_rate": 0.0002,
+      "loss": 1.5238,
+      "step": 360
+    },
+    {
+      "epoch": 0.2678248280854144,
+      "grad_norm": 0.6527034640312195,
+      "learning_rate": 0.0002,
+      "loss": 1.5175,
+      "step": 370
+    },
+    {
+      "epoch": 0.2750633369525878,
+      "grad_norm": 0.7672118544578552,
+      "learning_rate": 0.0002,
+      "loss": 1.627,
+      "step": 380
+    },
+    {
+      "epoch": 0.28230184581976114,
+      "grad_norm": 0.6035117506980896,
+      "learning_rate": 0.0002,
+      "loss": 1.5605,
+      "step": 390
+    },
+    {
+      "epoch": 0.2895403546869345,
+      "grad_norm": 0.5955103039741516,
+      "learning_rate": 0.0002,
+      "loss": 1.4603,
+      "step": 400
+    },
+    {
+      "epoch": 0.2967788635541079,
+      "grad_norm": 0.6015191674232483,
+      "learning_rate": 0.0002,
+      "loss": 1.558,
+      "step": 410
+    },
+    {
+      "epoch": 0.30401737242128124,
+      "grad_norm": 0.6380982398986816,
+      "learning_rate": 0.0002,
+      "loss": 1.6091,
+      "step": 420
+    },
+    {
+      "epoch": 0.3112558812884546,
+      "grad_norm": 0.6707863211631775,
+      "learning_rate": 0.0002,
+      "loss": 1.5292,
+      "step": 430
+    },
+    {
+      "epoch": 0.3184943901556279,
+      "grad_norm": 0.7010176777839661,
+      "learning_rate": 0.0002,
+      "loss": 1.4426,
+      "step": 440
+    },
+    {
+      "epoch": 0.3257328990228013,
+      "grad_norm": 0.8263739943504333,
+      "learning_rate": 0.0002,
+      "loss": 1.5572,
+      "step": 450
+    },
+    {
+      "epoch": 0.33297140788997465,
+      "grad_norm": 0.7253276109695435,
+      "learning_rate": 0.0002,
+      "loss": 1.5188,
+      "step": 460
+    },
+    {
+      "epoch": 0.340209916757148,
+      "grad_norm": 0.5238934755325317,
+      "learning_rate": 0.0002,
+      "loss": 1.584,
+      "step": 470
+    },
+    {
+      "epoch": 0.3474484256243214,
+      "grad_norm": 0.7869495749473572,
+      "learning_rate": 0.0002,
+      "loss": 1.7035,
+      "step": 480
+    },
+    {
+      "epoch": 0.35468693449149474,
+      "grad_norm": 0.7485215663909912,
+      "learning_rate": 0.0002,
+      "loss": 1.5776,
+      "step": 490
+    },
+    {
+      "epoch": 0.3619254433586681,
+      "grad_norm": 0.5413193106651306,
+      "learning_rate": 0.0002,
+      "loss": 1.6274,
+      "step": 500
+    },
+    {
+      "epoch": 0.3691639522258415,
+      "grad_norm": 0.7615048885345459,
+      "learning_rate": 0.0002,
+      "loss": 1.7323,
+      "step": 510
+    },
+    {
+      "epoch": 0.37640246109301484,
+      "grad_norm": 0.7685340046882629,
+      "learning_rate": 0.0002,
+      "loss": 1.532,
+      "step": 520
+    },
+    {
+      "epoch": 0.3836409699601882,
+      "grad_norm": 0.6379081010818481,
+      "learning_rate": 0.0002,
+      "loss": 1.6312,
+      "step": 530
+    },
+    {
+      "epoch": 0.39087947882736157,
+      "grad_norm": 0.7946939468383789,
+      "learning_rate": 0.0002,
+      "loss": 1.5645,
+      "step": 540
+    },
+    {
+      "epoch": 0.39811798769453494,
+      "grad_norm": 0.6287278532981873,
+      "learning_rate": 0.0002,
+      "loss": 1.4001,
+      "step": 550
+    },
+    {
+      "epoch": 0.4053564965617083,
+      "grad_norm": 0.6811642646789551,
+      "learning_rate": 0.0002,
+      "loss": 1.5982,
+      "step": 560
+    },
+    {
+      "epoch": 0.41259500542888167,
+      "grad_norm": 0.671073317527771,
+      "learning_rate": 0.0002,
+      "loss": 1.4953,
+      "step": 570
+    },
+    {
+      "epoch": 0.41983351429605503,
+      "grad_norm": 0.6313900351524353,
+      "learning_rate": 0.0002,
+      "loss": 1.6753,
+      "step": 580
+    },
+    {
+      "epoch": 0.4270720231632284,
+      "grad_norm": 0.5291772484779358,
+      "learning_rate": 0.0002,
+      "loss": 1.546,
+      "step": 590
+    },
+    {
+      "epoch": 0.43431053203040176,
+      "grad_norm": 0.62503582239151,
+      "learning_rate": 0.0002,
+      "loss": 1.5441,
+      "step": 600
+    },
+    {
+      "epoch": 0.4415490408975751,
+      "grad_norm": 0.5777305364608765,
+      "learning_rate": 0.0002,
+      "loss": 1.6276,
+      "step": 610
+    },
+    {
+      "epoch": 0.44878754976474844,
+      "grad_norm": 0.7013497352600098,
+      "learning_rate": 0.0002,
+      "loss": 1.4758,
+      "step": 620
+    },
+    {
+      "epoch": 0.4560260586319218,
+      "grad_norm": 0.8044822216033936,
+      "learning_rate": 0.0002,
+      "loss": 1.4029,
+      "step": 630
+    },
+    {
+      "epoch": 0.46326456749909517,
+      "grad_norm": 0.672531247138977,
+      "learning_rate": 0.0002,
+      "loss": 1.7195,
+      "step": 640
+    },
+    {
+      "epoch": 0.47050307636626854,
+      "grad_norm": 0.6233910322189331,
+      "learning_rate": 0.0002,
+      "loss": 1.614,
+      "step": 650
+    },
+    {
+      "epoch": 0.4777415852334419,
+      "grad_norm": 0.651524543762207,
+      "learning_rate": 0.0002,
+      "loss": 1.6041,
+      "step": 660
+    },
+    {
+      "epoch": 0.48498009410061527,
+      "grad_norm": 0.7213939428329468,
+      "learning_rate": 0.0002,
+      "loss": 1.5842,
+      "step": 670
+    },
+    {
+      "epoch": 0.49221860296778863,
+      "grad_norm": 0.6541454792022705,
+      "learning_rate": 0.0002,
+      "loss": 1.5453,
+      "step": 680
+    },
+    {
+      "epoch": 0.499457111834962,
+      "grad_norm": 0.6568936109542847,
+      "learning_rate": 0.0002,
+      "loss": 1.662,
+      "step": 690
+    },
+    {
+      "epoch": 0.5066956207021354,
+      "grad_norm": 0.7176415324211121,
+      "learning_rate": 0.0002,
+      "loss": 1.624,
+      "step": 700
+    },
+    {
+      "epoch": 0.5139341295693087,
+      "grad_norm": 0.6553855538368225,
+      "learning_rate": 0.0002,
+      "loss": 1.6099,
+      "step": 710
+    },
+    {
+      "epoch": 0.5211726384364821,
+      "grad_norm": 0.5654335618019104,
+      "learning_rate": 0.0002,
+      "loss": 1.5508,
+      "step": 720
+    },
+    {
+      "epoch": 0.5284111473036555,
+      "grad_norm": 0.5671001672744751,
+      "learning_rate": 0.0002,
+      "loss": 1.392,
+      "step": 730
+    },
+    {
+      "epoch": 0.5356496561708288,
+      "grad_norm": 0.7914412021636963,
+      "learning_rate": 0.0002,
+      "loss": 1.388,
+      "step": 740
+    },
+    {
+      "epoch": 0.5428881650380022,
+      "grad_norm": 0.6172138452529907,
+      "learning_rate": 0.0002,
+      "loss": 1.5931,
+      "step": 750
+    },
+    {
+      "epoch": 0.5501266739051756,
+      "grad_norm": 0.6132623553276062,
+      "learning_rate": 0.0002,
+      "loss": 1.4018,
+      "step": 760
+    },
+    {
+      "epoch": 0.5573651827723489,
+      "grad_norm": 0.654000461101532,
+      "learning_rate": 0.0002,
+      "loss": 1.513,
+      "step": 770
+    },
+    {
+      "epoch": 0.5646036916395223,
+      "grad_norm": 0.5691370964050293,
+      "learning_rate": 0.0002,
+      "loss": 1.5035,
+      "step": 780
+    },
+    {
+      "epoch": 0.5718422005066957,
+      "grad_norm": 0.7922580242156982,
+      "learning_rate": 0.0002,
+      "loss": 1.65,
+      "step": 790
+    },
+    {
+      "epoch": 0.579080709373869,
+      "grad_norm": 0.6831880211830139,
+      "learning_rate": 0.0002,
+      "loss": 1.4521,
+      "step": 800
+    },
+    {
+      "epoch": 0.5863192182410424,
+      "grad_norm": 0.6740124821662903,
+      "learning_rate": 0.0002,
+      "loss": 1.4734,
+      "step": 810
+    },
+    {
+      "epoch": 0.5935577271082157,
+      "grad_norm": 1.380016803741455,
+      "learning_rate": 0.0002,
+      "loss": 1.6498,
+      "step": 820
+    },
+    {
+      "epoch": 0.6007962359753891,
+      "grad_norm": 0.6552878022193909,
+      "learning_rate": 0.0002,
+      "loss": 1.4642,
+      "step": 830
+    },
+    {
+      "epoch": 0.6080347448425625,
+      "grad_norm": 0.6649535298347473,
+      "learning_rate": 0.0002,
+      "loss": 1.6271,
+      "step": 840
+    },
+    {
+      "epoch": 0.6152732537097358,
+      "grad_norm": 0.561738133430481,
+      "learning_rate": 0.0002,
+      "loss": 1.5886,
+      "step": 850
+    },
+    {
+      "epoch": 0.6225117625769092,
+      "grad_norm": 0.6133047938346863,
+      "learning_rate": 0.0002,
+      "loss": 1.5364,
+      "step": 860
+    },
+    {
+      "epoch": 0.6297502714440825,
+      "grad_norm": 0.559843122959137,
+      "learning_rate": 0.0002,
+      "loss": 1.3489,
+      "step": 870
+    },
+    {
+      "epoch": 0.6369887803112558,
+      "grad_norm": 0.6117811799049377,
+      "learning_rate": 0.0002,
+      "loss": 1.4878,
+      "step": 880
+    },
+    {
+      "epoch": 0.6442272891784292,
+      "grad_norm": 0.6209776401519775,
+      "learning_rate": 0.0002,
+      "loss": 1.56,
+      "step": 890
+    },
+    {
+      "epoch": 0.6514657980456026,
+      "grad_norm": 0.6234082579612732,
+      "learning_rate": 0.0002,
+      "loss": 1.6747,
+      "step": 900
+    },
+    {
+      "epoch": 0.6587043069127759,
+      "grad_norm": 0.7623258233070374,
+      "learning_rate": 0.0002,
+      "loss": 1.6963,
+      "step": 910
+    },
+    {
+      "epoch": 0.6659428157799493,
+      "grad_norm": 0.6148061752319336,
+      "learning_rate": 0.0002,
+      "loss": 1.2424,
+      "step": 920
+    },
+    {
+      "epoch": 0.6731813246471227,
+      "grad_norm": 0.6682973504066467,
+      "learning_rate": 0.0002,
+      "loss": 1.4319,
+      "step": 930
+    },
+    {
+      "epoch": 0.680419833514296,
+      "grad_norm": 0.5513041615486145,
+      "learning_rate": 0.0002,
+      "loss": 1.5377,
+      "step": 940
+    },
+    {
+      "epoch": 0.6876583423814694,
+      "grad_norm": 0.5197525024414062,
+      "learning_rate": 0.0002,
+      "loss": 1.3991,
+      "step": 950
+    },
+    {
+      "epoch": 0.6948968512486428,
+      "grad_norm": 0.6490758061408997,
+      "learning_rate": 0.0002,
+      "loss": 1.4398,
+      "step": 960
+    },
+    {
+      "epoch": 0.7021353601158161,
+      "grad_norm": 0.6450682878494263,
+      "learning_rate": 0.0002,
+      "loss": 1.5251,
+      "step": 970
+    },
+    {
+      "epoch": 0.7093738689829895,
+      "grad_norm": 0.6203766465187073,
+      "learning_rate": 0.0002,
+      "loss": 1.5417,
+      "step": 980
+    },
+    {
+      "epoch": 0.7166123778501629,
+      "grad_norm": 0.6023609638214111,
+      "learning_rate": 0.0002,
+      "loss": 1.4575,
+      "step": 990
+    },
+    {
+      "epoch": 0.7238508867173362,
+      "grad_norm": 0.5765255093574524,
+      "learning_rate": 0.0002,
+      "loss": 1.4973,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7310893955845096,
+      "grad_norm": 0.6650075316429138,
+      "learning_rate": 0.0002,
+      "loss": 1.483,
+      "step": 1010
+    },
+    {
+      "epoch": 0.738327904451683,
+      "grad_norm": 0.5610854029655457,
+      "learning_rate": 0.0002,
+      "loss": 1.5959,
+      "step": 1020
+    },
+    {
+      "epoch": 0.7455664133188563,
+      "grad_norm": 0.7072813510894775,
+      "learning_rate": 0.0002,
+      "loss": 1.5248,
+      "step": 1030
+    },
+    {
+      "epoch": 0.7528049221860297,
+      "grad_norm": 0.6815407872200012,
+      "learning_rate": 0.0002,
+      "loss": 1.5776,
+      "step": 1040
+    },
+    {
+      "epoch": 0.760043431053203,
+      "grad_norm": 0.7932390570640564,
+      "learning_rate": 0.0002,
+      "loss": 1.4577,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7672819399203764,
+      "grad_norm": 0.5798183083534241,
+      "learning_rate": 0.0002,
+      "loss": 1.4515,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7745204487875498,
+      "grad_norm": 0.7898504137992859,
+      "learning_rate": 0.0002,
+      "loss": 1.5053,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7817589576547231,
+      "grad_norm": 0.4983280301094055,
+      "learning_rate": 0.0002,
+      "loss": 1.4776,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7889974665218965,
+      "grad_norm": 0.691403329372406,
+      "learning_rate": 0.0002,
+      "loss": 1.5007,
+      "step": 1090
+    },
+    {
+      "epoch": 0.7962359753890699,
+      "grad_norm": 0.5394481420516968,
+      "learning_rate": 0.0002,
+      "loss": 1.5153,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8034744842562432,
+      "grad_norm": 0.5136822462081909,
+      "learning_rate": 0.0002,
+      "loss": 1.6892,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8107129931234166,
+      "grad_norm": 0.6828126907348633,
+      "learning_rate": 0.0002,
+      "loss": 1.4902,
+      "step": 1120
+    },
+    {
+      "epoch": 0.81795150199059,
+      "grad_norm": 0.6799656748771667,
+      "learning_rate": 0.0002,
+      "loss": 1.4346,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8251900108577633,
+      "grad_norm": 0.5428406000137329,
+      "learning_rate": 0.0002,
+      "loss": 1.2678,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8324285197249367,
+      "grad_norm": 0.4811290502548218,
+      "learning_rate": 0.0002,
+      "loss": 1.4072,
+      "step": 1150
+    },
+    {
+      "epoch": 0.8396670285921101,
+      "grad_norm": 0.5519434809684753,
+      "learning_rate": 0.0002,
+      "loss": 1.4512,
+      "step": 1160
+    },
+    {
+      "epoch": 0.8469055374592834,
+      "grad_norm": 0.9748060703277588,
+      "learning_rate": 0.0002,
+      "loss": 1.4072,
+      "step": 1170
+    },
+    {
+      "epoch": 0.8541440463264568,
+      "grad_norm": 0.712609589099884,
+      "learning_rate": 0.0002,
+      "loss": 1.4309,
+      "step": 1180
+    },
+    {
+      "epoch": 0.8613825551936302,
+      "grad_norm": 0.6866157054901123,
+      "learning_rate": 0.0002,
+      "loss": 1.434,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8686210640608035,
+      "grad_norm": 0.5068854093551636,
+      "learning_rate": 0.0002,
+      "loss": 1.3704,
+      "step": 1200
+    },
+    {
+      "epoch": 0.8758595729279768,
+      "grad_norm": 0.6333245038986206,
+      "learning_rate": 0.0002,
+      "loss": 1.5601,
+      "step": 1210
+    },
+    {
+      "epoch": 0.8830980817951501,
+      "grad_norm": 0.6424421072006226,
+      "learning_rate": 0.0002,
+      "loss": 1.4636,
+      "step": 1220
+    },
+    {
+      "epoch": 0.8903365906623235,
+      "grad_norm": 0.4771921932697296,
+      "learning_rate": 0.0002,
+      "loss": 1.4186,
+      "step": 1230
+    },
+    {
+      "epoch": 0.8975750995294969,
+      "grad_norm": 0.5191764235496521,
+      "learning_rate": 0.0002,
+      "loss": 1.6323,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9048136083966702,
+      "grad_norm": 0.756222128868103,
+      "learning_rate": 0.0002,
+      "loss": 1.6105,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9120521172638436,
+      "grad_norm": 0.623823881149292,
+      "learning_rate": 0.0002,
+      "loss": 1.4396,
+      "step": 1260
+    },
+    {
+      "epoch": 0.919290626131017,
+      "grad_norm": 0.8166571259498596,
+      "learning_rate": 0.0002,
+      "loss": 1.3097,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9265291349981903,
+      "grad_norm": 0.6059346795082092,
+      "learning_rate": 0.0002,
+      "loss": 1.4625,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9337676438653637,
+      "grad_norm": 0.5842690467834473,
+      "learning_rate": 0.0002,
+      "loss": 1.3555,
+      "step": 1290
+    },
+    {
+      "epoch": 0.9410061527325371,
+      "grad_norm": 0.7649800777435303,
+      "learning_rate": 0.0002,
+      "loss": 1.5859,
+      "step": 1300
+    },
+    {
+      "epoch": 0.9482446615997104,
+      "grad_norm": 0.6420919895172119,
+      "learning_rate": 0.0002,
+      "loss": 1.5915,
+      "step": 1310
+    },
+    {
+      "epoch": 0.9554831704668838,
+      "grad_norm": 0.7011452913284302,
+      "learning_rate": 0.0002,
+      "loss": 1.453,
+      "step": 1320
+    },
+    {
+      "epoch": 0.9627216793340572,
+      "grad_norm": 0.5783746242523193,
+      "learning_rate": 0.0002,
+      "loss": 1.6766,
+      "step": 1330
+    },
+    {
+      "epoch": 0.9699601882012305,
+      "grad_norm": 0.5973192453384399,
+      "learning_rate": 0.0002,
+      "loss": 1.6308,
+      "step": 1340
+    },
+    {
+      "epoch": 0.9771986970684039,
+      "grad_norm": 0.6181833744049072,
+      "learning_rate": 0.0002,
+      "loss": 1.5901,
+      "step": 1350
+    },
+    {
+      "epoch": 0.9844372059355773,
+      "grad_norm": 0.5563396215438843,
+      "learning_rate": 0.0002,
+      "loss": 1.5258,
+      "step": 1360
+    },
+    {
+      "epoch": 0.9916757148027506,
+      "grad_norm": 0.45723360776901245,
+      "learning_rate": 0.0002,
+      "loss": 1.4508,
+      "step": 1370
+    },
+    {
+      "epoch": 0.998914223669924,
+      "grad_norm": 0.5947498679161072,
+      "learning_rate": 0.0002,
+      "loss": 1.3291,
+      "step": 1380
+    },
+    {
+      "epoch": 0.9996380745566413,
+      "eval_loss": 1.480796456336975,
+      "eval_runtime": 27.3103,
+      "eval_samples_per_second": 15.965,
+      "eval_steps_per_second": 2.014,
+      "step": 1381
+    },
+    {
+      "epoch": 1.0061527325370974,
+      "grad_norm": 0.5599952936172485,
+      "learning_rate": 0.0002,
+      "loss": 1.3057,
+      "step": 1390
+    },
+    {
+      "epoch": 1.0133912414042707,
+      "grad_norm": 0.5932008028030396,
+      "learning_rate": 0.0002,
+      "loss": 1.4991,
+      "step": 1400
+    },
+    {
+      "epoch": 1.020629750271444,
+      "grad_norm": 0.6194121837615967,
+      "learning_rate": 0.0002,
+      "loss": 1.4506,
+      "step": 1410
+    },
+    {
+      "epoch": 1.0278682591386175,
+      "grad_norm": 0.6995621919631958,
+      "learning_rate": 0.0002,
+      "loss": 1.5966,
+      "step": 1420
+    },
+    {
+      "epoch": 1.0351067680057908,
+      "grad_norm": 0.7905810475349426,
+      "learning_rate": 0.0002,
+      "loss": 1.4153,
+      "step": 1430
+    },
+    {
+      "epoch": 1.0423452768729642,
+      "grad_norm": 0.7221615314483643,
+      "learning_rate": 0.0002,
+      "loss": 1.4414,
+      "step": 1440
+    },
+    {
+      "epoch": 1.0495837857401376,
+      "grad_norm": 0.6170642375946045,
+      "learning_rate": 0.0002,
+      "loss": 1.3859,
+      "step": 1450
+    },
+    {
+      "epoch": 1.056822294607311,
+      "grad_norm": 0.5844094753265381,
+      "learning_rate": 0.0002,
+      "loss": 1.3806,
+      "step": 1460
+    },
+    {
+      "epoch": 1.0640608034744843,
+      "grad_norm": 0.7731822729110718,
+      "learning_rate": 0.0002,
+      "loss": 1.4871,
+      "step": 1470
+    },
+    {
+      "epoch": 1.0712993123416577,
+      "grad_norm": 0.4554748237133026,
+      "learning_rate": 0.0002,
+      "loss": 1.4286,
+      "step": 1480
+    },
+    {
+      "epoch": 1.078537821208831,
+      "grad_norm": 0.6923259496688843,
+      "learning_rate": 0.0002,
+      "loss": 1.3977,
+      "step": 1490
+    },
+    {
+      "epoch": 1.0857763300760044,
+      "grad_norm": 0.6008219122886658,
+      "learning_rate": 0.0002,
+      "loss": 1.3936,
+      "step": 1500
+    },
+    {
+      "epoch": 1.0930148389431777,
+      "grad_norm": 0.6450045704841614,
+      "learning_rate": 0.0002,
+      "loss": 1.4821,
+      "step": 1510
+    },
+    {
+      "epoch": 1.1002533478103511,
+      "grad_norm": 0.7833753824234009,
+      "learning_rate": 0.0002,
+      "loss": 1.3295,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1074918566775245,
+      "grad_norm": 0.5076758861541748,
+      "learning_rate": 0.0002,
+      "loss": 1.3424,
+      "step": 1530
+    },
+    {
+      "epoch": 1.1147303655446978,
+      "grad_norm": 0.5661332011222839,
+      "learning_rate": 0.0002,
+      "loss": 1.4043,
+      "step": 1540
+    },
+    {
+      "epoch": 1.1219688744118712,
+      "grad_norm": 0.6526919603347778,
+      "learning_rate": 0.0002,
+      "loss": 1.4963,
+      "step": 1550
+    },
+    {
+      "epoch": 1.1292073832790446,
+      "grad_norm": 0.5613082647323608,
+      "learning_rate": 0.0002,
+      "loss": 1.3671,
+      "step": 1560
+    },
+    {
+      "epoch": 1.136445892146218,
+      "grad_norm": 0.6113885641098022,
+      "learning_rate": 0.0002,
+      "loss": 1.4458,
+      "step": 1570
+    },
+    {
+      "epoch": 1.1436844010133913,
+      "grad_norm": 0.6732510328292847,
+      "learning_rate": 0.0002,
+      "loss": 1.3552,
+      "step": 1580
+    },
+    {
+      "epoch": 1.1509229098805647,
+      "grad_norm": 0.6146392226219177,
+      "learning_rate": 0.0002,
+      "loss": 1.3114,
+      "step": 1590
+    },
+    {
+      "epoch": 1.158161418747738,
+      "grad_norm": 0.6766974329948425,
+      "learning_rate": 0.0002,
+      "loss": 1.411,
+      "step": 1600
+    },
+    {
+      "epoch": 1.1653999276149114,
+      "grad_norm": 0.7621957659721375,
+      "learning_rate": 0.0002,
+      "loss": 1.2401,
+      "step": 1610
+    },
+    {
+      "epoch": 1.1726384364820848,
+      "grad_norm": 0.6959581971168518,
+      "learning_rate": 0.0002,
+      "loss": 1.3758,
+      "step": 1620
+    },
+    {
+      "epoch": 1.1798769453492581,
+      "grad_norm": 0.6691278219223022,
+      "learning_rate": 0.0002,
+      "loss": 1.382,
+      "step": 1630
+    },
+    {
+      "epoch": 1.1871154542164315,
+      "grad_norm": 0.4927774965763092,
+      "learning_rate": 0.0002,
+      "loss": 1.4147,
+      "step": 1640
+    },
+    {
+      "epoch": 1.1943539630836049,
+      "grad_norm": 0.7724234461784363,
+      "learning_rate": 0.0002,
+      "loss": 1.449,
+      "step": 1650
+    },
+    {
+      "epoch": 1.2015924719507782,
+      "grad_norm": 0.6817787885665894,
+      "learning_rate": 0.0002,
+      "loss": 1.4778,
+      "step": 1660
+    },
+    {
+      "epoch": 1.2088309808179516,
+      "grad_norm": 0.6500699520111084,
+      "learning_rate": 0.0002,
+      "loss": 1.3776,
+      "step": 1670
+    },
+    {
+      "epoch": 1.216069489685125,
+      "grad_norm": 0.5703568458557129,
+      "learning_rate": 0.0002,
+      "loss": 1.3875,
+      "step": 1680
+    },
+    {
+      "epoch": 1.2233079985522983,
+      "grad_norm": 0.6261579990386963,
+      "learning_rate": 0.0002,
+      "loss": 1.4735,
+      "step": 1690
+    },
+    {
+      "epoch": 1.2305465074194717,
+      "grad_norm": 0.651713490486145,
+      "learning_rate": 0.0002,
+      "loss": 1.3898,
+      "step": 1700
+    },
+    {
+      "epoch": 1.237785016286645,
+      "grad_norm": 0.684399425983429,
+      "learning_rate": 0.0002,
+      "loss": 1.4002,
+      "step": 1710
+    },
+    {
+      "epoch": 1.2450235251538184,
+      "grad_norm": 0.6996857523918152,
+      "learning_rate": 0.0002,
+      "loss": 1.5027,
+      "step": 1720
+    },
+    {
+      "epoch": 1.2522620340209918,
+      "grad_norm": 0.7102537751197815,
+      "learning_rate": 0.0002,
+      "loss": 1.3326,
+      "step": 1730
+    },
+    {
+      "epoch": 1.2595005428881652,
+      "grad_norm": 0.45809897780418396,
+      "learning_rate": 0.0002,
+      "loss": 1.3675,
+      "step": 1740
+    },
+    {
+      "epoch": 1.2667390517553385,
+      "grad_norm": 0.6377046704292297,
+      "learning_rate": 0.0002,
+      "loss": 1.4175,
+      "step": 1750
+    },
+    {
+      "epoch": 1.2739775606225119,
+      "grad_norm": 0.6965704560279846,
+      "learning_rate": 0.0002,
+      "loss": 1.3479,
+      "step": 1760
+    },
+    {
+      "epoch": 1.2812160694896852,
+      "grad_norm": 0.5688214302062988,
+      "learning_rate": 0.0002,
+      "loss": 1.5647,
+      "step": 1770
+    },
+    {
+      "epoch": 1.2884545783568586,
+      "grad_norm": 0.6384190320968628,
+      "learning_rate": 0.0002,
+      "loss": 1.3967,
+      "step": 1780
+    },
+    {
+      "epoch": 1.295693087224032,
+      "grad_norm": 0.5629363656044006,
+      "learning_rate": 0.0002,
+      "loss": 1.3671,
+      "step": 1790
+    },
+    {
+      "epoch": 1.3029315960912053,
+      "grad_norm": 0.6148255467414856,
+      "learning_rate": 0.0002,
+      "loss": 1.2292,
+      "step": 1800
+    },
+    {
+      "epoch": 1.3101701049583787,
+      "grad_norm": 0.655580997467041,
+      "learning_rate": 0.0002,
+      "loss": 1.5806,
+      "step": 1810
+    },
+    {
+      "epoch": 1.3174086138255519,
+      "grad_norm": 0.5642657279968262,
+      "learning_rate": 0.0002,
+      "loss": 1.2398,
+      "step": 1820
+    },
+    {
+      "epoch": 1.3246471226927252,
+      "grad_norm": 0.59607994556427,
+      "learning_rate": 0.0002,
+      "loss": 1.3246,
+      "step": 1830
+    },
+    {
+      "epoch": 1.3318856315598986,
+      "grad_norm": 0.5564199090003967,
+      "learning_rate": 0.0002,
+      "loss": 1.3274,
+      "step": 1840
+    },
+    {
+      "epoch": 1.339124140427072,
+      "grad_norm": 0.6949955821037292,
+      "learning_rate": 0.0002,
+      "loss": 1.5834,
+      "step": 1850
+    },
+    {
+      "epoch": 1.3463626492942453,
+      "grad_norm": 0.7036856412887573,
+      "learning_rate": 0.0002,
+      "loss": 1.4722,
+      "step": 1860
+    },
+    {
+      "epoch": 1.3536011581614187,
+      "grad_norm": 0.722062885761261,
+      "learning_rate": 0.0002,
+      "loss": 1.333,
+      "step": 1870
+    },
+    {
+      "epoch": 1.360839667028592,
+      "grad_norm": 0.6098677515983582,
+      "learning_rate": 0.0002,
+      "loss": 1.4044,
+      "step": 1880
+    },
+    {
+      "epoch": 1.3680781758957654,
+      "grad_norm": 0.5376402735710144,
+      "learning_rate": 0.0002,
+      "loss": 1.6217,
+      "step": 1890
+    },
+    {
+      "epoch": 1.3753166847629388,
+      "grad_norm": 0.6974610090255737,
+      "learning_rate": 0.0002,
+      "loss": 1.5071,
+      "step": 1900
+    },
+    {
+      "epoch": 1.3825551936301121,
+      "grad_norm": 0.6520763635635376,
+      "learning_rate": 0.0002,
+      "loss": 1.5854,
+      "step": 1910
+    },
+    {
+      "epoch": 1.3897937024972855,
+      "grad_norm": 0.6604374647140503,
+      "learning_rate": 0.0002,
+      "loss": 1.4271,
+      "step": 1920
+    },
+    {
+      "epoch": 1.3970322113644589,
+      "grad_norm": 0.7364398241043091,
+      "learning_rate": 0.0002,
+      "loss": 1.419,
+      "step": 1930
+    },
+    {
+      "epoch": 1.4042707202316322,
+      "grad_norm": 0.6849475502967834,
+      "learning_rate": 0.0002,
+      "loss": 1.4585,
+      "step": 1940
+    },
+    {
+      "epoch": 1.4115092290988056,
+      "grad_norm": 0.6562670469284058,
+      "learning_rate": 0.0002,
+      "loss": 1.5577,
+      "step": 1950
+    },
+    {
+      "epoch": 1.418747737965979,
+      "grad_norm": 0.5695616006851196,
+      "learning_rate": 0.0002,
+      "loss": 1.4725,
+      "step": 1960
+    },
+    {
+      "epoch": 1.4259862468331523,
+      "grad_norm": 0.5244464874267578,
+      "learning_rate": 0.0002,
+      "loss": 1.3088,
+      "step": 1970
+    },
+    {
+      "epoch": 1.4332247557003257,
+      "grad_norm": 0.6347293257713318,
+      "learning_rate": 0.0002,
+      "loss": 1.5069,
+      "step": 1980
+    },
+    {
+      "epoch": 1.440463264567499,
+      "grad_norm": 0.5528361201286316,
+      "learning_rate": 0.0002,
+      "loss": 1.3502,
+      "step": 1990
+    },
+    {
+      "epoch": 1.4477017734346724,
+      "grad_norm": 0.6987585425376892,
+      "learning_rate": 0.0002,
+      "loss": 1.3978,
+      "step": 2000
+    },
+    {
+      "epoch": 1.4549402823018458,
+      "grad_norm": 0.6568987369537354,
+      "learning_rate": 0.0002,
+      "loss": 1.4262,
+      "step": 2010
+    },
+    {
+      "epoch": 1.4621787911690192,
+      "grad_norm": 0.7665994763374329,
+      "learning_rate": 0.0002,
+      "loss": 1.4175,
+      "step": 2020
+    },
+    {
+      "epoch": 1.4694173000361925,
+      "grad_norm": 0.5127707123756409,
+      "learning_rate": 0.0002,
+      "loss": 1.244,
+      "step": 2030
+    },
+    {
+      "epoch": 1.476655808903366,
+      "grad_norm": 0.5406824946403503,
+      "learning_rate": 0.0002,
+      "loss": 1.3699,
+      "step": 2040
+    },
+    {
+      "epoch": 1.4838943177705393,
+      "grad_norm": 0.5990166664123535,
+      "learning_rate": 0.0002,
+      "loss": 1.3353,
+      "step": 2050
+    },
+    {
+      "epoch": 1.4911328266377126,
+      "grad_norm": 0.6186193823814392,
+      "learning_rate": 0.0002,
+      "loss": 1.2454,
+      "step": 2060
+    },
+    {
+      "epoch": 1.498371335504886,
+      "grad_norm": 0.6154307126998901,
+      "learning_rate": 0.0002,
+      "loss": 1.428,
+      "step": 2070
+    },
+    {
+      "epoch": 1.5056098443720594,
+      "grad_norm": 0.5606056451797485,
+      "learning_rate": 0.0002,
+      "loss": 1.4528,
+      "step": 2080
+    },
+    {
+      "epoch": 1.5128483532392327,
+      "grad_norm": 0.5006417036056519,
+      "learning_rate": 0.0002,
+      "loss": 1.2405,
+      "step": 2090
+    },
+    {
+      "epoch": 1.520086862106406,
+      "grad_norm": 0.5968486070632935,
+      "learning_rate": 0.0002,
+      "loss": 1.4258,
+      "step": 2100
+    },
+    {
+      "epoch": 1.5273253709735795,
+      "grad_norm": 0.5835496187210083,
+      "learning_rate": 0.0002,
+      "loss": 1.2752,
+      "step": 2110
+    },
+    {
+      "epoch": 1.5345638798407528,
+      "grad_norm": 0.6753535270690918,
+      "learning_rate": 0.0002,
+      "loss": 1.5443,
+      "step": 2120
+    },
+    {
+      "epoch": 1.5418023887079262,
+      "grad_norm": 0.7299720644950867,
+      "learning_rate": 0.0002,
+      "loss": 1.2139,
+      "step": 2130
+    },
+    {
+      "epoch": 1.5490408975750996,
+      "grad_norm": 0.5105988383293152,
+      "learning_rate": 0.0002,
+      "loss": 1.2364,
+      "step": 2140
+    },
+    {
+      "epoch": 1.556279406442273,
+      "grad_norm": 0.5675431489944458,
+      "learning_rate": 0.0002,
+      "loss": 1.4528,
+      "step": 2150
+    },
+    {
+      "epoch": 1.5635179153094463,
+      "grad_norm": 0.6246723532676697,
+      "learning_rate": 0.0002,
+      "loss": 1.4563,
+      "step": 2160
+    },
+    {
+      "epoch": 1.5707564241766196,
+      "grad_norm": 0.7291720509529114,
+      "learning_rate": 0.0002,
+      "loss": 1.5255,
+      "step": 2170
+    },
+    {
+      "epoch": 1.577994933043793,
+      "grad_norm": 0.678114116191864,
+      "learning_rate": 0.0002,
+      "loss": 1.5432,
+      "step": 2180
+    },
+    {
+      "epoch": 1.5852334419109664,
+      "grad_norm": 0.5136260986328125,
+      "learning_rate": 0.0002,
+      "loss": 1.5212,
+      "step": 2190
+    },
+    {
+      "epoch": 1.5924719507781397,
+      "grad_norm": 0.6359935998916626,
+      "learning_rate": 0.0002,
+      "loss": 1.3271,
+      "step": 2200
+    },
+    {
+      "epoch": 1.599710459645313,
+      "grad_norm": 0.7650278806686401,
+      "learning_rate": 0.0002,
+      "loss": 1.4038,
+      "step": 2210
+    },
+    {
+      "epoch": 1.6069489685124865,
+      "grad_norm": 0.7256110906600952,
+      "learning_rate": 0.0002,
+      "loss": 1.5478,
+      "step": 2220
+    },
+    {
+      "epoch": 1.6141874773796598,
+      "grad_norm": 0.688689649105072,
+      "learning_rate": 0.0002,
+      "loss": 1.4387,
+      "step": 2230
+    },
+    {
+      "epoch": 1.6214259862468332,
+      "grad_norm": 0.6045311093330383,
+      "learning_rate": 0.0002,
+      "loss": 1.4096,
+      "step": 2240
+    },
+    {
+      "epoch": 1.6286644951140063,
+      "grad_norm": 0.7064604163169861,
+      "learning_rate": 0.0002,
+      "loss": 1.4097,
+      "step": 2250
+    },
+    {
+      "epoch": 1.6359030039811797,
+      "grad_norm": 0.5309562087059021,
+      "learning_rate": 0.0002,
+      "loss": 1.3477,
+      "step": 2260
+    },
+    {
+      "epoch": 1.643141512848353,
+      "grad_norm": 0.5687053203582764,
+      "learning_rate": 0.0002,
+      "loss": 1.4022,
+      "step": 2270
+    },
+    {
+      "epoch": 1.6503800217155264,
+      "grad_norm": 0.535872757434845,
+      "learning_rate": 0.0002,
+      "loss": 1.2977,
+      "step": 2280
+    },
+    {
+      "epoch": 1.6576185305826998,
+      "grad_norm": 0.5502381920814514,
+      "learning_rate": 0.0002,
+      "loss": 1.3844,
+      "step": 2290
+    },
+    {
+      "epoch": 1.6648570394498732,
+      "grad_norm": 0.6158602237701416,
+      "learning_rate": 0.0002,
+      "loss": 1.3764,
+      "step": 2300
+    },
+    {
+      "epoch": 1.6720955483170465,
+      "grad_norm": 0.5804675817489624,
+      "learning_rate": 0.0002,
+      "loss": 1.3515,
+      "step": 2310
+    },
+    {
+      "epoch": 1.67933405718422,
+      "grad_norm": 0.600742757320404,
+      "learning_rate": 0.0002,
+      "loss": 1.2532,
+      "step": 2320
+    },
+    {
+      "epoch": 1.6865725660513933,
+      "grad_norm": 0.7101941108703613,
+      "learning_rate": 0.0002,
+      "loss": 1.477,
+      "step": 2330
+    },
+    {
+      "epoch": 1.6938110749185666,
+      "grad_norm": 0.7507809996604919,
+      "learning_rate": 0.0002,
+      "loss": 1.4849,
+      "step": 2340
+    },
+    {
+      "epoch": 1.70104958378574,
+      "grad_norm": 0.768502414226532,
+      "learning_rate": 0.0002,
+      "loss": 1.2703,
+      "step": 2350
+    },
+    {
+      "epoch": 1.7082880926529134,
+      "grad_norm": 0.4801851212978363,
+      "learning_rate": 0.0002,
+      "loss": 1.3332,
+      "step": 2360
+    },
+    {
+      "epoch": 1.7155266015200867,
+      "grad_norm": 0.5322122573852539,
+      "learning_rate": 0.0002,
+      "loss": 1.4158,
+      "step": 2370
+    },
+    {
+      "epoch": 1.72276511038726,
+      "grad_norm": 0.587661862373352,
+      "learning_rate": 0.0002,
+      "loss": 1.4136,
+      "step": 2380
+    },
+    {
+      "epoch": 1.7300036192544335,
+      "grad_norm": 0.6073525547981262,
+      "learning_rate": 0.0002,
+      "loss": 1.3771,
+      "step": 2390
+    },
+    {
+      "epoch": 1.7372421281216068,
+      "grad_norm": 0.6950460076332092,
+      "learning_rate": 0.0002,
+      "loss": 1.2754,
+      "step": 2400
+    },
+    {
+      "epoch": 1.7444806369887802,
+      "grad_norm": 0.5981102585792542,
+      "learning_rate": 0.0002,
+      "loss": 1.3858,
+      "step": 2410
+    },
+    {
+      "epoch": 1.7517191458559536,
+      "grad_norm": 0.544570803642273,
+      "learning_rate": 0.0002,
+      "loss": 1.4075,
+      "step": 2420
+    },
+    {
+      "epoch": 1.758957654723127,
+      "grad_norm": 0.5304399728775024,
+      "learning_rate": 0.0002,
+      "loss": 1.3861,
+      "step": 2430
+    },
+    {
+      "epoch": 1.7661961635903003,
+      "grad_norm": 0.7921594977378845,
+      "learning_rate": 0.0002,
+      "loss": 1.4244,
+      "step": 2440
+    },
+    {
+      "epoch": 1.7734346724574737,
+      "grad_norm": 0.6084808707237244,
+      "learning_rate": 0.0002,
+      "loss": 1.3053,
+      "step": 2450
+    },
+    {
+      "epoch": 1.780673181324647,
+      "grad_norm": 0.8844701051712036,
+      "learning_rate": 0.0002,
+      "loss": 1.3781,
+      "step": 2460
+    },
+    {
+      "epoch": 1.7879116901918204,
+      "grad_norm": 0.5729258060455322,
+      "learning_rate": 0.0002,
+      "loss": 1.3227,
+      "step": 2470
+    },
+    {
+      "epoch": 1.7951501990589938,
+      "grad_norm": 0.6303611993789673,
+      "learning_rate": 0.0002,
+      "loss": 1.3422,
+      "step": 2480
+    },
+    {
+      "epoch": 1.8023887079261671,
+      "grad_norm": 0.5627942085266113,
+      "learning_rate": 0.0002,
+      "loss": 1.3926,
+      "step": 2490
+    },
+    {
+      "epoch": 1.8096272167933405,
+      "grad_norm": 0.6724274158477783,
+      "learning_rate": 0.0002,
+      "loss": 1.3816,
+      "step": 2500
+    },
+    {
+      "epoch": 1.8168657256605139,
+      "grad_norm": 0.5030826330184937,
+      "learning_rate": 0.0002,
+      "loss": 1.2951,
+      "step": 2510
+    },
+    {
+      "epoch": 1.8241042345276872,
+      "grad_norm": 0.5504099130630493,
+      "learning_rate": 0.0002,
+      "loss": 1.2839,
+      "step": 2520
+    },
+    {
+      "epoch": 1.8313427433948606,
+      "grad_norm": 0.6338945627212524,
+      "learning_rate": 0.0002,
+      "loss": 1.4264,
+      "step": 2530
+    },
+    {
+      "epoch": 1.838581252262034,
+      "grad_norm": 0.5902037620544434,
+      "learning_rate": 0.0002,
+      "loss": 1.563,
+      "step": 2540
+    },
+    {
+      "epoch": 1.8458197611292073,
+      "grad_norm": 0.48814457654953003,
+      "learning_rate": 0.0002,
+      "loss": 1.2961,
+      "step": 2550
+    },
+    {
+      "epoch": 1.8530582699963807,
+      "grad_norm": 0.6216312646865845,
+      "learning_rate": 0.0002,
+      "loss": 1.466,
+      "step": 2560
+    },
+    {
+      "epoch": 1.860296778863554,
+      "grad_norm": 0.635603666305542,
+      "learning_rate": 0.0002,
+      "loss": 1.5123,
+      "step": 2570
+    },
+    {
+      "epoch": 1.8675352877307274,
+      "grad_norm": 0.6938216090202332,
+      "learning_rate": 0.0002,
+      "loss": 1.372,
+      "step": 2580
+    },
+    {
+      "epoch": 1.8747737965979008,
+      "grad_norm": 0.599557638168335,
+      "learning_rate": 0.0002,
+      "loss": 1.5011,
+      "step": 2590
+    },
+    {
+      "epoch": 1.8820123054650741,
+      "grad_norm": 0.564424455165863,
+      "learning_rate": 0.0002,
+      "loss": 1.2714,
+      "step": 2600
+    },
+    {
+      "epoch": 1.8892508143322475,
+      "grad_norm": 0.5430700182914734,
+      "learning_rate": 0.0002,
+      "loss": 1.3403,
+      "step": 2610
+    },
+    {
+      "epoch": 1.8964893231994209,
+      "grad_norm": 0.6150169372558594,
+      "learning_rate": 0.0002,
+      "loss": 1.4347,
+      "step": 2620
+    },
+    {
+      "epoch": 1.9037278320665942,
+      "grad_norm": 0.48159119486808777,
+      "learning_rate": 0.0002,
+      "loss": 1.2474,
+      "step": 2630
+    },
+    {
+      "epoch": 1.9109663409337676,
+      "grad_norm": 0.5608997941017151,
+      "learning_rate": 0.0002,
+      "loss": 1.3716,
+      "step": 2640
+    },
+    {
+      "epoch": 1.918204849800941,
+      "grad_norm": 0.6454501748085022,
+      "learning_rate": 0.0002,
+      "loss": 1.5787,
+      "step": 2650
+    },
+    {
+      "epoch": 1.9254433586681143,
+      "grad_norm": 0.5458073616027832,
+      "learning_rate": 0.0002,
+      "loss": 1.3238,
+      "step": 2660
+    },
+    {
+      "epoch": 1.9326818675352877,
+      "grad_norm": 0.5328490734100342,
+      "learning_rate": 0.0002,
+      "loss": 1.3208,
+      "step": 2670
+    },
+    {
+      "epoch": 1.939920376402461,
+      "grad_norm": 0.6444696187973022,
+      "learning_rate": 0.0002,
+      "loss": 1.4971,
+      "step": 2680
+    },
+    {
+      "epoch": 1.9471588852696344,
+      "grad_norm": 0.7126023769378662,
+      "learning_rate": 0.0002,
+      "loss": 1.5387,
+      "step": 2690
+    },
+    {
+      "epoch": 1.9543973941368078,
+      "grad_norm": 0.5164045095443726,
+      "learning_rate": 0.0002,
+      "loss": 1.3637,
+      "step": 2700
+    },
+    {
+      "epoch": 1.9616359030039812,
+      "grad_norm": 0.5347061157226562,
+      "learning_rate": 0.0002,
+      "loss": 1.5303,
+      "step": 2710
+    },
+    {
+      "epoch": 1.9688744118711545,
+      "grad_norm": 0.5297950506210327,
+      "learning_rate": 0.0002,
+      "loss": 1.2815,
+      "step": 2720
+    },
+    {
+      "epoch": 1.976112920738328,
+      "grad_norm": 0.6537790298461914,
+      "learning_rate": 0.0002,
+      "loss": 1.3566,
+      "step": 2730
+    },
+    {
+      "epoch": 1.9833514296055013,
+      "grad_norm": 0.5536222457885742,
+      "learning_rate": 0.0002,
+      "loss": 1.332,
+      "step": 2740
+    },
+    {
+      "epoch": 1.9905899384726746,
+      "grad_norm": 0.4856105446815491,
+      "learning_rate": 0.0002,
+      "loss": 1.3333,
+      "step": 2750
+    },
+    {
+      "epoch": 1.997828447339848,
+      "grad_norm": 0.6642730832099915,
+      "learning_rate": 0.0002,
+      "loss": 1.3521,
+      "step": 2760
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.4366681575775146,
+      "eval_runtime": 27.3729,
+      "eval_samples_per_second": 15.928,
+      "eval_steps_per_second": 2.009,
+      "step": 2763
+    },
+    {
+      "epoch": 2.0050669562070214,
+      "grad_norm": 0.740253210067749,
+      "learning_rate": 0.0002,
+      "loss": 1.4322,
+      "step": 2770
+    },
+    {
+      "epoch": 2.0123054650741947,
+      "grad_norm": 0.5826276540756226,
+      "learning_rate": 0.0002,
+      "loss": 1.277,
+      "step": 2780
+    },
+    {
+      "epoch": 2.019543973941368,
+      "grad_norm": 0.607356071472168,
+      "learning_rate": 0.0002,
+      "loss": 1.2424,
+      "step": 2790
+    },
+    {
+      "epoch": 2.0267824828085415,
+      "grad_norm": 0.5918063521385193,
+      "learning_rate": 0.0002,
+      "loss": 1.2601,
+      "step": 2800
+    },
+    {
+      "epoch": 2.034020991675715,
+      "grad_norm": 0.5610089898109436,
+      "learning_rate": 0.0002,
+      "loss": 1.3715,
+      "step": 2810
+    },
+    {
+      "epoch": 2.041259500542888,
+      "grad_norm": 0.5869926810264587,
+      "learning_rate": 0.0002,
+      "loss": 1.2092,
+      "step": 2820
+    },
+    {
+      "epoch": 2.0484980094100615,
+      "grad_norm": 0.5753467679023743,
+      "learning_rate": 0.0002,
+      "loss": 1.1929,
+      "step": 2830
+    },
+    {
+      "epoch": 2.055736518277235,
+      "grad_norm": 0.7096508145332336,
+      "learning_rate": 0.0002,
+      "loss": 1.333,
+      "step": 2840
+    },
+    {
+      "epoch": 2.0629750271444083,
+      "grad_norm": 0.7653635144233704,
+      "learning_rate": 0.0002,
+      "loss": 1.1766,
+      "step": 2850
+    },
+    {
+      "epoch": 2.0702135360115816,
+      "grad_norm": 0.6202841997146606,
+      "learning_rate": 0.0002,
+      "loss": 1.2331,
+      "step": 2860
+    },
+    {
+      "epoch": 2.077452044878755,
+      "grad_norm": 0.6810227632522583,
+      "learning_rate": 0.0002,
+      "loss": 1.3298,
+      "step": 2870
+    },
+    {
+      "epoch": 2.0846905537459284,
+      "grad_norm": 0.7481493353843689,
+      "learning_rate": 0.0002,
+      "loss": 1.2505,
+      "step": 2880
+    },
+    {
+      "epoch": 2.0919290626131017,
+      "grad_norm": 0.7089637517929077,
+      "learning_rate": 0.0002,
+      "loss": 1.2484,
+      "step": 2890
+    },
+    {
+      "epoch": 2.099167571480275,
+      "grad_norm": 0.7472923398017883,
+      "learning_rate": 0.0002,
+      "loss": 1.3095,
+      "step": 2900
+    },
+    {
+      "epoch": 2.1064060803474485,
+      "grad_norm": 0.8135465979576111,
+      "learning_rate": 0.0002,
+      "loss": 1.304,
+      "step": 2910
+    },
+    {
+      "epoch": 2.113644589214622,
+      "grad_norm": 0.6097133159637451,
+      "learning_rate": 0.0002,
+      "loss": 1.273,
+      "step": 2920
+    },
+    {
+      "epoch": 2.120883098081795,
+      "grad_norm": 0.5970117449760437,
+      "learning_rate": 0.0002,
+      "loss": 1.3384,
+      "step": 2930
+    },
+    {
+      "epoch": 2.1281216069489686,
+      "grad_norm": 0.6169309616088867,
+      "learning_rate": 0.0002,
+      "loss": 1.3233,
+      "step": 2940
+    },
+    {
+      "epoch": 2.135360115816142,
+      "grad_norm": 0.9428738355636597,
+      "learning_rate": 0.0002,
+      "loss": 1.4246,
+      "step": 2950
+    },
+    {
+      "epoch": 2.1425986246833153,
+      "grad_norm": 0.5671679973602295,
+      "learning_rate": 0.0002,
+      "loss": 1.3527,
+      "step": 2960
+    },
+    {
+      "epoch": 2.1498371335504887,
+      "grad_norm": 0.7007262110710144,
+      "learning_rate": 0.0002,
+      "loss": 1.1375,
+      "step": 2970
+    },
+    {
+      "epoch": 2.157075642417662,
+      "grad_norm": 0.6294044256210327,
+      "learning_rate": 0.0002,
+      "loss": 1.2015,
+      "step": 2980
+    },
+    {
+      "epoch": 2.1643141512848354,
+      "grad_norm": 0.6105241775512695,
+      "learning_rate": 0.0002,
+      "loss": 1.2167,
+      "step": 2990
+    },
+    {
+      "epoch": 2.1715526601520088,
+      "grad_norm": 0.557124137878418,
+      "learning_rate": 0.0002,
+      "loss": 1.2065,
+      "step": 3000
+    },
+    {
+      "epoch": 2.178791169019182,
+      "grad_norm": 0.6250392198562622,
+      "learning_rate": 0.0002,
+      "loss": 1.2515,
+      "step": 3010
+    },
+    {
+      "epoch": 2.1860296778863555,
+      "grad_norm": 0.645218551158905,
+      "learning_rate": 0.0002,
+      "loss": 1.385,
+      "step": 3020
+    },
+    {
+      "epoch": 2.193268186753529,
+      "grad_norm": 0.9033605456352234,
+      "learning_rate": 0.0002,
+      "loss": 1.3928,
+      "step": 3030
+    },
+    {
+      "epoch": 2.2005066956207022,
+      "grad_norm": 0.5325747132301331,
+      "learning_rate": 0.0002,
+      "loss": 1.2458,
+      "step": 3040
+    },
+    {
+      "epoch": 2.2077452044878756,
+      "grad_norm": 0.6334700584411621,
+      "learning_rate": 0.0002,
+      "loss": 1.261,
+      "step": 3050
+    },
+    {
+      "epoch": 2.214983713355049,
+      "grad_norm": 0.5206325054168701,
+      "learning_rate": 0.0002,
+      "loss": 1.2385,
+      "step": 3060
+    },
+    {
+      "epoch": 2.2222222222222223,
+      "grad_norm": 0.5987200140953064,
+      "learning_rate": 0.0002,
+      "loss": 1.3103,
+      "step": 3070
+    },
+    {
+      "epoch": 2.2294607310893957,
+      "grad_norm": 0.5893264412879944,
+      "learning_rate": 0.0002,
+      "loss": 1.1756,
+      "step": 3080
+    },
+    {
+      "epoch": 2.236699239956569,
+      "grad_norm": 0.6869237422943115,
+      "learning_rate": 0.0002,
+      "loss": 1.235,
+      "step": 3090
+    },
+    {
+      "epoch": 2.2439377488237424,
+      "grad_norm": 0.5040048360824585,
+      "learning_rate": 0.0002,
+      "loss": 1.3285,
+      "step": 3100
+    },
+    {
+      "epoch": 2.251176257690916,
+      "grad_norm": 0.6660613417625427,
+      "learning_rate": 0.0002,
+      "loss": 1.3316,
+      "step": 3110
+    },
+    {
+      "epoch": 2.258414766558089,
+      "grad_norm": 0.5890918970108032,
+      "learning_rate": 0.0002,
+      "loss": 1.3108,
+      "step": 3120
+    },
+    {
+      "epoch": 2.2656532754252625,
+      "grad_norm": 0.6458896994590759,
+      "learning_rate": 0.0002,
+      "loss": 1.248,
+      "step": 3130
+    },
+    {
+      "epoch": 2.272891784292436,
+      "grad_norm": 0.6832690834999084,
+      "learning_rate": 0.0002,
+      "loss": 1.4151,
+      "step": 3140
+    },
+    {
+      "epoch": 2.2801302931596092,
+      "grad_norm": 0.833908200263977,
+      "learning_rate": 0.0002,
+      "loss": 1.4458,
+      "step": 3150
+    },
+    {
+      "epoch": 2.2873688020267826,
+      "grad_norm": 0.4596034586429596,
+      "learning_rate": 0.0002,
+      "loss": 1.2931,
+      "step": 3160
+    },
+    {
+      "epoch": 2.294607310893956,
+      "grad_norm": 0.9130966067314148,
+      "learning_rate": 0.0002,
+      "loss": 1.449,
+      "step": 3170
+    },
+    {
+      "epoch": 2.3018458197611293,
+      "grad_norm": 0.7143292427062988,
+      "learning_rate": 0.0002,
+      "loss": 1.3806,
+      "step": 3180
+    },
+    {
+      "epoch": 2.3090843286283027,
+      "grad_norm": 0.5388900637626648,
+      "learning_rate": 0.0002,
+      "loss": 1.2692,
+      "step": 3190
+    },
+    {
+      "epoch": 2.316322837495476,
+      "grad_norm": 0.5607513189315796,
+      "learning_rate": 0.0002,
+      "loss": 1.2402,
+      "step": 3200
+    },
+    {
+      "epoch": 2.3235613463626494,
+      "grad_norm": 0.6795142292976379,
+      "learning_rate": 0.0002,
+      "loss": 1.3874,
+      "step": 3210
+    },
+    {
+      "epoch": 2.330799855229823,
+      "grad_norm": 0.6561070680618286,
+      "learning_rate": 0.0002,
+      "loss": 1.3042,
+      "step": 3220
+    },
+    {
+      "epoch": 2.338038364096996,
+      "grad_norm": 0.8858118057250977,
+      "learning_rate": 0.0002,
+      "loss": 1.4636,
+      "step": 3230
+    },
+    {
+      "epoch": 2.3452768729641695,
+      "grad_norm": 0.6604151725769043,
+      "learning_rate": 0.0002,
+      "loss": 1.3214,
+      "step": 3240
+    },
+    {
+      "epoch": 2.352515381831343,
+      "grad_norm": 0.6755785346031189,
+      "learning_rate": 0.0002,
+      "loss": 1.4004,
+      "step": 3250
+    },
+    {
+      "epoch": 2.3597538906985163,
+      "grad_norm": 0.6981677412986755,
+      "learning_rate": 0.0002,
+      "loss": 1.2503,
+      "step": 3260
+    },
+    {
+      "epoch": 2.3669923995656896,
+      "grad_norm": 0.6338568329811096,
+      "learning_rate": 0.0002,
+      "loss": 1.3078,
+      "step": 3270
+    },
+    {
+      "epoch": 2.374230908432863,
+      "grad_norm": 0.5754265785217285,
+      "learning_rate": 0.0002,
+      "loss": 1.285,
+      "step": 3280
+    },
+    {
+      "epoch": 2.3814694173000364,
+      "grad_norm": 0.7533153295516968,
+      "learning_rate": 0.0002,
+      "loss": 1.2924,
+      "step": 3290
+    },
+    {
+      "epoch": 2.3887079261672097,
+      "grad_norm": 0.675065279006958,
+      "learning_rate": 0.0002,
+      "loss": 1.3711,
+      "step": 3300
+    },
+    {
+      "epoch": 2.395946435034383,
+      "grad_norm": 0.5686452984809875,
+      "learning_rate": 0.0002,
+      "loss": 1.3548,
+      "step": 3310
+    },
+    {
+      "epoch": 2.4031849439015565,
+      "grad_norm": 0.8129481673240662,
+      "learning_rate": 0.0002,
+      "loss": 1.1998,
+      "step": 3320
+    },
+    {
+      "epoch": 2.41042345276873,
+      "grad_norm": 0.6615934371948242,
+      "learning_rate": 0.0002,
+      "loss": 1.2584,
+      "step": 3330
+    },
+    {
+      "epoch": 2.417661961635903,
+      "grad_norm": 0.6678834557533264,
+      "learning_rate": 0.0002,
+      "loss": 1.3691,
+      "step": 3340
+    },
+    {
+      "epoch": 2.4249004705030766,
+      "grad_norm": 0.5581308007240295,
+      "learning_rate": 0.0002,
+      "loss": 1.2381,
+      "step": 3350
+    },
+    {
+      "epoch": 2.43213897937025,
+      "grad_norm": 0.6098920106887817,
+      "learning_rate": 0.0002,
+      "loss": 1.3853,
+      "step": 3360
+    },
+    {
+      "epoch": 2.4393774882374233,
+      "grad_norm": 0.8101736903190613,
+      "learning_rate": 0.0002,
+      "loss": 1.3692,
+      "step": 3370
+    },
+    {
+      "epoch": 2.4466159971045967,
+      "grad_norm": 0.6621488928794861,
+      "learning_rate": 0.0002,
+      "loss": 1.4418,
+      "step": 3380
+    },
+    {
+      "epoch": 2.45385450597177,
+      "grad_norm": 0.8693289160728455,
+      "learning_rate": 0.0002,
+      "loss": 1.4579,
+      "step": 3390
+    },
+    {
+      "epoch": 2.4610930148389434,
+      "grad_norm": 0.6724580526351929,
+      "learning_rate": 0.0002,
+      "loss": 1.3644,
+      "step": 3400
+    },
+    {
+      "epoch": 2.4683315237061167,
+      "grad_norm": 0.6776891946792603,
+      "learning_rate": 0.0002,
+      "loss": 1.2006,
+      "step": 3410
+    },
+    {
+      "epoch": 2.47557003257329,
+      "grad_norm": 0.7214453816413879,
+      "learning_rate": 0.0002,
+      "loss": 1.2937,
+      "step": 3420
+    },
+    {
+      "epoch": 2.4828085414404635,
+      "grad_norm": 0.8390451073646545,
+      "learning_rate": 0.0002,
+      "loss": 1.4051,
+      "step": 3430
+    },
+    {
+      "epoch": 2.490047050307637,
+      "grad_norm": 0.7130982279777527,
+      "learning_rate": 0.0002,
+      "loss": 1.25,
+      "step": 3440
+    },
+    {
+      "epoch": 2.49728555917481,
+      "grad_norm": 0.8873937129974365,
+      "learning_rate": 0.0002,
+      "loss": 1.2231,
+      "step": 3450
+    },
+    {
+      "epoch": 2.5045240680419836,
+      "grad_norm": 0.725185751914978,
+      "learning_rate": 0.0002,
+      "loss": 1.1429,
+      "step": 3460
+    },
+    {
+      "epoch": 2.511762576909157,
+      "grad_norm": 0.6120352149009705,
+      "learning_rate": 0.0002,
+      "loss": 1.2699,
+      "step": 3470
+    },
+    {
+      "epoch": 2.5190010857763303,
+      "grad_norm": 0.7713613510131836,
+      "learning_rate": 0.0002,
+      "loss": 1.2552,
+      "step": 3480
+    },
+    {
+      "epoch": 2.5262395946435037,
+      "grad_norm": 0.895309567451477,
+      "learning_rate": 0.0002,
+      "loss": 1.4648,
+      "step": 3490
+    },
+    {
+      "epoch": 2.533478103510677,
+      "grad_norm": 0.9631021022796631,
+      "learning_rate": 0.0002,
+      "loss": 1.3043,
+      "step": 3500
+    },
+    {
+      "epoch": 2.5407166123778504,
+      "grad_norm": 0.7475683093070984,
+      "learning_rate": 0.0002,
+      "loss": 1.3492,
+      "step": 3510
+    },
+    {
+      "epoch": 2.5479551212450238,
+      "grad_norm": 0.7271341681480408,
+      "learning_rate": 0.0002,
+      "loss": 1.3637,
+      "step": 3520
+    },
+    {
+      "epoch": 2.555193630112197,
+      "grad_norm": 0.6979510188102722,
+      "learning_rate": 0.0002,
+      "loss": 1.304,
+      "step": 3530
+    },
+    {
+      "epoch": 2.5624321389793705,
+      "grad_norm": 0.6504196524620056,
+      "learning_rate": 0.0002,
+      "loss": 1.2353,
+      "step": 3540
+    },
+    {
+      "epoch": 2.569670647846544,
+      "grad_norm": 0.7226675748825073,
+      "learning_rate": 0.0002,
+      "loss": 1.2699,
+      "step": 3550
+    },
+    {
+      "epoch": 2.5769091567137172,
+      "grad_norm": 0.6143222451210022,
+      "learning_rate": 0.0002,
+      "loss": 1.3002,
+      "step": 3560
+    },
+    {
+      "epoch": 2.5841476655808906,
+      "grad_norm": 0.7245154976844788,
+      "learning_rate": 0.0002,
+      "loss": 1.1585,
+      "step": 3570
+    },
+    {
+      "epoch": 2.591386174448064,
+      "grad_norm": 0.943540632724762,
+      "learning_rate": 0.0002,
+      "loss": 1.3651,
+      "step": 3580
+    },
+    {
+      "epoch": 2.5986246833152373,
+      "grad_norm": 0.7707241773605347,
+      "learning_rate": 0.0002,
+      "loss": 1.3034,
+      "step": 3590
+    },
+    {
+      "epoch": 2.6058631921824107,
+      "grad_norm": 0.6705001592636108,
+      "learning_rate": 0.0002,
+      "loss": 1.3063,
+      "step": 3600
+    },
+    {
+      "epoch": 2.613101701049584,
+      "grad_norm": 0.6360933780670166,
+      "learning_rate": 0.0002,
+      "loss": 1.2437,
+      "step": 3610
+    },
+    {
+      "epoch": 2.6203402099167574,
+      "grad_norm": 0.5846424698829651,
+      "learning_rate": 0.0002,
+      "loss": 1.1844,
+      "step": 3620
+    },
+    {
+      "epoch": 2.6275787187839303,
+      "grad_norm": 0.5958625674247742,
+      "learning_rate": 0.0002,
+      "loss": 1.3674,
+      "step": 3630
+    },
+    {
+      "epoch": 2.6348172276511037,
+      "grad_norm": 0.6819243431091309,
+      "learning_rate": 0.0002,
+      "loss": 1.3599,
+      "step": 3640
+    },
+    {
+      "epoch": 2.642055736518277,
+      "grad_norm": 0.7033445835113525,
+      "learning_rate": 0.0002,
+      "loss": 1.3884,
+      "step": 3650
+    },
+    {
+      "epoch": 2.6492942453854504,
+      "grad_norm": 0.6134849786758423,
+      "learning_rate": 0.0002,
+      "loss": 1.3392,
+      "step": 3660
+    },
+    {
+      "epoch": 2.656532754252624,
+      "grad_norm": 0.658009946346283,
+      "learning_rate": 0.0002,
+      "loss": 1.2661,
+      "step": 3670
+    },
+    {
+      "epoch": 2.663771263119797,
+      "grad_norm": 0.6280999779701233,
+      "learning_rate": 0.0002,
+      "loss": 1.3987,
+      "step": 3680
+    },
+    {
+      "epoch": 2.6710097719869705,
+      "grad_norm": 0.5536085963249207,
+      "learning_rate": 0.0002,
+      "loss": 1.2995,
+      "step": 3690
+    },
+    {
+      "epoch": 2.678248280854144,
+      "grad_norm": 0.8603981733322144,
+      "learning_rate": 0.0002,
+      "loss": 1.2044,
+      "step": 3700
+    },
+    {
+      "epoch": 2.6854867897213173,
+      "grad_norm": 0.5509994626045227,
+      "learning_rate": 0.0002,
+      "loss": 1.3879,
+      "step": 3710
+    },
+    {
+      "epoch": 2.6927252985884906,
+      "grad_norm": 0.9093621969223022,
+      "learning_rate": 0.0002,
+      "loss": 1.3253,
+      "step": 3720
+    },
+    {
+      "epoch": 2.699963807455664,
+      "grad_norm": 0.7525952458381653,
+      "learning_rate": 0.0002,
+      "loss": 1.2668,
+      "step": 3730
+    },
+    {
+      "epoch": 2.7072023163228374,
+      "grad_norm": 0.6737023591995239,
+      "learning_rate": 0.0002,
+      "loss": 1.248,
+      "step": 3740
+    },
+    {
+      "epoch": 2.7144408251900107,
+      "grad_norm": 0.8656924962997437,
+      "learning_rate": 0.0002,
+      "loss": 1.2981,
+      "step": 3750
+    },
+    {
+      "epoch": 2.721679334057184,
+      "grad_norm": 0.7494133114814758,
+      "learning_rate": 0.0002,
+      "loss": 1.2342,
+      "step": 3760
+    },
+    {
+      "epoch": 2.7289178429243575,
+      "grad_norm": 0.5725520849227905,
+      "learning_rate": 0.0002,
+      "loss": 1.2417,
+      "step": 3770
+    },
+    {
+      "epoch": 2.736156351791531,
+      "grad_norm": 0.836412787437439,
+      "learning_rate": 0.0002,
+      "loss": 1.28,
+      "step": 3780
+    },
+    {
+      "epoch": 2.743394860658704,
+      "grad_norm": 0.6893242597579956,
+      "learning_rate": 0.0002,
+      "loss": 1.3784,
+      "step": 3790
+    },
+    {
+      "epoch": 2.7506333695258776,
+      "grad_norm": 0.6696223020553589,
+      "learning_rate": 0.0002,
+      "loss": 1.2929,
+      "step": 3800
+    },
+    {
+      "epoch": 2.757871878393051,
+      "grad_norm": 0.6483015418052673,
+      "learning_rate": 0.0002,
+      "loss": 1.2449,
+      "step": 3810
+    },
+    {
+      "epoch": 2.7651103872602243,
+      "grad_norm": 0.8084456920623779,
+      "learning_rate": 0.0002,
+      "loss": 1.3282,
+      "step": 3820
+    },
+    {
+      "epoch": 2.7723488961273977,
+      "grad_norm": 0.6601949334144592,
+      "learning_rate": 0.0002,
+      "loss": 1.3694,
+      "step": 3830
+    },
+    {
+      "epoch": 2.779587404994571,
+      "grad_norm": 0.6905533671379089,
+      "learning_rate": 0.0002,
+      "loss": 1.3568,
+      "step": 3840
+    },
+    {
+      "epoch": 2.7868259138617444,
+      "grad_norm": 0.619318425655365,
+      "learning_rate": 0.0002,
+      "loss": 1.3854,
+      "step": 3850
+    },
+    {
+      "epoch": 2.7940644227289178,
+      "grad_norm": 0.5994023084640503,
+      "learning_rate": 0.0002,
+      "loss": 1.2551,
+      "step": 3860
+    },
+    {
+      "epoch": 2.801302931596091,
+      "grad_norm": 0.5627168416976929,
+      "learning_rate": 0.0002,
+      "loss": 1.2022,
+      "step": 3870
+    },
+    {
+      "epoch": 2.8085414404632645,
+      "grad_norm": 0.6001605987548828,
+      "learning_rate": 0.0002,
+      "loss": 1.3921,
+      "step": 3880
+    },
+    {
+      "epoch": 2.815779949330438,
+      "grad_norm": 0.6022412776947021,
+      "learning_rate": 0.0002,
+      "loss": 1.3026,
+      "step": 3890
+    },
+    {
+      "epoch": 2.823018458197611,
+      "grad_norm": 0.6832426190376282,
+      "learning_rate": 0.0002,
+      "loss": 1.2765,
+      "step": 3900
+    },
+    {
+      "epoch": 2.8302569670647846,
+      "grad_norm": 0.5936811566352844,
+      "learning_rate": 0.0002,
+      "loss": 1.1363,
+      "step": 3910
+    },
+    {
+      "epoch": 2.837495475931958,
+      "grad_norm": 0.6960572600364685,
+      "learning_rate": 0.0002,
+      "loss": 1.1707,
+      "step": 3920
+    },
+    {
+      "epoch": 2.8447339847991313,
+      "grad_norm": 0.5913406610488892,
+      "learning_rate": 0.0002,
+      "loss": 1.4063,
+      "step": 3930
+    },
+    {
+      "epoch": 2.8519724936663047,
+      "grad_norm": 0.678154706954956,
+      "learning_rate": 0.0002,
+      "loss": 1.3245,
+      "step": 3940
+    },
+    {
+      "epoch": 2.859211002533478,
+      "grad_norm": 0.7898936867713928,
+      "learning_rate": 0.0002,
+      "loss": 1.366,
+      "step": 3950
+    },
+    {
+      "epoch": 2.8664495114006514,
+      "grad_norm": 0.9234195351600647,
+      "learning_rate": 0.0002,
+      "loss": 1.3948,
+      "step": 3960
+    },
+    {
+      "epoch": 2.8736880202678248,
+      "grad_norm": 0.5960825085639954,
+      "learning_rate": 0.0002,
+      "loss": 1.2773,
+      "step": 3970
+    },
+    {
+      "epoch": 2.880926529134998,
+      "grad_norm": 0.677118182182312,
+      "learning_rate": 0.0002,
+      "loss": 1.3127,
+      "step": 3980
+    },
+    {
+      "epoch": 2.8881650380021715,
+      "grad_norm": 0.6505142450332642,
+      "learning_rate": 0.0002,
+      "loss": 1.2652,
+      "step": 3990
+    },
+    {
+      "epoch": 2.895403546869345,
+      "grad_norm": 0.550826907157898,
+      "learning_rate": 0.0002,
+      "loss": 1.2078,
+      "step": 4000
+    },
+    {
+      "epoch": 2.9026420557365182,
+      "grad_norm": 0.6209215521812439,
+      "learning_rate": 0.0002,
+      "loss": 1.1811,
+      "step": 4010
+    },
+    {
+      "epoch": 2.9098805646036916,
+      "grad_norm": 0.6549018025398254,
+      "learning_rate": 0.0002,
+      "loss": 1.4001,
+      "step": 4020
+    },
+    {
+      "epoch": 2.917119073470865,
+      "grad_norm": 0.570682168006897,
+      "learning_rate": 0.0002,
+      "loss": 1.2285,
+      "step": 4030
+    },
+    {
+      "epoch": 2.9243575823380383,
+      "grad_norm": 1.1807632446289062,
+      "learning_rate": 0.0002,
+      "loss": 1.0832,
+      "step": 4040
+    },
+    {
+      "epoch": 2.9315960912052117,
+      "grad_norm": 0.7058857679367065,
+      "learning_rate": 0.0002,
+      "loss": 1.2693,
+      "step": 4050
+    },
+    {
+      "epoch": 2.938834600072385,
+      "grad_norm": 0.5542812943458557,
+      "learning_rate": 0.0002,
+      "loss": 1.2905,
+      "step": 4060
+    },
+    {
+      "epoch": 2.9460731089395584,
+      "grad_norm": 0.63167804479599,
+      "learning_rate": 0.0002,
+      "loss": 1.33,
+      "step": 4070
+    },
+    {
+      "epoch": 2.953311617806732,
+      "grad_norm": 0.5702962279319763,
+      "learning_rate": 0.0002,
+      "loss": 1.3075,
+      "step": 4080
+    },
+    {
+      "epoch": 2.960550126673905,
+      "grad_norm": 0.620944082736969,
+      "learning_rate": 0.0002,
+      "loss": 1.2007,
+      "step": 4090
+    },
+    {
+      "epoch": 2.9677886355410785,
+      "grad_norm": 0.5866289734840393,
+      "learning_rate": 0.0002,
+      "loss": 1.2864,
+      "step": 4100
+    },
+    {
+      "epoch": 2.975027144408252,
+      "grad_norm": 0.560170590877533,
+      "learning_rate": 0.0002,
+      "loss": 1.3293,
+      "step": 4110
+    },
+    {
+      "epoch": 2.9822656532754253,
+      "grad_norm": 0.675082802772522,
+      "learning_rate": 0.0002,
+      "loss": 1.2071,
+      "step": 4120
+    },
+    {
+      "epoch": 2.9895041621425986,
+      "grad_norm": 0.62708580493927,
+      "learning_rate": 0.0002,
+      "loss": 1.2981,
+      "step": 4130
+    },
+    {
+      "epoch": 2.996742671009772,
+      "grad_norm": 0.7893929481506348,
+      "learning_rate": 0.0002,
+      "loss": 1.2758,
+      "step": 4140
+    },
+    {
+      "epoch": 2.9996380745566413,
+      "eval_loss": 1.4217946529388428,
+      "eval_runtime": 27.1596,
+      "eval_samples_per_second": 16.053,
+      "eval_steps_per_second": 2.025,
+      "step": 4144
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 11048,
+  "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": 5.065406381083853e+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-0.9-num-8828-sd-1/checkpoint-4144/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-0.9-num-8828-sd-1/checkpoint-4144/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..4d7b7431bbbe8c9bf29b925bca391a558af5ff8c
--- /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-0.9-num-8828-sd-1/checkpoint-4144/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ad613885e4f267fc04125f1a836d42cfa796bbe12e536f9ee60c955de02cdb5a
+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-0.9-num-8828-sd-1/checkpoint-5526/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-0.9-num-8828-sd-1/checkpoint-5526/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-0.9-num-8828-sd-1/checkpoint-5526/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-0.9-num-8828-sd-1/checkpoint-5526/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-0.9-num-8828-sd-1/checkpoint-5526/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e99bbcd43df1c19d98706c7e3be95c93844c5349
--- /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-0.9-num-8828-sd-1/checkpoint-5526/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-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-8828-sd-1/checkpoint-5526/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-0.9-num-8828-sd-1/checkpoint-5526/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..655eefd2eea5a13040896f1cd416fe42493622b4
--- /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-0.9-num-8828-sd-1/checkpoint-5526/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b98143cbd60c831e49f04f9f4ba0f3bbda73012695c7ecf5743a71cd76cf6e8
+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-0.9-num-8828-sd-1/checkpoint-5526/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-0.9-num-8828-sd-1/checkpoint-5526/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..072c0e2957726c5cc0328345945224af09bd3af6
--- /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-0.9-num-8828-sd-1/checkpoint-5526/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cf5bcbebb674c668272572527b65c1fe1f6778897cf32e232c0882f74c57876e
+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-0.9-num-8828-sd-1/checkpoint-5526/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-0.9-num-8828-sd-1/checkpoint-5526/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..fcf8222e07bdd4ee70690af4b5eb95ac29ec2451
--- /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-0.9-num-8828-sd-1/checkpoint-5526/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b535b07a56c32c1db908d2c4e8439bc752debb1e73d647fe63d6867f729e8329
+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-0.9-num-8828-sd-1/checkpoint-5526/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-0.9-num-8828-sd-1/checkpoint-5526/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..1c966bca080cbb78186c7353ae4cb971df996eed
--- /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-0.9-num-8828-sd-1/checkpoint-5526/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4473c852e8ab8679db533b88b697f77385cc25d705112cb6582681e17a7e16da
+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-0.9-num-8828-sd-1/checkpoint-5526/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-0.9-num-8828-sd-1/checkpoint-5526/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-0.9-num-8828-sd-1/checkpoint-5526/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-0.9-num-8828-sd-1/checkpoint-5526/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-0.9-num-8828-sd-1/checkpoint-5526/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-0.9-num-8828-sd-1/checkpoint-5526/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-0.9-num-8828-sd-1/checkpoint-5526/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-0.9-num-8828-sd-1/checkpoint-5526/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-0.9-num-8828-sd-1/checkpoint-5526/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-0.9-num-8828-sd-1/checkpoint-5526/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-0.9-num-8828-sd-1/checkpoint-5526/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-0.9-num-8828-sd-1/checkpoint-5526/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-0.9-num-8828-sd-1/checkpoint-5526/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-0.9-num-8828-sd-1/checkpoint-5526/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..b4e41babb7ba651c8dc0ba6615b5de4aed6a5c51
--- /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-0.9-num-8828-sd-1/checkpoint-5526/trainer_state.json
@@ -0,0 +1,3929 @@
+{
+  "best_metric": 1.4217946529388428,
+  "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-0.9-num-8828-sd-1/checkpoint-4144",
+  "epoch": 4.0,
+  "eval_steps": 10,
+  "global_step": 5526,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007238508867173362,
+      "grad_norm": 1.2523442506790161,
+      "learning_rate": 0.0002,
+      "loss": 4.7061,
+      "step": 10
+    },
+    {
+      "epoch": 0.014477017734346724,
+      "grad_norm": 1.8887330293655396,
+      "learning_rate": 0.0002,
+      "loss": 3.3493,
+      "step": 20
+    },
+    {
+      "epoch": 0.021715526601520086,
+      "grad_norm": 0.9668035507202148,
+      "learning_rate": 0.0002,
+      "loss": 2.7585,
+      "step": 30
+    },
+    {
+      "epoch": 0.028954035468693448,
+      "grad_norm": 2.9167306423187256,
+      "learning_rate": 0.0002,
+      "loss": 2.3699,
+      "step": 40
+    },
+    {
+      "epoch": 0.036192544335866814,
+      "grad_norm": 2.649867296218872,
+      "learning_rate": 0.0002,
+      "loss": 2.2679,
+      "step": 50
+    },
+    {
+      "epoch": 0.04343105320304017,
+      "grad_norm": 1.5120655298233032,
+      "learning_rate": 0.0002,
+      "loss": 2.2202,
+      "step": 60
+    },
+    {
+      "epoch": 0.05066956207021354,
+      "grad_norm": 0.7879868149757385,
+      "learning_rate": 0.0002,
+      "loss": 2.2026,
+      "step": 70
+    },
+    {
+      "epoch": 0.057908070937386896,
+      "grad_norm": 0.7616953253746033,
+      "learning_rate": 0.0002,
+      "loss": 1.9447,
+      "step": 80
+    },
+    {
+      "epoch": 0.06514657980456026,
+      "grad_norm": 1.8809149265289307,
+      "learning_rate": 0.0002,
+      "loss": 2.0112,
+      "step": 90
+    },
+    {
+      "epoch": 0.07238508867173363,
+      "grad_norm": 0.9294016361236572,
+      "learning_rate": 0.0002,
+      "loss": 1.8337,
+      "step": 100
+    },
+    {
+      "epoch": 0.07962359753890698,
+      "grad_norm": 0.7145281434059143,
+      "learning_rate": 0.0002,
+      "loss": 1.8419,
+      "step": 110
+    },
+    {
+      "epoch": 0.08686210640608034,
+      "grad_norm": 0.7564446330070496,
+      "learning_rate": 0.0002,
+      "loss": 2.0036,
+      "step": 120
+    },
+    {
+      "epoch": 0.09410061527325371,
+      "grad_norm": 1.1681925058364868,
+      "learning_rate": 0.0002,
+      "loss": 1.9306,
+      "step": 130
+    },
+    {
+      "epoch": 0.10133912414042708,
+      "grad_norm": 0.6708641648292542,
+      "learning_rate": 0.0002,
+      "loss": 1.7875,
+      "step": 140
+    },
+    {
+      "epoch": 0.10857763300760044,
+      "grad_norm": 0.7625647783279419,
+      "learning_rate": 0.0002,
+      "loss": 1.786,
+      "step": 150
+    },
+    {
+      "epoch": 0.11581614187477379,
+      "grad_norm": 0.8463464975357056,
+      "learning_rate": 0.0002,
+      "loss": 1.6687,
+      "step": 160
+    },
+    {
+      "epoch": 0.12305465074194716,
+      "grad_norm": 0.7502335906028748,
+      "learning_rate": 0.0002,
+      "loss": 1.6214,
+      "step": 170
+    },
+    {
+      "epoch": 0.13029315960912052,
+      "grad_norm": 0.6929958462715149,
+      "learning_rate": 0.0002,
+      "loss": 1.7433,
+      "step": 180
+    },
+    {
+      "epoch": 0.1375316684762939,
+      "grad_norm": 0.6798707842826843,
+      "learning_rate": 0.0002,
+      "loss": 1.6009,
+      "step": 190
+    },
+    {
+      "epoch": 0.14477017734346725,
+      "grad_norm": 0.7566508650779724,
+      "learning_rate": 0.0002,
+      "loss": 1.6208,
+      "step": 200
+    },
+    {
+      "epoch": 0.15200868621064062,
+      "grad_norm": 0.7196869850158691,
+      "learning_rate": 0.0002,
+      "loss": 1.5823,
+      "step": 210
+    },
+    {
+      "epoch": 0.15924719507781396,
+      "grad_norm": 0.8401045799255371,
+      "learning_rate": 0.0002,
+      "loss": 1.738,
+      "step": 220
+    },
+    {
+      "epoch": 0.16648570394498732,
+      "grad_norm": 0.8503773212432861,
+      "learning_rate": 0.0002,
+      "loss": 1.7574,
+      "step": 230
+    },
+    {
+      "epoch": 0.1737242128121607,
+      "grad_norm": 0.7183733582496643,
+      "learning_rate": 0.0002,
+      "loss": 1.7861,
+      "step": 240
+    },
+    {
+      "epoch": 0.18096272167933405,
+      "grad_norm": 0.7082605957984924,
+      "learning_rate": 0.0002,
+      "loss": 1.6693,
+      "step": 250
+    },
+    {
+      "epoch": 0.18820123054650742,
+      "grad_norm": 0.9386326670646667,
+      "learning_rate": 0.0002,
+      "loss": 1.619,
+      "step": 260
+    },
+    {
+      "epoch": 0.19543973941368079,
+      "grad_norm": 0.7332451939582825,
+      "learning_rate": 0.0002,
+      "loss": 1.6511,
+      "step": 270
+    },
+    {
+      "epoch": 0.20267824828085415,
+      "grad_norm": 0.7092869877815247,
+      "learning_rate": 0.0002,
+      "loss": 1.6353,
+      "step": 280
+    },
+    {
+      "epoch": 0.20991675714802752,
+      "grad_norm": 0.7256413698196411,
+      "learning_rate": 0.0002,
+      "loss": 1.5996,
+      "step": 290
+    },
+    {
+      "epoch": 0.21715526601520088,
+      "grad_norm": 0.6398681402206421,
+      "learning_rate": 0.0002,
+      "loss": 1.6754,
+      "step": 300
+    },
+    {
+      "epoch": 0.22439377488237422,
+      "grad_norm": 0.6273287534713745,
+      "learning_rate": 0.0002,
+      "loss": 1.397,
+      "step": 310
+    },
+    {
+      "epoch": 0.23163228374954759,
+      "grad_norm": 0.511648416519165,
+      "learning_rate": 0.0002,
+      "loss": 1.5115,
+      "step": 320
+    },
+    {
+      "epoch": 0.23887079261672095,
+      "grad_norm": 0.8677352070808411,
+      "learning_rate": 0.0002,
+      "loss": 1.5424,
+      "step": 330
+    },
+    {
+      "epoch": 0.24610930148389432,
+      "grad_norm": 0.6270743012428284,
+      "learning_rate": 0.0002,
+      "loss": 1.6779,
+      "step": 340
+    },
+    {
+      "epoch": 0.2533478103510677,
+      "grad_norm": 0.7980281114578247,
+      "learning_rate": 0.0002,
+      "loss": 1.626,
+      "step": 350
+    },
+    {
+      "epoch": 0.26058631921824105,
+      "grad_norm": 0.632486879825592,
+      "learning_rate": 0.0002,
+      "loss": 1.5238,
+      "step": 360
+    },
+    {
+      "epoch": 0.2678248280854144,
+      "grad_norm": 0.6527034640312195,
+      "learning_rate": 0.0002,
+      "loss": 1.5175,
+      "step": 370
+    },
+    {
+      "epoch": 0.2750633369525878,
+      "grad_norm": 0.7672118544578552,
+      "learning_rate": 0.0002,
+      "loss": 1.627,
+      "step": 380
+    },
+    {
+      "epoch": 0.28230184581976114,
+      "grad_norm": 0.6035117506980896,
+      "learning_rate": 0.0002,
+      "loss": 1.5605,
+      "step": 390
+    },
+    {
+      "epoch": 0.2895403546869345,
+      "grad_norm": 0.5955103039741516,
+      "learning_rate": 0.0002,
+      "loss": 1.4603,
+      "step": 400
+    },
+    {
+      "epoch": 0.2967788635541079,
+      "grad_norm": 0.6015191674232483,
+      "learning_rate": 0.0002,
+      "loss": 1.558,
+      "step": 410
+    },
+    {
+      "epoch": 0.30401737242128124,
+      "grad_norm": 0.6380982398986816,
+      "learning_rate": 0.0002,
+      "loss": 1.6091,
+      "step": 420
+    },
+    {
+      "epoch": 0.3112558812884546,
+      "grad_norm": 0.6707863211631775,
+      "learning_rate": 0.0002,
+      "loss": 1.5292,
+      "step": 430
+    },
+    {
+      "epoch": 0.3184943901556279,
+      "grad_norm": 0.7010176777839661,
+      "learning_rate": 0.0002,
+      "loss": 1.4426,
+      "step": 440
+    },
+    {
+      "epoch": 0.3257328990228013,
+      "grad_norm": 0.8263739943504333,
+      "learning_rate": 0.0002,
+      "loss": 1.5572,
+      "step": 450
+    },
+    {
+      "epoch": 0.33297140788997465,
+      "grad_norm": 0.7253276109695435,
+      "learning_rate": 0.0002,
+      "loss": 1.5188,
+      "step": 460
+    },
+    {
+      "epoch": 0.340209916757148,
+      "grad_norm": 0.5238934755325317,
+      "learning_rate": 0.0002,
+      "loss": 1.584,
+      "step": 470
+    },
+    {
+      "epoch": 0.3474484256243214,
+      "grad_norm": 0.7869495749473572,
+      "learning_rate": 0.0002,
+      "loss": 1.7035,
+      "step": 480
+    },
+    {
+      "epoch": 0.35468693449149474,
+      "grad_norm": 0.7485215663909912,
+      "learning_rate": 0.0002,
+      "loss": 1.5776,
+      "step": 490
+    },
+    {
+      "epoch": 0.3619254433586681,
+      "grad_norm": 0.5413193106651306,
+      "learning_rate": 0.0002,
+      "loss": 1.6274,
+      "step": 500
+    },
+    {
+      "epoch": 0.3691639522258415,
+      "grad_norm": 0.7615048885345459,
+      "learning_rate": 0.0002,
+      "loss": 1.7323,
+      "step": 510
+    },
+    {
+      "epoch": 0.37640246109301484,
+      "grad_norm": 0.7685340046882629,
+      "learning_rate": 0.0002,
+      "loss": 1.532,
+      "step": 520
+    },
+    {
+      "epoch": 0.3836409699601882,
+      "grad_norm": 0.6379081010818481,
+      "learning_rate": 0.0002,
+      "loss": 1.6312,
+      "step": 530
+    },
+    {
+      "epoch": 0.39087947882736157,
+      "grad_norm": 0.7946939468383789,
+      "learning_rate": 0.0002,
+      "loss": 1.5645,
+      "step": 540
+    },
+    {
+      "epoch": 0.39811798769453494,
+      "grad_norm": 0.6287278532981873,
+      "learning_rate": 0.0002,
+      "loss": 1.4001,
+      "step": 550
+    },
+    {
+      "epoch": 0.4053564965617083,
+      "grad_norm": 0.6811642646789551,
+      "learning_rate": 0.0002,
+      "loss": 1.5982,
+      "step": 560
+    },
+    {
+      "epoch": 0.41259500542888167,
+      "grad_norm": 0.671073317527771,
+      "learning_rate": 0.0002,
+      "loss": 1.4953,
+      "step": 570
+    },
+    {
+      "epoch": 0.41983351429605503,
+      "grad_norm": 0.6313900351524353,
+      "learning_rate": 0.0002,
+      "loss": 1.6753,
+      "step": 580
+    },
+    {
+      "epoch": 0.4270720231632284,
+      "grad_norm": 0.5291772484779358,
+      "learning_rate": 0.0002,
+      "loss": 1.546,
+      "step": 590
+    },
+    {
+      "epoch": 0.43431053203040176,
+      "grad_norm": 0.62503582239151,
+      "learning_rate": 0.0002,
+      "loss": 1.5441,
+      "step": 600
+    },
+    {
+      "epoch": 0.4415490408975751,
+      "grad_norm": 0.5777305364608765,
+      "learning_rate": 0.0002,
+      "loss": 1.6276,
+      "step": 610
+    },
+    {
+      "epoch": 0.44878754976474844,
+      "grad_norm": 0.7013497352600098,
+      "learning_rate": 0.0002,
+      "loss": 1.4758,
+      "step": 620
+    },
+    {
+      "epoch": 0.4560260586319218,
+      "grad_norm": 0.8044822216033936,
+      "learning_rate": 0.0002,
+      "loss": 1.4029,
+      "step": 630
+    },
+    {
+      "epoch": 0.46326456749909517,
+      "grad_norm": 0.672531247138977,
+      "learning_rate": 0.0002,
+      "loss": 1.7195,
+      "step": 640
+    },
+    {
+      "epoch": 0.47050307636626854,
+      "grad_norm": 0.6233910322189331,
+      "learning_rate": 0.0002,
+      "loss": 1.614,
+      "step": 650
+    },
+    {
+      "epoch": 0.4777415852334419,
+      "grad_norm": 0.651524543762207,
+      "learning_rate": 0.0002,
+      "loss": 1.6041,
+      "step": 660
+    },
+    {
+      "epoch": 0.48498009410061527,
+      "grad_norm": 0.7213939428329468,
+      "learning_rate": 0.0002,
+      "loss": 1.5842,
+      "step": 670
+    },
+    {
+      "epoch": 0.49221860296778863,
+      "grad_norm": 0.6541454792022705,
+      "learning_rate": 0.0002,
+      "loss": 1.5453,
+      "step": 680
+    },
+    {
+      "epoch": 0.499457111834962,
+      "grad_norm": 0.6568936109542847,
+      "learning_rate": 0.0002,
+      "loss": 1.662,
+      "step": 690
+    },
+    {
+      "epoch": 0.5066956207021354,
+      "grad_norm": 0.7176415324211121,
+      "learning_rate": 0.0002,
+      "loss": 1.624,
+      "step": 700
+    },
+    {
+      "epoch": 0.5139341295693087,
+      "grad_norm": 0.6553855538368225,
+      "learning_rate": 0.0002,
+      "loss": 1.6099,
+      "step": 710
+    },
+    {
+      "epoch": 0.5211726384364821,
+      "grad_norm": 0.5654335618019104,
+      "learning_rate": 0.0002,
+      "loss": 1.5508,
+      "step": 720
+    },
+    {
+      "epoch": 0.5284111473036555,
+      "grad_norm": 0.5671001672744751,
+      "learning_rate": 0.0002,
+      "loss": 1.392,
+      "step": 730
+    },
+    {
+      "epoch": 0.5356496561708288,
+      "grad_norm": 0.7914412021636963,
+      "learning_rate": 0.0002,
+      "loss": 1.388,
+      "step": 740
+    },
+    {
+      "epoch": 0.5428881650380022,
+      "grad_norm": 0.6172138452529907,
+      "learning_rate": 0.0002,
+      "loss": 1.5931,
+      "step": 750
+    },
+    {
+      "epoch": 0.5501266739051756,
+      "grad_norm": 0.6132623553276062,
+      "learning_rate": 0.0002,
+      "loss": 1.4018,
+      "step": 760
+    },
+    {
+      "epoch": 0.5573651827723489,
+      "grad_norm": 0.654000461101532,
+      "learning_rate": 0.0002,
+      "loss": 1.513,
+      "step": 770
+    },
+    {
+      "epoch": 0.5646036916395223,
+      "grad_norm": 0.5691370964050293,
+      "learning_rate": 0.0002,
+      "loss": 1.5035,
+      "step": 780
+    },
+    {
+      "epoch": 0.5718422005066957,
+      "grad_norm": 0.7922580242156982,
+      "learning_rate": 0.0002,
+      "loss": 1.65,
+      "step": 790
+    },
+    {
+      "epoch": 0.579080709373869,
+      "grad_norm": 0.6831880211830139,
+      "learning_rate": 0.0002,
+      "loss": 1.4521,
+      "step": 800
+    },
+    {
+      "epoch": 0.5863192182410424,
+      "grad_norm": 0.6740124821662903,
+      "learning_rate": 0.0002,
+      "loss": 1.4734,
+      "step": 810
+    },
+    {
+      "epoch": 0.5935577271082157,
+      "grad_norm": 1.380016803741455,
+      "learning_rate": 0.0002,
+      "loss": 1.6498,
+      "step": 820
+    },
+    {
+      "epoch": 0.6007962359753891,
+      "grad_norm": 0.6552878022193909,
+      "learning_rate": 0.0002,
+      "loss": 1.4642,
+      "step": 830
+    },
+    {
+      "epoch": 0.6080347448425625,
+      "grad_norm": 0.6649535298347473,
+      "learning_rate": 0.0002,
+      "loss": 1.6271,
+      "step": 840
+    },
+    {
+      "epoch": 0.6152732537097358,
+      "grad_norm": 0.561738133430481,
+      "learning_rate": 0.0002,
+      "loss": 1.5886,
+      "step": 850
+    },
+    {
+      "epoch": 0.6225117625769092,
+      "grad_norm": 0.6133047938346863,
+      "learning_rate": 0.0002,
+      "loss": 1.5364,
+      "step": 860
+    },
+    {
+      "epoch": 0.6297502714440825,
+      "grad_norm": 0.559843122959137,
+      "learning_rate": 0.0002,
+      "loss": 1.3489,
+      "step": 870
+    },
+    {
+      "epoch": 0.6369887803112558,
+      "grad_norm": 0.6117811799049377,
+      "learning_rate": 0.0002,
+      "loss": 1.4878,
+      "step": 880
+    },
+    {
+      "epoch": 0.6442272891784292,
+      "grad_norm": 0.6209776401519775,
+      "learning_rate": 0.0002,
+      "loss": 1.56,
+      "step": 890
+    },
+    {
+      "epoch": 0.6514657980456026,
+      "grad_norm": 0.6234082579612732,
+      "learning_rate": 0.0002,
+      "loss": 1.6747,
+      "step": 900
+    },
+    {
+      "epoch": 0.6587043069127759,
+      "grad_norm": 0.7623258233070374,
+      "learning_rate": 0.0002,
+      "loss": 1.6963,
+      "step": 910
+    },
+    {
+      "epoch": 0.6659428157799493,
+      "grad_norm": 0.6148061752319336,
+      "learning_rate": 0.0002,
+      "loss": 1.2424,
+      "step": 920
+    },
+    {
+      "epoch": 0.6731813246471227,
+      "grad_norm": 0.6682973504066467,
+      "learning_rate": 0.0002,
+      "loss": 1.4319,
+      "step": 930
+    },
+    {
+      "epoch": 0.680419833514296,
+      "grad_norm": 0.5513041615486145,
+      "learning_rate": 0.0002,
+      "loss": 1.5377,
+      "step": 940
+    },
+    {
+      "epoch": 0.6876583423814694,
+      "grad_norm": 0.5197525024414062,
+      "learning_rate": 0.0002,
+      "loss": 1.3991,
+      "step": 950
+    },
+    {
+      "epoch": 0.6948968512486428,
+      "grad_norm": 0.6490758061408997,
+      "learning_rate": 0.0002,
+      "loss": 1.4398,
+      "step": 960
+    },
+    {
+      "epoch": 0.7021353601158161,
+      "grad_norm": 0.6450682878494263,
+      "learning_rate": 0.0002,
+      "loss": 1.5251,
+      "step": 970
+    },
+    {
+      "epoch": 0.7093738689829895,
+      "grad_norm": 0.6203766465187073,
+      "learning_rate": 0.0002,
+      "loss": 1.5417,
+      "step": 980
+    },
+    {
+      "epoch": 0.7166123778501629,
+      "grad_norm": 0.6023609638214111,
+      "learning_rate": 0.0002,
+      "loss": 1.4575,
+      "step": 990
+    },
+    {
+      "epoch": 0.7238508867173362,
+      "grad_norm": 0.5765255093574524,
+      "learning_rate": 0.0002,
+      "loss": 1.4973,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7310893955845096,
+      "grad_norm": 0.6650075316429138,
+      "learning_rate": 0.0002,
+      "loss": 1.483,
+      "step": 1010
+    },
+    {
+      "epoch": 0.738327904451683,
+      "grad_norm": 0.5610854029655457,
+      "learning_rate": 0.0002,
+      "loss": 1.5959,
+      "step": 1020
+    },
+    {
+      "epoch": 0.7455664133188563,
+      "grad_norm": 0.7072813510894775,
+      "learning_rate": 0.0002,
+      "loss": 1.5248,
+      "step": 1030
+    },
+    {
+      "epoch": 0.7528049221860297,
+      "grad_norm": 0.6815407872200012,
+      "learning_rate": 0.0002,
+      "loss": 1.5776,
+      "step": 1040
+    },
+    {
+      "epoch": 0.760043431053203,
+      "grad_norm": 0.7932390570640564,
+      "learning_rate": 0.0002,
+      "loss": 1.4577,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7672819399203764,
+      "grad_norm": 0.5798183083534241,
+      "learning_rate": 0.0002,
+      "loss": 1.4515,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7745204487875498,
+      "grad_norm": 0.7898504137992859,
+      "learning_rate": 0.0002,
+      "loss": 1.5053,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7817589576547231,
+      "grad_norm": 0.4983280301094055,
+      "learning_rate": 0.0002,
+      "loss": 1.4776,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7889974665218965,
+      "grad_norm": 0.691403329372406,
+      "learning_rate": 0.0002,
+      "loss": 1.5007,
+      "step": 1090
+    },
+    {
+      "epoch": 0.7962359753890699,
+      "grad_norm": 0.5394481420516968,
+      "learning_rate": 0.0002,
+      "loss": 1.5153,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8034744842562432,
+      "grad_norm": 0.5136822462081909,
+      "learning_rate": 0.0002,
+      "loss": 1.6892,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8107129931234166,
+      "grad_norm": 0.6828126907348633,
+      "learning_rate": 0.0002,
+      "loss": 1.4902,
+      "step": 1120
+    },
+    {
+      "epoch": 0.81795150199059,
+      "grad_norm": 0.6799656748771667,
+      "learning_rate": 0.0002,
+      "loss": 1.4346,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8251900108577633,
+      "grad_norm": 0.5428406000137329,
+      "learning_rate": 0.0002,
+      "loss": 1.2678,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8324285197249367,
+      "grad_norm": 0.4811290502548218,
+      "learning_rate": 0.0002,
+      "loss": 1.4072,
+      "step": 1150
+    },
+    {
+      "epoch": 0.8396670285921101,
+      "grad_norm": 0.5519434809684753,
+      "learning_rate": 0.0002,
+      "loss": 1.4512,
+      "step": 1160
+    },
+    {
+      "epoch": 0.8469055374592834,
+      "grad_norm": 0.9748060703277588,
+      "learning_rate": 0.0002,
+      "loss": 1.4072,
+      "step": 1170
+    },
+    {
+      "epoch": 0.8541440463264568,
+      "grad_norm": 0.712609589099884,
+      "learning_rate": 0.0002,
+      "loss": 1.4309,
+      "step": 1180
+    },
+    {
+      "epoch": 0.8613825551936302,
+      "grad_norm": 0.6866157054901123,
+      "learning_rate": 0.0002,
+      "loss": 1.434,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8686210640608035,
+      "grad_norm": 0.5068854093551636,
+      "learning_rate": 0.0002,
+      "loss": 1.3704,
+      "step": 1200
+    },
+    {
+      "epoch": 0.8758595729279768,
+      "grad_norm": 0.6333245038986206,
+      "learning_rate": 0.0002,
+      "loss": 1.5601,
+      "step": 1210
+    },
+    {
+      "epoch": 0.8830980817951501,
+      "grad_norm": 0.6424421072006226,
+      "learning_rate": 0.0002,
+      "loss": 1.4636,
+      "step": 1220
+    },
+    {
+      "epoch": 0.8903365906623235,
+      "grad_norm": 0.4771921932697296,
+      "learning_rate": 0.0002,
+      "loss": 1.4186,
+      "step": 1230
+    },
+    {
+      "epoch": 0.8975750995294969,
+      "grad_norm": 0.5191764235496521,
+      "learning_rate": 0.0002,
+      "loss": 1.6323,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9048136083966702,
+      "grad_norm": 0.756222128868103,
+      "learning_rate": 0.0002,
+      "loss": 1.6105,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9120521172638436,
+      "grad_norm": 0.623823881149292,
+      "learning_rate": 0.0002,
+      "loss": 1.4396,
+      "step": 1260
+    },
+    {
+      "epoch": 0.919290626131017,
+      "grad_norm": 0.8166571259498596,
+      "learning_rate": 0.0002,
+      "loss": 1.3097,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9265291349981903,
+      "grad_norm": 0.6059346795082092,
+      "learning_rate": 0.0002,
+      "loss": 1.4625,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9337676438653637,
+      "grad_norm": 0.5842690467834473,
+      "learning_rate": 0.0002,
+      "loss": 1.3555,
+      "step": 1290
+    },
+    {
+      "epoch": 0.9410061527325371,
+      "grad_norm": 0.7649800777435303,
+      "learning_rate": 0.0002,
+      "loss": 1.5859,
+      "step": 1300
+    },
+    {
+      "epoch": 0.9482446615997104,
+      "grad_norm": 0.6420919895172119,
+      "learning_rate": 0.0002,
+      "loss": 1.5915,
+      "step": 1310
+    },
+    {
+      "epoch": 0.9554831704668838,
+      "grad_norm": 0.7011452913284302,
+      "learning_rate": 0.0002,
+      "loss": 1.453,
+      "step": 1320
+    },
+    {
+      "epoch": 0.9627216793340572,
+      "grad_norm": 0.5783746242523193,
+      "learning_rate": 0.0002,
+      "loss": 1.6766,
+      "step": 1330
+    },
+    {
+      "epoch": 0.9699601882012305,
+      "grad_norm": 0.5973192453384399,
+      "learning_rate": 0.0002,
+      "loss": 1.6308,
+      "step": 1340
+    },
+    {
+      "epoch": 0.9771986970684039,
+      "grad_norm": 0.6181833744049072,
+      "learning_rate": 0.0002,
+      "loss": 1.5901,
+      "step": 1350
+    },
+    {
+      "epoch": 0.9844372059355773,
+      "grad_norm": 0.5563396215438843,
+      "learning_rate": 0.0002,
+      "loss": 1.5258,
+      "step": 1360
+    },
+    {
+      "epoch": 0.9916757148027506,
+      "grad_norm": 0.45723360776901245,
+      "learning_rate": 0.0002,
+      "loss": 1.4508,
+      "step": 1370
+    },
+    {
+      "epoch": 0.998914223669924,
+      "grad_norm": 0.5947498679161072,
+      "learning_rate": 0.0002,
+      "loss": 1.3291,
+      "step": 1380
+    },
+    {
+      "epoch": 0.9996380745566413,
+      "eval_loss": 1.480796456336975,
+      "eval_runtime": 27.3103,
+      "eval_samples_per_second": 15.965,
+      "eval_steps_per_second": 2.014,
+      "step": 1381
+    },
+    {
+      "epoch": 1.0061527325370974,
+      "grad_norm": 0.5599952936172485,
+      "learning_rate": 0.0002,
+      "loss": 1.3057,
+      "step": 1390
+    },
+    {
+      "epoch": 1.0133912414042707,
+      "grad_norm": 0.5932008028030396,
+      "learning_rate": 0.0002,
+      "loss": 1.4991,
+      "step": 1400
+    },
+    {
+      "epoch": 1.020629750271444,
+      "grad_norm": 0.6194121837615967,
+      "learning_rate": 0.0002,
+      "loss": 1.4506,
+      "step": 1410
+    },
+    {
+      "epoch": 1.0278682591386175,
+      "grad_norm": 0.6995621919631958,
+      "learning_rate": 0.0002,
+      "loss": 1.5966,
+      "step": 1420
+    },
+    {
+      "epoch": 1.0351067680057908,
+      "grad_norm": 0.7905810475349426,
+      "learning_rate": 0.0002,
+      "loss": 1.4153,
+      "step": 1430
+    },
+    {
+      "epoch": 1.0423452768729642,
+      "grad_norm": 0.7221615314483643,
+      "learning_rate": 0.0002,
+      "loss": 1.4414,
+      "step": 1440
+    },
+    {
+      "epoch": 1.0495837857401376,
+      "grad_norm": 0.6170642375946045,
+      "learning_rate": 0.0002,
+      "loss": 1.3859,
+      "step": 1450
+    },
+    {
+      "epoch": 1.056822294607311,
+      "grad_norm": 0.5844094753265381,
+      "learning_rate": 0.0002,
+      "loss": 1.3806,
+      "step": 1460
+    },
+    {
+      "epoch": 1.0640608034744843,
+      "grad_norm": 0.7731822729110718,
+      "learning_rate": 0.0002,
+      "loss": 1.4871,
+      "step": 1470
+    },
+    {
+      "epoch": 1.0712993123416577,
+      "grad_norm": 0.4554748237133026,
+      "learning_rate": 0.0002,
+      "loss": 1.4286,
+      "step": 1480
+    },
+    {
+      "epoch": 1.078537821208831,
+      "grad_norm": 0.6923259496688843,
+      "learning_rate": 0.0002,
+      "loss": 1.3977,
+      "step": 1490
+    },
+    {
+      "epoch": 1.0857763300760044,
+      "grad_norm": 0.6008219122886658,
+      "learning_rate": 0.0002,
+      "loss": 1.3936,
+      "step": 1500
+    },
+    {
+      "epoch": 1.0930148389431777,
+      "grad_norm": 0.6450045704841614,
+      "learning_rate": 0.0002,
+      "loss": 1.4821,
+      "step": 1510
+    },
+    {
+      "epoch": 1.1002533478103511,
+      "grad_norm": 0.7833753824234009,
+      "learning_rate": 0.0002,
+      "loss": 1.3295,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1074918566775245,
+      "grad_norm": 0.5076758861541748,
+      "learning_rate": 0.0002,
+      "loss": 1.3424,
+      "step": 1530
+    },
+    {
+      "epoch": 1.1147303655446978,
+      "grad_norm": 0.5661332011222839,
+      "learning_rate": 0.0002,
+      "loss": 1.4043,
+      "step": 1540
+    },
+    {
+      "epoch": 1.1219688744118712,
+      "grad_norm": 0.6526919603347778,
+      "learning_rate": 0.0002,
+      "loss": 1.4963,
+      "step": 1550
+    },
+    {
+      "epoch": 1.1292073832790446,
+      "grad_norm": 0.5613082647323608,
+      "learning_rate": 0.0002,
+      "loss": 1.3671,
+      "step": 1560
+    },
+    {
+      "epoch": 1.136445892146218,
+      "grad_norm": 0.6113885641098022,
+      "learning_rate": 0.0002,
+      "loss": 1.4458,
+      "step": 1570
+    },
+    {
+      "epoch": 1.1436844010133913,
+      "grad_norm": 0.6732510328292847,
+      "learning_rate": 0.0002,
+      "loss": 1.3552,
+      "step": 1580
+    },
+    {
+      "epoch": 1.1509229098805647,
+      "grad_norm": 0.6146392226219177,
+      "learning_rate": 0.0002,
+      "loss": 1.3114,
+      "step": 1590
+    },
+    {
+      "epoch": 1.158161418747738,
+      "grad_norm": 0.6766974329948425,
+      "learning_rate": 0.0002,
+      "loss": 1.411,
+      "step": 1600
+    },
+    {
+      "epoch": 1.1653999276149114,
+      "grad_norm": 0.7621957659721375,
+      "learning_rate": 0.0002,
+      "loss": 1.2401,
+      "step": 1610
+    },
+    {
+      "epoch": 1.1726384364820848,
+      "grad_norm": 0.6959581971168518,
+      "learning_rate": 0.0002,
+      "loss": 1.3758,
+      "step": 1620
+    },
+    {
+      "epoch": 1.1798769453492581,
+      "grad_norm": 0.6691278219223022,
+      "learning_rate": 0.0002,
+      "loss": 1.382,
+      "step": 1630
+    },
+    {
+      "epoch": 1.1871154542164315,
+      "grad_norm": 0.4927774965763092,
+      "learning_rate": 0.0002,
+      "loss": 1.4147,
+      "step": 1640
+    },
+    {
+      "epoch": 1.1943539630836049,
+      "grad_norm": 0.7724234461784363,
+      "learning_rate": 0.0002,
+      "loss": 1.449,
+      "step": 1650
+    },
+    {
+      "epoch": 1.2015924719507782,
+      "grad_norm": 0.6817787885665894,
+      "learning_rate": 0.0002,
+      "loss": 1.4778,
+      "step": 1660
+    },
+    {
+      "epoch": 1.2088309808179516,
+      "grad_norm": 0.6500699520111084,
+      "learning_rate": 0.0002,
+      "loss": 1.3776,
+      "step": 1670
+    },
+    {
+      "epoch": 1.216069489685125,
+      "grad_norm": 0.5703568458557129,
+      "learning_rate": 0.0002,
+      "loss": 1.3875,
+      "step": 1680
+    },
+    {
+      "epoch": 1.2233079985522983,
+      "grad_norm": 0.6261579990386963,
+      "learning_rate": 0.0002,
+      "loss": 1.4735,
+      "step": 1690
+    },
+    {
+      "epoch": 1.2305465074194717,
+      "grad_norm": 0.651713490486145,
+      "learning_rate": 0.0002,
+      "loss": 1.3898,
+      "step": 1700
+    },
+    {
+      "epoch": 1.237785016286645,
+      "grad_norm": 0.684399425983429,
+      "learning_rate": 0.0002,
+      "loss": 1.4002,
+      "step": 1710
+    },
+    {
+      "epoch": 1.2450235251538184,
+      "grad_norm": 0.6996857523918152,
+      "learning_rate": 0.0002,
+      "loss": 1.5027,
+      "step": 1720
+    },
+    {
+      "epoch": 1.2522620340209918,
+      "grad_norm": 0.7102537751197815,
+      "learning_rate": 0.0002,
+      "loss": 1.3326,
+      "step": 1730
+    },
+    {
+      "epoch": 1.2595005428881652,
+      "grad_norm": 0.45809897780418396,
+      "learning_rate": 0.0002,
+      "loss": 1.3675,
+      "step": 1740
+    },
+    {
+      "epoch": 1.2667390517553385,
+      "grad_norm": 0.6377046704292297,
+      "learning_rate": 0.0002,
+      "loss": 1.4175,
+      "step": 1750
+    },
+    {
+      "epoch": 1.2739775606225119,
+      "grad_norm": 0.6965704560279846,
+      "learning_rate": 0.0002,
+      "loss": 1.3479,
+      "step": 1760
+    },
+    {
+      "epoch": 1.2812160694896852,
+      "grad_norm": 0.5688214302062988,
+      "learning_rate": 0.0002,
+      "loss": 1.5647,
+      "step": 1770
+    },
+    {
+      "epoch": 1.2884545783568586,
+      "grad_norm": 0.6384190320968628,
+      "learning_rate": 0.0002,
+      "loss": 1.3967,
+      "step": 1780
+    },
+    {
+      "epoch": 1.295693087224032,
+      "grad_norm": 0.5629363656044006,
+      "learning_rate": 0.0002,
+      "loss": 1.3671,
+      "step": 1790
+    },
+    {
+      "epoch": 1.3029315960912053,
+      "grad_norm": 0.6148255467414856,
+      "learning_rate": 0.0002,
+      "loss": 1.2292,
+      "step": 1800
+    },
+    {
+      "epoch": 1.3101701049583787,
+      "grad_norm": 0.655580997467041,
+      "learning_rate": 0.0002,
+      "loss": 1.5806,
+      "step": 1810
+    },
+    {
+      "epoch": 1.3174086138255519,
+      "grad_norm": 0.5642657279968262,
+      "learning_rate": 0.0002,
+      "loss": 1.2398,
+      "step": 1820
+    },
+    {
+      "epoch": 1.3246471226927252,
+      "grad_norm": 0.59607994556427,
+      "learning_rate": 0.0002,
+      "loss": 1.3246,
+      "step": 1830
+    },
+    {
+      "epoch": 1.3318856315598986,
+      "grad_norm": 0.5564199090003967,
+      "learning_rate": 0.0002,
+      "loss": 1.3274,
+      "step": 1840
+    },
+    {
+      "epoch": 1.339124140427072,
+      "grad_norm": 0.6949955821037292,
+      "learning_rate": 0.0002,
+      "loss": 1.5834,
+      "step": 1850
+    },
+    {
+      "epoch": 1.3463626492942453,
+      "grad_norm": 0.7036856412887573,
+      "learning_rate": 0.0002,
+      "loss": 1.4722,
+      "step": 1860
+    },
+    {
+      "epoch": 1.3536011581614187,
+      "grad_norm": 0.722062885761261,
+      "learning_rate": 0.0002,
+      "loss": 1.333,
+      "step": 1870
+    },
+    {
+      "epoch": 1.360839667028592,
+      "grad_norm": 0.6098677515983582,
+      "learning_rate": 0.0002,
+      "loss": 1.4044,
+      "step": 1880
+    },
+    {
+      "epoch": 1.3680781758957654,
+      "grad_norm": 0.5376402735710144,
+      "learning_rate": 0.0002,
+      "loss": 1.6217,
+      "step": 1890
+    },
+    {
+      "epoch": 1.3753166847629388,
+      "grad_norm": 0.6974610090255737,
+      "learning_rate": 0.0002,
+      "loss": 1.5071,
+      "step": 1900
+    },
+    {
+      "epoch": 1.3825551936301121,
+      "grad_norm": 0.6520763635635376,
+      "learning_rate": 0.0002,
+      "loss": 1.5854,
+      "step": 1910
+    },
+    {
+      "epoch": 1.3897937024972855,
+      "grad_norm": 0.6604374647140503,
+      "learning_rate": 0.0002,
+      "loss": 1.4271,
+      "step": 1920
+    },
+    {
+      "epoch": 1.3970322113644589,
+      "grad_norm": 0.7364398241043091,
+      "learning_rate": 0.0002,
+      "loss": 1.419,
+      "step": 1930
+    },
+    {
+      "epoch": 1.4042707202316322,
+      "grad_norm": 0.6849475502967834,
+      "learning_rate": 0.0002,
+      "loss": 1.4585,
+      "step": 1940
+    },
+    {
+      "epoch": 1.4115092290988056,
+      "grad_norm": 0.6562670469284058,
+      "learning_rate": 0.0002,
+      "loss": 1.5577,
+      "step": 1950
+    },
+    {
+      "epoch": 1.418747737965979,
+      "grad_norm": 0.5695616006851196,
+      "learning_rate": 0.0002,
+      "loss": 1.4725,
+      "step": 1960
+    },
+    {
+      "epoch": 1.4259862468331523,
+      "grad_norm": 0.5244464874267578,
+      "learning_rate": 0.0002,
+      "loss": 1.3088,
+      "step": 1970
+    },
+    {
+      "epoch": 1.4332247557003257,
+      "grad_norm": 0.6347293257713318,
+      "learning_rate": 0.0002,
+      "loss": 1.5069,
+      "step": 1980
+    },
+    {
+      "epoch": 1.440463264567499,
+      "grad_norm": 0.5528361201286316,
+      "learning_rate": 0.0002,
+      "loss": 1.3502,
+      "step": 1990
+    },
+    {
+      "epoch": 1.4477017734346724,
+      "grad_norm": 0.6987585425376892,
+      "learning_rate": 0.0002,
+      "loss": 1.3978,
+      "step": 2000
+    },
+    {
+      "epoch": 1.4549402823018458,
+      "grad_norm": 0.6568987369537354,
+      "learning_rate": 0.0002,
+      "loss": 1.4262,
+      "step": 2010
+    },
+    {
+      "epoch": 1.4621787911690192,
+      "grad_norm": 0.7665994763374329,
+      "learning_rate": 0.0002,
+      "loss": 1.4175,
+      "step": 2020
+    },
+    {
+      "epoch": 1.4694173000361925,
+      "grad_norm": 0.5127707123756409,
+      "learning_rate": 0.0002,
+      "loss": 1.244,
+      "step": 2030
+    },
+    {
+      "epoch": 1.476655808903366,
+      "grad_norm": 0.5406824946403503,
+      "learning_rate": 0.0002,
+      "loss": 1.3699,
+      "step": 2040
+    },
+    {
+      "epoch": 1.4838943177705393,
+      "grad_norm": 0.5990166664123535,
+      "learning_rate": 0.0002,
+      "loss": 1.3353,
+      "step": 2050
+    },
+    {
+      "epoch": 1.4911328266377126,
+      "grad_norm": 0.6186193823814392,
+      "learning_rate": 0.0002,
+      "loss": 1.2454,
+      "step": 2060
+    },
+    {
+      "epoch": 1.498371335504886,
+      "grad_norm": 0.6154307126998901,
+      "learning_rate": 0.0002,
+      "loss": 1.428,
+      "step": 2070
+    },
+    {
+      "epoch": 1.5056098443720594,
+      "grad_norm": 0.5606056451797485,
+      "learning_rate": 0.0002,
+      "loss": 1.4528,
+      "step": 2080
+    },
+    {
+      "epoch": 1.5128483532392327,
+      "grad_norm": 0.5006417036056519,
+      "learning_rate": 0.0002,
+      "loss": 1.2405,
+      "step": 2090
+    },
+    {
+      "epoch": 1.520086862106406,
+      "grad_norm": 0.5968486070632935,
+      "learning_rate": 0.0002,
+      "loss": 1.4258,
+      "step": 2100
+    },
+    {
+      "epoch": 1.5273253709735795,
+      "grad_norm": 0.5835496187210083,
+      "learning_rate": 0.0002,
+      "loss": 1.2752,
+      "step": 2110
+    },
+    {
+      "epoch": 1.5345638798407528,
+      "grad_norm": 0.6753535270690918,
+      "learning_rate": 0.0002,
+      "loss": 1.5443,
+      "step": 2120
+    },
+    {
+      "epoch": 1.5418023887079262,
+      "grad_norm": 0.7299720644950867,
+      "learning_rate": 0.0002,
+      "loss": 1.2139,
+      "step": 2130
+    },
+    {
+      "epoch": 1.5490408975750996,
+      "grad_norm": 0.5105988383293152,
+      "learning_rate": 0.0002,
+      "loss": 1.2364,
+      "step": 2140
+    },
+    {
+      "epoch": 1.556279406442273,
+      "grad_norm": 0.5675431489944458,
+      "learning_rate": 0.0002,
+      "loss": 1.4528,
+      "step": 2150
+    },
+    {
+      "epoch": 1.5635179153094463,
+      "grad_norm": 0.6246723532676697,
+      "learning_rate": 0.0002,
+      "loss": 1.4563,
+      "step": 2160
+    },
+    {
+      "epoch": 1.5707564241766196,
+      "grad_norm": 0.7291720509529114,
+      "learning_rate": 0.0002,
+      "loss": 1.5255,
+      "step": 2170
+    },
+    {
+      "epoch": 1.577994933043793,
+      "grad_norm": 0.678114116191864,
+      "learning_rate": 0.0002,
+      "loss": 1.5432,
+      "step": 2180
+    },
+    {
+      "epoch": 1.5852334419109664,
+      "grad_norm": 0.5136260986328125,
+      "learning_rate": 0.0002,
+      "loss": 1.5212,
+      "step": 2190
+    },
+    {
+      "epoch": 1.5924719507781397,
+      "grad_norm": 0.6359935998916626,
+      "learning_rate": 0.0002,
+      "loss": 1.3271,
+      "step": 2200
+    },
+    {
+      "epoch": 1.599710459645313,
+      "grad_norm": 0.7650278806686401,
+      "learning_rate": 0.0002,
+      "loss": 1.4038,
+      "step": 2210
+    },
+    {
+      "epoch": 1.6069489685124865,
+      "grad_norm": 0.7256110906600952,
+      "learning_rate": 0.0002,
+      "loss": 1.5478,
+      "step": 2220
+    },
+    {
+      "epoch": 1.6141874773796598,
+      "grad_norm": 0.688689649105072,
+      "learning_rate": 0.0002,
+      "loss": 1.4387,
+      "step": 2230
+    },
+    {
+      "epoch": 1.6214259862468332,
+      "grad_norm": 0.6045311093330383,
+      "learning_rate": 0.0002,
+      "loss": 1.4096,
+      "step": 2240
+    },
+    {
+      "epoch": 1.6286644951140063,
+      "grad_norm": 0.7064604163169861,
+      "learning_rate": 0.0002,
+      "loss": 1.4097,
+      "step": 2250
+    },
+    {
+      "epoch": 1.6359030039811797,
+      "grad_norm": 0.5309562087059021,
+      "learning_rate": 0.0002,
+      "loss": 1.3477,
+      "step": 2260
+    },
+    {
+      "epoch": 1.643141512848353,
+      "grad_norm": 0.5687053203582764,
+      "learning_rate": 0.0002,
+      "loss": 1.4022,
+      "step": 2270
+    },
+    {
+      "epoch": 1.6503800217155264,
+      "grad_norm": 0.535872757434845,
+      "learning_rate": 0.0002,
+      "loss": 1.2977,
+      "step": 2280
+    },
+    {
+      "epoch": 1.6576185305826998,
+      "grad_norm": 0.5502381920814514,
+      "learning_rate": 0.0002,
+      "loss": 1.3844,
+      "step": 2290
+    },
+    {
+      "epoch": 1.6648570394498732,
+      "grad_norm": 0.6158602237701416,
+      "learning_rate": 0.0002,
+      "loss": 1.3764,
+      "step": 2300
+    },
+    {
+      "epoch": 1.6720955483170465,
+      "grad_norm": 0.5804675817489624,
+      "learning_rate": 0.0002,
+      "loss": 1.3515,
+      "step": 2310
+    },
+    {
+      "epoch": 1.67933405718422,
+      "grad_norm": 0.600742757320404,
+      "learning_rate": 0.0002,
+      "loss": 1.2532,
+      "step": 2320
+    },
+    {
+      "epoch": 1.6865725660513933,
+      "grad_norm": 0.7101941108703613,
+      "learning_rate": 0.0002,
+      "loss": 1.477,
+      "step": 2330
+    },
+    {
+      "epoch": 1.6938110749185666,
+      "grad_norm": 0.7507809996604919,
+      "learning_rate": 0.0002,
+      "loss": 1.4849,
+      "step": 2340
+    },
+    {
+      "epoch": 1.70104958378574,
+      "grad_norm": 0.768502414226532,
+      "learning_rate": 0.0002,
+      "loss": 1.2703,
+      "step": 2350
+    },
+    {
+      "epoch": 1.7082880926529134,
+      "grad_norm": 0.4801851212978363,
+      "learning_rate": 0.0002,
+      "loss": 1.3332,
+      "step": 2360
+    },
+    {
+      "epoch": 1.7155266015200867,
+      "grad_norm": 0.5322122573852539,
+      "learning_rate": 0.0002,
+      "loss": 1.4158,
+      "step": 2370
+    },
+    {
+      "epoch": 1.72276511038726,
+      "grad_norm": 0.587661862373352,
+      "learning_rate": 0.0002,
+      "loss": 1.4136,
+      "step": 2380
+    },
+    {
+      "epoch": 1.7300036192544335,
+      "grad_norm": 0.6073525547981262,
+      "learning_rate": 0.0002,
+      "loss": 1.3771,
+      "step": 2390
+    },
+    {
+      "epoch": 1.7372421281216068,
+      "grad_norm": 0.6950460076332092,
+      "learning_rate": 0.0002,
+      "loss": 1.2754,
+      "step": 2400
+    },
+    {
+      "epoch": 1.7444806369887802,
+      "grad_norm": 0.5981102585792542,
+      "learning_rate": 0.0002,
+      "loss": 1.3858,
+      "step": 2410
+    },
+    {
+      "epoch": 1.7517191458559536,
+      "grad_norm": 0.544570803642273,
+      "learning_rate": 0.0002,
+      "loss": 1.4075,
+      "step": 2420
+    },
+    {
+      "epoch": 1.758957654723127,
+      "grad_norm": 0.5304399728775024,
+      "learning_rate": 0.0002,
+      "loss": 1.3861,
+      "step": 2430
+    },
+    {
+      "epoch": 1.7661961635903003,
+      "grad_norm": 0.7921594977378845,
+      "learning_rate": 0.0002,
+      "loss": 1.4244,
+      "step": 2440
+    },
+    {
+      "epoch": 1.7734346724574737,
+      "grad_norm": 0.6084808707237244,
+      "learning_rate": 0.0002,
+      "loss": 1.3053,
+      "step": 2450
+    },
+    {
+      "epoch": 1.780673181324647,
+      "grad_norm": 0.8844701051712036,
+      "learning_rate": 0.0002,
+      "loss": 1.3781,
+      "step": 2460
+    },
+    {
+      "epoch": 1.7879116901918204,
+      "grad_norm": 0.5729258060455322,
+      "learning_rate": 0.0002,
+      "loss": 1.3227,
+      "step": 2470
+    },
+    {
+      "epoch": 1.7951501990589938,
+      "grad_norm": 0.6303611993789673,
+      "learning_rate": 0.0002,
+      "loss": 1.3422,
+      "step": 2480
+    },
+    {
+      "epoch": 1.8023887079261671,
+      "grad_norm": 0.5627942085266113,
+      "learning_rate": 0.0002,
+      "loss": 1.3926,
+      "step": 2490
+    },
+    {
+      "epoch": 1.8096272167933405,
+      "grad_norm": 0.6724274158477783,
+      "learning_rate": 0.0002,
+      "loss": 1.3816,
+      "step": 2500
+    },
+    {
+      "epoch": 1.8168657256605139,
+      "grad_norm": 0.5030826330184937,
+      "learning_rate": 0.0002,
+      "loss": 1.2951,
+      "step": 2510
+    },
+    {
+      "epoch": 1.8241042345276872,
+      "grad_norm": 0.5504099130630493,
+      "learning_rate": 0.0002,
+      "loss": 1.2839,
+      "step": 2520
+    },
+    {
+      "epoch": 1.8313427433948606,
+      "grad_norm": 0.6338945627212524,
+      "learning_rate": 0.0002,
+      "loss": 1.4264,
+      "step": 2530
+    },
+    {
+      "epoch": 1.838581252262034,
+      "grad_norm": 0.5902037620544434,
+      "learning_rate": 0.0002,
+      "loss": 1.563,
+      "step": 2540
+    },
+    {
+      "epoch": 1.8458197611292073,
+      "grad_norm": 0.48814457654953003,
+      "learning_rate": 0.0002,
+      "loss": 1.2961,
+      "step": 2550
+    },
+    {
+      "epoch": 1.8530582699963807,
+      "grad_norm": 0.6216312646865845,
+      "learning_rate": 0.0002,
+      "loss": 1.466,
+      "step": 2560
+    },
+    {
+      "epoch": 1.860296778863554,
+      "grad_norm": 0.635603666305542,
+      "learning_rate": 0.0002,
+      "loss": 1.5123,
+      "step": 2570
+    },
+    {
+      "epoch": 1.8675352877307274,
+      "grad_norm": 0.6938216090202332,
+      "learning_rate": 0.0002,
+      "loss": 1.372,
+      "step": 2580
+    },
+    {
+      "epoch": 1.8747737965979008,
+      "grad_norm": 0.599557638168335,
+      "learning_rate": 0.0002,
+      "loss": 1.5011,
+      "step": 2590
+    },
+    {
+      "epoch": 1.8820123054650741,
+      "grad_norm": 0.564424455165863,
+      "learning_rate": 0.0002,
+      "loss": 1.2714,
+      "step": 2600
+    },
+    {
+      "epoch": 1.8892508143322475,
+      "grad_norm": 0.5430700182914734,
+      "learning_rate": 0.0002,
+      "loss": 1.3403,
+      "step": 2610
+    },
+    {
+      "epoch": 1.8964893231994209,
+      "grad_norm": 0.6150169372558594,
+      "learning_rate": 0.0002,
+      "loss": 1.4347,
+      "step": 2620
+    },
+    {
+      "epoch": 1.9037278320665942,
+      "grad_norm": 0.48159119486808777,
+      "learning_rate": 0.0002,
+      "loss": 1.2474,
+      "step": 2630
+    },
+    {
+      "epoch": 1.9109663409337676,
+      "grad_norm": 0.5608997941017151,
+      "learning_rate": 0.0002,
+      "loss": 1.3716,
+      "step": 2640
+    },
+    {
+      "epoch": 1.918204849800941,
+      "grad_norm": 0.6454501748085022,
+      "learning_rate": 0.0002,
+      "loss": 1.5787,
+      "step": 2650
+    },
+    {
+      "epoch": 1.9254433586681143,
+      "grad_norm": 0.5458073616027832,
+      "learning_rate": 0.0002,
+      "loss": 1.3238,
+      "step": 2660
+    },
+    {
+      "epoch": 1.9326818675352877,
+      "grad_norm": 0.5328490734100342,
+      "learning_rate": 0.0002,
+      "loss": 1.3208,
+      "step": 2670
+    },
+    {
+      "epoch": 1.939920376402461,
+      "grad_norm": 0.6444696187973022,
+      "learning_rate": 0.0002,
+      "loss": 1.4971,
+      "step": 2680
+    },
+    {
+      "epoch": 1.9471588852696344,
+      "grad_norm": 0.7126023769378662,
+      "learning_rate": 0.0002,
+      "loss": 1.5387,
+      "step": 2690
+    },
+    {
+      "epoch": 1.9543973941368078,
+      "grad_norm": 0.5164045095443726,
+      "learning_rate": 0.0002,
+      "loss": 1.3637,
+      "step": 2700
+    },
+    {
+      "epoch": 1.9616359030039812,
+      "grad_norm": 0.5347061157226562,
+      "learning_rate": 0.0002,
+      "loss": 1.5303,
+      "step": 2710
+    },
+    {
+      "epoch": 1.9688744118711545,
+      "grad_norm": 0.5297950506210327,
+      "learning_rate": 0.0002,
+      "loss": 1.2815,
+      "step": 2720
+    },
+    {
+      "epoch": 1.976112920738328,
+      "grad_norm": 0.6537790298461914,
+      "learning_rate": 0.0002,
+      "loss": 1.3566,
+      "step": 2730
+    },
+    {
+      "epoch": 1.9833514296055013,
+      "grad_norm": 0.5536222457885742,
+      "learning_rate": 0.0002,
+      "loss": 1.332,
+      "step": 2740
+    },
+    {
+      "epoch": 1.9905899384726746,
+      "grad_norm": 0.4856105446815491,
+      "learning_rate": 0.0002,
+      "loss": 1.3333,
+      "step": 2750
+    },
+    {
+      "epoch": 1.997828447339848,
+      "grad_norm": 0.6642730832099915,
+      "learning_rate": 0.0002,
+      "loss": 1.3521,
+      "step": 2760
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.4366681575775146,
+      "eval_runtime": 27.3729,
+      "eval_samples_per_second": 15.928,
+      "eval_steps_per_second": 2.009,
+      "step": 2763
+    },
+    {
+      "epoch": 2.0050669562070214,
+      "grad_norm": 0.740253210067749,
+      "learning_rate": 0.0002,
+      "loss": 1.4322,
+      "step": 2770
+    },
+    {
+      "epoch": 2.0123054650741947,
+      "grad_norm": 0.5826276540756226,
+      "learning_rate": 0.0002,
+      "loss": 1.277,
+      "step": 2780
+    },
+    {
+      "epoch": 2.019543973941368,
+      "grad_norm": 0.607356071472168,
+      "learning_rate": 0.0002,
+      "loss": 1.2424,
+      "step": 2790
+    },
+    {
+      "epoch": 2.0267824828085415,
+      "grad_norm": 0.5918063521385193,
+      "learning_rate": 0.0002,
+      "loss": 1.2601,
+      "step": 2800
+    },
+    {
+      "epoch": 2.034020991675715,
+      "grad_norm": 0.5610089898109436,
+      "learning_rate": 0.0002,
+      "loss": 1.3715,
+      "step": 2810
+    },
+    {
+      "epoch": 2.041259500542888,
+      "grad_norm": 0.5869926810264587,
+      "learning_rate": 0.0002,
+      "loss": 1.2092,
+      "step": 2820
+    },
+    {
+      "epoch": 2.0484980094100615,
+      "grad_norm": 0.5753467679023743,
+      "learning_rate": 0.0002,
+      "loss": 1.1929,
+      "step": 2830
+    },
+    {
+      "epoch": 2.055736518277235,
+      "grad_norm": 0.7096508145332336,
+      "learning_rate": 0.0002,
+      "loss": 1.333,
+      "step": 2840
+    },
+    {
+      "epoch": 2.0629750271444083,
+      "grad_norm": 0.7653635144233704,
+      "learning_rate": 0.0002,
+      "loss": 1.1766,
+      "step": 2850
+    },
+    {
+      "epoch": 2.0702135360115816,
+      "grad_norm": 0.6202841997146606,
+      "learning_rate": 0.0002,
+      "loss": 1.2331,
+      "step": 2860
+    },
+    {
+      "epoch": 2.077452044878755,
+      "grad_norm": 0.6810227632522583,
+      "learning_rate": 0.0002,
+      "loss": 1.3298,
+      "step": 2870
+    },
+    {
+      "epoch": 2.0846905537459284,
+      "grad_norm": 0.7481493353843689,
+      "learning_rate": 0.0002,
+      "loss": 1.2505,
+      "step": 2880
+    },
+    {
+      "epoch": 2.0919290626131017,
+      "grad_norm": 0.7089637517929077,
+      "learning_rate": 0.0002,
+      "loss": 1.2484,
+      "step": 2890
+    },
+    {
+      "epoch": 2.099167571480275,
+      "grad_norm": 0.7472923398017883,
+      "learning_rate": 0.0002,
+      "loss": 1.3095,
+      "step": 2900
+    },
+    {
+      "epoch": 2.1064060803474485,
+      "grad_norm": 0.8135465979576111,
+      "learning_rate": 0.0002,
+      "loss": 1.304,
+      "step": 2910
+    },
+    {
+      "epoch": 2.113644589214622,
+      "grad_norm": 0.6097133159637451,
+      "learning_rate": 0.0002,
+      "loss": 1.273,
+      "step": 2920
+    },
+    {
+      "epoch": 2.120883098081795,
+      "grad_norm": 0.5970117449760437,
+      "learning_rate": 0.0002,
+      "loss": 1.3384,
+      "step": 2930
+    },
+    {
+      "epoch": 2.1281216069489686,
+      "grad_norm": 0.6169309616088867,
+      "learning_rate": 0.0002,
+      "loss": 1.3233,
+      "step": 2940
+    },
+    {
+      "epoch": 2.135360115816142,
+      "grad_norm": 0.9428738355636597,
+      "learning_rate": 0.0002,
+      "loss": 1.4246,
+      "step": 2950
+    },
+    {
+      "epoch": 2.1425986246833153,
+      "grad_norm": 0.5671679973602295,
+      "learning_rate": 0.0002,
+      "loss": 1.3527,
+      "step": 2960
+    },
+    {
+      "epoch": 2.1498371335504887,
+      "grad_norm": 0.7007262110710144,
+      "learning_rate": 0.0002,
+      "loss": 1.1375,
+      "step": 2970
+    },
+    {
+      "epoch": 2.157075642417662,
+      "grad_norm": 0.6294044256210327,
+      "learning_rate": 0.0002,
+      "loss": 1.2015,
+      "step": 2980
+    },
+    {
+      "epoch": 2.1643141512848354,
+      "grad_norm": 0.6105241775512695,
+      "learning_rate": 0.0002,
+      "loss": 1.2167,
+      "step": 2990
+    },
+    {
+      "epoch": 2.1715526601520088,
+      "grad_norm": 0.557124137878418,
+      "learning_rate": 0.0002,
+      "loss": 1.2065,
+      "step": 3000
+    },
+    {
+      "epoch": 2.178791169019182,
+      "grad_norm": 0.6250392198562622,
+      "learning_rate": 0.0002,
+      "loss": 1.2515,
+      "step": 3010
+    },
+    {
+      "epoch": 2.1860296778863555,
+      "grad_norm": 0.645218551158905,
+      "learning_rate": 0.0002,
+      "loss": 1.385,
+      "step": 3020
+    },
+    {
+      "epoch": 2.193268186753529,
+      "grad_norm": 0.9033605456352234,
+      "learning_rate": 0.0002,
+      "loss": 1.3928,
+      "step": 3030
+    },
+    {
+      "epoch": 2.2005066956207022,
+      "grad_norm": 0.5325747132301331,
+      "learning_rate": 0.0002,
+      "loss": 1.2458,
+      "step": 3040
+    },
+    {
+      "epoch": 2.2077452044878756,
+      "grad_norm": 0.6334700584411621,
+      "learning_rate": 0.0002,
+      "loss": 1.261,
+      "step": 3050
+    },
+    {
+      "epoch": 2.214983713355049,
+      "grad_norm": 0.5206325054168701,
+      "learning_rate": 0.0002,
+      "loss": 1.2385,
+      "step": 3060
+    },
+    {
+      "epoch": 2.2222222222222223,
+      "grad_norm": 0.5987200140953064,
+      "learning_rate": 0.0002,
+      "loss": 1.3103,
+      "step": 3070
+    },
+    {
+      "epoch": 2.2294607310893957,
+      "grad_norm": 0.5893264412879944,
+      "learning_rate": 0.0002,
+      "loss": 1.1756,
+      "step": 3080
+    },
+    {
+      "epoch": 2.236699239956569,
+      "grad_norm": 0.6869237422943115,
+      "learning_rate": 0.0002,
+      "loss": 1.235,
+      "step": 3090
+    },
+    {
+      "epoch": 2.2439377488237424,
+      "grad_norm": 0.5040048360824585,
+      "learning_rate": 0.0002,
+      "loss": 1.3285,
+      "step": 3100
+    },
+    {
+      "epoch": 2.251176257690916,
+      "grad_norm": 0.6660613417625427,
+      "learning_rate": 0.0002,
+      "loss": 1.3316,
+      "step": 3110
+    },
+    {
+      "epoch": 2.258414766558089,
+      "grad_norm": 0.5890918970108032,
+      "learning_rate": 0.0002,
+      "loss": 1.3108,
+      "step": 3120
+    },
+    {
+      "epoch": 2.2656532754252625,
+      "grad_norm": 0.6458896994590759,
+      "learning_rate": 0.0002,
+      "loss": 1.248,
+      "step": 3130
+    },
+    {
+      "epoch": 2.272891784292436,
+      "grad_norm": 0.6832690834999084,
+      "learning_rate": 0.0002,
+      "loss": 1.4151,
+      "step": 3140
+    },
+    {
+      "epoch": 2.2801302931596092,
+      "grad_norm": 0.833908200263977,
+      "learning_rate": 0.0002,
+      "loss": 1.4458,
+      "step": 3150
+    },
+    {
+      "epoch": 2.2873688020267826,
+      "grad_norm": 0.4596034586429596,
+      "learning_rate": 0.0002,
+      "loss": 1.2931,
+      "step": 3160
+    },
+    {
+      "epoch": 2.294607310893956,
+      "grad_norm": 0.9130966067314148,
+      "learning_rate": 0.0002,
+      "loss": 1.449,
+      "step": 3170
+    },
+    {
+      "epoch": 2.3018458197611293,
+      "grad_norm": 0.7143292427062988,
+      "learning_rate": 0.0002,
+      "loss": 1.3806,
+      "step": 3180
+    },
+    {
+      "epoch": 2.3090843286283027,
+      "grad_norm": 0.5388900637626648,
+      "learning_rate": 0.0002,
+      "loss": 1.2692,
+      "step": 3190
+    },
+    {
+      "epoch": 2.316322837495476,
+      "grad_norm": 0.5607513189315796,
+      "learning_rate": 0.0002,
+      "loss": 1.2402,
+      "step": 3200
+    },
+    {
+      "epoch": 2.3235613463626494,
+      "grad_norm": 0.6795142292976379,
+      "learning_rate": 0.0002,
+      "loss": 1.3874,
+      "step": 3210
+    },
+    {
+      "epoch": 2.330799855229823,
+      "grad_norm": 0.6561070680618286,
+      "learning_rate": 0.0002,
+      "loss": 1.3042,
+      "step": 3220
+    },
+    {
+      "epoch": 2.338038364096996,
+      "grad_norm": 0.8858118057250977,
+      "learning_rate": 0.0002,
+      "loss": 1.4636,
+      "step": 3230
+    },
+    {
+      "epoch": 2.3452768729641695,
+      "grad_norm": 0.6604151725769043,
+      "learning_rate": 0.0002,
+      "loss": 1.3214,
+      "step": 3240
+    },
+    {
+      "epoch": 2.352515381831343,
+      "grad_norm": 0.6755785346031189,
+      "learning_rate": 0.0002,
+      "loss": 1.4004,
+      "step": 3250
+    },
+    {
+      "epoch": 2.3597538906985163,
+      "grad_norm": 0.6981677412986755,
+      "learning_rate": 0.0002,
+      "loss": 1.2503,
+      "step": 3260
+    },
+    {
+      "epoch": 2.3669923995656896,
+      "grad_norm": 0.6338568329811096,
+      "learning_rate": 0.0002,
+      "loss": 1.3078,
+      "step": 3270
+    },
+    {
+      "epoch": 2.374230908432863,
+      "grad_norm": 0.5754265785217285,
+      "learning_rate": 0.0002,
+      "loss": 1.285,
+      "step": 3280
+    },
+    {
+      "epoch": 2.3814694173000364,
+      "grad_norm": 0.7533153295516968,
+      "learning_rate": 0.0002,
+      "loss": 1.2924,
+      "step": 3290
+    },
+    {
+      "epoch": 2.3887079261672097,
+      "grad_norm": 0.675065279006958,
+      "learning_rate": 0.0002,
+      "loss": 1.3711,
+      "step": 3300
+    },
+    {
+      "epoch": 2.395946435034383,
+      "grad_norm": 0.5686452984809875,
+      "learning_rate": 0.0002,
+      "loss": 1.3548,
+      "step": 3310
+    },
+    {
+      "epoch": 2.4031849439015565,
+      "grad_norm": 0.8129481673240662,
+      "learning_rate": 0.0002,
+      "loss": 1.1998,
+      "step": 3320
+    },
+    {
+      "epoch": 2.41042345276873,
+      "grad_norm": 0.6615934371948242,
+      "learning_rate": 0.0002,
+      "loss": 1.2584,
+      "step": 3330
+    },
+    {
+      "epoch": 2.417661961635903,
+      "grad_norm": 0.6678834557533264,
+      "learning_rate": 0.0002,
+      "loss": 1.3691,
+      "step": 3340
+    },
+    {
+      "epoch": 2.4249004705030766,
+      "grad_norm": 0.5581308007240295,
+      "learning_rate": 0.0002,
+      "loss": 1.2381,
+      "step": 3350
+    },
+    {
+      "epoch": 2.43213897937025,
+      "grad_norm": 0.6098920106887817,
+      "learning_rate": 0.0002,
+      "loss": 1.3853,
+      "step": 3360
+    },
+    {
+      "epoch": 2.4393774882374233,
+      "grad_norm": 0.8101736903190613,
+      "learning_rate": 0.0002,
+      "loss": 1.3692,
+      "step": 3370
+    },
+    {
+      "epoch": 2.4466159971045967,
+      "grad_norm": 0.6621488928794861,
+      "learning_rate": 0.0002,
+      "loss": 1.4418,
+      "step": 3380
+    },
+    {
+      "epoch": 2.45385450597177,
+      "grad_norm": 0.8693289160728455,
+      "learning_rate": 0.0002,
+      "loss": 1.4579,
+      "step": 3390
+    },
+    {
+      "epoch": 2.4610930148389434,
+      "grad_norm": 0.6724580526351929,
+      "learning_rate": 0.0002,
+      "loss": 1.3644,
+      "step": 3400
+    },
+    {
+      "epoch": 2.4683315237061167,
+      "grad_norm": 0.6776891946792603,
+      "learning_rate": 0.0002,
+      "loss": 1.2006,
+      "step": 3410
+    },
+    {
+      "epoch": 2.47557003257329,
+      "grad_norm": 0.7214453816413879,
+      "learning_rate": 0.0002,
+      "loss": 1.2937,
+      "step": 3420
+    },
+    {
+      "epoch": 2.4828085414404635,
+      "grad_norm": 0.8390451073646545,
+      "learning_rate": 0.0002,
+      "loss": 1.4051,
+      "step": 3430
+    },
+    {
+      "epoch": 2.490047050307637,
+      "grad_norm": 0.7130982279777527,
+      "learning_rate": 0.0002,
+      "loss": 1.25,
+      "step": 3440
+    },
+    {
+      "epoch": 2.49728555917481,
+      "grad_norm": 0.8873937129974365,
+      "learning_rate": 0.0002,
+      "loss": 1.2231,
+      "step": 3450
+    },
+    {
+      "epoch": 2.5045240680419836,
+      "grad_norm": 0.725185751914978,
+      "learning_rate": 0.0002,
+      "loss": 1.1429,
+      "step": 3460
+    },
+    {
+      "epoch": 2.511762576909157,
+      "grad_norm": 0.6120352149009705,
+      "learning_rate": 0.0002,
+      "loss": 1.2699,
+      "step": 3470
+    },
+    {
+      "epoch": 2.5190010857763303,
+      "grad_norm": 0.7713613510131836,
+      "learning_rate": 0.0002,
+      "loss": 1.2552,
+      "step": 3480
+    },
+    {
+      "epoch": 2.5262395946435037,
+      "grad_norm": 0.895309567451477,
+      "learning_rate": 0.0002,
+      "loss": 1.4648,
+      "step": 3490
+    },
+    {
+      "epoch": 2.533478103510677,
+      "grad_norm": 0.9631021022796631,
+      "learning_rate": 0.0002,
+      "loss": 1.3043,
+      "step": 3500
+    },
+    {
+      "epoch": 2.5407166123778504,
+      "grad_norm": 0.7475683093070984,
+      "learning_rate": 0.0002,
+      "loss": 1.3492,
+      "step": 3510
+    },
+    {
+      "epoch": 2.5479551212450238,
+      "grad_norm": 0.7271341681480408,
+      "learning_rate": 0.0002,
+      "loss": 1.3637,
+      "step": 3520
+    },
+    {
+      "epoch": 2.555193630112197,
+      "grad_norm": 0.6979510188102722,
+      "learning_rate": 0.0002,
+      "loss": 1.304,
+      "step": 3530
+    },
+    {
+      "epoch": 2.5624321389793705,
+      "grad_norm": 0.6504196524620056,
+      "learning_rate": 0.0002,
+      "loss": 1.2353,
+      "step": 3540
+    },
+    {
+      "epoch": 2.569670647846544,
+      "grad_norm": 0.7226675748825073,
+      "learning_rate": 0.0002,
+      "loss": 1.2699,
+      "step": 3550
+    },
+    {
+      "epoch": 2.5769091567137172,
+      "grad_norm": 0.6143222451210022,
+      "learning_rate": 0.0002,
+      "loss": 1.3002,
+      "step": 3560
+    },
+    {
+      "epoch": 2.5841476655808906,
+      "grad_norm": 0.7245154976844788,
+      "learning_rate": 0.0002,
+      "loss": 1.1585,
+      "step": 3570
+    },
+    {
+      "epoch": 2.591386174448064,
+      "grad_norm": 0.943540632724762,
+      "learning_rate": 0.0002,
+      "loss": 1.3651,
+      "step": 3580
+    },
+    {
+      "epoch": 2.5986246833152373,
+      "grad_norm": 0.7707241773605347,
+      "learning_rate": 0.0002,
+      "loss": 1.3034,
+      "step": 3590
+    },
+    {
+      "epoch": 2.6058631921824107,
+      "grad_norm": 0.6705001592636108,
+      "learning_rate": 0.0002,
+      "loss": 1.3063,
+      "step": 3600
+    },
+    {
+      "epoch": 2.613101701049584,
+      "grad_norm": 0.6360933780670166,
+      "learning_rate": 0.0002,
+      "loss": 1.2437,
+      "step": 3610
+    },
+    {
+      "epoch": 2.6203402099167574,
+      "grad_norm": 0.5846424698829651,
+      "learning_rate": 0.0002,
+      "loss": 1.1844,
+      "step": 3620
+    },
+    {
+      "epoch": 2.6275787187839303,
+      "grad_norm": 0.5958625674247742,
+      "learning_rate": 0.0002,
+      "loss": 1.3674,
+      "step": 3630
+    },
+    {
+      "epoch": 2.6348172276511037,
+      "grad_norm": 0.6819243431091309,
+      "learning_rate": 0.0002,
+      "loss": 1.3599,
+      "step": 3640
+    },
+    {
+      "epoch": 2.642055736518277,
+      "grad_norm": 0.7033445835113525,
+      "learning_rate": 0.0002,
+      "loss": 1.3884,
+      "step": 3650
+    },
+    {
+      "epoch": 2.6492942453854504,
+      "grad_norm": 0.6134849786758423,
+      "learning_rate": 0.0002,
+      "loss": 1.3392,
+      "step": 3660
+    },
+    {
+      "epoch": 2.656532754252624,
+      "grad_norm": 0.658009946346283,
+      "learning_rate": 0.0002,
+      "loss": 1.2661,
+      "step": 3670
+    },
+    {
+      "epoch": 2.663771263119797,
+      "grad_norm": 0.6280999779701233,
+      "learning_rate": 0.0002,
+      "loss": 1.3987,
+      "step": 3680
+    },
+    {
+      "epoch": 2.6710097719869705,
+      "grad_norm": 0.5536085963249207,
+      "learning_rate": 0.0002,
+      "loss": 1.2995,
+      "step": 3690
+    },
+    {
+      "epoch": 2.678248280854144,
+      "grad_norm": 0.8603981733322144,
+      "learning_rate": 0.0002,
+      "loss": 1.2044,
+      "step": 3700
+    },
+    {
+      "epoch": 2.6854867897213173,
+      "grad_norm": 0.5509994626045227,
+      "learning_rate": 0.0002,
+      "loss": 1.3879,
+      "step": 3710
+    },
+    {
+      "epoch": 2.6927252985884906,
+      "grad_norm": 0.9093621969223022,
+      "learning_rate": 0.0002,
+      "loss": 1.3253,
+      "step": 3720
+    },
+    {
+      "epoch": 2.699963807455664,
+      "grad_norm": 0.7525952458381653,
+      "learning_rate": 0.0002,
+      "loss": 1.2668,
+      "step": 3730
+    },
+    {
+      "epoch": 2.7072023163228374,
+      "grad_norm": 0.6737023591995239,
+      "learning_rate": 0.0002,
+      "loss": 1.248,
+      "step": 3740
+    },
+    {
+      "epoch": 2.7144408251900107,
+      "grad_norm": 0.8656924962997437,
+      "learning_rate": 0.0002,
+      "loss": 1.2981,
+      "step": 3750
+    },
+    {
+      "epoch": 2.721679334057184,
+      "grad_norm": 0.7494133114814758,
+      "learning_rate": 0.0002,
+      "loss": 1.2342,
+      "step": 3760
+    },
+    {
+      "epoch": 2.7289178429243575,
+      "grad_norm": 0.5725520849227905,
+      "learning_rate": 0.0002,
+      "loss": 1.2417,
+      "step": 3770
+    },
+    {
+      "epoch": 2.736156351791531,
+      "grad_norm": 0.836412787437439,
+      "learning_rate": 0.0002,
+      "loss": 1.28,
+      "step": 3780
+    },
+    {
+      "epoch": 2.743394860658704,
+      "grad_norm": 0.6893242597579956,
+      "learning_rate": 0.0002,
+      "loss": 1.3784,
+      "step": 3790
+    },
+    {
+      "epoch": 2.7506333695258776,
+      "grad_norm": 0.6696223020553589,
+      "learning_rate": 0.0002,
+      "loss": 1.2929,
+      "step": 3800
+    },
+    {
+      "epoch": 2.757871878393051,
+      "grad_norm": 0.6483015418052673,
+      "learning_rate": 0.0002,
+      "loss": 1.2449,
+      "step": 3810
+    },
+    {
+      "epoch": 2.7651103872602243,
+      "grad_norm": 0.8084456920623779,
+      "learning_rate": 0.0002,
+      "loss": 1.3282,
+      "step": 3820
+    },
+    {
+      "epoch": 2.7723488961273977,
+      "grad_norm": 0.6601949334144592,
+      "learning_rate": 0.0002,
+      "loss": 1.3694,
+      "step": 3830
+    },
+    {
+      "epoch": 2.779587404994571,
+      "grad_norm": 0.6905533671379089,
+      "learning_rate": 0.0002,
+      "loss": 1.3568,
+      "step": 3840
+    },
+    {
+      "epoch": 2.7868259138617444,
+      "grad_norm": 0.619318425655365,
+      "learning_rate": 0.0002,
+      "loss": 1.3854,
+      "step": 3850
+    },
+    {
+      "epoch": 2.7940644227289178,
+      "grad_norm": 0.5994023084640503,
+      "learning_rate": 0.0002,
+      "loss": 1.2551,
+      "step": 3860
+    },
+    {
+      "epoch": 2.801302931596091,
+      "grad_norm": 0.5627168416976929,
+      "learning_rate": 0.0002,
+      "loss": 1.2022,
+      "step": 3870
+    },
+    {
+      "epoch": 2.8085414404632645,
+      "grad_norm": 0.6001605987548828,
+      "learning_rate": 0.0002,
+      "loss": 1.3921,
+      "step": 3880
+    },
+    {
+      "epoch": 2.815779949330438,
+      "grad_norm": 0.6022412776947021,
+      "learning_rate": 0.0002,
+      "loss": 1.3026,
+      "step": 3890
+    },
+    {
+      "epoch": 2.823018458197611,
+      "grad_norm": 0.6832426190376282,
+      "learning_rate": 0.0002,
+      "loss": 1.2765,
+      "step": 3900
+    },
+    {
+      "epoch": 2.8302569670647846,
+      "grad_norm": 0.5936811566352844,
+      "learning_rate": 0.0002,
+      "loss": 1.1363,
+      "step": 3910
+    },
+    {
+      "epoch": 2.837495475931958,
+      "grad_norm": 0.6960572600364685,
+      "learning_rate": 0.0002,
+      "loss": 1.1707,
+      "step": 3920
+    },
+    {
+      "epoch": 2.8447339847991313,
+      "grad_norm": 0.5913406610488892,
+      "learning_rate": 0.0002,
+      "loss": 1.4063,
+      "step": 3930
+    },
+    {
+      "epoch": 2.8519724936663047,
+      "grad_norm": 0.678154706954956,
+      "learning_rate": 0.0002,
+      "loss": 1.3245,
+      "step": 3940
+    },
+    {
+      "epoch": 2.859211002533478,
+      "grad_norm": 0.7898936867713928,
+      "learning_rate": 0.0002,
+      "loss": 1.366,
+      "step": 3950
+    },
+    {
+      "epoch": 2.8664495114006514,
+      "grad_norm": 0.9234195351600647,
+      "learning_rate": 0.0002,
+      "loss": 1.3948,
+      "step": 3960
+    },
+    {
+      "epoch": 2.8736880202678248,
+      "grad_norm": 0.5960825085639954,
+      "learning_rate": 0.0002,
+      "loss": 1.2773,
+      "step": 3970
+    },
+    {
+      "epoch": 2.880926529134998,
+      "grad_norm": 0.677118182182312,
+      "learning_rate": 0.0002,
+      "loss": 1.3127,
+      "step": 3980
+    },
+    {
+      "epoch": 2.8881650380021715,
+      "grad_norm": 0.6505142450332642,
+      "learning_rate": 0.0002,
+      "loss": 1.2652,
+      "step": 3990
+    },
+    {
+      "epoch": 2.895403546869345,
+      "grad_norm": 0.550826907157898,
+      "learning_rate": 0.0002,
+      "loss": 1.2078,
+      "step": 4000
+    },
+    {
+      "epoch": 2.9026420557365182,
+      "grad_norm": 0.6209215521812439,
+      "learning_rate": 0.0002,
+      "loss": 1.1811,
+      "step": 4010
+    },
+    {
+      "epoch": 2.9098805646036916,
+      "grad_norm": 0.6549018025398254,
+      "learning_rate": 0.0002,
+      "loss": 1.4001,
+      "step": 4020
+    },
+    {
+      "epoch": 2.917119073470865,
+      "grad_norm": 0.570682168006897,
+      "learning_rate": 0.0002,
+      "loss": 1.2285,
+      "step": 4030
+    },
+    {
+      "epoch": 2.9243575823380383,
+      "grad_norm": 1.1807632446289062,
+      "learning_rate": 0.0002,
+      "loss": 1.0832,
+      "step": 4040
+    },
+    {
+      "epoch": 2.9315960912052117,
+      "grad_norm": 0.7058857679367065,
+      "learning_rate": 0.0002,
+      "loss": 1.2693,
+      "step": 4050
+    },
+    {
+      "epoch": 2.938834600072385,
+      "grad_norm": 0.5542812943458557,
+      "learning_rate": 0.0002,
+      "loss": 1.2905,
+      "step": 4060
+    },
+    {
+      "epoch": 2.9460731089395584,
+      "grad_norm": 0.63167804479599,
+      "learning_rate": 0.0002,
+      "loss": 1.33,
+      "step": 4070
+    },
+    {
+      "epoch": 2.953311617806732,
+      "grad_norm": 0.5702962279319763,
+      "learning_rate": 0.0002,
+      "loss": 1.3075,
+      "step": 4080
+    },
+    {
+      "epoch": 2.960550126673905,
+      "grad_norm": 0.620944082736969,
+      "learning_rate": 0.0002,
+      "loss": 1.2007,
+      "step": 4090
+    },
+    {
+      "epoch": 2.9677886355410785,
+      "grad_norm": 0.5866289734840393,
+      "learning_rate": 0.0002,
+      "loss": 1.2864,
+      "step": 4100
+    },
+    {
+      "epoch": 2.975027144408252,
+      "grad_norm": 0.560170590877533,
+      "learning_rate": 0.0002,
+      "loss": 1.3293,
+      "step": 4110
+    },
+    {
+      "epoch": 2.9822656532754253,
+      "grad_norm": 0.675082802772522,
+      "learning_rate": 0.0002,
+      "loss": 1.2071,
+      "step": 4120
+    },
+    {
+      "epoch": 2.9895041621425986,
+      "grad_norm": 0.62708580493927,
+      "learning_rate": 0.0002,
+      "loss": 1.2981,
+      "step": 4130
+    },
+    {
+      "epoch": 2.996742671009772,
+      "grad_norm": 0.7893929481506348,
+      "learning_rate": 0.0002,
+      "loss": 1.2758,
+      "step": 4140
+    },
+    {
+      "epoch": 2.9996380745566413,
+      "eval_loss": 1.4217946529388428,
+      "eval_runtime": 27.1596,
+      "eval_samples_per_second": 16.053,
+      "eval_steps_per_second": 2.025,
+      "step": 4144
+    },
+    {
+      "epoch": 3.0039811798769454,
+      "grad_norm": 0.7043836116790771,
+      "learning_rate": 0.0002,
+      "loss": 1.2152,
+      "step": 4150
+    },
+    {
+      "epoch": 3.0112196887441187,
+      "grad_norm": 0.6806283593177795,
+      "learning_rate": 0.0002,
+      "loss": 1.1664,
+      "step": 4160
+    },
+    {
+      "epoch": 3.018458197611292,
+      "grad_norm": 0.7684550285339355,
+      "learning_rate": 0.0002,
+      "loss": 1.292,
+      "step": 4170
+    },
+    {
+      "epoch": 3.0256967064784654,
+      "grad_norm": 0.7895237803459167,
+      "learning_rate": 0.0002,
+      "loss": 1.3467,
+      "step": 4180
+    },
+    {
+      "epoch": 3.032935215345639,
+      "grad_norm": 0.7464531064033508,
+      "learning_rate": 0.0002,
+      "loss": 1.1324,
+      "step": 4190
+    },
+    {
+      "epoch": 3.040173724212812,
+      "grad_norm": 0.9358500838279724,
+      "learning_rate": 0.0002,
+      "loss": 1.1614,
+      "step": 4200
+    },
+    {
+      "epoch": 3.0474122330799855,
+      "grad_norm": 1.1066628694534302,
+      "learning_rate": 0.0002,
+      "loss": 1.1834,
+      "step": 4210
+    },
+    {
+      "epoch": 3.054650741947159,
+      "grad_norm": 0.6663267612457275,
+      "learning_rate": 0.0002,
+      "loss": 1.1557,
+      "step": 4220
+    },
+    {
+      "epoch": 3.0618892508143323,
+      "grad_norm": 0.6669464707374573,
+      "learning_rate": 0.0002,
+      "loss": 1.1707,
+      "step": 4230
+    },
+    {
+      "epoch": 3.0691277596815056,
+      "grad_norm": 0.7052164077758789,
+      "learning_rate": 0.0002,
+      "loss": 1.1841,
+      "step": 4240
+    },
+    {
+      "epoch": 3.076366268548679,
+      "grad_norm": 0.6118432879447937,
+      "learning_rate": 0.0002,
+      "loss": 1.2913,
+      "step": 4250
+    },
+    {
+      "epoch": 3.0836047774158524,
+      "grad_norm": 0.6915903687477112,
+      "learning_rate": 0.0002,
+      "loss": 1.1526,
+      "step": 4260
+    },
+    {
+      "epoch": 3.0908432862830257,
+      "grad_norm": 0.7441644668579102,
+      "learning_rate": 0.0002,
+      "loss": 1.1348,
+      "step": 4270
+    },
+    {
+      "epoch": 3.098081795150199,
+      "grad_norm": 0.823850691318512,
+      "learning_rate": 0.0002,
+      "loss": 1.1672,
+      "step": 4280
+    },
+    {
+      "epoch": 3.1053203040173725,
+      "grad_norm": 0.9677883386611938,
+      "learning_rate": 0.0002,
+      "loss": 1.2655,
+      "step": 4290
+    },
+    {
+      "epoch": 3.112558812884546,
+      "grad_norm": 0.7002579569816589,
+      "learning_rate": 0.0002,
+      "loss": 1.1794,
+      "step": 4300
+    },
+    {
+      "epoch": 3.119797321751719,
+      "grad_norm": 0.778789758682251,
+      "learning_rate": 0.0002,
+      "loss": 1.135,
+      "step": 4310
+    },
+    {
+      "epoch": 3.1270358306188926,
+      "grad_norm": 0.7236007452011108,
+      "learning_rate": 0.0002,
+      "loss": 1.0818,
+      "step": 4320
+    },
+    {
+      "epoch": 3.134274339486066,
+      "grad_norm": 0.8809133768081665,
+      "learning_rate": 0.0002,
+      "loss": 1.1803,
+      "step": 4330
+    },
+    {
+      "epoch": 3.1415128483532393,
+      "grad_norm": 0.7924913167953491,
+      "learning_rate": 0.0002,
+      "loss": 1.2571,
+      "step": 4340
+    },
+    {
+      "epoch": 3.1487513572204127,
+      "grad_norm": 0.7437422275543213,
+      "learning_rate": 0.0002,
+      "loss": 1.1413,
+      "step": 4350
+    },
+    {
+      "epoch": 3.155989866087586,
+      "grad_norm": 0.6428450345993042,
+      "learning_rate": 0.0002,
+      "loss": 1.2088,
+      "step": 4360
+    },
+    {
+      "epoch": 3.1632283749547594,
+      "grad_norm": 0.7922873497009277,
+      "learning_rate": 0.0002,
+      "loss": 1.3032,
+      "step": 4370
+    },
+    {
+      "epoch": 3.1704668838219328,
+      "grad_norm": 0.5252506732940674,
+      "learning_rate": 0.0002,
+      "loss": 1.216,
+      "step": 4380
+    },
+    {
+      "epoch": 3.177705392689106,
+      "grad_norm": 0.8570457696914673,
+      "learning_rate": 0.0002,
+      "loss": 1.1297,
+      "step": 4390
+    },
+    {
+      "epoch": 3.1849439015562795,
+      "grad_norm": 0.7218987345695496,
+      "learning_rate": 0.0002,
+      "loss": 1.0994,
+      "step": 4400
+    },
+    {
+      "epoch": 3.192182410423453,
+      "grad_norm": 0.6921393275260925,
+      "learning_rate": 0.0002,
+      "loss": 1.2891,
+      "step": 4410
+    },
+    {
+      "epoch": 3.199420919290626,
+      "grad_norm": 0.7386137843132019,
+      "learning_rate": 0.0002,
+      "loss": 1.2668,
+      "step": 4420
+    },
+    {
+      "epoch": 3.2066594281577996,
+      "grad_norm": 0.6227759122848511,
+      "learning_rate": 0.0002,
+      "loss": 1.1654,
+      "step": 4430
+    },
+    {
+      "epoch": 3.213897937024973,
+      "grad_norm": 0.7180278897285461,
+      "learning_rate": 0.0002,
+      "loss": 1.1752,
+      "step": 4440
+    },
+    {
+      "epoch": 3.2211364458921463,
+      "grad_norm": 0.745830774307251,
+      "learning_rate": 0.0002,
+      "loss": 1.1757,
+      "step": 4450
+    },
+    {
+      "epoch": 3.2283749547593197,
+      "grad_norm": 0.6766072511672974,
+      "learning_rate": 0.0002,
+      "loss": 1.234,
+      "step": 4460
+    },
+    {
+      "epoch": 3.235613463626493,
+      "grad_norm": 0.8325067162513733,
+      "learning_rate": 0.0002,
+      "loss": 1.1999,
+      "step": 4470
+    },
+    {
+      "epoch": 3.2428519724936664,
+      "grad_norm": 0.7148305177688599,
+      "learning_rate": 0.0002,
+      "loss": 1.1606,
+      "step": 4480
+    },
+    {
+      "epoch": 3.25009048136084,
+      "grad_norm": 0.7752676010131836,
+      "learning_rate": 0.0002,
+      "loss": 1.1383,
+      "step": 4490
+    },
+    {
+      "epoch": 3.257328990228013,
+      "grad_norm": 0.6776860952377319,
+      "learning_rate": 0.0002,
+      "loss": 1.3006,
+      "step": 4500
+    },
+    {
+      "epoch": 3.2645674990951865,
+      "grad_norm": 0.704359769821167,
+      "learning_rate": 0.0002,
+      "loss": 1.0796,
+      "step": 4510
+    },
+    {
+      "epoch": 3.27180600796236,
+      "grad_norm": 0.6880282163619995,
+      "learning_rate": 0.0002,
+      "loss": 1.2496,
+      "step": 4520
+    },
+    {
+      "epoch": 3.2790445168295332,
+      "grad_norm": 0.8179270029067993,
+      "learning_rate": 0.0002,
+      "loss": 1.0947,
+      "step": 4530
+    },
+    {
+      "epoch": 3.2862830256967066,
+      "grad_norm": 0.6718448996543884,
+      "learning_rate": 0.0002,
+      "loss": 1.1909,
+      "step": 4540
+    },
+    {
+      "epoch": 3.29352153456388,
+      "grad_norm": 0.8300657868385315,
+      "learning_rate": 0.0002,
+      "loss": 1.2708,
+      "step": 4550
+    },
+    {
+      "epoch": 3.3007600434310533,
+      "grad_norm": 0.6433690786361694,
+      "learning_rate": 0.0002,
+      "loss": 1.2594,
+      "step": 4560
+    },
+    {
+      "epoch": 3.3079985522982267,
+      "grad_norm": 0.690262496471405,
+      "learning_rate": 0.0002,
+      "loss": 1.2479,
+      "step": 4570
+    },
+    {
+      "epoch": 3.3152370611654,
+      "grad_norm": 0.7022852301597595,
+      "learning_rate": 0.0002,
+      "loss": 1.1342,
+      "step": 4580
+    },
+    {
+      "epoch": 3.3224755700325734,
+      "grad_norm": 0.6438387632369995,
+      "learning_rate": 0.0002,
+      "loss": 1.0844,
+      "step": 4590
+    },
+    {
+      "epoch": 3.329714078899747,
+      "grad_norm": 0.6866899132728577,
+      "learning_rate": 0.0002,
+      "loss": 1.17,
+      "step": 4600
+    },
+    {
+      "epoch": 3.33695258776692,
+      "grad_norm": 0.8233968019485474,
+      "learning_rate": 0.0002,
+      "loss": 1.1289,
+      "step": 4610
+    },
+    {
+      "epoch": 3.3441910966340935,
+      "grad_norm": 0.7251574993133545,
+      "learning_rate": 0.0002,
+      "loss": 1.1855,
+      "step": 4620
+    },
+    {
+      "epoch": 3.351429605501267,
+      "grad_norm": 0.7855110168457031,
+      "learning_rate": 0.0002,
+      "loss": 1.3403,
+      "step": 4630
+    },
+    {
+      "epoch": 3.3586681143684403,
+      "grad_norm": 0.8487356305122375,
+      "learning_rate": 0.0002,
+      "loss": 1.2922,
+      "step": 4640
+    },
+    {
+      "epoch": 3.3659066232356136,
+      "grad_norm": 0.6429011225700378,
+      "learning_rate": 0.0002,
+      "loss": 1.2462,
+      "step": 4650
+    },
+    {
+      "epoch": 3.373145132102787,
+      "grad_norm": 0.7095270156860352,
+      "learning_rate": 0.0002,
+      "loss": 1.129,
+      "step": 4660
+    },
+    {
+      "epoch": 3.3803836409699604,
+      "grad_norm": 0.6792303323745728,
+      "learning_rate": 0.0002,
+      "loss": 1.262,
+      "step": 4670
+    },
+    {
+      "epoch": 3.3876221498371337,
+      "grad_norm": 0.6784825921058655,
+      "learning_rate": 0.0002,
+      "loss": 1.256,
+      "step": 4680
+    },
+    {
+      "epoch": 3.394860658704307,
+      "grad_norm": 0.6362888216972351,
+      "learning_rate": 0.0002,
+      "loss": 1.0838,
+      "step": 4690
+    },
+    {
+      "epoch": 3.4020991675714805,
+      "grad_norm": 0.7794778943061829,
+      "learning_rate": 0.0002,
+      "loss": 1.2165,
+      "step": 4700
+    },
+    {
+      "epoch": 3.409337676438654,
+      "grad_norm": 0.7287485003471375,
+      "learning_rate": 0.0002,
+      "loss": 1.0644,
+      "step": 4710
+    },
+    {
+      "epoch": 3.416576185305827,
+      "grad_norm": 0.6481451392173767,
+      "learning_rate": 0.0002,
+      "loss": 1.2925,
+      "step": 4720
+    },
+    {
+      "epoch": 3.4238146941730006,
+      "grad_norm": 0.9200371503829956,
+      "learning_rate": 0.0002,
+      "loss": 1.2121,
+      "step": 4730
+    },
+    {
+      "epoch": 3.431053203040174,
+      "grad_norm": 1.074180245399475,
+      "learning_rate": 0.0002,
+      "loss": 1.072,
+      "step": 4740
+    },
+    {
+      "epoch": 3.438291711907347,
+      "grad_norm": 0.6722986698150635,
+      "learning_rate": 0.0002,
+      "loss": 1.0421,
+      "step": 4750
+    },
+    {
+      "epoch": 3.44553022077452,
+      "grad_norm": 0.7945933938026428,
+      "learning_rate": 0.0002,
+      "loss": 1.2258,
+      "step": 4760
+    },
+    {
+      "epoch": 3.4527687296416936,
+      "grad_norm": 0.7624640464782715,
+      "learning_rate": 0.0002,
+      "loss": 1.0927,
+      "step": 4770
+    },
+    {
+      "epoch": 3.460007238508867,
+      "grad_norm": 0.7763656377792358,
+      "learning_rate": 0.0002,
+      "loss": 1.2428,
+      "step": 4780
+    },
+    {
+      "epoch": 3.4672457473760403,
+      "grad_norm": 0.7736947536468506,
+      "learning_rate": 0.0002,
+      "loss": 1.2584,
+      "step": 4790
+    },
+    {
+      "epoch": 3.4744842562432137,
+      "grad_norm": 0.8450354933738708,
+      "learning_rate": 0.0002,
+      "loss": 1.1953,
+      "step": 4800
+    },
+    {
+      "epoch": 3.481722765110387,
+      "grad_norm": 0.6480133533477783,
+      "learning_rate": 0.0002,
+      "loss": 1.1362,
+      "step": 4810
+    },
+    {
+      "epoch": 3.4889612739775604,
+      "grad_norm": 0.8437445759773254,
+      "learning_rate": 0.0002,
+      "loss": 1.1882,
+      "step": 4820
+    },
+    {
+      "epoch": 3.4961997828447338,
+      "grad_norm": 0.7781730890274048,
+      "learning_rate": 0.0002,
+      "loss": 1.1519,
+      "step": 4830
+    },
+    {
+      "epoch": 3.503438291711907,
+      "grad_norm": 0.8523228168487549,
+      "learning_rate": 0.0002,
+      "loss": 1.1836,
+      "step": 4840
+    },
+    {
+      "epoch": 3.5106768005790805,
+      "grad_norm": 0.6236732006072998,
+      "learning_rate": 0.0002,
+      "loss": 1.1672,
+      "step": 4850
+    },
+    {
+      "epoch": 3.517915309446254,
+      "grad_norm": 0.7500787377357483,
+      "learning_rate": 0.0002,
+      "loss": 1.1926,
+      "step": 4860
+    },
+    {
+      "epoch": 3.5251538183134272,
+      "grad_norm": 0.7665374875068665,
+      "learning_rate": 0.0002,
+      "loss": 1.1998,
+      "step": 4870
+    },
+    {
+      "epoch": 3.5323923271806006,
+      "grad_norm": 0.787857711315155,
+      "learning_rate": 0.0002,
+      "loss": 1.1551,
+      "step": 4880
+    },
+    {
+      "epoch": 3.539630836047774,
+      "grad_norm": 0.970595121383667,
+      "learning_rate": 0.0002,
+      "loss": 1.2758,
+      "step": 4890
+    },
+    {
+      "epoch": 3.5468693449149473,
+      "grad_norm": 0.6409347057342529,
+      "learning_rate": 0.0002,
+      "loss": 1.1274,
+      "step": 4900
+    },
+    {
+      "epoch": 3.5541078537821207,
+      "grad_norm": 0.888551652431488,
+      "learning_rate": 0.0002,
+      "loss": 1.1596,
+      "step": 4910
+    },
+    {
+      "epoch": 3.561346362649294,
+      "grad_norm": 1.0808377265930176,
+      "learning_rate": 0.0002,
+      "loss": 1.1644,
+      "step": 4920
+    },
+    {
+      "epoch": 3.5685848715164674,
+      "grad_norm": 0.7501053214073181,
+      "learning_rate": 0.0002,
+      "loss": 1.2564,
+      "step": 4930
+    },
+    {
+      "epoch": 3.575823380383641,
+      "grad_norm": 0.7375240325927734,
+      "learning_rate": 0.0002,
+      "loss": 1.2351,
+      "step": 4940
+    },
+    {
+      "epoch": 3.583061889250814,
+      "grad_norm": 0.7075039744377136,
+      "learning_rate": 0.0002,
+      "loss": 1.3568,
+      "step": 4950
+    },
+    {
+      "epoch": 3.5903003981179875,
+      "grad_norm": 0.939337432384491,
+      "learning_rate": 0.0002,
+      "loss": 1.3355,
+      "step": 4960
+    },
+    {
+      "epoch": 3.597538906985161,
+      "grad_norm": 0.6717396974563599,
+      "learning_rate": 0.0002,
+      "loss": 1.1722,
+      "step": 4970
+    },
+    {
+      "epoch": 3.6047774158523342,
+      "grad_norm": 0.7141643762588501,
+      "learning_rate": 0.0002,
+      "loss": 1.1186,
+      "step": 4980
+    },
+    {
+      "epoch": 3.6120159247195076,
+      "grad_norm": 0.7109216451644897,
+      "learning_rate": 0.0002,
+      "loss": 1.1011,
+      "step": 4990
+    },
+    {
+      "epoch": 3.619254433586681,
+      "grad_norm": 0.7020776867866516,
+      "learning_rate": 0.0002,
+      "loss": 1.2178,
+      "step": 5000
+    },
+    {
+      "epoch": 3.6264929424538543,
+      "grad_norm": 0.7158873677253723,
+      "learning_rate": 0.0002,
+      "loss": 1.1939,
+      "step": 5010
+    },
+    {
+      "epoch": 3.6337314513210277,
+      "grad_norm": 0.7062035202980042,
+      "learning_rate": 0.0002,
+      "loss": 1.2624,
+      "step": 5020
+    },
+    {
+      "epoch": 3.640969960188201,
+      "grad_norm": 0.7081155776977539,
+      "learning_rate": 0.0002,
+      "loss": 1.0224,
+      "step": 5030
+    },
+    {
+      "epoch": 3.6482084690553744,
+      "grad_norm": 1.2210607528686523,
+      "learning_rate": 0.0002,
+      "loss": 1.2195,
+      "step": 5040
+    },
+    {
+      "epoch": 3.655446977922548,
+      "grad_norm": 0.6650236248970032,
+      "learning_rate": 0.0002,
+      "loss": 1.2596,
+      "step": 5050
+    },
+    {
+      "epoch": 3.662685486789721,
+      "grad_norm": 0.6884829998016357,
+      "learning_rate": 0.0002,
+      "loss": 1.1072,
+      "step": 5060
+    },
+    {
+      "epoch": 3.6699239956568945,
+      "grad_norm": 0.7317819595336914,
+      "learning_rate": 0.0002,
+      "loss": 1.2292,
+      "step": 5070
+    },
+    {
+      "epoch": 3.677162504524068,
+      "grad_norm": 0.7406691908836365,
+      "learning_rate": 0.0002,
+      "loss": 1.1917,
+      "step": 5080
+    },
+    {
+      "epoch": 3.6844010133912413,
+      "grad_norm": 0.9009454250335693,
+      "learning_rate": 0.0002,
+      "loss": 1.2949,
+      "step": 5090
+    },
+    {
+      "epoch": 3.6916395222584146,
+      "grad_norm": 0.8189385533332825,
+      "learning_rate": 0.0002,
+      "loss": 1.1528,
+      "step": 5100
+    },
+    {
+      "epoch": 3.698878031125588,
+      "grad_norm": 1.0793628692626953,
+      "learning_rate": 0.0002,
+      "loss": 1.3408,
+      "step": 5110
+    },
+    {
+      "epoch": 3.7061165399927614,
+      "grad_norm": 0.8593027591705322,
+      "learning_rate": 0.0002,
+      "loss": 1.2417,
+      "step": 5120
+    },
+    {
+      "epoch": 3.7133550488599347,
+      "grad_norm": 0.8481812477111816,
+      "learning_rate": 0.0002,
+      "loss": 1.2141,
+      "step": 5130
+    },
+    {
+      "epoch": 3.720593557727108,
+      "grad_norm": 0.6527451276779175,
+      "learning_rate": 0.0002,
+      "loss": 1.125,
+      "step": 5140
+    },
+    {
+      "epoch": 3.7278320665942815,
+      "grad_norm": 0.9220114350318909,
+      "learning_rate": 0.0002,
+      "loss": 1.1584,
+      "step": 5150
+    },
+    {
+      "epoch": 3.735070575461455,
+      "grad_norm": 1.0842019319534302,
+      "learning_rate": 0.0002,
+      "loss": 1.2267,
+      "step": 5160
+    },
+    {
+      "epoch": 3.742309084328628,
+      "grad_norm": 0.965453565120697,
+      "learning_rate": 0.0002,
+      "loss": 1.3083,
+      "step": 5170
+    },
+    {
+      "epoch": 3.7495475931958016,
+      "grad_norm": 0.9903319478034973,
+      "learning_rate": 0.0002,
+      "loss": 1.1772,
+      "step": 5180
+    },
+    {
+      "epoch": 3.756786102062975,
+      "grad_norm": 0.7434818148612976,
+      "learning_rate": 0.0002,
+      "loss": 1.2515,
+      "step": 5190
+    },
+    {
+      "epoch": 3.7640246109301483,
+      "grad_norm": 0.6717280745506287,
+      "learning_rate": 0.0002,
+      "loss": 1.2631,
+      "step": 5200
+    },
+    {
+      "epoch": 3.7712631197973217,
+      "grad_norm": 0.7754665613174438,
+      "learning_rate": 0.0002,
+      "loss": 1.2012,
+      "step": 5210
+    },
+    {
+      "epoch": 3.778501628664495,
+      "grad_norm": 1.028374433517456,
+      "learning_rate": 0.0002,
+      "loss": 1.305,
+      "step": 5220
+    },
+    {
+      "epoch": 3.7857401375316684,
+      "grad_norm": 0.6026996374130249,
+      "learning_rate": 0.0002,
+      "loss": 1.1866,
+      "step": 5230
+    },
+    {
+      "epoch": 3.7929786463988417,
+      "grad_norm": 0.6978490948677063,
+      "learning_rate": 0.0002,
+      "loss": 1.1901,
+      "step": 5240
+    },
+    {
+      "epoch": 3.800217155266015,
+      "grad_norm": 0.7303446531295776,
+      "learning_rate": 0.0002,
+      "loss": 1.2576,
+      "step": 5250
+    },
+    {
+      "epoch": 3.8074556641331885,
+      "grad_norm": 1.0734210014343262,
+      "learning_rate": 0.0002,
+      "loss": 1.3173,
+      "step": 5260
+    },
+    {
+      "epoch": 3.814694173000362,
+      "grad_norm": 0.6383201479911804,
+      "learning_rate": 0.0002,
+      "loss": 1.1137,
+      "step": 5270
+    },
+    {
+      "epoch": 3.821932681867535,
+      "grad_norm": 0.7742630243301392,
+      "learning_rate": 0.0002,
+      "loss": 1.0904,
+      "step": 5280
+    },
+    {
+      "epoch": 3.8291711907347086,
+      "grad_norm": 0.8477074503898621,
+      "learning_rate": 0.0002,
+      "loss": 1.2232,
+      "step": 5290
+    },
+    {
+      "epoch": 3.836409699601882,
+      "grad_norm": 0.6675317883491516,
+      "learning_rate": 0.0002,
+      "loss": 1.2047,
+      "step": 5300
+    },
+    {
+      "epoch": 3.8436482084690553,
+      "grad_norm": 0.7515445351600647,
+      "learning_rate": 0.0002,
+      "loss": 1.2275,
+      "step": 5310
+    },
+    {
+      "epoch": 3.8508867173362287,
+      "grad_norm": 1.1441220045089722,
+      "learning_rate": 0.0002,
+      "loss": 1.2569,
+      "step": 5320
+    },
+    {
+      "epoch": 3.858125226203402,
+      "grad_norm": 0.7968795895576477,
+      "learning_rate": 0.0002,
+      "loss": 1.1512,
+      "step": 5330
+    },
+    {
+      "epoch": 3.8653637350705754,
+      "grad_norm": 0.7842824459075928,
+      "learning_rate": 0.0002,
+      "loss": 1.232,
+      "step": 5340
+    },
+    {
+      "epoch": 3.8726022439377488,
+      "grad_norm": 0.8272225260734558,
+      "learning_rate": 0.0002,
+      "loss": 1.1847,
+      "step": 5350
+    },
+    {
+      "epoch": 3.879840752804922,
+      "grad_norm": 0.8413397669792175,
+      "learning_rate": 0.0002,
+      "loss": 1.1381,
+      "step": 5360
+    },
+    {
+      "epoch": 3.8870792616720955,
+      "grad_norm": 1.141764760017395,
+      "learning_rate": 0.0002,
+      "loss": 1.2349,
+      "step": 5370
+    },
+    {
+      "epoch": 3.894317770539269,
+      "grad_norm": 0.9826975464820862,
+      "learning_rate": 0.0002,
+      "loss": 1.212,
+      "step": 5380
+    },
+    {
+      "epoch": 3.9015562794064422,
+      "grad_norm": 0.8598255515098572,
+      "learning_rate": 0.0002,
+      "loss": 1.1833,
+      "step": 5390
+    },
+    {
+      "epoch": 3.9087947882736156,
+      "grad_norm": 0.6271058320999146,
+      "learning_rate": 0.0002,
+      "loss": 1.1247,
+      "step": 5400
+    },
+    {
+      "epoch": 3.916033297140789,
+      "grad_norm": 0.6379870772361755,
+      "learning_rate": 0.0002,
+      "loss": 1.2212,
+      "step": 5410
+    },
+    {
+      "epoch": 3.9232718060079623,
+      "grad_norm": 1.0313376188278198,
+      "learning_rate": 0.0002,
+      "loss": 1.2481,
+      "step": 5420
+    },
+    {
+      "epoch": 3.9305103148751357,
+      "grad_norm": 0.8220619559288025,
+      "learning_rate": 0.0002,
+      "loss": 1.1872,
+      "step": 5430
+    },
+    {
+      "epoch": 3.937748823742309,
+      "grad_norm": 0.7576116919517517,
+      "learning_rate": 0.0002,
+      "loss": 1.2006,
+      "step": 5440
+    },
+    {
+      "epoch": 3.9449873326094824,
+      "grad_norm": 1.226235032081604,
+      "learning_rate": 0.0002,
+      "loss": 1.1969,
+      "step": 5450
+    },
+    {
+      "epoch": 3.952225841476656,
+      "grad_norm": 0.7979229688644409,
+      "learning_rate": 0.0002,
+      "loss": 1.2945,
+      "step": 5460
+    },
+    {
+      "epoch": 3.959464350343829,
+      "grad_norm": 0.9911929965019226,
+      "learning_rate": 0.0002,
+      "loss": 1.1922,
+      "step": 5470
+    },
+    {
+      "epoch": 3.9667028592110025,
+      "grad_norm": 0.643738865852356,
+      "learning_rate": 0.0002,
+      "loss": 1.0924,
+      "step": 5480
+    },
+    {
+      "epoch": 3.973941368078176,
+      "grad_norm": 0.682305634021759,
+      "learning_rate": 0.0002,
+      "loss": 1.0607,
+      "step": 5490
+    },
+    {
+      "epoch": 3.9811798769453492,
+      "grad_norm": 1.18373441696167,
+      "learning_rate": 0.0002,
+      "loss": 1.2908,
+      "step": 5500
+    },
+    {
+      "epoch": 3.9884183858125226,
+      "grad_norm": 0.7190203070640564,
+      "learning_rate": 0.0002,
+      "loss": 1.0889,
+      "step": 5510
+    },
+    {
+      "epoch": 3.995656894679696,
+      "grad_norm": 0.7516948580741882,
+      "learning_rate": 0.0002,
+      "loss": 1.2745,
+      "step": 5520
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.4252897500991821,
+      "eval_runtime": 27.235,
+      "eval_samples_per_second": 16.009,
+      "eval_steps_per_second": 2.019,
+      "step": 5526
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 11048,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 6.75387517477847e+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-0.9-num-8828-sd-1/checkpoint-5526/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-0.9-num-8828-sd-1/checkpoint-5526/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..4d7b7431bbbe8c9bf29b925bca391a558af5ff8c
--- /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-0.9-num-8828-sd-1/checkpoint-5526/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ad613885e4f267fc04125f1a836d42cfa796bbe12e536f9ee60c955de02cdb5a
+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-0.9-num-8828-sd-1/checkpoint-6907/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-0.9-num-8828-sd-1/checkpoint-6907/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-0.9-num-8828-sd-1/checkpoint-6907/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-0.9-num-8828-sd-1/checkpoint-6907/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-0.9-num-8828-sd-1/checkpoint-6907/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e99bbcd43df1c19d98706c7e3be95c93844c5349
--- /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-0.9-num-8828-sd-1/checkpoint-6907/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-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-8828-sd-1/checkpoint-6907/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-0.9-num-8828-sd-1/checkpoint-6907/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..93f5b07d9711413b652e622cedfedc2581a14bad
--- /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-0.9-num-8828-sd-1/checkpoint-6907/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a1006d59ff8e2780d4cfe18bd27cdb3f74a23ccb598b0a59612a9fdafd67b6f
+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-0.9-num-8828-sd-1/checkpoint-6907/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-0.9-num-8828-sd-1/checkpoint-6907/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..ba234c4a424a38b21457b82999e51477d6a086ae
--- /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-0.9-num-8828-sd-1/checkpoint-6907/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:34c12c0218085fe8586d9e929d59be855fc8893669fdddf6a47dce5c47b438eb
+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-0.9-num-8828-sd-1/checkpoint-6907/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-0.9-num-8828-sd-1/checkpoint-6907/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..330e758ffbee81e4ab59edde8112c68541222f1b
--- /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-0.9-num-8828-sd-1/checkpoint-6907/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fdac20adea3f3dbdec3792643a10befb0d02333cfe6b985cab6bfc8304efd581
+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-0.9-num-8828-sd-1/checkpoint-6907/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-0.9-num-8828-sd-1/checkpoint-6907/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..fd1fb4610159d98fa0d65c1ddc6944cc68fd2f16
--- /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-0.9-num-8828-sd-1/checkpoint-6907/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8708b5027417181075c393e1f571c167fa59f46aebfcf6dfe740a8679bfb73b7
+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-0.9-num-8828-sd-1/checkpoint-6907/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-0.9-num-8828-sd-1/checkpoint-6907/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-0.9-num-8828-sd-1/checkpoint-6907/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-0.9-num-8828-sd-1/checkpoint-6907/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-0.9-num-8828-sd-1/checkpoint-6907/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-0.9-num-8828-sd-1/checkpoint-6907/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-0.9-num-8828-sd-1/checkpoint-6907/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-0.9-num-8828-sd-1/checkpoint-6907/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-0.9-num-8828-sd-1/checkpoint-6907/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-0.9-num-8828-sd-1/checkpoint-6907/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-0.9-num-8828-sd-1/checkpoint-6907/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-0.9-num-8828-sd-1/checkpoint-6907/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-0.9-num-8828-sd-1/checkpoint-6907/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-0.9-num-8828-sd-1/checkpoint-6907/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..2bfc3a30615d20a2ab2f8a721f687253c4aa449a
--- /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-0.9-num-8828-sd-1/checkpoint-6907/trainer_state.json
@@ -0,0 +1,4903 @@
+{
+  "best_metric": 1.4217946529388428,
+  "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-0.9-num-8828-sd-1/checkpoint-4144",
+  "epoch": 4.999638074556641,
+  "eval_steps": 10,
+  "global_step": 6907,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007238508867173362,
+      "grad_norm": 1.2523442506790161,
+      "learning_rate": 0.0002,
+      "loss": 4.7061,
+      "step": 10
+    },
+    {
+      "epoch": 0.014477017734346724,
+      "grad_norm": 1.8887330293655396,
+      "learning_rate": 0.0002,
+      "loss": 3.3493,
+      "step": 20
+    },
+    {
+      "epoch": 0.021715526601520086,
+      "grad_norm": 0.9668035507202148,
+      "learning_rate": 0.0002,
+      "loss": 2.7585,
+      "step": 30
+    },
+    {
+      "epoch": 0.028954035468693448,
+      "grad_norm": 2.9167306423187256,
+      "learning_rate": 0.0002,
+      "loss": 2.3699,
+      "step": 40
+    },
+    {
+      "epoch": 0.036192544335866814,
+      "grad_norm": 2.649867296218872,
+      "learning_rate": 0.0002,
+      "loss": 2.2679,
+      "step": 50
+    },
+    {
+      "epoch": 0.04343105320304017,
+      "grad_norm": 1.5120655298233032,
+      "learning_rate": 0.0002,
+      "loss": 2.2202,
+      "step": 60
+    },
+    {
+      "epoch": 0.05066956207021354,
+      "grad_norm": 0.7879868149757385,
+      "learning_rate": 0.0002,
+      "loss": 2.2026,
+      "step": 70
+    },
+    {
+      "epoch": 0.057908070937386896,
+      "grad_norm": 0.7616953253746033,
+      "learning_rate": 0.0002,
+      "loss": 1.9447,
+      "step": 80
+    },
+    {
+      "epoch": 0.06514657980456026,
+      "grad_norm": 1.8809149265289307,
+      "learning_rate": 0.0002,
+      "loss": 2.0112,
+      "step": 90
+    },
+    {
+      "epoch": 0.07238508867173363,
+      "grad_norm": 0.9294016361236572,
+      "learning_rate": 0.0002,
+      "loss": 1.8337,
+      "step": 100
+    },
+    {
+      "epoch": 0.07962359753890698,
+      "grad_norm": 0.7145281434059143,
+      "learning_rate": 0.0002,
+      "loss": 1.8419,
+      "step": 110
+    },
+    {
+      "epoch": 0.08686210640608034,
+      "grad_norm": 0.7564446330070496,
+      "learning_rate": 0.0002,
+      "loss": 2.0036,
+      "step": 120
+    },
+    {
+      "epoch": 0.09410061527325371,
+      "grad_norm": 1.1681925058364868,
+      "learning_rate": 0.0002,
+      "loss": 1.9306,
+      "step": 130
+    },
+    {
+      "epoch": 0.10133912414042708,
+      "grad_norm": 0.6708641648292542,
+      "learning_rate": 0.0002,
+      "loss": 1.7875,
+      "step": 140
+    },
+    {
+      "epoch": 0.10857763300760044,
+      "grad_norm": 0.7625647783279419,
+      "learning_rate": 0.0002,
+      "loss": 1.786,
+      "step": 150
+    },
+    {
+      "epoch": 0.11581614187477379,
+      "grad_norm": 0.8463464975357056,
+      "learning_rate": 0.0002,
+      "loss": 1.6687,
+      "step": 160
+    },
+    {
+      "epoch": 0.12305465074194716,
+      "grad_norm": 0.7502335906028748,
+      "learning_rate": 0.0002,
+      "loss": 1.6214,
+      "step": 170
+    },
+    {
+      "epoch": 0.13029315960912052,
+      "grad_norm": 0.6929958462715149,
+      "learning_rate": 0.0002,
+      "loss": 1.7433,
+      "step": 180
+    },
+    {
+      "epoch": 0.1375316684762939,
+      "grad_norm": 0.6798707842826843,
+      "learning_rate": 0.0002,
+      "loss": 1.6009,
+      "step": 190
+    },
+    {
+      "epoch": 0.14477017734346725,
+      "grad_norm": 0.7566508650779724,
+      "learning_rate": 0.0002,
+      "loss": 1.6208,
+      "step": 200
+    },
+    {
+      "epoch": 0.15200868621064062,
+      "grad_norm": 0.7196869850158691,
+      "learning_rate": 0.0002,
+      "loss": 1.5823,
+      "step": 210
+    },
+    {
+      "epoch": 0.15924719507781396,
+      "grad_norm": 0.8401045799255371,
+      "learning_rate": 0.0002,
+      "loss": 1.738,
+      "step": 220
+    },
+    {
+      "epoch": 0.16648570394498732,
+      "grad_norm": 0.8503773212432861,
+      "learning_rate": 0.0002,
+      "loss": 1.7574,
+      "step": 230
+    },
+    {
+      "epoch": 0.1737242128121607,
+      "grad_norm": 0.7183733582496643,
+      "learning_rate": 0.0002,
+      "loss": 1.7861,
+      "step": 240
+    },
+    {
+      "epoch": 0.18096272167933405,
+      "grad_norm": 0.7082605957984924,
+      "learning_rate": 0.0002,
+      "loss": 1.6693,
+      "step": 250
+    },
+    {
+      "epoch": 0.18820123054650742,
+      "grad_norm": 0.9386326670646667,
+      "learning_rate": 0.0002,
+      "loss": 1.619,
+      "step": 260
+    },
+    {
+      "epoch": 0.19543973941368079,
+      "grad_norm": 0.7332451939582825,
+      "learning_rate": 0.0002,
+      "loss": 1.6511,
+      "step": 270
+    },
+    {
+      "epoch": 0.20267824828085415,
+      "grad_norm": 0.7092869877815247,
+      "learning_rate": 0.0002,
+      "loss": 1.6353,
+      "step": 280
+    },
+    {
+      "epoch": 0.20991675714802752,
+      "grad_norm": 0.7256413698196411,
+      "learning_rate": 0.0002,
+      "loss": 1.5996,
+      "step": 290
+    },
+    {
+      "epoch": 0.21715526601520088,
+      "grad_norm": 0.6398681402206421,
+      "learning_rate": 0.0002,
+      "loss": 1.6754,
+      "step": 300
+    },
+    {
+      "epoch": 0.22439377488237422,
+      "grad_norm": 0.6273287534713745,
+      "learning_rate": 0.0002,
+      "loss": 1.397,
+      "step": 310
+    },
+    {
+      "epoch": 0.23163228374954759,
+      "grad_norm": 0.511648416519165,
+      "learning_rate": 0.0002,
+      "loss": 1.5115,
+      "step": 320
+    },
+    {
+      "epoch": 0.23887079261672095,
+      "grad_norm": 0.8677352070808411,
+      "learning_rate": 0.0002,
+      "loss": 1.5424,
+      "step": 330
+    },
+    {
+      "epoch": 0.24610930148389432,
+      "grad_norm": 0.6270743012428284,
+      "learning_rate": 0.0002,
+      "loss": 1.6779,
+      "step": 340
+    },
+    {
+      "epoch": 0.2533478103510677,
+      "grad_norm": 0.7980281114578247,
+      "learning_rate": 0.0002,
+      "loss": 1.626,
+      "step": 350
+    },
+    {
+      "epoch": 0.26058631921824105,
+      "grad_norm": 0.632486879825592,
+      "learning_rate": 0.0002,
+      "loss": 1.5238,
+      "step": 360
+    },
+    {
+      "epoch": 0.2678248280854144,
+      "grad_norm": 0.6527034640312195,
+      "learning_rate": 0.0002,
+      "loss": 1.5175,
+      "step": 370
+    },
+    {
+      "epoch": 0.2750633369525878,
+      "grad_norm": 0.7672118544578552,
+      "learning_rate": 0.0002,
+      "loss": 1.627,
+      "step": 380
+    },
+    {
+      "epoch": 0.28230184581976114,
+      "grad_norm": 0.6035117506980896,
+      "learning_rate": 0.0002,
+      "loss": 1.5605,
+      "step": 390
+    },
+    {
+      "epoch": 0.2895403546869345,
+      "grad_norm": 0.5955103039741516,
+      "learning_rate": 0.0002,
+      "loss": 1.4603,
+      "step": 400
+    },
+    {
+      "epoch": 0.2967788635541079,
+      "grad_norm": 0.6015191674232483,
+      "learning_rate": 0.0002,
+      "loss": 1.558,
+      "step": 410
+    },
+    {
+      "epoch": 0.30401737242128124,
+      "grad_norm": 0.6380982398986816,
+      "learning_rate": 0.0002,
+      "loss": 1.6091,
+      "step": 420
+    },
+    {
+      "epoch": 0.3112558812884546,
+      "grad_norm": 0.6707863211631775,
+      "learning_rate": 0.0002,
+      "loss": 1.5292,
+      "step": 430
+    },
+    {
+      "epoch": 0.3184943901556279,
+      "grad_norm": 0.7010176777839661,
+      "learning_rate": 0.0002,
+      "loss": 1.4426,
+      "step": 440
+    },
+    {
+      "epoch": 0.3257328990228013,
+      "grad_norm": 0.8263739943504333,
+      "learning_rate": 0.0002,
+      "loss": 1.5572,
+      "step": 450
+    },
+    {
+      "epoch": 0.33297140788997465,
+      "grad_norm": 0.7253276109695435,
+      "learning_rate": 0.0002,
+      "loss": 1.5188,
+      "step": 460
+    },
+    {
+      "epoch": 0.340209916757148,
+      "grad_norm": 0.5238934755325317,
+      "learning_rate": 0.0002,
+      "loss": 1.584,
+      "step": 470
+    },
+    {
+      "epoch": 0.3474484256243214,
+      "grad_norm": 0.7869495749473572,
+      "learning_rate": 0.0002,
+      "loss": 1.7035,
+      "step": 480
+    },
+    {
+      "epoch": 0.35468693449149474,
+      "grad_norm": 0.7485215663909912,
+      "learning_rate": 0.0002,
+      "loss": 1.5776,
+      "step": 490
+    },
+    {
+      "epoch": 0.3619254433586681,
+      "grad_norm": 0.5413193106651306,
+      "learning_rate": 0.0002,
+      "loss": 1.6274,
+      "step": 500
+    },
+    {
+      "epoch": 0.3691639522258415,
+      "grad_norm": 0.7615048885345459,
+      "learning_rate": 0.0002,
+      "loss": 1.7323,
+      "step": 510
+    },
+    {
+      "epoch": 0.37640246109301484,
+      "grad_norm": 0.7685340046882629,
+      "learning_rate": 0.0002,
+      "loss": 1.532,
+      "step": 520
+    },
+    {
+      "epoch": 0.3836409699601882,
+      "grad_norm": 0.6379081010818481,
+      "learning_rate": 0.0002,
+      "loss": 1.6312,
+      "step": 530
+    },
+    {
+      "epoch": 0.39087947882736157,
+      "grad_norm": 0.7946939468383789,
+      "learning_rate": 0.0002,
+      "loss": 1.5645,
+      "step": 540
+    },
+    {
+      "epoch": 0.39811798769453494,
+      "grad_norm": 0.6287278532981873,
+      "learning_rate": 0.0002,
+      "loss": 1.4001,
+      "step": 550
+    },
+    {
+      "epoch": 0.4053564965617083,
+      "grad_norm": 0.6811642646789551,
+      "learning_rate": 0.0002,
+      "loss": 1.5982,
+      "step": 560
+    },
+    {
+      "epoch": 0.41259500542888167,
+      "grad_norm": 0.671073317527771,
+      "learning_rate": 0.0002,
+      "loss": 1.4953,
+      "step": 570
+    },
+    {
+      "epoch": 0.41983351429605503,
+      "grad_norm": 0.6313900351524353,
+      "learning_rate": 0.0002,
+      "loss": 1.6753,
+      "step": 580
+    },
+    {
+      "epoch": 0.4270720231632284,
+      "grad_norm": 0.5291772484779358,
+      "learning_rate": 0.0002,
+      "loss": 1.546,
+      "step": 590
+    },
+    {
+      "epoch": 0.43431053203040176,
+      "grad_norm": 0.62503582239151,
+      "learning_rate": 0.0002,
+      "loss": 1.5441,
+      "step": 600
+    },
+    {
+      "epoch": 0.4415490408975751,
+      "grad_norm": 0.5777305364608765,
+      "learning_rate": 0.0002,
+      "loss": 1.6276,
+      "step": 610
+    },
+    {
+      "epoch": 0.44878754976474844,
+      "grad_norm": 0.7013497352600098,
+      "learning_rate": 0.0002,
+      "loss": 1.4758,
+      "step": 620
+    },
+    {
+      "epoch": 0.4560260586319218,
+      "grad_norm": 0.8044822216033936,
+      "learning_rate": 0.0002,
+      "loss": 1.4029,
+      "step": 630
+    },
+    {
+      "epoch": 0.46326456749909517,
+      "grad_norm": 0.672531247138977,
+      "learning_rate": 0.0002,
+      "loss": 1.7195,
+      "step": 640
+    },
+    {
+      "epoch": 0.47050307636626854,
+      "grad_norm": 0.6233910322189331,
+      "learning_rate": 0.0002,
+      "loss": 1.614,
+      "step": 650
+    },
+    {
+      "epoch": 0.4777415852334419,
+      "grad_norm": 0.651524543762207,
+      "learning_rate": 0.0002,
+      "loss": 1.6041,
+      "step": 660
+    },
+    {
+      "epoch": 0.48498009410061527,
+      "grad_norm": 0.7213939428329468,
+      "learning_rate": 0.0002,
+      "loss": 1.5842,
+      "step": 670
+    },
+    {
+      "epoch": 0.49221860296778863,
+      "grad_norm": 0.6541454792022705,
+      "learning_rate": 0.0002,
+      "loss": 1.5453,
+      "step": 680
+    },
+    {
+      "epoch": 0.499457111834962,
+      "grad_norm": 0.6568936109542847,
+      "learning_rate": 0.0002,
+      "loss": 1.662,
+      "step": 690
+    },
+    {
+      "epoch": 0.5066956207021354,
+      "grad_norm": 0.7176415324211121,
+      "learning_rate": 0.0002,
+      "loss": 1.624,
+      "step": 700
+    },
+    {
+      "epoch": 0.5139341295693087,
+      "grad_norm": 0.6553855538368225,
+      "learning_rate": 0.0002,
+      "loss": 1.6099,
+      "step": 710
+    },
+    {
+      "epoch": 0.5211726384364821,
+      "grad_norm": 0.5654335618019104,
+      "learning_rate": 0.0002,
+      "loss": 1.5508,
+      "step": 720
+    },
+    {
+      "epoch": 0.5284111473036555,
+      "grad_norm": 0.5671001672744751,
+      "learning_rate": 0.0002,
+      "loss": 1.392,
+      "step": 730
+    },
+    {
+      "epoch": 0.5356496561708288,
+      "grad_norm": 0.7914412021636963,
+      "learning_rate": 0.0002,
+      "loss": 1.388,
+      "step": 740
+    },
+    {
+      "epoch": 0.5428881650380022,
+      "grad_norm": 0.6172138452529907,
+      "learning_rate": 0.0002,
+      "loss": 1.5931,
+      "step": 750
+    },
+    {
+      "epoch": 0.5501266739051756,
+      "grad_norm": 0.6132623553276062,
+      "learning_rate": 0.0002,
+      "loss": 1.4018,
+      "step": 760
+    },
+    {
+      "epoch": 0.5573651827723489,
+      "grad_norm": 0.654000461101532,
+      "learning_rate": 0.0002,
+      "loss": 1.513,
+      "step": 770
+    },
+    {
+      "epoch": 0.5646036916395223,
+      "grad_norm": 0.5691370964050293,
+      "learning_rate": 0.0002,
+      "loss": 1.5035,
+      "step": 780
+    },
+    {
+      "epoch": 0.5718422005066957,
+      "grad_norm": 0.7922580242156982,
+      "learning_rate": 0.0002,
+      "loss": 1.65,
+      "step": 790
+    },
+    {
+      "epoch": 0.579080709373869,
+      "grad_norm": 0.6831880211830139,
+      "learning_rate": 0.0002,
+      "loss": 1.4521,
+      "step": 800
+    },
+    {
+      "epoch": 0.5863192182410424,
+      "grad_norm": 0.6740124821662903,
+      "learning_rate": 0.0002,
+      "loss": 1.4734,
+      "step": 810
+    },
+    {
+      "epoch": 0.5935577271082157,
+      "grad_norm": 1.380016803741455,
+      "learning_rate": 0.0002,
+      "loss": 1.6498,
+      "step": 820
+    },
+    {
+      "epoch": 0.6007962359753891,
+      "grad_norm": 0.6552878022193909,
+      "learning_rate": 0.0002,
+      "loss": 1.4642,
+      "step": 830
+    },
+    {
+      "epoch": 0.6080347448425625,
+      "grad_norm": 0.6649535298347473,
+      "learning_rate": 0.0002,
+      "loss": 1.6271,
+      "step": 840
+    },
+    {
+      "epoch": 0.6152732537097358,
+      "grad_norm": 0.561738133430481,
+      "learning_rate": 0.0002,
+      "loss": 1.5886,
+      "step": 850
+    },
+    {
+      "epoch": 0.6225117625769092,
+      "grad_norm": 0.6133047938346863,
+      "learning_rate": 0.0002,
+      "loss": 1.5364,
+      "step": 860
+    },
+    {
+      "epoch": 0.6297502714440825,
+      "grad_norm": 0.559843122959137,
+      "learning_rate": 0.0002,
+      "loss": 1.3489,
+      "step": 870
+    },
+    {
+      "epoch": 0.6369887803112558,
+      "grad_norm": 0.6117811799049377,
+      "learning_rate": 0.0002,
+      "loss": 1.4878,
+      "step": 880
+    },
+    {
+      "epoch": 0.6442272891784292,
+      "grad_norm": 0.6209776401519775,
+      "learning_rate": 0.0002,
+      "loss": 1.56,
+      "step": 890
+    },
+    {
+      "epoch": 0.6514657980456026,
+      "grad_norm": 0.6234082579612732,
+      "learning_rate": 0.0002,
+      "loss": 1.6747,
+      "step": 900
+    },
+    {
+      "epoch": 0.6587043069127759,
+      "grad_norm": 0.7623258233070374,
+      "learning_rate": 0.0002,
+      "loss": 1.6963,
+      "step": 910
+    },
+    {
+      "epoch": 0.6659428157799493,
+      "grad_norm": 0.6148061752319336,
+      "learning_rate": 0.0002,
+      "loss": 1.2424,
+      "step": 920
+    },
+    {
+      "epoch": 0.6731813246471227,
+      "grad_norm": 0.6682973504066467,
+      "learning_rate": 0.0002,
+      "loss": 1.4319,
+      "step": 930
+    },
+    {
+      "epoch": 0.680419833514296,
+      "grad_norm": 0.5513041615486145,
+      "learning_rate": 0.0002,
+      "loss": 1.5377,
+      "step": 940
+    },
+    {
+      "epoch": 0.6876583423814694,
+      "grad_norm": 0.5197525024414062,
+      "learning_rate": 0.0002,
+      "loss": 1.3991,
+      "step": 950
+    },
+    {
+      "epoch": 0.6948968512486428,
+      "grad_norm": 0.6490758061408997,
+      "learning_rate": 0.0002,
+      "loss": 1.4398,
+      "step": 960
+    },
+    {
+      "epoch": 0.7021353601158161,
+      "grad_norm": 0.6450682878494263,
+      "learning_rate": 0.0002,
+      "loss": 1.5251,
+      "step": 970
+    },
+    {
+      "epoch": 0.7093738689829895,
+      "grad_norm": 0.6203766465187073,
+      "learning_rate": 0.0002,
+      "loss": 1.5417,
+      "step": 980
+    },
+    {
+      "epoch": 0.7166123778501629,
+      "grad_norm": 0.6023609638214111,
+      "learning_rate": 0.0002,
+      "loss": 1.4575,
+      "step": 990
+    },
+    {
+      "epoch": 0.7238508867173362,
+      "grad_norm": 0.5765255093574524,
+      "learning_rate": 0.0002,
+      "loss": 1.4973,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7310893955845096,
+      "grad_norm": 0.6650075316429138,
+      "learning_rate": 0.0002,
+      "loss": 1.483,
+      "step": 1010
+    },
+    {
+      "epoch": 0.738327904451683,
+      "grad_norm": 0.5610854029655457,
+      "learning_rate": 0.0002,
+      "loss": 1.5959,
+      "step": 1020
+    },
+    {
+      "epoch": 0.7455664133188563,
+      "grad_norm": 0.7072813510894775,
+      "learning_rate": 0.0002,
+      "loss": 1.5248,
+      "step": 1030
+    },
+    {
+      "epoch": 0.7528049221860297,
+      "grad_norm": 0.6815407872200012,
+      "learning_rate": 0.0002,
+      "loss": 1.5776,
+      "step": 1040
+    },
+    {
+      "epoch": 0.760043431053203,
+      "grad_norm": 0.7932390570640564,
+      "learning_rate": 0.0002,
+      "loss": 1.4577,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7672819399203764,
+      "grad_norm": 0.5798183083534241,
+      "learning_rate": 0.0002,
+      "loss": 1.4515,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7745204487875498,
+      "grad_norm": 0.7898504137992859,
+      "learning_rate": 0.0002,
+      "loss": 1.5053,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7817589576547231,
+      "grad_norm": 0.4983280301094055,
+      "learning_rate": 0.0002,
+      "loss": 1.4776,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7889974665218965,
+      "grad_norm": 0.691403329372406,
+      "learning_rate": 0.0002,
+      "loss": 1.5007,
+      "step": 1090
+    },
+    {
+      "epoch": 0.7962359753890699,
+      "grad_norm": 0.5394481420516968,
+      "learning_rate": 0.0002,
+      "loss": 1.5153,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8034744842562432,
+      "grad_norm": 0.5136822462081909,
+      "learning_rate": 0.0002,
+      "loss": 1.6892,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8107129931234166,
+      "grad_norm": 0.6828126907348633,
+      "learning_rate": 0.0002,
+      "loss": 1.4902,
+      "step": 1120
+    },
+    {
+      "epoch": 0.81795150199059,
+      "grad_norm": 0.6799656748771667,
+      "learning_rate": 0.0002,
+      "loss": 1.4346,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8251900108577633,
+      "grad_norm": 0.5428406000137329,
+      "learning_rate": 0.0002,
+      "loss": 1.2678,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8324285197249367,
+      "grad_norm": 0.4811290502548218,
+      "learning_rate": 0.0002,
+      "loss": 1.4072,
+      "step": 1150
+    },
+    {
+      "epoch": 0.8396670285921101,
+      "grad_norm": 0.5519434809684753,
+      "learning_rate": 0.0002,
+      "loss": 1.4512,
+      "step": 1160
+    },
+    {
+      "epoch": 0.8469055374592834,
+      "grad_norm": 0.9748060703277588,
+      "learning_rate": 0.0002,
+      "loss": 1.4072,
+      "step": 1170
+    },
+    {
+      "epoch": 0.8541440463264568,
+      "grad_norm": 0.712609589099884,
+      "learning_rate": 0.0002,
+      "loss": 1.4309,
+      "step": 1180
+    },
+    {
+      "epoch": 0.8613825551936302,
+      "grad_norm": 0.6866157054901123,
+      "learning_rate": 0.0002,
+      "loss": 1.434,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8686210640608035,
+      "grad_norm": 0.5068854093551636,
+      "learning_rate": 0.0002,
+      "loss": 1.3704,
+      "step": 1200
+    },
+    {
+      "epoch": 0.8758595729279768,
+      "grad_norm": 0.6333245038986206,
+      "learning_rate": 0.0002,
+      "loss": 1.5601,
+      "step": 1210
+    },
+    {
+      "epoch": 0.8830980817951501,
+      "grad_norm": 0.6424421072006226,
+      "learning_rate": 0.0002,
+      "loss": 1.4636,
+      "step": 1220
+    },
+    {
+      "epoch": 0.8903365906623235,
+      "grad_norm": 0.4771921932697296,
+      "learning_rate": 0.0002,
+      "loss": 1.4186,
+      "step": 1230
+    },
+    {
+      "epoch": 0.8975750995294969,
+      "grad_norm": 0.5191764235496521,
+      "learning_rate": 0.0002,
+      "loss": 1.6323,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9048136083966702,
+      "grad_norm": 0.756222128868103,
+      "learning_rate": 0.0002,
+      "loss": 1.6105,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9120521172638436,
+      "grad_norm": 0.623823881149292,
+      "learning_rate": 0.0002,
+      "loss": 1.4396,
+      "step": 1260
+    },
+    {
+      "epoch": 0.919290626131017,
+      "grad_norm": 0.8166571259498596,
+      "learning_rate": 0.0002,
+      "loss": 1.3097,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9265291349981903,
+      "grad_norm": 0.6059346795082092,
+      "learning_rate": 0.0002,
+      "loss": 1.4625,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9337676438653637,
+      "grad_norm": 0.5842690467834473,
+      "learning_rate": 0.0002,
+      "loss": 1.3555,
+      "step": 1290
+    },
+    {
+      "epoch": 0.9410061527325371,
+      "grad_norm": 0.7649800777435303,
+      "learning_rate": 0.0002,
+      "loss": 1.5859,
+      "step": 1300
+    },
+    {
+      "epoch": 0.9482446615997104,
+      "grad_norm": 0.6420919895172119,
+      "learning_rate": 0.0002,
+      "loss": 1.5915,
+      "step": 1310
+    },
+    {
+      "epoch": 0.9554831704668838,
+      "grad_norm": 0.7011452913284302,
+      "learning_rate": 0.0002,
+      "loss": 1.453,
+      "step": 1320
+    },
+    {
+      "epoch": 0.9627216793340572,
+      "grad_norm": 0.5783746242523193,
+      "learning_rate": 0.0002,
+      "loss": 1.6766,
+      "step": 1330
+    },
+    {
+      "epoch": 0.9699601882012305,
+      "grad_norm": 0.5973192453384399,
+      "learning_rate": 0.0002,
+      "loss": 1.6308,
+      "step": 1340
+    },
+    {
+      "epoch": 0.9771986970684039,
+      "grad_norm": 0.6181833744049072,
+      "learning_rate": 0.0002,
+      "loss": 1.5901,
+      "step": 1350
+    },
+    {
+      "epoch": 0.9844372059355773,
+      "grad_norm": 0.5563396215438843,
+      "learning_rate": 0.0002,
+      "loss": 1.5258,
+      "step": 1360
+    },
+    {
+      "epoch": 0.9916757148027506,
+      "grad_norm": 0.45723360776901245,
+      "learning_rate": 0.0002,
+      "loss": 1.4508,
+      "step": 1370
+    },
+    {
+      "epoch": 0.998914223669924,
+      "grad_norm": 0.5947498679161072,
+      "learning_rate": 0.0002,
+      "loss": 1.3291,
+      "step": 1380
+    },
+    {
+      "epoch": 0.9996380745566413,
+      "eval_loss": 1.480796456336975,
+      "eval_runtime": 27.3103,
+      "eval_samples_per_second": 15.965,
+      "eval_steps_per_second": 2.014,
+      "step": 1381
+    },
+    {
+      "epoch": 1.0061527325370974,
+      "grad_norm": 0.5599952936172485,
+      "learning_rate": 0.0002,
+      "loss": 1.3057,
+      "step": 1390
+    },
+    {
+      "epoch": 1.0133912414042707,
+      "grad_norm": 0.5932008028030396,
+      "learning_rate": 0.0002,
+      "loss": 1.4991,
+      "step": 1400
+    },
+    {
+      "epoch": 1.020629750271444,
+      "grad_norm": 0.6194121837615967,
+      "learning_rate": 0.0002,
+      "loss": 1.4506,
+      "step": 1410
+    },
+    {
+      "epoch": 1.0278682591386175,
+      "grad_norm": 0.6995621919631958,
+      "learning_rate": 0.0002,
+      "loss": 1.5966,
+      "step": 1420
+    },
+    {
+      "epoch": 1.0351067680057908,
+      "grad_norm": 0.7905810475349426,
+      "learning_rate": 0.0002,
+      "loss": 1.4153,
+      "step": 1430
+    },
+    {
+      "epoch": 1.0423452768729642,
+      "grad_norm": 0.7221615314483643,
+      "learning_rate": 0.0002,
+      "loss": 1.4414,
+      "step": 1440
+    },
+    {
+      "epoch": 1.0495837857401376,
+      "grad_norm": 0.6170642375946045,
+      "learning_rate": 0.0002,
+      "loss": 1.3859,
+      "step": 1450
+    },
+    {
+      "epoch": 1.056822294607311,
+      "grad_norm": 0.5844094753265381,
+      "learning_rate": 0.0002,
+      "loss": 1.3806,
+      "step": 1460
+    },
+    {
+      "epoch": 1.0640608034744843,
+      "grad_norm": 0.7731822729110718,
+      "learning_rate": 0.0002,
+      "loss": 1.4871,
+      "step": 1470
+    },
+    {
+      "epoch": 1.0712993123416577,
+      "grad_norm": 0.4554748237133026,
+      "learning_rate": 0.0002,
+      "loss": 1.4286,
+      "step": 1480
+    },
+    {
+      "epoch": 1.078537821208831,
+      "grad_norm": 0.6923259496688843,
+      "learning_rate": 0.0002,
+      "loss": 1.3977,
+      "step": 1490
+    },
+    {
+      "epoch": 1.0857763300760044,
+      "grad_norm": 0.6008219122886658,
+      "learning_rate": 0.0002,
+      "loss": 1.3936,
+      "step": 1500
+    },
+    {
+      "epoch": 1.0930148389431777,
+      "grad_norm": 0.6450045704841614,
+      "learning_rate": 0.0002,
+      "loss": 1.4821,
+      "step": 1510
+    },
+    {
+      "epoch": 1.1002533478103511,
+      "grad_norm": 0.7833753824234009,
+      "learning_rate": 0.0002,
+      "loss": 1.3295,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1074918566775245,
+      "grad_norm": 0.5076758861541748,
+      "learning_rate": 0.0002,
+      "loss": 1.3424,
+      "step": 1530
+    },
+    {
+      "epoch": 1.1147303655446978,
+      "grad_norm": 0.5661332011222839,
+      "learning_rate": 0.0002,
+      "loss": 1.4043,
+      "step": 1540
+    },
+    {
+      "epoch": 1.1219688744118712,
+      "grad_norm": 0.6526919603347778,
+      "learning_rate": 0.0002,
+      "loss": 1.4963,
+      "step": 1550
+    },
+    {
+      "epoch": 1.1292073832790446,
+      "grad_norm": 0.5613082647323608,
+      "learning_rate": 0.0002,
+      "loss": 1.3671,
+      "step": 1560
+    },
+    {
+      "epoch": 1.136445892146218,
+      "grad_norm": 0.6113885641098022,
+      "learning_rate": 0.0002,
+      "loss": 1.4458,
+      "step": 1570
+    },
+    {
+      "epoch": 1.1436844010133913,
+      "grad_norm": 0.6732510328292847,
+      "learning_rate": 0.0002,
+      "loss": 1.3552,
+      "step": 1580
+    },
+    {
+      "epoch": 1.1509229098805647,
+      "grad_norm": 0.6146392226219177,
+      "learning_rate": 0.0002,
+      "loss": 1.3114,
+      "step": 1590
+    },
+    {
+      "epoch": 1.158161418747738,
+      "grad_norm": 0.6766974329948425,
+      "learning_rate": 0.0002,
+      "loss": 1.411,
+      "step": 1600
+    },
+    {
+      "epoch": 1.1653999276149114,
+      "grad_norm": 0.7621957659721375,
+      "learning_rate": 0.0002,
+      "loss": 1.2401,
+      "step": 1610
+    },
+    {
+      "epoch": 1.1726384364820848,
+      "grad_norm": 0.6959581971168518,
+      "learning_rate": 0.0002,
+      "loss": 1.3758,
+      "step": 1620
+    },
+    {
+      "epoch": 1.1798769453492581,
+      "grad_norm": 0.6691278219223022,
+      "learning_rate": 0.0002,
+      "loss": 1.382,
+      "step": 1630
+    },
+    {
+      "epoch": 1.1871154542164315,
+      "grad_norm": 0.4927774965763092,
+      "learning_rate": 0.0002,
+      "loss": 1.4147,
+      "step": 1640
+    },
+    {
+      "epoch": 1.1943539630836049,
+      "grad_norm": 0.7724234461784363,
+      "learning_rate": 0.0002,
+      "loss": 1.449,
+      "step": 1650
+    },
+    {
+      "epoch": 1.2015924719507782,
+      "grad_norm": 0.6817787885665894,
+      "learning_rate": 0.0002,
+      "loss": 1.4778,
+      "step": 1660
+    },
+    {
+      "epoch": 1.2088309808179516,
+      "grad_norm": 0.6500699520111084,
+      "learning_rate": 0.0002,
+      "loss": 1.3776,
+      "step": 1670
+    },
+    {
+      "epoch": 1.216069489685125,
+      "grad_norm": 0.5703568458557129,
+      "learning_rate": 0.0002,
+      "loss": 1.3875,
+      "step": 1680
+    },
+    {
+      "epoch": 1.2233079985522983,
+      "grad_norm": 0.6261579990386963,
+      "learning_rate": 0.0002,
+      "loss": 1.4735,
+      "step": 1690
+    },
+    {
+      "epoch": 1.2305465074194717,
+      "grad_norm": 0.651713490486145,
+      "learning_rate": 0.0002,
+      "loss": 1.3898,
+      "step": 1700
+    },
+    {
+      "epoch": 1.237785016286645,
+      "grad_norm": 0.684399425983429,
+      "learning_rate": 0.0002,
+      "loss": 1.4002,
+      "step": 1710
+    },
+    {
+      "epoch": 1.2450235251538184,
+      "grad_norm": 0.6996857523918152,
+      "learning_rate": 0.0002,
+      "loss": 1.5027,
+      "step": 1720
+    },
+    {
+      "epoch": 1.2522620340209918,
+      "grad_norm": 0.7102537751197815,
+      "learning_rate": 0.0002,
+      "loss": 1.3326,
+      "step": 1730
+    },
+    {
+      "epoch": 1.2595005428881652,
+      "grad_norm": 0.45809897780418396,
+      "learning_rate": 0.0002,
+      "loss": 1.3675,
+      "step": 1740
+    },
+    {
+      "epoch": 1.2667390517553385,
+      "grad_norm": 0.6377046704292297,
+      "learning_rate": 0.0002,
+      "loss": 1.4175,
+      "step": 1750
+    },
+    {
+      "epoch": 1.2739775606225119,
+      "grad_norm": 0.6965704560279846,
+      "learning_rate": 0.0002,
+      "loss": 1.3479,
+      "step": 1760
+    },
+    {
+      "epoch": 1.2812160694896852,
+      "grad_norm": 0.5688214302062988,
+      "learning_rate": 0.0002,
+      "loss": 1.5647,
+      "step": 1770
+    },
+    {
+      "epoch": 1.2884545783568586,
+      "grad_norm": 0.6384190320968628,
+      "learning_rate": 0.0002,
+      "loss": 1.3967,
+      "step": 1780
+    },
+    {
+      "epoch": 1.295693087224032,
+      "grad_norm": 0.5629363656044006,
+      "learning_rate": 0.0002,
+      "loss": 1.3671,
+      "step": 1790
+    },
+    {
+      "epoch": 1.3029315960912053,
+      "grad_norm": 0.6148255467414856,
+      "learning_rate": 0.0002,
+      "loss": 1.2292,
+      "step": 1800
+    },
+    {
+      "epoch": 1.3101701049583787,
+      "grad_norm": 0.655580997467041,
+      "learning_rate": 0.0002,
+      "loss": 1.5806,
+      "step": 1810
+    },
+    {
+      "epoch": 1.3174086138255519,
+      "grad_norm": 0.5642657279968262,
+      "learning_rate": 0.0002,
+      "loss": 1.2398,
+      "step": 1820
+    },
+    {
+      "epoch": 1.3246471226927252,
+      "grad_norm": 0.59607994556427,
+      "learning_rate": 0.0002,
+      "loss": 1.3246,
+      "step": 1830
+    },
+    {
+      "epoch": 1.3318856315598986,
+      "grad_norm": 0.5564199090003967,
+      "learning_rate": 0.0002,
+      "loss": 1.3274,
+      "step": 1840
+    },
+    {
+      "epoch": 1.339124140427072,
+      "grad_norm": 0.6949955821037292,
+      "learning_rate": 0.0002,
+      "loss": 1.5834,
+      "step": 1850
+    },
+    {
+      "epoch": 1.3463626492942453,
+      "grad_norm": 0.7036856412887573,
+      "learning_rate": 0.0002,
+      "loss": 1.4722,
+      "step": 1860
+    },
+    {
+      "epoch": 1.3536011581614187,
+      "grad_norm": 0.722062885761261,
+      "learning_rate": 0.0002,
+      "loss": 1.333,
+      "step": 1870
+    },
+    {
+      "epoch": 1.360839667028592,
+      "grad_norm": 0.6098677515983582,
+      "learning_rate": 0.0002,
+      "loss": 1.4044,
+      "step": 1880
+    },
+    {
+      "epoch": 1.3680781758957654,
+      "grad_norm": 0.5376402735710144,
+      "learning_rate": 0.0002,
+      "loss": 1.6217,
+      "step": 1890
+    },
+    {
+      "epoch": 1.3753166847629388,
+      "grad_norm": 0.6974610090255737,
+      "learning_rate": 0.0002,
+      "loss": 1.5071,
+      "step": 1900
+    },
+    {
+      "epoch": 1.3825551936301121,
+      "grad_norm": 0.6520763635635376,
+      "learning_rate": 0.0002,
+      "loss": 1.5854,
+      "step": 1910
+    },
+    {
+      "epoch": 1.3897937024972855,
+      "grad_norm": 0.6604374647140503,
+      "learning_rate": 0.0002,
+      "loss": 1.4271,
+      "step": 1920
+    },
+    {
+      "epoch": 1.3970322113644589,
+      "grad_norm": 0.7364398241043091,
+      "learning_rate": 0.0002,
+      "loss": 1.419,
+      "step": 1930
+    },
+    {
+      "epoch": 1.4042707202316322,
+      "grad_norm": 0.6849475502967834,
+      "learning_rate": 0.0002,
+      "loss": 1.4585,
+      "step": 1940
+    },
+    {
+      "epoch": 1.4115092290988056,
+      "grad_norm": 0.6562670469284058,
+      "learning_rate": 0.0002,
+      "loss": 1.5577,
+      "step": 1950
+    },
+    {
+      "epoch": 1.418747737965979,
+      "grad_norm": 0.5695616006851196,
+      "learning_rate": 0.0002,
+      "loss": 1.4725,
+      "step": 1960
+    },
+    {
+      "epoch": 1.4259862468331523,
+      "grad_norm": 0.5244464874267578,
+      "learning_rate": 0.0002,
+      "loss": 1.3088,
+      "step": 1970
+    },
+    {
+      "epoch": 1.4332247557003257,
+      "grad_norm": 0.6347293257713318,
+      "learning_rate": 0.0002,
+      "loss": 1.5069,
+      "step": 1980
+    },
+    {
+      "epoch": 1.440463264567499,
+      "grad_norm": 0.5528361201286316,
+      "learning_rate": 0.0002,
+      "loss": 1.3502,
+      "step": 1990
+    },
+    {
+      "epoch": 1.4477017734346724,
+      "grad_norm": 0.6987585425376892,
+      "learning_rate": 0.0002,
+      "loss": 1.3978,
+      "step": 2000
+    },
+    {
+      "epoch": 1.4549402823018458,
+      "grad_norm": 0.6568987369537354,
+      "learning_rate": 0.0002,
+      "loss": 1.4262,
+      "step": 2010
+    },
+    {
+      "epoch": 1.4621787911690192,
+      "grad_norm": 0.7665994763374329,
+      "learning_rate": 0.0002,
+      "loss": 1.4175,
+      "step": 2020
+    },
+    {
+      "epoch": 1.4694173000361925,
+      "grad_norm": 0.5127707123756409,
+      "learning_rate": 0.0002,
+      "loss": 1.244,
+      "step": 2030
+    },
+    {
+      "epoch": 1.476655808903366,
+      "grad_norm": 0.5406824946403503,
+      "learning_rate": 0.0002,
+      "loss": 1.3699,
+      "step": 2040
+    },
+    {
+      "epoch": 1.4838943177705393,
+      "grad_norm": 0.5990166664123535,
+      "learning_rate": 0.0002,
+      "loss": 1.3353,
+      "step": 2050
+    },
+    {
+      "epoch": 1.4911328266377126,
+      "grad_norm": 0.6186193823814392,
+      "learning_rate": 0.0002,
+      "loss": 1.2454,
+      "step": 2060
+    },
+    {
+      "epoch": 1.498371335504886,
+      "grad_norm": 0.6154307126998901,
+      "learning_rate": 0.0002,
+      "loss": 1.428,
+      "step": 2070
+    },
+    {
+      "epoch": 1.5056098443720594,
+      "grad_norm": 0.5606056451797485,
+      "learning_rate": 0.0002,
+      "loss": 1.4528,
+      "step": 2080
+    },
+    {
+      "epoch": 1.5128483532392327,
+      "grad_norm": 0.5006417036056519,
+      "learning_rate": 0.0002,
+      "loss": 1.2405,
+      "step": 2090
+    },
+    {
+      "epoch": 1.520086862106406,
+      "grad_norm": 0.5968486070632935,
+      "learning_rate": 0.0002,
+      "loss": 1.4258,
+      "step": 2100
+    },
+    {
+      "epoch": 1.5273253709735795,
+      "grad_norm": 0.5835496187210083,
+      "learning_rate": 0.0002,
+      "loss": 1.2752,
+      "step": 2110
+    },
+    {
+      "epoch": 1.5345638798407528,
+      "grad_norm": 0.6753535270690918,
+      "learning_rate": 0.0002,
+      "loss": 1.5443,
+      "step": 2120
+    },
+    {
+      "epoch": 1.5418023887079262,
+      "grad_norm": 0.7299720644950867,
+      "learning_rate": 0.0002,
+      "loss": 1.2139,
+      "step": 2130
+    },
+    {
+      "epoch": 1.5490408975750996,
+      "grad_norm": 0.5105988383293152,
+      "learning_rate": 0.0002,
+      "loss": 1.2364,
+      "step": 2140
+    },
+    {
+      "epoch": 1.556279406442273,
+      "grad_norm": 0.5675431489944458,
+      "learning_rate": 0.0002,
+      "loss": 1.4528,
+      "step": 2150
+    },
+    {
+      "epoch": 1.5635179153094463,
+      "grad_norm": 0.6246723532676697,
+      "learning_rate": 0.0002,
+      "loss": 1.4563,
+      "step": 2160
+    },
+    {
+      "epoch": 1.5707564241766196,
+      "grad_norm": 0.7291720509529114,
+      "learning_rate": 0.0002,
+      "loss": 1.5255,
+      "step": 2170
+    },
+    {
+      "epoch": 1.577994933043793,
+      "grad_norm": 0.678114116191864,
+      "learning_rate": 0.0002,
+      "loss": 1.5432,
+      "step": 2180
+    },
+    {
+      "epoch": 1.5852334419109664,
+      "grad_norm": 0.5136260986328125,
+      "learning_rate": 0.0002,
+      "loss": 1.5212,
+      "step": 2190
+    },
+    {
+      "epoch": 1.5924719507781397,
+      "grad_norm": 0.6359935998916626,
+      "learning_rate": 0.0002,
+      "loss": 1.3271,
+      "step": 2200
+    },
+    {
+      "epoch": 1.599710459645313,
+      "grad_norm": 0.7650278806686401,
+      "learning_rate": 0.0002,
+      "loss": 1.4038,
+      "step": 2210
+    },
+    {
+      "epoch": 1.6069489685124865,
+      "grad_norm": 0.7256110906600952,
+      "learning_rate": 0.0002,
+      "loss": 1.5478,
+      "step": 2220
+    },
+    {
+      "epoch": 1.6141874773796598,
+      "grad_norm": 0.688689649105072,
+      "learning_rate": 0.0002,
+      "loss": 1.4387,
+      "step": 2230
+    },
+    {
+      "epoch": 1.6214259862468332,
+      "grad_norm": 0.6045311093330383,
+      "learning_rate": 0.0002,
+      "loss": 1.4096,
+      "step": 2240
+    },
+    {
+      "epoch": 1.6286644951140063,
+      "grad_norm": 0.7064604163169861,
+      "learning_rate": 0.0002,
+      "loss": 1.4097,
+      "step": 2250
+    },
+    {
+      "epoch": 1.6359030039811797,
+      "grad_norm": 0.5309562087059021,
+      "learning_rate": 0.0002,
+      "loss": 1.3477,
+      "step": 2260
+    },
+    {
+      "epoch": 1.643141512848353,
+      "grad_norm": 0.5687053203582764,
+      "learning_rate": 0.0002,
+      "loss": 1.4022,
+      "step": 2270
+    },
+    {
+      "epoch": 1.6503800217155264,
+      "grad_norm": 0.535872757434845,
+      "learning_rate": 0.0002,
+      "loss": 1.2977,
+      "step": 2280
+    },
+    {
+      "epoch": 1.6576185305826998,
+      "grad_norm": 0.5502381920814514,
+      "learning_rate": 0.0002,
+      "loss": 1.3844,
+      "step": 2290
+    },
+    {
+      "epoch": 1.6648570394498732,
+      "grad_norm": 0.6158602237701416,
+      "learning_rate": 0.0002,
+      "loss": 1.3764,
+      "step": 2300
+    },
+    {
+      "epoch": 1.6720955483170465,
+      "grad_norm": 0.5804675817489624,
+      "learning_rate": 0.0002,
+      "loss": 1.3515,
+      "step": 2310
+    },
+    {
+      "epoch": 1.67933405718422,
+      "grad_norm": 0.600742757320404,
+      "learning_rate": 0.0002,
+      "loss": 1.2532,
+      "step": 2320
+    },
+    {
+      "epoch": 1.6865725660513933,
+      "grad_norm": 0.7101941108703613,
+      "learning_rate": 0.0002,
+      "loss": 1.477,
+      "step": 2330
+    },
+    {
+      "epoch": 1.6938110749185666,
+      "grad_norm": 0.7507809996604919,
+      "learning_rate": 0.0002,
+      "loss": 1.4849,
+      "step": 2340
+    },
+    {
+      "epoch": 1.70104958378574,
+      "grad_norm": 0.768502414226532,
+      "learning_rate": 0.0002,
+      "loss": 1.2703,
+      "step": 2350
+    },
+    {
+      "epoch": 1.7082880926529134,
+      "grad_norm": 0.4801851212978363,
+      "learning_rate": 0.0002,
+      "loss": 1.3332,
+      "step": 2360
+    },
+    {
+      "epoch": 1.7155266015200867,
+      "grad_norm": 0.5322122573852539,
+      "learning_rate": 0.0002,
+      "loss": 1.4158,
+      "step": 2370
+    },
+    {
+      "epoch": 1.72276511038726,
+      "grad_norm": 0.587661862373352,
+      "learning_rate": 0.0002,
+      "loss": 1.4136,
+      "step": 2380
+    },
+    {
+      "epoch": 1.7300036192544335,
+      "grad_norm": 0.6073525547981262,
+      "learning_rate": 0.0002,
+      "loss": 1.3771,
+      "step": 2390
+    },
+    {
+      "epoch": 1.7372421281216068,
+      "grad_norm": 0.6950460076332092,
+      "learning_rate": 0.0002,
+      "loss": 1.2754,
+      "step": 2400
+    },
+    {
+      "epoch": 1.7444806369887802,
+      "grad_norm": 0.5981102585792542,
+      "learning_rate": 0.0002,
+      "loss": 1.3858,
+      "step": 2410
+    },
+    {
+      "epoch": 1.7517191458559536,
+      "grad_norm": 0.544570803642273,
+      "learning_rate": 0.0002,
+      "loss": 1.4075,
+      "step": 2420
+    },
+    {
+      "epoch": 1.758957654723127,
+      "grad_norm": 0.5304399728775024,
+      "learning_rate": 0.0002,
+      "loss": 1.3861,
+      "step": 2430
+    },
+    {
+      "epoch": 1.7661961635903003,
+      "grad_norm": 0.7921594977378845,
+      "learning_rate": 0.0002,
+      "loss": 1.4244,
+      "step": 2440
+    },
+    {
+      "epoch": 1.7734346724574737,
+      "grad_norm": 0.6084808707237244,
+      "learning_rate": 0.0002,
+      "loss": 1.3053,
+      "step": 2450
+    },
+    {
+      "epoch": 1.780673181324647,
+      "grad_norm": 0.8844701051712036,
+      "learning_rate": 0.0002,
+      "loss": 1.3781,
+      "step": 2460
+    },
+    {
+      "epoch": 1.7879116901918204,
+      "grad_norm": 0.5729258060455322,
+      "learning_rate": 0.0002,
+      "loss": 1.3227,
+      "step": 2470
+    },
+    {
+      "epoch": 1.7951501990589938,
+      "grad_norm": 0.6303611993789673,
+      "learning_rate": 0.0002,
+      "loss": 1.3422,
+      "step": 2480
+    },
+    {
+      "epoch": 1.8023887079261671,
+      "grad_norm": 0.5627942085266113,
+      "learning_rate": 0.0002,
+      "loss": 1.3926,
+      "step": 2490
+    },
+    {
+      "epoch": 1.8096272167933405,
+      "grad_norm": 0.6724274158477783,
+      "learning_rate": 0.0002,
+      "loss": 1.3816,
+      "step": 2500
+    },
+    {
+      "epoch": 1.8168657256605139,
+      "grad_norm": 0.5030826330184937,
+      "learning_rate": 0.0002,
+      "loss": 1.2951,
+      "step": 2510
+    },
+    {
+      "epoch": 1.8241042345276872,
+      "grad_norm": 0.5504099130630493,
+      "learning_rate": 0.0002,
+      "loss": 1.2839,
+      "step": 2520
+    },
+    {
+      "epoch": 1.8313427433948606,
+      "grad_norm": 0.6338945627212524,
+      "learning_rate": 0.0002,
+      "loss": 1.4264,
+      "step": 2530
+    },
+    {
+      "epoch": 1.838581252262034,
+      "grad_norm": 0.5902037620544434,
+      "learning_rate": 0.0002,
+      "loss": 1.563,
+      "step": 2540
+    },
+    {
+      "epoch": 1.8458197611292073,
+      "grad_norm": 0.48814457654953003,
+      "learning_rate": 0.0002,
+      "loss": 1.2961,
+      "step": 2550
+    },
+    {
+      "epoch": 1.8530582699963807,
+      "grad_norm": 0.6216312646865845,
+      "learning_rate": 0.0002,
+      "loss": 1.466,
+      "step": 2560
+    },
+    {
+      "epoch": 1.860296778863554,
+      "grad_norm": 0.635603666305542,
+      "learning_rate": 0.0002,
+      "loss": 1.5123,
+      "step": 2570
+    },
+    {
+      "epoch": 1.8675352877307274,
+      "grad_norm": 0.6938216090202332,
+      "learning_rate": 0.0002,
+      "loss": 1.372,
+      "step": 2580
+    },
+    {
+      "epoch": 1.8747737965979008,
+      "grad_norm": 0.599557638168335,
+      "learning_rate": 0.0002,
+      "loss": 1.5011,
+      "step": 2590
+    },
+    {
+      "epoch": 1.8820123054650741,
+      "grad_norm": 0.564424455165863,
+      "learning_rate": 0.0002,
+      "loss": 1.2714,
+      "step": 2600
+    },
+    {
+      "epoch": 1.8892508143322475,
+      "grad_norm": 0.5430700182914734,
+      "learning_rate": 0.0002,
+      "loss": 1.3403,
+      "step": 2610
+    },
+    {
+      "epoch": 1.8964893231994209,
+      "grad_norm": 0.6150169372558594,
+      "learning_rate": 0.0002,
+      "loss": 1.4347,
+      "step": 2620
+    },
+    {
+      "epoch": 1.9037278320665942,
+      "grad_norm": 0.48159119486808777,
+      "learning_rate": 0.0002,
+      "loss": 1.2474,
+      "step": 2630
+    },
+    {
+      "epoch": 1.9109663409337676,
+      "grad_norm": 0.5608997941017151,
+      "learning_rate": 0.0002,
+      "loss": 1.3716,
+      "step": 2640
+    },
+    {
+      "epoch": 1.918204849800941,
+      "grad_norm": 0.6454501748085022,
+      "learning_rate": 0.0002,
+      "loss": 1.5787,
+      "step": 2650
+    },
+    {
+      "epoch": 1.9254433586681143,
+      "grad_norm": 0.5458073616027832,
+      "learning_rate": 0.0002,
+      "loss": 1.3238,
+      "step": 2660
+    },
+    {
+      "epoch": 1.9326818675352877,
+      "grad_norm": 0.5328490734100342,
+      "learning_rate": 0.0002,
+      "loss": 1.3208,
+      "step": 2670
+    },
+    {
+      "epoch": 1.939920376402461,
+      "grad_norm": 0.6444696187973022,
+      "learning_rate": 0.0002,
+      "loss": 1.4971,
+      "step": 2680
+    },
+    {
+      "epoch": 1.9471588852696344,
+      "grad_norm": 0.7126023769378662,
+      "learning_rate": 0.0002,
+      "loss": 1.5387,
+      "step": 2690
+    },
+    {
+      "epoch": 1.9543973941368078,
+      "grad_norm": 0.5164045095443726,
+      "learning_rate": 0.0002,
+      "loss": 1.3637,
+      "step": 2700
+    },
+    {
+      "epoch": 1.9616359030039812,
+      "grad_norm": 0.5347061157226562,
+      "learning_rate": 0.0002,
+      "loss": 1.5303,
+      "step": 2710
+    },
+    {
+      "epoch": 1.9688744118711545,
+      "grad_norm": 0.5297950506210327,
+      "learning_rate": 0.0002,
+      "loss": 1.2815,
+      "step": 2720
+    },
+    {
+      "epoch": 1.976112920738328,
+      "grad_norm": 0.6537790298461914,
+      "learning_rate": 0.0002,
+      "loss": 1.3566,
+      "step": 2730
+    },
+    {
+      "epoch": 1.9833514296055013,
+      "grad_norm": 0.5536222457885742,
+      "learning_rate": 0.0002,
+      "loss": 1.332,
+      "step": 2740
+    },
+    {
+      "epoch": 1.9905899384726746,
+      "grad_norm": 0.4856105446815491,
+      "learning_rate": 0.0002,
+      "loss": 1.3333,
+      "step": 2750
+    },
+    {
+      "epoch": 1.997828447339848,
+      "grad_norm": 0.6642730832099915,
+      "learning_rate": 0.0002,
+      "loss": 1.3521,
+      "step": 2760
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.4366681575775146,
+      "eval_runtime": 27.3729,
+      "eval_samples_per_second": 15.928,
+      "eval_steps_per_second": 2.009,
+      "step": 2763
+    },
+    {
+      "epoch": 2.0050669562070214,
+      "grad_norm": 0.740253210067749,
+      "learning_rate": 0.0002,
+      "loss": 1.4322,
+      "step": 2770
+    },
+    {
+      "epoch": 2.0123054650741947,
+      "grad_norm": 0.5826276540756226,
+      "learning_rate": 0.0002,
+      "loss": 1.277,
+      "step": 2780
+    },
+    {
+      "epoch": 2.019543973941368,
+      "grad_norm": 0.607356071472168,
+      "learning_rate": 0.0002,
+      "loss": 1.2424,
+      "step": 2790
+    },
+    {
+      "epoch": 2.0267824828085415,
+      "grad_norm": 0.5918063521385193,
+      "learning_rate": 0.0002,
+      "loss": 1.2601,
+      "step": 2800
+    },
+    {
+      "epoch": 2.034020991675715,
+      "grad_norm": 0.5610089898109436,
+      "learning_rate": 0.0002,
+      "loss": 1.3715,
+      "step": 2810
+    },
+    {
+      "epoch": 2.041259500542888,
+      "grad_norm": 0.5869926810264587,
+      "learning_rate": 0.0002,
+      "loss": 1.2092,
+      "step": 2820
+    },
+    {
+      "epoch": 2.0484980094100615,
+      "grad_norm": 0.5753467679023743,
+      "learning_rate": 0.0002,
+      "loss": 1.1929,
+      "step": 2830
+    },
+    {
+      "epoch": 2.055736518277235,
+      "grad_norm": 0.7096508145332336,
+      "learning_rate": 0.0002,
+      "loss": 1.333,
+      "step": 2840
+    },
+    {
+      "epoch": 2.0629750271444083,
+      "grad_norm": 0.7653635144233704,
+      "learning_rate": 0.0002,
+      "loss": 1.1766,
+      "step": 2850
+    },
+    {
+      "epoch": 2.0702135360115816,
+      "grad_norm": 0.6202841997146606,
+      "learning_rate": 0.0002,
+      "loss": 1.2331,
+      "step": 2860
+    },
+    {
+      "epoch": 2.077452044878755,
+      "grad_norm": 0.6810227632522583,
+      "learning_rate": 0.0002,
+      "loss": 1.3298,
+      "step": 2870
+    },
+    {
+      "epoch": 2.0846905537459284,
+      "grad_norm": 0.7481493353843689,
+      "learning_rate": 0.0002,
+      "loss": 1.2505,
+      "step": 2880
+    },
+    {
+      "epoch": 2.0919290626131017,
+      "grad_norm": 0.7089637517929077,
+      "learning_rate": 0.0002,
+      "loss": 1.2484,
+      "step": 2890
+    },
+    {
+      "epoch": 2.099167571480275,
+      "grad_norm": 0.7472923398017883,
+      "learning_rate": 0.0002,
+      "loss": 1.3095,
+      "step": 2900
+    },
+    {
+      "epoch": 2.1064060803474485,
+      "grad_norm": 0.8135465979576111,
+      "learning_rate": 0.0002,
+      "loss": 1.304,
+      "step": 2910
+    },
+    {
+      "epoch": 2.113644589214622,
+      "grad_norm": 0.6097133159637451,
+      "learning_rate": 0.0002,
+      "loss": 1.273,
+      "step": 2920
+    },
+    {
+      "epoch": 2.120883098081795,
+      "grad_norm": 0.5970117449760437,
+      "learning_rate": 0.0002,
+      "loss": 1.3384,
+      "step": 2930
+    },
+    {
+      "epoch": 2.1281216069489686,
+      "grad_norm": 0.6169309616088867,
+      "learning_rate": 0.0002,
+      "loss": 1.3233,
+      "step": 2940
+    },
+    {
+      "epoch": 2.135360115816142,
+      "grad_norm": 0.9428738355636597,
+      "learning_rate": 0.0002,
+      "loss": 1.4246,
+      "step": 2950
+    },
+    {
+      "epoch": 2.1425986246833153,
+      "grad_norm": 0.5671679973602295,
+      "learning_rate": 0.0002,
+      "loss": 1.3527,
+      "step": 2960
+    },
+    {
+      "epoch": 2.1498371335504887,
+      "grad_norm": 0.7007262110710144,
+      "learning_rate": 0.0002,
+      "loss": 1.1375,
+      "step": 2970
+    },
+    {
+      "epoch": 2.157075642417662,
+      "grad_norm": 0.6294044256210327,
+      "learning_rate": 0.0002,
+      "loss": 1.2015,
+      "step": 2980
+    },
+    {
+      "epoch": 2.1643141512848354,
+      "grad_norm": 0.6105241775512695,
+      "learning_rate": 0.0002,
+      "loss": 1.2167,
+      "step": 2990
+    },
+    {
+      "epoch": 2.1715526601520088,
+      "grad_norm": 0.557124137878418,
+      "learning_rate": 0.0002,
+      "loss": 1.2065,
+      "step": 3000
+    },
+    {
+      "epoch": 2.178791169019182,
+      "grad_norm": 0.6250392198562622,
+      "learning_rate": 0.0002,
+      "loss": 1.2515,
+      "step": 3010
+    },
+    {
+      "epoch": 2.1860296778863555,
+      "grad_norm": 0.645218551158905,
+      "learning_rate": 0.0002,
+      "loss": 1.385,
+      "step": 3020
+    },
+    {
+      "epoch": 2.193268186753529,
+      "grad_norm": 0.9033605456352234,
+      "learning_rate": 0.0002,
+      "loss": 1.3928,
+      "step": 3030
+    },
+    {
+      "epoch": 2.2005066956207022,
+      "grad_norm": 0.5325747132301331,
+      "learning_rate": 0.0002,
+      "loss": 1.2458,
+      "step": 3040
+    },
+    {
+      "epoch": 2.2077452044878756,
+      "grad_norm": 0.6334700584411621,
+      "learning_rate": 0.0002,
+      "loss": 1.261,
+      "step": 3050
+    },
+    {
+      "epoch": 2.214983713355049,
+      "grad_norm": 0.5206325054168701,
+      "learning_rate": 0.0002,
+      "loss": 1.2385,
+      "step": 3060
+    },
+    {
+      "epoch": 2.2222222222222223,
+      "grad_norm": 0.5987200140953064,
+      "learning_rate": 0.0002,
+      "loss": 1.3103,
+      "step": 3070
+    },
+    {
+      "epoch": 2.2294607310893957,
+      "grad_norm": 0.5893264412879944,
+      "learning_rate": 0.0002,
+      "loss": 1.1756,
+      "step": 3080
+    },
+    {
+      "epoch": 2.236699239956569,
+      "grad_norm": 0.6869237422943115,
+      "learning_rate": 0.0002,
+      "loss": 1.235,
+      "step": 3090
+    },
+    {
+      "epoch": 2.2439377488237424,
+      "grad_norm": 0.5040048360824585,
+      "learning_rate": 0.0002,
+      "loss": 1.3285,
+      "step": 3100
+    },
+    {
+      "epoch": 2.251176257690916,
+      "grad_norm": 0.6660613417625427,
+      "learning_rate": 0.0002,
+      "loss": 1.3316,
+      "step": 3110
+    },
+    {
+      "epoch": 2.258414766558089,
+      "grad_norm": 0.5890918970108032,
+      "learning_rate": 0.0002,
+      "loss": 1.3108,
+      "step": 3120
+    },
+    {
+      "epoch": 2.2656532754252625,
+      "grad_norm": 0.6458896994590759,
+      "learning_rate": 0.0002,
+      "loss": 1.248,
+      "step": 3130
+    },
+    {
+      "epoch": 2.272891784292436,
+      "grad_norm": 0.6832690834999084,
+      "learning_rate": 0.0002,
+      "loss": 1.4151,
+      "step": 3140
+    },
+    {
+      "epoch": 2.2801302931596092,
+      "grad_norm": 0.833908200263977,
+      "learning_rate": 0.0002,
+      "loss": 1.4458,
+      "step": 3150
+    },
+    {
+      "epoch": 2.2873688020267826,
+      "grad_norm": 0.4596034586429596,
+      "learning_rate": 0.0002,
+      "loss": 1.2931,
+      "step": 3160
+    },
+    {
+      "epoch": 2.294607310893956,
+      "grad_norm": 0.9130966067314148,
+      "learning_rate": 0.0002,
+      "loss": 1.449,
+      "step": 3170
+    },
+    {
+      "epoch": 2.3018458197611293,
+      "grad_norm": 0.7143292427062988,
+      "learning_rate": 0.0002,
+      "loss": 1.3806,
+      "step": 3180
+    },
+    {
+      "epoch": 2.3090843286283027,
+      "grad_norm": 0.5388900637626648,
+      "learning_rate": 0.0002,
+      "loss": 1.2692,
+      "step": 3190
+    },
+    {
+      "epoch": 2.316322837495476,
+      "grad_norm": 0.5607513189315796,
+      "learning_rate": 0.0002,
+      "loss": 1.2402,
+      "step": 3200
+    },
+    {
+      "epoch": 2.3235613463626494,
+      "grad_norm": 0.6795142292976379,
+      "learning_rate": 0.0002,
+      "loss": 1.3874,
+      "step": 3210
+    },
+    {
+      "epoch": 2.330799855229823,
+      "grad_norm": 0.6561070680618286,
+      "learning_rate": 0.0002,
+      "loss": 1.3042,
+      "step": 3220
+    },
+    {
+      "epoch": 2.338038364096996,
+      "grad_norm": 0.8858118057250977,
+      "learning_rate": 0.0002,
+      "loss": 1.4636,
+      "step": 3230
+    },
+    {
+      "epoch": 2.3452768729641695,
+      "grad_norm": 0.6604151725769043,
+      "learning_rate": 0.0002,
+      "loss": 1.3214,
+      "step": 3240
+    },
+    {
+      "epoch": 2.352515381831343,
+      "grad_norm": 0.6755785346031189,
+      "learning_rate": 0.0002,
+      "loss": 1.4004,
+      "step": 3250
+    },
+    {
+      "epoch": 2.3597538906985163,
+      "grad_norm": 0.6981677412986755,
+      "learning_rate": 0.0002,
+      "loss": 1.2503,
+      "step": 3260
+    },
+    {
+      "epoch": 2.3669923995656896,
+      "grad_norm": 0.6338568329811096,
+      "learning_rate": 0.0002,
+      "loss": 1.3078,
+      "step": 3270
+    },
+    {
+      "epoch": 2.374230908432863,
+      "grad_norm": 0.5754265785217285,
+      "learning_rate": 0.0002,
+      "loss": 1.285,
+      "step": 3280
+    },
+    {
+      "epoch": 2.3814694173000364,
+      "grad_norm": 0.7533153295516968,
+      "learning_rate": 0.0002,
+      "loss": 1.2924,
+      "step": 3290
+    },
+    {
+      "epoch": 2.3887079261672097,
+      "grad_norm": 0.675065279006958,
+      "learning_rate": 0.0002,
+      "loss": 1.3711,
+      "step": 3300
+    },
+    {
+      "epoch": 2.395946435034383,
+      "grad_norm": 0.5686452984809875,
+      "learning_rate": 0.0002,
+      "loss": 1.3548,
+      "step": 3310
+    },
+    {
+      "epoch": 2.4031849439015565,
+      "grad_norm": 0.8129481673240662,
+      "learning_rate": 0.0002,
+      "loss": 1.1998,
+      "step": 3320
+    },
+    {
+      "epoch": 2.41042345276873,
+      "grad_norm": 0.6615934371948242,
+      "learning_rate": 0.0002,
+      "loss": 1.2584,
+      "step": 3330
+    },
+    {
+      "epoch": 2.417661961635903,
+      "grad_norm": 0.6678834557533264,
+      "learning_rate": 0.0002,
+      "loss": 1.3691,
+      "step": 3340
+    },
+    {
+      "epoch": 2.4249004705030766,
+      "grad_norm": 0.5581308007240295,
+      "learning_rate": 0.0002,
+      "loss": 1.2381,
+      "step": 3350
+    },
+    {
+      "epoch": 2.43213897937025,
+      "grad_norm": 0.6098920106887817,
+      "learning_rate": 0.0002,
+      "loss": 1.3853,
+      "step": 3360
+    },
+    {
+      "epoch": 2.4393774882374233,
+      "grad_norm": 0.8101736903190613,
+      "learning_rate": 0.0002,
+      "loss": 1.3692,
+      "step": 3370
+    },
+    {
+      "epoch": 2.4466159971045967,
+      "grad_norm": 0.6621488928794861,
+      "learning_rate": 0.0002,
+      "loss": 1.4418,
+      "step": 3380
+    },
+    {
+      "epoch": 2.45385450597177,
+      "grad_norm": 0.8693289160728455,
+      "learning_rate": 0.0002,
+      "loss": 1.4579,
+      "step": 3390
+    },
+    {
+      "epoch": 2.4610930148389434,
+      "grad_norm": 0.6724580526351929,
+      "learning_rate": 0.0002,
+      "loss": 1.3644,
+      "step": 3400
+    },
+    {
+      "epoch": 2.4683315237061167,
+      "grad_norm": 0.6776891946792603,
+      "learning_rate": 0.0002,
+      "loss": 1.2006,
+      "step": 3410
+    },
+    {
+      "epoch": 2.47557003257329,
+      "grad_norm": 0.7214453816413879,
+      "learning_rate": 0.0002,
+      "loss": 1.2937,
+      "step": 3420
+    },
+    {
+      "epoch": 2.4828085414404635,
+      "grad_norm": 0.8390451073646545,
+      "learning_rate": 0.0002,
+      "loss": 1.4051,
+      "step": 3430
+    },
+    {
+      "epoch": 2.490047050307637,
+      "grad_norm": 0.7130982279777527,
+      "learning_rate": 0.0002,
+      "loss": 1.25,
+      "step": 3440
+    },
+    {
+      "epoch": 2.49728555917481,
+      "grad_norm": 0.8873937129974365,
+      "learning_rate": 0.0002,
+      "loss": 1.2231,
+      "step": 3450
+    },
+    {
+      "epoch": 2.5045240680419836,
+      "grad_norm": 0.725185751914978,
+      "learning_rate": 0.0002,
+      "loss": 1.1429,
+      "step": 3460
+    },
+    {
+      "epoch": 2.511762576909157,
+      "grad_norm": 0.6120352149009705,
+      "learning_rate": 0.0002,
+      "loss": 1.2699,
+      "step": 3470
+    },
+    {
+      "epoch": 2.5190010857763303,
+      "grad_norm": 0.7713613510131836,
+      "learning_rate": 0.0002,
+      "loss": 1.2552,
+      "step": 3480
+    },
+    {
+      "epoch": 2.5262395946435037,
+      "grad_norm": 0.895309567451477,
+      "learning_rate": 0.0002,
+      "loss": 1.4648,
+      "step": 3490
+    },
+    {
+      "epoch": 2.533478103510677,
+      "grad_norm": 0.9631021022796631,
+      "learning_rate": 0.0002,
+      "loss": 1.3043,
+      "step": 3500
+    },
+    {
+      "epoch": 2.5407166123778504,
+      "grad_norm": 0.7475683093070984,
+      "learning_rate": 0.0002,
+      "loss": 1.3492,
+      "step": 3510
+    },
+    {
+      "epoch": 2.5479551212450238,
+      "grad_norm": 0.7271341681480408,
+      "learning_rate": 0.0002,
+      "loss": 1.3637,
+      "step": 3520
+    },
+    {
+      "epoch": 2.555193630112197,
+      "grad_norm": 0.6979510188102722,
+      "learning_rate": 0.0002,
+      "loss": 1.304,
+      "step": 3530
+    },
+    {
+      "epoch": 2.5624321389793705,
+      "grad_norm": 0.6504196524620056,
+      "learning_rate": 0.0002,
+      "loss": 1.2353,
+      "step": 3540
+    },
+    {
+      "epoch": 2.569670647846544,
+      "grad_norm": 0.7226675748825073,
+      "learning_rate": 0.0002,
+      "loss": 1.2699,
+      "step": 3550
+    },
+    {
+      "epoch": 2.5769091567137172,
+      "grad_norm": 0.6143222451210022,
+      "learning_rate": 0.0002,
+      "loss": 1.3002,
+      "step": 3560
+    },
+    {
+      "epoch": 2.5841476655808906,
+      "grad_norm": 0.7245154976844788,
+      "learning_rate": 0.0002,
+      "loss": 1.1585,
+      "step": 3570
+    },
+    {
+      "epoch": 2.591386174448064,
+      "grad_norm": 0.943540632724762,
+      "learning_rate": 0.0002,
+      "loss": 1.3651,
+      "step": 3580
+    },
+    {
+      "epoch": 2.5986246833152373,
+      "grad_norm": 0.7707241773605347,
+      "learning_rate": 0.0002,
+      "loss": 1.3034,
+      "step": 3590
+    },
+    {
+      "epoch": 2.6058631921824107,
+      "grad_norm": 0.6705001592636108,
+      "learning_rate": 0.0002,
+      "loss": 1.3063,
+      "step": 3600
+    },
+    {
+      "epoch": 2.613101701049584,
+      "grad_norm": 0.6360933780670166,
+      "learning_rate": 0.0002,
+      "loss": 1.2437,
+      "step": 3610
+    },
+    {
+      "epoch": 2.6203402099167574,
+      "grad_norm": 0.5846424698829651,
+      "learning_rate": 0.0002,
+      "loss": 1.1844,
+      "step": 3620
+    },
+    {
+      "epoch": 2.6275787187839303,
+      "grad_norm": 0.5958625674247742,
+      "learning_rate": 0.0002,
+      "loss": 1.3674,
+      "step": 3630
+    },
+    {
+      "epoch": 2.6348172276511037,
+      "grad_norm": 0.6819243431091309,
+      "learning_rate": 0.0002,
+      "loss": 1.3599,
+      "step": 3640
+    },
+    {
+      "epoch": 2.642055736518277,
+      "grad_norm": 0.7033445835113525,
+      "learning_rate": 0.0002,
+      "loss": 1.3884,
+      "step": 3650
+    },
+    {
+      "epoch": 2.6492942453854504,
+      "grad_norm": 0.6134849786758423,
+      "learning_rate": 0.0002,
+      "loss": 1.3392,
+      "step": 3660
+    },
+    {
+      "epoch": 2.656532754252624,
+      "grad_norm": 0.658009946346283,
+      "learning_rate": 0.0002,
+      "loss": 1.2661,
+      "step": 3670
+    },
+    {
+      "epoch": 2.663771263119797,
+      "grad_norm": 0.6280999779701233,
+      "learning_rate": 0.0002,
+      "loss": 1.3987,
+      "step": 3680
+    },
+    {
+      "epoch": 2.6710097719869705,
+      "grad_norm": 0.5536085963249207,
+      "learning_rate": 0.0002,
+      "loss": 1.2995,
+      "step": 3690
+    },
+    {
+      "epoch": 2.678248280854144,
+      "grad_norm": 0.8603981733322144,
+      "learning_rate": 0.0002,
+      "loss": 1.2044,
+      "step": 3700
+    },
+    {
+      "epoch": 2.6854867897213173,
+      "grad_norm": 0.5509994626045227,
+      "learning_rate": 0.0002,
+      "loss": 1.3879,
+      "step": 3710
+    },
+    {
+      "epoch": 2.6927252985884906,
+      "grad_norm": 0.9093621969223022,
+      "learning_rate": 0.0002,
+      "loss": 1.3253,
+      "step": 3720
+    },
+    {
+      "epoch": 2.699963807455664,
+      "grad_norm": 0.7525952458381653,
+      "learning_rate": 0.0002,
+      "loss": 1.2668,
+      "step": 3730
+    },
+    {
+      "epoch": 2.7072023163228374,
+      "grad_norm": 0.6737023591995239,
+      "learning_rate": 0.0002,
+      "loss": 1.248,
+      "step": 3740
+    },
+    {
+      "epoch": 2.7144408251900107,
+      "grad_norm": 0.8656924962997437,
+      "learning_rate": 0.0002,
+      "loss": 1.2981,
+      "step": 3750
+    },
+    {
+      "epoch": 2.721679334057184,
+      "grad_norm": 0.7494133114814758,
+      "learning_rate": 0.0002,
+      "loss": 1.2342,
+      "step": 3760
+    },
+    {
+      "epoch": 2.7289178429243575,
+      "grad_norm": 0.5725520849227905,
+      "learning_rate": 0.0002,
+      "loss": 1.2417,
+      "step": 3770
+    },
+    {
+      "epoch": 2.736156351791531,
+      "grad_norm": 0.836412787437439,
+      "learning_rate": 0.0002,
+      "loss": 1.28,
+      "step": 3780
+    },
+    {
+      "epoch": 2.743394860658704,
+      "grad_norm": 0.6893242597579956,
+      "learning_rate": 0.0002,
+      "loss": 1.3784,
+      "step": 3790
+    },
+    {
+      "epoch": 2.7506333695258776,
+      "grad_norm": 0.6696223020553589,
+      "learning_rate": 0.0002,
+      "loss": 1.2929,
+      "step": 3800
+    },
+    {
+      "epoch": 2.757871878393051,
+      "grad_norm": 0.6483015418052673,
+      "learning_rate": 0.0002,
+      "loss": 1.2449,
+      "step": 3810
+    },
+    {
+      "epoch": 2.7651103872602243,
+      "grad_norm": 0.8084456920623779,
+      "learning_rate": 0.0002,
+      "loss": 1.3282,
+      "step": 3820
+    },
+    {
+      "epoch": 2.7723488961273977,
+      "grad_norm": 0.6601949334144592,
+      "learning_rate": 0.0002,
+      "loss": 1.3694,
+      "step": 3830
+    },
+    {
+      "epoch": 2.779587404994571,
+      "grad_norm": 0.6905533671379089,
+      "learning_rate": 0.0002,
+      "loss": 1.3568,
+      "step": 3840
+    },
+    {
+      "epoch": 2.7868259138617444,
+      "grad_norm": 0.619318425655365,
+      "learning_rate": 0.0002,
+      "loss": 1.3854,
+      "step": 3850
+    },
+    {
+      "epoch": 2.7940644227289178,
+      "grad_norm": 0.5994023084640503,
+      "learning_rate": 0.0002,
+      "loss": 1.2551,
+      "step": 3860
+    },
+    {
+      "epoch": 2.801302931596091,
+      "grad_norm": 0.5627168416976929,
+      "learning_rate": 0.0002,
+      "loss": 1.2022,
+      "step": 3870
+    },
+    {
+      "epoch": 2.8085414404632645,
+      "grad_norm": 0.6001605987548828,
+      "learning_rate": 0.0002,
+      "loss": 1.3921,
+      "step": 3880
+    },
+    {
+      "epoch": 2.815779949330438,
+      "grad_norm": 0.6022412776947021,
+      "learning_rate": 0.0002,
+      "loss": 1.3026,
+      "step": 3890
+    },
+    {
+      "epoch": 2.823018458197611,
+      "grad_norm": 0.6832426190376282,
+      "learning_rate": 0.0002,
+      "loss": 1.2765,
+      "step": 3900
+    },
+    {
+      "epoch": 2.8302569670647846,
+      "grad_norm": 0.5936811566352844,
+      "learning_rate": 0.0002,
+      "loss": 1.1363,
+      "step": 3910
+    },
+    {
+      "epoch": 2.837495475931958,
+      "grad_norm": 0.6960572600364685,
+      "learning_rate": 0.0002,
+      "loss": 1.1707,
+      "step": 3920
+    },
+    {
+      "epoch": 2.8447339847991313,
+      "grad_norm": 0.5913406610488892,
+      "learning_rate": 0.0002,
+      "loss": 1.4063,
+      "step": 3930
+    },
+    {
+      "epoch": 2.8519724936663047,
+      "grad_norm": 0.678154706954956,
+      "learning_rate": 0.0002,
+      "loss": 1.3245,
+      "step": 3940
+    },
+    {
+      "epoch": 2.859211002533478,
+      "grad_norm": 0.7898936867713928,
+      "learning_rate": 0.0002,
+      "loss": 1.366,
+      "step": 3950
+    },
+    {
+      "epoch": 2.8664495114006514,
+      "grad_norm": 0.9234195351600647,
+      "learning_rate": 0.0002,
+      "loss": 1.3948,
+      "step": 3960
+    },
+    {
+      "epoch": 2.8736880202678248,
+      "grad_norm": 0.5960825085639954,
+      "learning_rate": 0.0002,
+      "loss": 1.2773,
+      "step": 3970
+    },
+    {
+      "epoch": 2.880926529134998,
+      "grad_norm": 0.677118182182312,
+      "learning_rate": 0.0002,
+      "loss": 1.3127,
+      "step": 3980
+    },
+    {
+      "epoch": 2.8881650380021715,
+      "grad_norm": 0.6505142450332642,
+      "learning_rate": 0.0002,
+      "loss": 1.2652,
+      "step": 3990
+    },
+    {
+      "epoch": 2.895403546869345,
+      "grad_norm": 0.550826907157898,
+      "learning_rate": 0.0002,
+      "loss": 1.2078,
+      "step": 4000
+    },
+    {
+      "epoch": 2.9026420557365182,
+      "grad_norm": 0.6209215521812439,
+      "learning_rate": 0.0002,
+      "loss": 1.1811,
+      "step": 4010
+    },
+    {
+      "epoch": 2.9098805646036916,
+      "grad_norm": 0.6549018025398254,
+      "learning_rate": 0.0002,
+      "loss": 1.4001,
+      "step": 4020
+    },
+    {
+      "epoch": 2.917119073470865,
+      "grad_norm": 0.570682168006897,
+      "learning_rate": 0.0002,
+      "loss": 1.2285,
+      "step": 4030
+    },
+    {
+      "epoch": 2.9243575823380383,
+      "grad_norm": 1.1807632446289062,
+      "learning_rate": 0.0002,
+      "loss": 1.0832,
+      "step": 4040
+    },
+    {
+      "epoch": 2.9315960912052117,
+      "grad_norm": 0.7058857679367065,
+      "learning_rate": 0.0002,
+      "loss": 1.2693,
+      "step": 4050
+    },
+    {
+      "epoch": 2.938834600072385,
+      "grad_norm": 0.5542812943458557,
+      "learning_rate": 0.0002,
+      "loss": 1.2905,
+      "step": 4060
+    },
+    {
+      "epoch": 2.9460731089395584,
+      "grad_norm": 0.63167804479599,
+      "learning_rate": 0.0002,
+      "loss": 1.33,
+      "step": 4070
+    },
+    {
+      "epoch": 2.953311617806732,
+      "grad_norm": 0.5702962279319763,
+      "learning_rate": 0.0002,
+      "loss": 1.3075,
+      "step": 4080
+    },
+    {
+      "epoch": 2.960550126673905,
+      "grad_norm": 0.620944082736969,
+      "learning_rate": 0.0002,
+      "loss": 1.2007,
+      "step": 4090
+    },
+    {
+      "epoch": 2.9677886355410785,
+      "grad_norm": 0.5866289734840393,
+      "learning_rate": 0.0002,
+      "loss": 1.2864,
+      "step": 4100
+    },
+    {
+      "epoch": 2.975027144408252,
+      "grad_norm": 0.560170590877533,
+      "learning_rate": 0.0002,
+      "loss": 1.3293,
+      "step": 4110
+    },
+    {
+      "epoch": 2.9822656532754253,
+      "grad_norm": 0.675082802772522,
+      "learning_rate": 0.0002,
+      "loss": 1.2071,
+      "step": 4120
+    },
+    {
+      "epoch": 2.9895041621425986,
+      "grad_norm": 0.62708580493927,
+      "learning_rate": 0.0002,
+      "loss": 1.2981,
+      "step": 4130
+    },
+    {
+      "epoch": 2.996742671009772,
+      "grad_norm": 0.7893929481506348,
+      "learning_rate": 0.0002,
+      "loss": 1.2758,
+      "step": 4140
+    },
+    {
+      "epoch": 2.9996380745566413,
+      "eval_loss": 1.4217946529388428,
+      "eval_runtime": 27.1596,
+      "eval_samples_per_second": 16.053,
+      "eval_steps_per_second": 2.025,
+      "step": 4144
+    },
+    {
+      "epoch": 3.0039811798769454,
+      "grad_norm": 0.7043836116790771,
+      "learning_rate": 0.0002,
+      "loss": 1.2152,
+      "step": 4150
+    },
+    {
+      "epoch": 3.0112196887441187,
+      "grad_norm": 0.6806283593177795,
+      "learning_rate": 0.0002,
+      "loss": 1.1664,
+      "step": 4160
+    },
+    {
+      "epoch": 3.018458197611292,
+      "grad_norm": 0.7684550285339355,
+      "learning_rate": 0.0002,
+      "loss": 1.292,
+      "step": 4170
+    },
+    {
+      "epoch": 3.0256967064784654,
+      "grad_norm": 0.7895237803459167,
+      "learning_rate": 0.0002,
+      "loss": 1.3467,
+      "step": 4180
+    },
+    {
+      "epoch": 3.032935215345639,
+      "grad_norm": 0.7464531064033508,
+      "learning_rate": 0.0002,
+      "loss": 1.1324,
+      "step": 4190
+    },
+    {
+      "epoch": 3.040173724212812,
+      "grad_norm": 0.9358500838279724,
+      "learning_rate": 0.0002,
+      "loss": 1.1614,
+      "step": 4200
+    },
+    {
+      "epoch": 3.0474122330799855,
+      "grad_norm": 1.1066628694534302,
+      "learning_rate": 0.0002,
+      "loss": 1.1834,
+      "step": 4210
+    },
+    {
+      "epoch": 3.054650741947159,
+      "grad_norm": 0.6663267612457275,
+      "learning_rate": 0.0002,
+      "loss": 1.1557,
+      "step": 4220
+    },
+    {
+      "epoch": 3.0618892508143323,
+      "grad_norm": 0.6669464707374573,
+      "learning_rate": 0.0002,
+      "loss": 1.1707,
+      "step": 4230
+    },
+    {
+      "epoch": 3.0691277596815056,
+      "grad_norm": 0.7052164077758789,
+      "learning_rate": 0.0002,
+      "loss": 1.1841,
+      "step": 4240
+    },
+    {
+      "epoch": 3.076366268548679,
+      "grad_norm": 0.6118432879447937,
+      "learning_rate": 0.0002,
+      "loss": 1.2913,
+      "step": 4250
+    },
+    {
+      "epoch": 3.0836047774158524,
+      "grad_norm": 0.6915903687477112,
+      "learning_rate": 0.0002,
+      "loss": 1.1526,
+      "step": 4260
+    },
+    {
+      "epoch": 3.0908432862830257,
+      "grad_norm": 0.7441644668579102,
+      "learning_rate": 0.0002,
+      "loss": 1.1348,
+      "step": 4270
+    },
+    {
+      "epoch": 3.098081795150199,
+      "grad_norm": 0.823850691318512,
+      "learning_rate": 0.0002,
+      "loss": 1.1672,
+      "step": 4280
+    },
+    {
+      "epoch": 3.1053203040173725,
+      "grad_norm": 0.9677883386611938,
+      "learning_rate": 0.0002,
+      "loss": 1.2655,
+      "step": 4290
+    },
+    {
+      "epoch": 3.112558812884546,
+      "grad_norm": 0.7002579569816589,
+      "learning_rate": 0.0002,
+      "loss": 1.1794,
+      "step": 4300
+    },
+    {
+      "epoch": 3.119797321751719,
+      "grad_norm": 0.778789758682251,
+      "learning_rate": 0.0002,
+      "loss": 1.135,
+      "step": 4310
+    },
+    {
+      "epoch": 3.1270358306188926,
+      "grad_norm": 0.7236007452011108,
+      "learning_rate": 0.0002,
+      "loss": 1.0818,
+      "step": 4320
+    },
+    {
+      "epoch": 3.134274339486066,
+      "grad_norm": 0.8809133768081665,
+      "learning_rate": 0.0002,
+      "loss": 1.1803,
+      "step": 4330
+    },
+    {
+      "epoch": 3.1415128483532393,
+      "grad_norm": 0.7924913167953491,
+      "learning_rate": 0.0002,
+      "loss": 1.2571,
+      "step": 4340
+    },
+    {
+      "epoch": 3.1487513572204127,
+      "grad_norm": 0.7437422275543213,
+      "learning_rate": 0.0002,
+      "loss": 1.1413,
+      "step": 4350
+    },
+    {
+      "epoch": 3.155989866087586,
+      "grad_norm": 0.6428450345993042,
+      "learning_rate": 0.0002,
+      "loss": 1.2088,
+      "step": 4360
+    },
+    {
+      "epoch": 3.1632283749547594,
+      "grad_norm": 0.7922873497009277,
+      "learning_rate": 0.0002,
+      "loss": 1.3032,
+      "step": 4370
+    },
+    {
+      "epoch": 3.1704668838219328,
+      "grad_norm": 0.5252506732940674,
+      "learning_rate": 0.0002,
+      "loss": 1.216,
+      "step": 4380
+    },
+    {
+      "epoch": 3.177705392689106,
+      "grad_norm": 0.8570457696914673,
+      "learning_rate": 0.0002,
+      "loss": 1.1297,
+      "step": 4390
+    },
+    {
+      "epoch": 3.1849439015562795,
+      "grad_norm": 0.7218987345695496,
+      "learning_rate": 0.0002,
+      "loss": 1.0994,
+      "step": 4400
+    },
+    {
+      "epoch": 3.192182410423453,
+      "grad_norm": 0.6921393275260925,
+      "learning_rate": 0.0002,
+      "loss": 1.2891,
+      "step": 4410
+    },
+    {
+      "epoch": 3.199420919290626,
+      "grad_norm": 0.7386137843132019,
+      "learning_rate": 0.0002,
+      "loss": 1.2668,
+      "step": 4420
+    },
+    {
+      "epoch": 3.2066594281577996,
+      "grad_norm": 0.6227759122848511,
+      "learning_rate": 0.0002,
+      "loss": 1.1654,
+      "step": 4430
+    },
+    {
+      "epoch": 3.213897937024973,
+      "grad_norm": 0.7180278897285461,
+      "learning_rate": 0.0002,
+      "loss": 1.1752,
+      "step": 4440
+    },
+    {
+      "epoch": 3.2211364458921463,
+      "grad_norm": 0.745830774307251,
+      "learning_rate": 0.0002,
+      "loss": 1.1757,
+      "step": 4450
+    },
+    {
+      "epoch": 3.2283749547593197,
+      "grad_norm": 0.6766072511672974,
+      "learning_rate": 0.0002,
+      "loss": 1.234,
+      "step": 4460
+    },
+    {
+      "epoch": 3.235613463626493,
+      "grad_norm": 0.8325067162513733,
+      "learning_rate": 0.0002,
+      "loss": 1.1999,
+      "step": 4470
+    },
+    {
+      "epoch": 3.2428519724936664,
+      "grad_norm": 0.7148305177688599,
+      "learning_rate": 0.0002,
+      "loss": 1.1606,
+      "step": 4480
+    },
+    {
+      "epoch": 3.25009048136084,
+      "grad_norm": 0.7752676010131836,
+      "learning_rate": 0.0002,
+      "loss": 1.1383,
+      "step": 4490
+    },
+    {
+      "epoch": 3.257328990228013,
+      "grad_norm": 0.6776860952377319,
+      "learning_rate": 0.0002,
+      "loss": 1.3006,
+      "step": 4500
+    },
+    {
+      "epoch": 3.2645674990951865,
+      "grad_norm": 0.704359769821167,
+      "learning_rate": 0.0002,
+      "loss": 1.0796,
+      "step": 4510
+    },
+    {
+      "epoch": 3.27180600796236,
+      "grad_norm": 0.6880282163619995,
+      "learning_rate": 0.0002,
+      "loss": 1.2496,
+      "step": 4520
+    },
+    {
+      "epoch": 3.2790445168295332,
+      "grad_norm": 0.8179270029067993,
+      "learning_rate": 0.0002,
+      "loss": 1.0947,
+      "step": 4530
+    },
+    {
+      "epoch": 3.2862830256967066,
+      "grad_norm": 0.6718448996543884,
+      "learning_rate": 0.0002,
+      "loss": 1.1909,
+      "step": 4540
+    },
+    {
+      "epoch": 3.29352153456388,
+      "grad_norm": 0.8300657868385315,
+      "learning_rate": 0.0002,
+      "loss": 1.2708,
+      "step": 4550
+    },
+    {
+      "epoch": 3.3007600434310533,
+      "grad_norm": 0.6433690786361694,
+      "learning_rate": 0.0002,
+      "loss": 1.2594,
+      "step": 4560
+    },
+    {
+      "epoch": 3.3079985522982267,
+      "grad_norm": 0.690262496471405,
+      "learning_rate": 0.0002,
+      "loss": 1.2479,
+      "step": 4570
+    },
+    {
+      "epoch": 3.3152370611654,
+      "grad_norm": 0.7022852301597595,
+      "learning_rate": 0.0002,
+      "loss": 1.1342,
+      "step": 4580
+    },
+    {
+      "epoch": 3.3224755700325734,
+      "grad_norm": 0.6438387632369995,
+      "learning_rate": 0.0002,
+      "loss": 1.0844,
+      "step": 4590
+    },
+    {
+      "epoch": 3.329714078899747,
+      "grad_norm": 0.6866899132728577,
+      "learning_rate": 0.0002,
+      "loss": 1.17,
+      "step": 4600
+    },
+    {
+      "epoch": 3.33695258776692,
+      "grad_norm": 0.8233968019485474,
+      "learning_rate": 0.0002,
+      "loss": 1.1289,
+      "step": 4610
+    },
+    {
+      "epoch": 3.3441910966340935,
+      "grad_norm": 0.7251574993133545,
+      "learning_rate": 0.0002,
+      "loss": 1.1855,
+      "step": 4620
+    },
+    {
+      "epoch": 3.351429605501267,
+      "grad_norm": 0.7855110168457031,
+      "learning_rate": 0.0002,
+      "loss": 1.3403,
+      "step": 4630
+    },
+    {
+      "epoch": 3.3586681143684403,
+      "grad_norm": 0.8487356305122375,
+      "learning_rate": 0.0002,
+      "loss": 1.2922,
+      "step": 4640
+    },
+    {
+      "epoch": 3.3659066232356136,
+      "grad_norm": 0.6429011225700378,
+      "learning_rate": 0.0002,
+      "loss": 1.2462,
+      "step": 4650
+    },
+    {
+      "epoch": 3.373145132102787,
+      "grad_norm": 0.7095270156860352,
+      "learning_rate": 0.0002,
+      "loss": 1.129,
+      "step": 4660
+    },
+    {
+      "epoch": 3.3803836409699604,
+      "grad_norm": 0.6792303323745728,
+      "learning_rate": 0.0002,
+      "loss": 1.262,
+      "step": 4670
+    },
+    {
+      "epoch": 3.3876221498371337,
+      "grad_norm": 0.6784825921058655,
+      "learning_rate": 0.0002,
+      "loss": 1.256,
+      "step": 4680
+    },
+    {
+      "epoch": 3.394860658704307,
+      "grad_norm": 0.6362888216972351,
+      "learning_rate": 0.0002,
+      "loss": 1.0838,
+      "step": 4690
+    },
+    {
+      "epoch": 3.4020991675714805,
+      "grad_norm": 0.7794778943061829,
+      "learning_rate": 0.0002,
+      "loss": 1.2165,
+      "step": 4700
+    },
+    {
+      "epoch": 3.409337676438654,
+      "grad_norm": 0.7287485003471375,
+      "learning_rate": 0.0002,
+      "loss": 1.0644,
+      "step": 4710
+    },
+    {
+      "epoch": 3.416576185305827,
+      "grad_norm": 0.6481451392173767,
+      "learning_rate": 0.0002,
+      "loss": 1.2925,
+      "step": 4720
+    },
+    {
+      "epoch": 3.4238146941730006,
+      "grad_norm": 0.9200371503829956,
+      "learning_rate": 0.0002,
+      "loss": 1.2121,
+      "step": 4730
+    },
+    {
+      "epoch": 3.431053203040174,
+      "grad_norm": 1.074180245399475,
+      "learning_rate": 0.0002,
+      "loss": 1.072,
+      "step": 4740
+    },
+    {
+      "epoch": 3.438291711907347,
+      "grad_norm": 0.6722986698150635,
+      "learning_rate": 0.0002,
+      "loss": 1.0421,
+      "step": 4750
+    },
+    {
+      "epoch": 3.44553022077452,
+      "grad_norm": 0.7945933938026428,
+      "learning_rate": 0.0002,
+      "loss": 1.2258,
+      "step": 4760
+    },
+    {
+      "epoch": 3.4527687296416936,
+      "grad_norm": 0.7624640464782715,
+      "learning_rate": 0.0002,
+      "loss": 1.0927,
+      "step": 4770
+    },
+    {
+      "epoch": 3.460007238508867,
+      "grad_norm": 0.7763656377792358,
+      "learning_rate": 0.0002,
+      "loss": 1.2428,
+      "step": 4780
+    },
+    {
+      "epoch": 3.4672457473760403,
+      "grad_norm": 0.7736947536468506,
+      "learning_rate": 0.0002,
+      "loss": 1.2584,
+      "step": 4790
+    },
+    {
+      "epoch": 3.4744842562432137,
+      "grad_norm": 0.8450354933738708,
+      "learning_rate": 0.0002,
+      "loss": 1.1953,
+      "step": 4800
+    },
+    {
+      "epoch": 3.481722765110387,
+      "grad_norm": 0.6480133533477783,
+      "learning_rate": 0.0002,
+      "loss": 1.1362,
+      "step": 4810
+    },
+    {
+      "epoch": 3.4889612739775604,
+      "grad_norm": 0.8437445759773254,
+      "learning_rate": 0.0002,
+      "loss": 1.1882,
+      "step": 4820
+    },
+    {
+      "epoch": 3.4961997828447338,
+      "grad_norm": 0.7781730890274048,
+      "learning_rate": 0.0002,
+      "loss": 1.1519,
+      "step": 4830
+    },
+    {
+      "epoch": 3.503438291711907,
+      "grad_norm": 0.8523228168487549,
+      "learning_rate": 0.0002,
+      "loss": 1.1836,
+      "step": 4840
+    },
+    {
+      "epoch": 3.5106768005790805,
+      "grad_norm": 0.6236732006072998,
+      "learning_rate": 0.0002,
+      "loss": 1.1672,
+      "step": 4850
+    },
+    {
+      "epoch": 3.517915309446254,
+      "grad_norm": 0.7500787377357483,
+      "learning_rate": 0.0002,
+      "loss": 1.1926,
+      "step": 4860
+    },
+    {
+      "epoch": 3.5251538183134272,
+      "grad_norm": 0.7665374875068665,
+      "learning_rate": 0.0002,
+      "loss": 1.1998,
+      "step": 4870
+    },
+    {
+      "epoch": 3.5323923271806006,
+      "grad_norm": 0.787857711315155,
+      "learning_rate": 0.0002,
+      "loss": 1.1551,
+      "step": 4880
+    },
+    {
+      "epoch": 3.539630836047774,
+      "grad_norm": 0.970595121383667,
+      "learning_rate": 0.0002,
+      "loss": 1.2758,
+      "step": 4890
+    },
+    {
+      "epoch": 3.5468693449149473,
+      "grad_norm": 0.6409347057342529,
+      "learning_rate": 0.0002,
+      "loss": 1.1274,
+      "step": 4900
+    },
+    {
+      "epoch": 3.5541078537821207,
+      "grad_norm": 0.888551652431488,
+      "learning_rate": 0.0002,
+      "loss": 1.1596,
+      "step": 4910
+    },
+    {
+      "epoch": 3.561346362649294,
+      "grad_norm": 1.0808377265930176,
+      "learning_rate": 0.0002,
+      "loss": 1.1644,
+      "step": 4920
+    },
+    {
+      "epoch": 3.5685848715164674,
+      "grad_norm": 0.7501053214073181,
+      "learning_rate": 0.0002,
+      "loss": 1.2564,
+      "step": 4930
+    },
+    {
+      "epoch": 3.575823380383641,
+      "grad_norm": 0.7375240325927734,
+      "learning_rate": 0.0002,
+      "loss": 1.2351,
+      "step": 4940
+    },
+    {
+      "epoch": 3.583061889250814,
+      "grad_norm": 0.7075039744377136,
+      "learning_rate": 0.0002,
+      "loss": 1.3568,
+      "step": 4950
+    },
+    {
+      "epoch": 3.5903003981179875,
+      "grad_norm": 0.939337432384491,
+      "learning_rate": 0.0002,
+      "loss": 1.3355,
+      "step": 4960
+    },
+    {
+      "epoch": 3.597538906985161,
+      "grad_norm": 0.6717396974563599,
+      "learning_rate": 0.0002,
+      "loss": 1.1722,
+      "step": 4970
+    },
+    {
+      "epoch": 3.6047774158523342,
+      "grad_norm": 0.7141643762588501,
+      "learning_rate": 0.0002,
+      "loss": 1.1186,
+      "step": 4980
+    },
+    {
+      "epoch": 3.6120159247195076,
+      "grad_norm": 0.7109216451644897,
+      "learning_rate": 0.0002,
+      "loss": 1.1011,
+      "step": 4990
+    },
+    {
+      "epoch": 3.619254433586681,
+      "grad_norm": 0.7020776867866516,
+      "learning_rate": 0.0002,
+      "loss": 1.2178,
+      "step": 5000
+    },
+    {
+      "epoch": 3.6264929424538543,
+      "grad_norm": 0.7158873677253723,
+      "learning_rate": 0.0002,
+      "loss": 1.1939,
+      "step": 5010
+    },
+    {
+      "epoch": 3.6337314513210277,
+      "grad_norm": 0.7062035202980042,
+      "learning_rate": 0.0002,
+      "loss": 1.2624,
+      "step": 5020
+    },
+    {
+      "epoch": 3.640969960188201,
+      "grad_norm": 0.7081155776977539,
+      "learning_rate": 0.0002,
+      "loss": 1.0224,
+      "step": 5030
+    },
+    {
+      "epoch": 3.6482084690553744,
+      "grad_norm": 1.2210607528686523,
+      "learning_rate": 0.0002,
+      "loss": 1.2195,
+      "step": 5040
+    },
+    {
+      "epoch": 3.655446977922548,
+      "grad_norm": 0.6650236248970032,
+      "learning_rate": 0.0002,
+      "loss": 1.2596,
+      "step": 5050
+    },
+    {
+      "epoch": 3.662685486789721,
+      "grad_norm": 0.6884829998016357,
+      "learning_rate": 0.0002,
+      "loss": 1.1072,
+      "step": 5060
+    },
+    {
+      "epoch": 3.6699239956568945,
+      "grad_norm": 0.7317819595336914,
+      "learning_rate": 0.0002,
+      "loss": 1.2292,
+      "step": 5070
+    },
+    {
+      "epoch": 3.677162504524068,
+      "grad_norm": 0.7406691908836365,
+      "learning_rate": 0.0002,
+      "loss": 1.1917,
+      "step": 5080
+    },
+    {
+      "epoch": 3.6844010133912413,
+      "grad_norm": 0.9009454250335693,
+      "learning_rate": 0.0002,
+      "loss": 1.2949,
+      "step": 5090
+    },
+    {
+      "epoch": 3.6916395222584146,
+      "grad_norm": 0.8189385533332825,
+      "learning_rate": 0.0002,
+      "loss": 1.1528,
+      "step": 5100
+    },
+    {
+      "epoch": 3.698878031125588,
+      "grad_norm": 1.0793628692626953,
+      "learning_rate": 0.0002,
+      "loss": 1.3408,
+      "step": 5110
+    },
+    {
+      "epoch": 3.7061165399927614,
+      "grad_norm": 0.8593027591705322,
+      "learning_rate": 0.0002,
+      "loss": 1.2417,
+      "step": 5120
+    },
+    {
+      "epoch": 3.7133550488599347,
+      "grad_norm": 0.8481812477111816,
+      "learning_rate": 0.0002,
+      "loss": 1.2141,
+      "step": 5130
+    },
+    {
+      "epoch": 3.720593557727108,
+      "grad_norm": 0.6527451276779175,
+      "learning_rate": 0.0002,
+      "loss": 1.125,
+      "step": 5140
+    },
+    {
+      "epoch": 3.7278320665942815,
+      "grad_norm": 0.9220114350318909,
+      "learning_rate": 0.0002,
+      "loss": 1.1584,
+      "step": 5150
+    },
+    {
+      "epoch": 3.735070575461455,
+      "grad_norm": 1.0842019319534302,
+      "learning_rate": 0.0002,
+      "loss": 1.2267,
+      "step": 5160
+    },
+    {
+      "epoch": 3.742309084328628,
+      "grad_norm": 0.965453565120697,
+      "learning_rate": 0.0002,
+      "loss": 1.3083,
+      "step": 5170
+    },
+    {
+      "epoch": 3.7495475931958016,
+      "grad_norm": 0.9903319478034973,
+      "learning_rate": 0.0002,
+      "loss": 1.1772,
+      "step": 5180
+    },
+    {
+      "epoch": 3.756786102062975,
+      "grad_norm": 0.7434818148612976,
+      "learning_rate": 0.0002,
+      "loss": 1.2515,
+      "step": 5190
+    },
+    {
+      "epoch": 3.7640246109301483,
+      "grad_norm": 0.6717280745506287,
+      "learning_rate": 0.0002,
+      "loss": 1.2631,
+      "step": 5200
+    },
+    {
+      "epoch": 3.7712631197973217,
+      "grad_norm": 0.7754665613174438,
+      "learning_rate": 0.0002,
+      "loss": 1.2012,
+      "step": 5210
+    },
+    {
+      "epoch": 3.778501628664495,
+      "grad_norm": 1.028374433517456,
+      "learning_rate": 0.0002,
+      "loss": 1.305,
+      "step": 5220
+    },
+    {
+      "epoch": 3.7857401375316684,
+      "grad_norm": 0.6026996374130249,
+      "learning_rate": 0.0002,
+      "loss": 1.1866,
+      "step": 5230
+    },
+    {
+      "epoch": 3.7929786463988417,
+      "grad_norm": 0.6978490948677063,
+      "learning_rate": 0.0002,
+      "loss": 1.1901,
+      "step": 5240
+    },
+    {
+      "epoch": 3.800217155266015,
+      "grad_norm": 0.7303446531295776,
+      "learning_rate": 0.0002,
+      "loss": 1.2576,
+      "step": 5250
+    },
+    {
+      "epoch": 3.8074556641331885,
+      "grad_norm": 1.0734210014343262,
+      "learning_rate": 0.0002,
+      "loss": 1.3173,
+      "step": 5260
+    },
+    {
+      "epoch": 3.814694173000362,
+      "grad_norm": 0.6383201479911804,
+      "learning_rate": 0.0002,
+      "loss": 1.1137,
+      "step": 5270
+    },
+    {
+      "epoch": 3.821932681867535,
+      "grad_norm": 0.7742630243301392,
+      "learning_rate": 0.0002,
+      "loss": 1.0904,
+      "step": 5280
+    },
+    {
+      "epoch": 3.8291711907347086,
+      "grad_norm": 0.8477074503898621,
+      "learning_rate": 0.0002,
+      "loss": 1.2232,
+      "step": 5290
+    },
+    {
+      "epoch": 3.836409699601882,
+      "grad_norm": 0.6675317883491516,
+      "learning_rate": 0.0002,
+      "loss": 1.2047,
+      "step": 5300
+    },
+    {
+      "epoch": 3.8436482084690553,
+      "grad_norm": 0.7515445351600647,
+      "learning_rate": 0.0002,
+      "loss": 1.2275,
+      "step": 5310
+    },
+    {
+      "epoch": 3.8508867173362287,
+      "grad_norm": 1.1441220045089722,
+      "learning_rate": 0.0002,
+      "loss": 1.2569,
+      "step": 5320
+    },
+    {
+      "epoch": 3.858125226203402,
+      "grad_norm": 0.7968795895576477,
+      "learning_rate": 0.0002,
+      "loss": 1.1512,
+      "step": 5330
+    },
+    {
+      "epoch": 3.8653637350705754,
+      "grad_norm": 0.7842824459075928,
+      "learning_rate": 0.0002,
+      "loss": 1.232,
+      "step": 5340
+    },
+    {
+      "epoch": 3.8726022439377488,
+      "grad_norm": 0.8272225260734558,
+      "learning_rate": 0.0002,
+      "loss": 1.1847,
+      "step": 5350
+    },
+    {
+      "epoch": 3.879840752804922,
+      "grad_norm": 0.8413397669792175,
+      "learning_rate": 0.0002,
+      "loss": 1.1381,
+      "step": 5360
+    },
+    {
+      "epoch": 3.8870792616720955,
+      "grad_norm": 1.141764760017395,
+      "learning_rate": 0.0002,
+      "loss": 1.2349,
+      "step": 5370
+    },
+    {
+      "epoch": 3.894317770539269,
+      "grad_norm": 0.9826975464820862,
+      "learning_rate": 0.0002,
+      "loss": 1.212,
+      "step": 5380
+    },
+    {
+      "epoch": 3.9015562794064422,
+      "grad_norm": 0.8598255515098572,
+      "learning_rate": 0.0002,
+      "loss": 1.1833,
+      "step": 5390
+    },
+    {
+      "epoch": 3.9087947882736156,
+      "grad_norm": 0.6271058320999146,
+      "learning_rate": 0.0002,
+      "loss": 1.1247,
+      "step": 5400
+    },
+    {
+      "epoch": 3.916033297140789,
+      "grad_norm": 0.6379870772361755,
+      "learning_rate": 0.0002,
+      "loss": 1.2212,
+      "step": 5410
+    },
+    {
+      "epoch": 3.9232718060079623,
+      "grad_norm": 1.0313376188278198,
+      "learning_rate": 0.0002,
+      "loss": 1.2481,
+      "step": 5420
+    },
+    {
+      "epoch": 3.9305103148751357,
+      "grad_norm": 0.8220619559288025,
+      "learning_rate": 0.0002,
+      "loss": 1.1872,
+      "step": 5430
+    },
+    {
+      "epoch": 3.937748823742309,
+      "grad_norm": 0.7576116919517517,
+      "learning_rate": 0.0002,
+      "loss": 1.2006,
+      "step": 5440
+    },
+    {
+      "epoch": 3.9449873326094824,
+      "grad_norm": 1.226235032081604,
+      "learning_rate": 0.0002,
+      "loss": 1.1969,
+      "step": 5450
+    },
+    {
+      "epoch": 3.952225841476656,
+      "grad_norm": 0.7979229688644409,
+      "learning_rate": 0.0002,
+      "loss": 1.2945,
+      "step": 5460
+    },
+    {
+      "epoch": 3.959464350343829,
+      "grad_norm": 0.9911929965019226,
+      "learning_rate": 0.0002,
+      "loss": 1.1922,
+      "step": 5470
+    },
+    {
+      "epoch": 3.9667028592110025,
+      "grad_norm": 0.643738865852356,
+      "learning_rate": 0.0002,
+      "loss": 1.0924,
+      "step": 5480
+    },
+    {
+      "epoch": 3.973941368078176,
+      "grad_norm": 0.682305634021759,
+      "learning_rate": 0.0002,
+      "loss": 1.0607,
+      "step": 5490
+    },
+    {
+      "epoch": 3.9811798769453492,
+      "grad_norm": 1.18373441696167,
+      "learning_rate": 0.0002,
+      "loss": 1.2908,
+      "step": 5500
+    },
+    {
+      "epoch": 3.9884183858125226,
+      "grad_norm": 0.7190203070640564,
+      "learning_rate": 0.0002,
+      "loss": 1.0889,
+      "step": 5510
+    },
+    {
+      "epoch": 3.995656894679696,
+      "grad_norm": 0.7516948580741882,
+      "learning_rate": 0.0002,
+      "loss": 1.2745,
+      "step": 5520
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.4252897500991821,
+      "eval_runtime": 27.235,
+      "eval_samples_per_second": 16.009,
+      "eval_steps_per_second": 2.019,
+      "step": 5526
+    },
+    {
+      "epoch": 4.002895403546869,
+      "grad_norm": 0.6353074312210083,
+      "learning_rate": 0.0002,
+      "loss": 1.0088,
+      "step": 5530
+    },
+    {
+      "epoch": 4.010133912414043,
+      "grad_norm": 0.7424906492233276,
+      "learning_rate": 0.0002,
+      "loss": 1.0326,
+      "step": 5540
+    },
+    {
+      "epoch": 4.017372421281216,
+      "grad_norm": 0.8856638073921204,
+      "learning_rate": 0.0002,
+      "loss": 1.0667,
+      "step": 5550
+    },
+    {
+      "epoch": 4.024610930148389,
+      "grad_norm": 0.9627974033355713,
+      "learning_rate": 0.0002,
+      "loss": 1.0905,
+      "step": 5560
+    },
+    {
+      "epoch": 4.031849439015563,
+      "grad_norm": 0.9048978686332703,
+      "learning_rate": 0.0002,
+      "loss": 1.0965,
+      "step": 5570
+    },
+    {
+      "epoch": 4.039087947882736,
+      "grad_norm": 0.921119213104248,
+      "learning_rate": 0.0002,
+      "loss": 1.1108,
+      "step": 5580
+    },
+    {
+      "epoch": 4.0463264567499095,
+      "grad_norm": 0.8654361963272095,
+      "learning_rate": 0.0002,
+      "loss": 1.1235,
+      "step": 5590
+    },
+    {
+      "epoch": 4.053564965617083,
+      "grad_norm": 0.7947945594787598,
+      "learning_rate": 0.0002,
+      "loss": 1.0794,
+      "step": 5600
+    },
+    {
+      "epoch": 4.060803474484256,
+      "grad_norm": 0.8307326436042786,
+      "learning_rate": 0.0002,
+      "loss": 1.0674,
+      "step": 5610
+    },
+    {
+      "epoch": 4.06804198335143,
+      "grad_norm": 0.793273389339447,
+      "learning_rate": 0.0002,
+      "loss": 1.0076,
+      "step": 5620
+    },
+    {
+      "epoch": 4.075280492218603,
+      "grad_norm": 0.8748673796653748,
+      "learning_rate": 0.0002,
+      "loss": 1.0651,
+      "step": 5630
+    },
+    {
+      "epoch": 4.082519001085776,
+      "grad_norm": 0.7926856279373169,
+      "learning_rate": 0.0002,
+      "loss": 1.111,
+      "step": 5640
+    },
+    {
+      "epoch": 4.08975750995295,
+      "grad_norm": 0.922645092010498,
+      "learning_rate": 0.0002,
+      "loss": 1.044,
+      "step": 5650
+    },
+    {
+      "epoch": 4.096996018820123,
+      "grad_norm": 0.9539641737937927,
+      "learning_rate": 0.0002,
+      "loss": 1.109,
+      "step": 5660
+    },
+    {
+      "epoch": 4.1042345276872965,
+      "grad_norm": 0.8674443364143372,
+      "learning_rate": 0.0002,
+      "loss": 1.0788,
+      "step": 5670
+    },
+    {
+      "epoch": 4.11147303655447,
+      "grad_norm": 0.7097609043121338,
+      "learning_rate": 0.0002,
+      "loss": 0.9867,
+      "step": 5680
+    },
+    {
+      "epoch": 4.118711545421643,
+      "grad_norm": 0.8875522613525391,
+      "learning_rate": 0.0002,
+      "loss": 1.1154,
+      "step": 5690
+    },
+    {
+      "epoch": 4.125950054288817,
+      "grad_norm": 0.8583634495735168,
+      "learning_rate": 0.0002,
+      "loss": 1.1217,
+      "step": 5700
+    },
+    {
+      "epoch": 4.13318856315599,
+      "grad_norm": 0.6736377477645874,
+      "learning_rate": 0.0002,
+      "loss": 1.0973,
+      "step": 5710
+    },
+    {
+      "epoch": 4.140427072023163,
+      "grad_norm": 0.9349062442779541,
+      "learning_rate": 0.0002,
+      "loss": 1.1199,
+      "step": 5720
+    },
+    {
+      "epoch": 4.147665580890337,
+      "grad_norm": 1.0610365867614746,
+      "learning_rate": 0.0002,
+      "loss": 1.0508,
+      "step": 5730
+    },
+    {
+      "epoch": 4.15490408975751,
+      "grad_norm": 1.5838189125061035,
+      "learning_rate": 0.0002,
+      "loss": 1.1146,
+      "step": 5740
+    },
+    {
+      "epoch": 4.162142598624683,
+      "grad_norm": 0.747522234916687,
+      "learning_rate": 0.0002,
+      "loss": 1.0222,
+      "step": 5750
+    },
+    {
+      "epoch": 4.169381107491857,
+      "grad_norm": 1.3247915506362915,
+      "learning_rate": 0.0002,
+      "loss": 1.1328,
+      "step": 5760
+    },
+    {
+      "epoch": 4.17661961635903,
+      "grad_norm": 0.8750247955322266,
+      "learning_rate": 0.0002,
+      "loss": 1.1655,
+      "step": 5770
+    },
+    {
+      "epoch": 4.1838581252262035,
+      "grad_norm": 0.7914144992828369,
+      "learning_rate": 0.0002,
+      "loss": 1.199,
+      "step": 5780
+    },
+    {
+      "epoch": 4.191096634093377,
+      "grad_norm": 0.9493299126625061,
+      "learning_rate": 0.0002,
+      "loss": 1.1213,
+      "step": 5790
+    },
+    {
+      "epoch": 4.19833514296055,
+      "grad_norm": 0.7802295088768005,
+      "learning_rate": 0.0002,
+      "loss": 1.1515,
+      "step": 5800
+    },
+    {
+      "epoch": 4.205573651827724,
+      "grad_norm": 0.6987314820289612,
+      "learning_rate": 0.0002,
+      "loss": 1.0704,
+      "step": 5810
+    },
+    {
+      "epoch": 4.212812160694897,
+      "grad_norm": 0.9220341444015503,
+      "learning_rate": 0.0002,
+      "loss": 1.1699,
+      "step": 5820
+    },
+    {
+      "epoch": 4.22005066956207,
+      "grad_norm": 0.8932939767837524,
+      "learning_rate": 0.0002,
+      "loss": 1.1394,
+      "step": 5830
+    },
+    {
+      "epoch": 4.227289178429244,
+      "grad_norm": 0.920002818107605,
+      "learning_rate": 0.0002,
+      "loss": 1.0048,
+      "step": 5840
+    },
+    {
+      "epoch": 4.234527687296417,
+      "grad_norm": 0.6662752032279968,
+      "learning_rate": 0.0002,
+      "loss": 0.964,
+      "step": 5850
+    },
+    {
+      "epoch": 4.24176619616359,
+      "grad_norm": 0.8679718971252441,
+      "learning_rate": 0.0002,
+      "loss": 0.986,
+      "step": 5860
+    },
+    {
+      "epoch": 4.249004705030764,
+      "grad_norm": 0.7020887732505798,
+      "learning_rate": 0.0002,
+      "loss": 0.8991,
+      "step": 5870
+    },
+    {
+      "epoch": 4.256243213897937,
+      "grad_norm": 0.869611382484436,
+      "learning_rate": 0.0002,
+      "loss": 1.1132,
+      "step": 5880
+    },
+    {
+      "epoch": 4.2634817227651105,
+      "grad_norm": 0.7796585559844971,
+      "learning_rate": 0.0002,
+      "loss": 1.1026,
+      "step": 5890
+    },
+    {
+      "epoch": 4.270720231632284,
+      "grad_norm": 0.8978819251060486,
+      "learning_rate": 0.0002,
+      "loss": 1.0957,
+      "step": 5900
+    },
+    {
+      "epoch": 4.277958740499457,
+      "grad_norm": 1.0837205648422241,
+      "learning_rate": 0.0002,
+      "loss": 1.1325,
+      "step": 5910
+    },
+    {
+      "epoch": 4.285197249366631,
+      "grad_norm": 0.7584353089332581,
+      "learning_rate": 0.0002,
+      "loss": 1.1279,
+      "step": 5920
+    },
+    {
+      "epoch": 4.292435758233804,
+      "grad_norm": 0.7313185334205627,
+      "learning_rate": 0.0002,
+      "loss": 1.0513,
+      "step": 5930
+    },
+    {
+      "epoch": 4.299674267100977,
+      "grad_norm": 0.8004671335220337,
+      "learning_rate": 0.0002,
+      "loss": 1.1101,
+      "step": 5940
+    },
+    {
+      "epoch": 4.306912775968151,
+      "grad_norm": 2.154958724975586,
+      "learning_rate": 0.0002,
+      "loss": 1.14,
+      "step": 5950
+    },
+    {
+      "epoch": 4.314151284835324,
+      "grad_norm": 0.9163479804992676,
+      "learning_rate": 0.0002,
+      "loss": 1.1206,
+      "step": 5960
+    },
+    {
+      "epoch": 4.321389793702497,
+      "grad_norm": 0.9151589274406433,
+      "learning_rate": 0.0002,
+      "loss": 0.9941,
+      "step": 5970
+    },
+    {
+      "epoch": 4.328628302569671,
+      "grad_norm": 0.8624112010002136,
+      "learning_rate": 0.0002,
+      "loss": 1.0606,
+      "step": 5980
+    },
+    {
+      "epoch": 4.335866811436844,
+      "grad_norm": 0.9357741475105286,
+      "learning_rate": 0.0002,
+      "loss": 1.1625,
+      "step": 5990
+    },
+    {
+      "epoch": 4.3431053203040175,
+      "grad_norm": 1.3482335805892944,
+      "learning_rate": 0.0002,
+      "loss": 1.0712,
+      "step": 6000
+    },
+    {
+      "epoch": 4.350343829171191,
+      "grad_norm": 0.7156149744987488,
+      "learning_rate": 0.0002,
+      "loss": 1.1224,
+      "step": 6010
+    },
+    {
+      "epoch": 4.357582338038364,
+      "grad_norm": 0.8480049967765808,
+      "learning_rate": 0.0002,
+      "loss": 1.0753,
+      "step": 6020
+    },
+    {
+      "epoch": 4.364820846905538,
+      "grad_norm": 0.8262244462966919,
+      "learning_rate": 0.0002,
+      "loss": 1.051,
+      "step": 6030
+    },
+    {
+      "epoch": 4.372059355772711,
+      "grad_norm": 0.7733905911445618,
+      "learning_rate": 0.0002,
+      "loss": 0.9966,
+      "step": 6040
+    },
+    {
+      "epoch": 4.379297864639884,
+      "grad_norm": 0.8553919792175293,
+      "learning_rate": 0.0002,
+      "loss": 1.1008,
+      "step": 6050
+    },
+    {
+      "epoch": 4.386536373507058,
+      "grad_norm": 0.8666832447052002,
+      "learning_rate": 0.0002,
+      "loss": 1.1777,
+      "step": 6060
+    },
+    {
+      "epoch": 4.393774882374231,
+      "grad_norm": 0.9168295860290527,
+      "learning_rate": 0.0002,
+      "loss": 1.1934,
+      "step": 6070
+    },
+    {
+      "epoch": 4.4010133912414044,
+      "grad_norm": 0.7315238118171692,
+      "learning_rate": 0.0002,
+      "loss": 1.0988,
+      "step": 6080
+    },
+    {
+      "epoch": 4.408251900108578,
+      "grad_norm": 1.020263433456421,
+      "learning_rate": 0.0002,
+      "loss": 1.1599,
+      "step": 6090
+    },
+    {
+      "epoch": 4.415490408975751,
+      "grad_norm": 0.9978243708610535,
+      "learning_rate": 0.0002,
+      "loss": 1.133,
+      "step": 6100
+    },
+    {
+      "epoch": 4.4227289178429245,
+      "grad_norm": 0.995453953742981,
+      "learning_rate": 0.0002,
+      "loss": 1.1324,
+      "step": 6110
+    },
+    {
+      "epoch": 4.429967426710098,
+      "grad_norm": 0.9360884428024292,
+      "learning_rate": 0.0002,
+      "loss": 1.0957,
+      "step": 6120
+    },
+    {
+      "epoch": 4.437205935577271,
+      "grad_norm": 0.8099448084831238,
+      "learning_rate": 0.0002,
+      "loss": 0.9506,
+      "step": 6130
+    },
+    {
+      "epoch": 4.444444444444445,
+      "grad_norm": 0.8173841238021851,
+      "learning_rate": 0.0002,
+      "loss": 1.0887,
+      "step": 6140
+    },
+    {
+      "epoch": 4.451682953311618,
+      "grad_norm": 0.7972666025161743,
+      "learning_rate": 0.0002,
+      "loss": 1.1219,
+      "step": 6150
+    },
+    {
+      "epoch": 4.458921462178791,
+      "grad_norm": 0.7685779333114624,
+      "learning_rate": 0.0002,
+      "loss": 1.0226,
+      "step": 6160
+    },
+    {
+      "epoch": 4.466159971045965,
+      "grad_norm": 0.7872623801231384,
+      "learning_rate": 0.0002,
+      "loss": 1.0732,
+      "step": 6170
+    },
+    {
+      "epoch": 4.473398479913138,
+      "grad_norm": 0.7677070498466492,
+      "learning_rate": 0.0002,
+      "loss": 0.9911,
+      "step": 6180
+    },
+    {
+      "epoch": 4.4806369887803115,
+      "grad_norm": 0.7878316044807434,
+      "learning_rate": 0.0002,
+      "loss": 1.0919,
+      "step": 6190
+    },
+    {
+      "epoch": 4.487875497647485,
+      "grad_norm": 0.8178079724311829,
+      "learning_rate": 0.0002,
+      "loss": 1.018,
+      "step": 6200
+    },
+    {
+      "epoch": 4.495114006514658,
+      "grad_norm": 1.2820082902908325,
+      "learning_rate": 0.0002,
+      "loss": 1.0517,
+      "step": 6210
+    },
+    {
+      "epoch": 4.502352515381832,
+      "grad_norm": 0.9380832314491272,
+      "learning_rate": 0.0002,
+      "loss": 1.3101,
+      "step": 6220
+    },
+    {
+      "epoch": 4.509591024249005,
+      "grad_norm": 0.7810422778129578,
+      "learning_rate": 0.0002,
+      "loss": 0.9818,
+      "step": 6230
+    },
+    {
+      "epoch": 4.516829533116178,
+      "grad_norm": 1.1022917032241821,
+      "learning_rate": 0.0002,
+      "loss": 1.1677,
+      "step": 6240
+    },
+    {
+      "epoch": 4.524068041983352,
+      "grad_norm": 1.4275553226470947,
+      "learning_rate": 0.0002,
+      "loss": 1.1579,
+      "step": 6250
+    },
+    {
+      "epoch": 4.531306550850525,
+      "grad_norm": 0.7597777247428894,
+      "learning_rate": 0.0002,
+      "loss": 1.3237,
+      "step": 6260
+    },
+    {
+      "epoch": 4.538545059717698,
+      "grad_norm": 1.10992431640625,
+      "learning_rate": 0.0002,
+      "loss": 1.1529,
+      "step": 6270
+    },
+    {
+      "epoch": 4.545783568584872,
+      "grad_norm": 0.8981178998947144,
+      "learning_rate": 0.0002,
+      "loss": 1.0732,
+      "step": 6280
+    },
+    {
+      "epoch": 4.553022077452045,
+      "grad_norm": 0.7863979339599609,
+      "learning_rate": 0.0002,
+      "loss": 1.086,
+      "step": 6290
+    },
+    {
+      "epoch": 4.5602605863192185,
+      "grad_norm": 0.9071474671363831,
+      "learning_rate": 0.0002,
+      "loss": 1.2008,
+      "step": 6300
+    },
+    {
+      "epoch": 4.567499095186392,
+      "grad_norm": 0.7429424524307251,
+      "learning_rate": 0.0002,
+      "loss": 1.0916,
+      "step": 6310
+    },
+    {
+      "epoch": 4.574737604053565,
+      "grad_norm": 1.0767850875854492,
+      "learning_rate": 0.0002,
+      "loss": 1.095,
+      "step": 6320
+    },
+    {
+      "epoch": 4.581976112920739,
+      "grad_norm": 0.7885915637016296,
+      "learning_rate": 0.0002,
+      "loss": 1.1023,
+      "step": 6330
+    },
+    {
+      "epoch": 4.589214621787912,
+      "grad_norm": 0.8350457549095154,
+      "learning_rate": 0.0002,
+      "loss": 1.1131,
+      "step": 6340
+    },
+    {
+      "epoch": 4.596453130655085,
+      "grad_norm": 0.7853530645370483,
+      "learning_rate": 0.0002,
+      "loss": 1.0743,
+      "step": 6350
+    },
+    {
+      "epoch": 4.603691639522259,
+      "grad_norm": 1.1220661401748657,
+      "learning_rate": 0.0002,
+      "loss": 1.1912,
+      "step": 6360
+    },
+    {
+      "epoch": 4.610930148389432,
+      "grad_norm": 0.7959423065185547,
+      "learning_rate": 0.0002,
+      "loss": 1.0927,
+      "step": 6370
+    },
+    {
+      "epoch": 4.618168657256605,
+      "grad_norm": 0.7782652378082275,
+      "learning_rate": 0.0002,
+      "loss": 1.1542,
+      "step": 6380
+    },
+    {
+      "epoch": 4.625407166123779,
+      "grad_norm": 0.7882203459739685,
+      "learning_rate": 0.0002,
+      "loss": 1.0753,
+      "step": 6390
+    },
+    {
+      "epoch": 4.632645674990952,
+      "grad_norm": 0.8841899037361145,
+      "learning_rate": 0.0002,
+      "loss": 1.0676,
+      "step": 6400
+    },
+    {
+      "epoch": 4.6398841838581255,
+      "grad_norm": 0.7936127781867981,
+      "learning_rate": 0.0002,
+      "loss": 1.0815,
+      "step": 6410
+    },
+    {
+      "epoch": 4.647122692725299,
+      "grad_norm": 0.9213966131210327,
+      "learning_rate": 0.0002,
+      "loss": 1.0198,
+      "step": 6420
+    },
+    {
+      "epoch": 4.654361201592472,
+      "grad_norm": 0.9246473908424377,
+      "learning_rate": 0.0002,
+      "loss": 0.9872,
+      "step": 6430
+    },
+    {
+      "epoch": 4.661599710459646,
+      "grad_norm": 0.766572892665863,
+      "learning_rate": 0.0002,
+      "loss": 1.1309,
+      "step": 6440
+    },
+    {
+      "epoch": 4.668838219326819,
+      "grad_norm": 0.8596171736717224,
+      "learning_rate": 0.0002,
+      "loss": 1.1095,
+      "step": 6450
+    },
+    {
+      "epoch": 4.676076728193992,
+      "grad_norm": 0.8482751846313477,
+      "learning_rate": 0.0002,
+      "loss": 1.1869,
+      "step": 6460
+    },
+    {
+      "epoch": 4.683315237061166,
+      "grad_norm": 1.0826905965805054,
+      "learning_rate": 0.0002,
+      "loss": 1.0622,
+      "step": 6470
+    },
+    {
+      "epoch": 4.690553745928339,
+      "grad_norm": 1.1048457622528076,
+      "learning_rate": 0.0002,
+      "loss": 1.0256,
+      "step": 6480
+    },
+    {
+      "epoch": 4.697792254795512,
+      "grad_norm": 0.9429134726524353,
+      "learning_rate": 0.0002,
+      "loss": 1.0514,
+      "step": 6490
+    },
+    {
+      "epoch": 4.705030763662686,
+      "grad_norm": 0.8587502837181091,
+      "learning_rate": 0.0002,
+      "loss": 1.1351,
+      "step": 6500
+    },
+    {
+      "epoch": 4.712269272529859,
+      "grad_norm": 1.0387083292007446,
+      "learning_rate": 0.0002,
+      "loss": 1.0969,
+      "step": 6510
+    },
+    {
+      "epoch": 4.7195077813970325,
+      "grad_norm": 0.7471951842308044,
+      "learning_rate": 0.0002,
+      "loss": 1.0493,
+      "step": 6520
+    },
+    {
+      "epoch": 4.726746290264206,
+      "grad_norm": 0.8800424933433533,
+      "learning_rate": 0.0002,
+      "loss": 1.2632,
+      "step": 6530
+    },
+    {
+      "epoch": 4.733984799131379,
+      "grad_norm": 0.8136811852455139,
+      "learning_rate": 0.0002,
+      "loss": 1.2126,
+      "step": 6540
+    },
+    {
+      "epoch": 4.741223307998553,
+      "grad_norm": 0.9910339713096619,
+      "learning_rate": 0.0002,
+      "loss": 1.195,
+      "step": 6550
+    },
+    {
+      "epoch": 4.748461816865726,
+      "grad_norm": 1.0679163932800293,
+      "learning_rate": 0.0002,
+      "loss": 1.1201,
+      "step": 6560
+    },
+    {
+      "epoch": 4.755700325732899,
+      "grad_norm": 0.8468248248100281,
+      "learning_rate": 0.0002,
+      "loss": 1.0297,
+      "step": 6570
+    },
+    {
+      "epoch": 4.762938834600073,
+      "grad_norm": 0.8771235942840576,
+      "learning_rate": 0.0002,
+      "loss": 1.0858,
+      "step": 6580
+    },
+    {
+      "epoch": 4.770177343467246,
+      "grad_norm": 0.7024846076965332,
+      "learning_rate": 0.0002,
+      "loss": 1.077,
+      "step": 6590
+    },
+    {
+      "epoch": 4.7774158523344195,
+      "grad_norm": 0.7836683392524719,
+      "learning_rate": 0.0002,
+      "loss": 1.0876,
+      "step": 6600
+    },
+    {
+      "epoch": 4.784654361201593,
+      "grad_norm": 0.7717288136482239,
+      "learning_rate": 0.0002,
+      "loss": 1.1006,
+      "step": 6610
+    },
+    {
+      "epoch": 4.791892870068766,
+      "grad_norm": 0.884183943271637,
+      "learning_rate": 0.0002,
+      "loss": 1.0376,
+      "step": 6620
+    },
+    {
+      "epoch": 4.7991313789359396,
+      "grad_norm": 1.383867621421814,
+      "learning_rate": 0.0002,
+      "loss": 1.1757,
+      "step": 6630
+    },
+    {
+      "epoch": 4.806369887803113,
+      "grad_norm": 0.9741523861885071,
+      "learning_rate": 0.0002,
+      "loss": 1.0861,
+      "step": 6640
+    },
+    {
+      "epoch": 4.813608396670286,
+      "grad_norm": 0.9723693132400513,
+      "learning_rate": 0.0002,
+      "loss": 1.0884,
+      "step": 6650
+    },
+    {
+      "epoch": 4.82084690553746,
+      "grad_norm": 1.8324809074401855,
+      "learning_rate": 0.0002,
+      "loss": 1.2203,
+      "step": 6660
+    },
+    {
+      "epoch": 4.828085414404633,
+      "grad_norm": 0.904909074306488,
+      "learning_rate": 0.0002,
+      "loss": 1.0292,
+      "step": 6670
+    },
+    {
+      "epoch": 4.835323923271806,
+      "grad_norm": 0.7355411648750305,
+      "learning_rate": 0.0002,
+      "loss": 1.0349,
+      "step": 6680
+    },
+    {
+      "epoch": 4.84256243213898,
+      "grad_norm": 0.8934960961341858,
+      "learning_rate": 0.0002,
+      "loss": 1.0793,
+      "step": 6690
+    },
+    {
+      "epoch": 4.849800941006153,
+      "grad_norm": 1.4596954584121704,
+      "learning_rate": 0.0002,
+      "loss": 1.0375,
+      "step": 6700
+    },
+    {
+      "epoch": 4.8570394498733265,
+      "grad_norm": 0.8310341238975525,
+      "learning_rate": 0.0002,
+      "loss": 1.1065,
+      "step": 6710
+    },
+    {
+      "epoch": 4.8642779587405,
+      "grad_norm": 0.9709894061088562,
+      "learning_rate": 0.0002,
+      "loss": 1.1089,
+      "step": 6720
+    },
+    {
+      "epoch": 4.871516467607673,
+      "grad_norm": 0.852142333984375,
+      "learning_rate": 0.0002,
+      "loss": 1.0069,
+      "step": 6730
+    },
+    {
+      "epoch": 4.878754976474847,
+      "grad_norm": 1.0643625259399414,
+      "learning_rate": 0.0002,
+      "loss": 1.0507,
+      "step": 6740
+    },
+    {
+      "epoch": 4.88599348534202,
+      "grad_norm": 0.9419508576393127,
+      "learning_rate": 0.0002,
+      "loss": 1.056,
+      "step": 6750
+    },
+    {
+      "epoch": 4.893231994209193,
+      "grad_norm": 1.1818498373031616,
+      "learning_rate": 0.0002,
+      "loss": 1.1995,
+      "step": 6760
+    },
+    {
+      "epoch": 4.900470503076367,
+      "grad_norm": 0.9369569420814514,
+      "learning_rate": 0.0002,
+      "loss": 1.0925,
+      "step": 6770
+    },
+    {
+      "epoch": 4.90770901194354,
+      "grad_norm": 0.7012579441070557,
+      "learning_rate": 0.0002,
+      "loss": 1.1648,
+      "step": 6780
+    },
+    {
+      "epoch": 4.914947520810713,
+      "grad_norm": 0.9109319448471069,
+      "learning_rate": 0.0002,
+      "loss": 1.0926,
+      "step": 6790
+    },
+    {
+      "epoch": 4.922186029677887,
+      "grad_norm": 0.8077534437179565,
+      "learning_rate": 0.0002,
+      "loss": 1.0358,
+      "step": 6800
+    },
+    {
+      "epoch": 4.92942453854506,
+      "grad_norm": 0.7571148872375488,
+      "learning_rate": 0.0002,
+      "loss": 1.2549,
+      "step": 6810
+    },
+    {
+      "epoch": 4.9366630474122335,
+      "grad_norm": 0.7325633764266968,
+      "learning_rate": 0.0002,
+      "loss": 0.9638,
+      "step": 6820
+    },
+    {
+      "epoch": 4.943901556279407,
+      "grad_norm": 0.8465084433555603,
+      "learning_rate": 0.0002,
+      "loss": 1.0128,
+      "step": 6830
+    },
+    {
+      "epoch": 4.95114006514658,
+      "grad_norm": 0.8753737807273865,
+      "learning_rate": 0.0002,
+      "loss": 1.153,
+      "step": 6840
+    },
+    {
+      "epoch": 4.958378574013754,
+      "grad_norm": 0.9421748518943787,
+      "learning_rate": 0.0002,
+      "loss": 1.0247,
+      "step": 6850
+    },
+    {
+      "epoch": 4.965617082880927,
+      "grad_norm": 0.8245896697044373,
+      "learning_rate": 0.0002,
+      "loss": 1.1483,
+      "step": 6860
+    },
+    {
+      "epoch": 4.9728555917481,
+      "grad_norm": 0.8823089599609375,
+      "learning_rate": 0.0002,
+      "loss": 0.9905,
+      "step": 6870
+    },
+    {
+      "epoch": 4.980094100615274,
+      "grad_norm": 0.8406389355659485,
+      "learning_rate": 0.0002,
+      "loss": 1.1664,
+      "step": 6880
+    },
+    {
+      "epoch": 4.987332609482447,
+      "grad_norm": 0.9732868075370789,
+      "learning_rate": 0.0002,
+      "loss": 1.0944,
+      "step": 6890
+    },
+    {
+      "epoch": 4.99457111834962,
+      "grad_norm": 2.125141143798828,
+      "learning_rate": 0.0002,
+      "loss": 1.1776,
+      "step": 6900
+    },
+    {
+      "epoch": 4.999638074556641,
+      "eval_loss": 1.445176601409912,
+      "eval_runtime": 27.2351,
+      "eval_samples_per_second": 16.009,
+      "eval_steps_per_second": 2.019,
+      "step": 6907
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 11048,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 8.442343968473088e+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-0.9-num-8828-sd-1/checkpoint-6907/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-0.9-num-8828-sd-1/checkpoint-6907/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..4d7b7431bbbe8c9bf29b925bca391a558af5ff8c
--- /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-0.9-num-8828-sd-1/checkpoint-6907/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ad613885e4f267fc04125f1a836d42cfa796bbe12e536f9ee60c955de02cdb5a
+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-0.9-num-8828-sd-1/checkpoint-8289/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-0.9-num-8828-sd-1/checkpoint-8289/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-0.9-num-8828-sd-1/checkpoint-8289/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-0.9-num-8828-sd-1/checkpoint-8289/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-0.9-num-8828-sd-1/checkpoint-8289/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e99bbcd43df1c19d98706c7e3be95c93844c5349
--- /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-0.9-num-8828-sd-1/checkpoint-8289/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-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-8828-sd-1/checkpoint-8289/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-0.9-num-8828-sd-1/checkpoint-8289/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..cb32d048077f7d36cfe26b386f4c5c7c8584ba1d
--- /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-0.9-num-8828-sd-1/checkpoint-8289/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b4e29c954dbe5c796580cbf124ef520cc237dd5b537fb2b50a51fcecc829d770
+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-0.9-num-8828-sd-1/checkpoint-8289/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-0.9-num-8828-sd-1/checkpoint-8289/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..5ecc859b0d2dee022050e34694df22248804ffd1
--- /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-0.9-num-8828-sd-1/checkpoint-8289/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:408bb73e1e7320415d913dbf2d032b37fee48d77d984e0a4b275d0aa8a937b1b
+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-0.9-num-8828-sd-1/checkpoint-8289/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-0.9-num-8828-sd-1/checkpoint-8289/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..91dd0cfc4765c5fba307f74c5667ab094ea9162c
--- /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-0.9-num-8828-sd-1/checkpoint-8289/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:477d843dcd5d8c3d2382e25e5ea3a1d9e518dc0db552a7775a088d3410a82d91
+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-0.9-num-8828-sd-1/checkpoint-8289/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-0.9-num-8828-sd-1/checkpoint-8289/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..2113435227a989284d1334d01f058630811bf130
--- /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-0.9-num-8828-sd-1/checkpoint-8289/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6faa4422eb7d977fd89c4cc4434e03971817a911178543ca18703285022fa252
+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-0.9-num-8828-sd-1/checkpoint-8289/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-0.9-num-8828-sd-1/checkpoint-8289/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-0.9-num-8828-sd-1/checkpoint-8289/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-0.9-num-8828-sd-1/checkpoint-8289/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-0.9-num-8828-sd-1/checkpoint-8289/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-0.9-num-8828-sd-1/checkpoint-8289/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-0.9-num-8828-sd-1/checkpoint-8289/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-0.9-num-8828-sd-1/checkpoint-8289/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-0.9-num-8828-sd-1/checkpoint-8289/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-0.9-num-8828-sd-1/checkpoint-8289/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-0.9-num-8828-sd-1/checkpoint-8289/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-0.9-num-8828-sd-1/checkpoint-8289/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-0.9-num-8828-sd-1/checkpoint-8289/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-0.9-num-8828-sd-1/checkpoint-8289/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..33541a1db43cf7eae2af7f68b07ea57eb9740b96
--- /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-0.9-num-8828-sd-1/checkpoint-8289/trainer_state.json
@@ -0,0 +1,5877 @@
+{
+  "best_metric": 1.4217946529388428,
+  "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-0.9-num-8828-sd-1/checkpoint-4144",
+  "epoch": 6.0,
+  "eval_steps": 10,
+  "global_step": 8289,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007238508867173362,
+      "grad_norm": 1.2523442506790161,
+      "learning_rate": 0.0002,
+      "loss": 4.7061,
+      "step": 10
+    },
+    {
+      "epoch": 0.014477017734346724,
+      "grad_norm": 1.8887330293655396,
+      "learning_rate": 0.0002,
+      "loss": 3.3493,
+      "step": 20
+    },
+    {
+      "epoch": 0.021715526601520086,
+      "grad_norm": 0.9668035507202148,
+      "learning_rate": 0.0002,
+      "loss": 2.7585,
+      "step": 30
+    },
+    {
+      "epoch": 0.028954035468693448,
+      "grad_norm": 2.9167306423187256,
+      "learning_rate": 0.0002,
+      "loss": 2.3699,
+      "step": 40
+    },
+    {
+      "epoch": 0.036192544335866814,
+      "grad_norm": 2.649867296218872,
+      "learning_rate": 0.0002,
+      "loss": 2.2679,
+      "step": 50
+    },
+    {
+      "epoch": 0.04343105320304017,
+      "grad_norm": 1.5120655298233032,
+      "learning_rate": 0.0002,
+      "loss": 2.2202,
+      "step": 60
+    },
+    {
+      "epoch": 0.05066956207021354,
+      "grad_norm": 0.7879868149757385,
+      "learning_rate": 0.0002,
+      "loss": 2.2026,
+      "step": 70
+    },
+    {
+      "epoch": 0.057908070937386896,
+      "grad_norm": 0.7616953253746033,
+      "learning_rate": 0.0002,
+      "loss": 1.9447,
+      "step": 80
+    },
+    {
+      "epoch": 0.06514657980456026,
+      "grad_norm": 1.8809149265289307,
+      "learning_rate": 0.0002,
+      "loss": 2.0112,
+      "step": 90
+    },
+    {
+      "epoch": 0.07238508867173363,
+      "grad_norm": 0.9294016361236572,
+      "learning_rate": 0.0002,
+      "loss": 1.8337,
+      "step": 100
+    },
+    {
+      "epoch": 0.07962359753890698,
+      "grad_norm": 0.7145281434059143,
+      "learning_rate": 0.0002,
+      "loss": 1.8419,
+      "step": 110
+    },
+    {
+      "epoch": 0.08686210640608034,
+      "grad_norm": 0.7564446330070496,
+      "learning_rate": 0.0002,
+      "loss": 2.0036,
+      "step": 120
+    },
+    {
+      "epoch": 0.09410061527325371,
+      "grad_norm": 1.1681925058364868,
+      "learning_rate": 0.0002,
+      "loss": 1.9306,
+      "step": 130
+    },
+    {
+      "epoch": 0.10133912414042708,
+      "grad_norm": 0.6708641648292542,
+      "learning_rate": 0.0002,
+      "loss": 1.7875,
+      "step": 140
+    },
+    {
+      "epoch": 0.10857763300760044,
+      "grad_norm": 0.7625647783279419,
+      "learning_rate": 0.0002,
+      "loss": 1.786,
+      "step": 150
+    },
+    {
+      "epoch": 0.11581614187477379,
+      "grad_norm": 0.8463464975357056,
+      "learning_rate": 0.0002,
+      "loss": 1.6687,
+      "step": 160
+    },
+    {
+      "epoch": 0.12305465074194716,
+      "grad_norm": 0.7502335906028748,
+      "learning_rate": 0.0002,
+      "loss": 1.6214,
+      "step": 170
+    },
+    {
+      "epoch": 0.13029315960912052,
+      "grad_norm": 0.6929958462715149,
+      "learning_rate": 0.0002,
+      "loss": 1.7433,
+      "step": 180
+    },
+    {
+      "epoch": 0.1375316684762939,
+      "grad_norm": 0.6798707842826843,
+      "learning_rate": 0.0002,
+      "loss": 1.6009,
+      "step": 190
+    },
+    {
+      "epoch": 0.14477017734346725,
+      "grad_norm": 0.7566508650779724,
+      "learning_rate": 0.0002,
+      "loss": 1.6208,
+      "step": 200
+    },
+    {
+      "epoch": 0.15200868621064062,
+      "grad_norm": 0.7196869850158691,
+      "learning_rate": 0.0002,
+      "loss": 1.5823,
+      "step": 210
+    },
+    {
+      "epoch": 0.15924719507781396,
+      "grad_norm": 0.8401045799255371,
+      "learning_rate": 0.0002,
+      "loss": 1.738,
+      "step": 220
+    },
+    {
+      "epoch": 0.16648570394498732,
+      "grad_norm": 0.8503773212432861,
+      "learning_rate": 0.0002,
+      "loss": 1.7574,
+      "step": 230
+    },
+    {
+      "epoch": 0.1737242128121607,
+      "grad_norm": 0.7183733582496643,
+      "learning_rate": 0.0002,
+      "loss": 1.7861,
+      "step": 240
+    },
+    {
+      "epoch": 0.18096272167933405,
+      "grad_norm": 0.7082605957984924,
+      "learning_rate": 0.0002,
+      "loss": 1.6693,
+      "step": 250
+    },
+    {
+      "epoch": 0.18820123054650742,
+      "grad_norm": 0.9386326670646667,
+      "learning_rate": 0.0002,
+      "loss": 1.619,
+      "step": 260
+    },
+    {
+      "epoch": 0.19543973941368079,
+      "grad_norm": 0.7332451939582825,
+      "learning_rate": 0.0002,
+      "loss": 1.6511,
+      "step": 270
+    },
+    {
+      "epoch": 0.20267824828085415,
+      "grad_norm": 0.7092869877815247,
+      "learning_rate": 0.0002,
+      "loss": 1.6353,
+      "step": 280
+    },
+    {
+      "epoch": 0.20991675714802752,
+      "grad_norm": 0.7256413698196411,
+      "learning_rate": 0.0002,
+      "loss": 1.5996,
+      "step": 290
+    },
+    {
+      "epoch": 0.21715526601520088,
+      "grad_norm": 0.6398681402206421,
+      "learning_rate": 0.0002,
+      "loss": 1.6754,
+      "step": 300
+    },
+    {
+      "epoch": 0.22439377488237422,
+      "grad_norm": 0.6273287534713745,
+      "learning_rate": 0.0002,
+      "loss": 1.397,
+      "step": 310
+    },
+    {
+      "epoch": 0.23163228374954759,
+      "grad_norm": 0.511648416519165,
+      "learning_rate": 0.0002,
+      "loss": 1.5115,
+      "step": 320
+    },
+    {
+      "epoch": 0.23887079261672095,
+      "grad_norm": 0.8677352070808411,
+      "learning_rate": 0.0002,
+      "loss": 1.5424,
+      "step": 330
+    },
+    {
+      "epoch": 0.24610930148389432,
+      "grad_norm": 0.6270743012428284,
+      "learning_rate": 0.0002,
+      "loss": 1.6779,
+      "step": 340
+    },
+    {
+      "epoch": 0.2533478103510677,
+      "grad_norm": 0.7980281114578247,
+      "learning_rate": 0.0002,
+      "loss": 1.626,
+      "step": 350
+    },
+    {
+      "epoch": 0.26058631921824105,
+      "grad_norm": 0.632486879825592,
+      "learning_rate": 0.0002,
+      "loss": 1.5238,
+      "step": 360
+    },
+    {
+      "epoch": 0.2678248280854144,
+      "grad_norm": 0.6527034640312195,
+      "learning_rate": 0.0002,
+      "loss": 1.5175,
+      "step": 370
+    },
+    {
+      "epoch": 0.2750633369525878,
+      "grad_norm": 0.7672118544578552,
+      "learning_rate": 0.0002,
+      "loss": 1.627,
+      "step": 380
+    },
+    {
+      "epoch": 0.28230184581976114,
+      "grad_norm": 0.6035117506980896,
+      "learning_rate": 0.0002,
+      "loss": 1.5605,
+      "step": 390
+    },
+    {
+      "epoch": 0.2895403546869345,
+      "grad_norm": 0.5955103039741516,
+      "learning_rate": 0.0002,
+      "loss": 1.4603,
+      "step": 400
+    },
+    {
+      "epoch": 0.2967788635541079,
+      "grad_norm": 0.6015191674232483,
+      "learning_rate": 0.0002,
+      "loss": 1.558,
+      "step": 410
+    },
+    {
+      "epoch": 0.30401737242128124,
+      "grad_norm": 0.6380982398986816,
+      "learning_rate": 0.0002,
+      "loss": 1.6091,
+      "step": 420
+    },
+    {
+      "epoch": 0.3112558812884546,
+      "grad_norm": 0.6707863211631775,
+      "learning_rate": 0.0002,
+      "loss": 1.5292,
+      "step": 430
+    },
+    {
+      "epoch": 0.3184943901556279,
+      "grad_norm": 0.7010176777839661,
+      "learning_rate": 0.0002,
+      "loss": 1.4426,
+      "step": 440
+    },
+    {
+      "epoch": 0.3257328990228013,
+      "grad_norm": 0.8263739943504333,
+      "learning_rate": 0.0002,
+      "loss": 1.5572,
+      "step": 450
+    },
+    {
+      "epoch": 0.33297140788997465,
+      "grad_norm": 0.7253276109695435,
+      "learning_rate": 0.0002,
+      "loss": 1.5188,
+      "step": 460
+    },
+    {
+      "epoch": 0.340209916757148,
+      "grad_norm": 0.5238934755325317,
+      "learning_rate": 0.0002,
+      "loss": 1.584,
+      "step": 470
+    },
+    {
+      "epoch": 0.3474484256243214,
+      "grad_norm": 0.7869495749473572,
+      "learning_rate": 0.0002,
+      "loss": 1.7035,
+      "step": 480
+    },
+    {
+      "epoch": 0.35468693449149474,
+      "grad_norm": 0.7485215663909912,
+      "learning_rate": 0.0002,
+      "loss": 1.5776,
+      "step": 490
+    },
+    {
+      "epoch": 0.3619254433586681,
+      "grad_norm": 0.5413193106651306,
+      "learning_rate": 0.0002,
+      "loss": 1.6274,
+      "step": 500
+    },
+    {
+      "epoch": 0.3691639522258415,
+      "grad_norm": 0.7615048885345459,
+      "learning_rate": 0.0002,
+      "loss": 1.7323,
+      "step": 510
+    },
+    {
+      "epoch": 0.37640246109301484,
+      "grad_norm": 0.7685340046882629,
+      "learning_rate": 0.0002,
+      "loss": 1.532,
+      "step": 520
+    },
+    {
+      "epoch": 0.3836409699601882,
+      "grad_norm": 0.6379081010818481,
+      "learning_rate": 0.0002,
+      "loss": 1.6312,
+      "step": 530
+    },
+    {
+      "epoch": 0.39087947882736157,
+      "grad_norm": 0.7946939468383789,
+      "learning_rate": 0.0002,
+      "loss": 1.5645,
+      "step": 540
+    },
+    {
+      "epoch": 0.39811798769453494,
+      "grad_norm": 0.6287278532981873,
+      "learning_rate": 0.0002,
+      "loss": 1.4001,
+      "step": 550
+    },
+    {
+      "epoch": 0.4053564965617083,
+      "grad_norm": 0.6811642646789551,
+      "learning_rate": 0.0002,
+      "loss": 1.5982,
+      "step": 560
+    },
+    {
+      "epoch": 0.41259500542888167,
+      "grad_norm": 0.671073317527771,
+      "learning_rate": 0.0002,
+      "loss": 1.4953,
+      "step": 570
+    },
+    {
+      "epoch": 0.41983351429605503,
+      "grad_norm": 0.6313900351524353,
+      "learning_rate": 0.0002,
+      "loss": 1.6753,
+      "step": 580
+    },
+    {
+      "epoch": 0.4270720231632284,
+      "grad_norm": 0.5291772484779358,
+      "learning_rate": 0.0002,
+      "loss": 1.546,
+      "step": 590
+    },
+    {
+      "epoch": 0.43431053203040176,
+      "grad_norm": 0.62503582239151,
+      "learning_rate": 0.0002,
+      "loss": 1.5441,
+      "step": 600
+    },
+    {
+      "epoch": 0.4415490408975751,
+      "grad_norm": 0.5777305364608765,
+      "learning_rate": 0.0002,
+      "loss": 1.6276,
+      "step": 610
+    },
+    {
+      "epoch": 0.44878754976474844,
+      "grad_norm": 0.7013497352600098,
+      "learning_rate": 0.0002,
+      "loss": 1.4758,
+      "step": 620
+    },
+    {
+      "epoch": 0.4560260586319218,
+      "grad_norm": 0.8044822216033936,
+      "learning_rate": 0.0002,
+      "loss": 1.4029,
+      "step": 630
+    },
+    {
+      "epoch": 0.46326456749909517,
+      "grad_norm": 0.672531247138977,
+      "learning_rate": 0.0002,
+      "loss": 1.7195,
+      "step": 640
+    },
+    {
+      "epoch": 0.47050307636626854,
+      "grad_norm": 0.6233910322189331,
+      "learning_rate": 0.0002,
+      "loss": 1.614,
+      "step": 650
+    },
+    {
+      "epoch": 0.4777415852334419,
+      "grad_norm": 0.651524543762207,
+      "learning_rate": 0.0002,
+      "loss": 1.6041,
+      "step": 660
+    },
+    {
+      "epoch": 0.48498009410061527,
+      "grad_norm": 0.7213939428329468,
+      "learning_rate": 0.0002,
+      "loss": 1.5842,
+      "step": 670
+    },
+    {
+      "epoch": 0.49221860296778863,
+      "grad_norm": 0.6541454792022705,
+      "learning_rate": 0.0002,
+      "loss": 1.5453,
+      "step": 680
+    },
+    {
+      "epoch": 0.499457111834962,
+      "grad_norm": 0.6568936109542847,
+      "learning_rate": 0.0002,
+      "loss": 1.662,
+      "step": 690
+    },
+    {
+      "epoch": 0.5066956207021354,
+      "grad_norm": 0.7176415324211121,
+      "learning_rate": 0.0002,
+      "loss": 1.624,
+      "step": 700
+    },
+    {
+      "epoch": 0.5139341295693087,
+      "grad_norm": 0.6553855538368225,
+      "learning_rate": 0.0002,
+      "loss": 1.6099,
+      "step": 710
+    },
+    {
+      "epoch": 0.5211726384364821,
+      "grad_norm": 0.5654335618019104,
+      "learning_rate": 0.0002,
+      "loss": 1.5508,
+      "step": 720
+    },
+    {
+      "epoch": 0.5284111473036555,
+      "grad_norm": 0.5671001672744751,
+      "learning_rate": 0.0002,
+      "loss": 1.392,
+      "step": 730
+    },
+    {
+      "epoch": 0.5356496561708288,
+      "grad_norm": 0.7914412021636963,
+      "learning_rate": 0.0002,
+      "loss": 1.388,
+      "step": 740
+    },
+    {
+      "epoch": 0.5428881650380022,
+      "grad_norm": 0.6172138452529907,
+      "learning_rate": 0.0002,
+      "loss": 1.5931,
+      "step": 750
+    },
+    {
+      "epoch": 0.5501266739051756,
+      "grad_norm": 0.6132623553276062,
+      "learning_rate": 0.0002,
+      "loss": 1.4018,
+      "step": 760
+    },
+    {
+      "epoch": 0.5573651827723489,
+      "grad_norm": 0.654000461101532,
+      "learning_rate": 0.0002,
+      "loss": 1.513,
+      "step": 770
+    },
+    {
+      "epoch": 0.5646036916395223,
+      "grad_norm": 0.5691370964050293,
+      "learning_rate": 0.0002,
+      "loss": 1.5035,
+      "step": 780
+    },
+    {
+      "epoch": 0.5718422005066957,
+      "grad_norm": 0.7922580242156982,
+      "learning_rate": 0.0002,
+      "loss": 1.65,
+      "step": 790
+    },
+    {
+      "epoch": 0.579080709373869,
+      "grad_norm": 0.6831880211830139,
+      "learning_rate": 0.0002,
+      "loss": 1.4521,
+      "step": 800
+    },
+    {
+      "epoch": 0.5863192182410424,
+      "grad_norm": 0.6740124821662903,
+      "learning_rate": 0.0002,
+      "loss": 1.4734,
+      "step": 810
+    },
+    {
+      "epoch": 0.5935577271082157,
+      "grad_norm": 1.380016803741455,
+      "learning_rate": 0.0002,
+      "loss": 1.6498,
+      "step": 820
+    },
+    {
+      "epoch": 0.6007962359753891,
+      "grad_norm": 0.6552878022193909,
+      "learning_rate": 0.0002,
+      "loss": 1.4642,
+      "step": 830
+    },
+    {
+      "epoch": 0.6080347448425625,
+      "grad_norm": 0.6649535298347473,
+      "learning_rate": 0.0002,
+      "loss": 1.6271,
+      "step": 840
+    },
+    {
+      "epoch": 0.6152732537097358,
+      "grad_norm": 0.561738133430481,
+      "learning_rate": 0.0002,
+      "loss": 1.5886,
+      "step": 850
+    },
+    {
+      "epoch": 0.6225117625769092,
+      "grad_norm": 0.6133047938346863,
+      "learning_rate": 0.0002,
+      "loss": 1.5364,
+      "step": 860
+    },
+    {
+      "epoch": 0.6297502714440825,
+      "grad_norm": 0.559843122959137,
+      "learning_rate": 0.0002,
+      "loss": 1.3489,
+      "step": 870
+    },
+    {
+      "epoch": 0.6369887803112558,
+      "grad_norm": 0.6117811799049377,
+      "learning_rate": 0.0002,
+      "loss": 1.4878,
+      "step": 880
+    },
+    {
+      "epoch": 0.6442272891784292,
+      "grad_norm": 0.6209776401519775,
+      "learning_rate": 0.0002,
+      "loss": 1.56,
+      "step": 890
+    },
+    {
+      "epoch": 0.6514657980456026,
+      "grad_norm": 0.6234082579612732,
+      "learning_rate": 0.0002,
+      "loss": 1.6747,
+      "step": 900
+    },
+    {
+      "epoch": 0.6587043069127759,
+      "grad_norm": 0.7623258233070374,
+      "learning_rate": 0.0002,
+      "loss": 1.6963,
+      "step": 910
+    },
+    {
+      "epoch": 0.6659428157799493,
+      "grad_norm": 0.6148061752319336,
+      "learning_rate": 0.0002,
+      "loss": 1.2424,
+      "step": 920
+    },
+    {
+      "epoch": 0.6731813246471227,
+      "grad_norm": 0.6682973504066467,
+      "learning_rate": 0.0002,
+      "loss": 1.4319,
+      "step": 930
+    },
+    {
+      "epoch": 0.680419833514296,
+      "grad_norm": 0.5513041615486145,
+      "learning_rate": 0.0002,
+      "loss": 1.5377,
+      "step": 940
+    },
+    {
+      "epoch": 0.6876583423814694,
+      "grad_norm": 0.5197525024414062,
+      "learning_rate": 0.0002,
+      "loss": 1.3991,
+      "step": 950
+    },
+    {
+      "epoch": 0.6948968512486428,
+      "grad_norm": 0.6490758061408997,
+      "learning_rate": 0.0002,
+      "loss": 1.4398,
+      "step": 960
+    },
+    {
+      "epoch": 0.7021353601158161,
+      "grad_norm": 0.6450682878494263,
+      "learning_rate": 0.0002,
+      "loss": 1.5251,
+      "step": 970
+    },
+    {
+      "epoch": 0.7093738689829895,
+      "grad_norm": 0.6203766465187073,
+      "learning_rate": 0.0002,
+      "loss": 1.5417,
+      "step": 980
+    },
+    {
+      "epoch": 0.7166123778501629,
+      "grad_norm": 0.6023609638214111,
+      "learning_rate": 0.0002,
+      "loss": 1.4575,
+      "step": 990
+    },
+    {
+      "epoch": 0.7238508867173362,
+      "grad_norm": 0.5765255093574524,
+      "learning_rate": 0.0002,
+      "loss": 1.4973,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7310893955845096,
+      "grad_norm": 0.6650075316429138,
+      "learning_rate": 0.0002,
+      "loss": 1.483,
+      "step": 1010
+    },
+    {
+      "epoch": 0.738327904451683,
+      "grad_norm": 0.5610854029655457,
+      "learning_rate": 0.0002,
+      "loss": 1.5959,
+      "step": 1020
+    },
+    {
+      "epoch": 0.7455664133188563,
+      "grad_norm": 0.7072813510894775,
+      "learning_rate": 0.0002,
+      "loss": 1.5248,
+      "step": 1030
+    },
+    {
+      "epoch": 0.7528049221860297,
+      "grad_norm": 0.6815407872200012,
+      "learning_rate": 0.0002,
+      "loss": 1.5776,
+      "step": 1040
+    },
+    {
+      "epoch": 0.760043431053203,
+      "grad_norm": 0.7932390570640564,
+      "learning_rate": 0.0002,
+      "loss": 1.4577,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7672819399203764,
+      "grad_norm": 0.5798183083534241,
+      "learning_rate": 0.0002,
+      "loss": 1.4515,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7745204487875498,
+      "grad_norm": 0.7898504137992859,
+      "learning_rate": 0.0002,
+      "loss": 1.5053,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7817589576547231,
+      "grad_norm": 0.4983280301094055,
+      "learning_rate": 0.0002,
+      "loss": 1.4776,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7889974665218965,
+      "grad_norm": 0.691403329372406,
+      "learning_rate": 0.0002,
+      "loss": 1.5007,
+      "step": 1090
+    },
+    {
+      "epoch": 0.7962359753890699,
+      "grad_norm": 0.5394481420516968,
+      "learning_rate": 0.0002,
+      "loss": 1.5153,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8034744842562432,
+      "grad_norm": 0.5136822462081909,
+      "learning_rate": 0.0002,
+      "loss": 1.6892,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8107129931234166,
+      "grad_norm": 0.6828126907348633,
+      "learning_rate": 0.0002,
+      "loss": 1.4902,
+      "step": 1120
+    },
+    {
+      "epoch": 0.81795150199059,
+      "grad_norm": 0.6799656748771667,
+      "learning_rate": 0.0002,
+      "loss": 1.4346,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8251900108577633,
+      "grad_norm": 0.5428406000137329,
+      "learning_rate": 0.0002,
+      "loss": 1.2678,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8324285197249367,
+      "grad_norm": 0.4811290502548218,
+      "learning_rate": 0.0002,
+      "loss": 1.4072,
+      "step": 1150
+    },
+    {
+      "epoch": 0.8396670285921101,
+      "grad_norm": 0.5519434809684753,
+      "learning_rate": 0.0002,
+      "loss": 1.4512,
+      "step": 1160
+    },
+    {
+      "epoch": 0.8469055374592834,
+      "grad_norm": 0.9748060703277588,
+      "learning_rate": 0.0002,
+      "loss": 1.4072,
+      "step": 1170
+    },
+    {
+      "epoch": 0.8541440463264568,
+      "grad_norm": 0.712609589099884,
+      "learning_rate": 0.0002,
+      "loss": 1.4309,
+      "step": 1180
+    },
+    {
+      "epoch": 0.8613825551936302,
+      "grad_norm": 0.6866157054901123,
+      "learning_rate": 0.0002,
+      "loss": 1.434,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8686210640608035,
+      "grad_norm": 0.5068854093551636,
+      "learning_rate": 0.0002,
+      "loss": 1.3704,
+      "step": 1200
+    },
+    {
+      "epoch": 0.8758595729279768,
+      "grad_norm": 0.6333245038986206,
+      "learning_rate": 0.0002,
+      "loss": 1.5601,
+      "step": 1210
+    },
+    {
+      "epoch": 0.8830980817951501,
+      "grad_norm": 0.6424421072006226,
+      "learning_rate": 0.0002,
+      "loss": 1.4636,
+      "step": 1220
+    },
+    {
+      "epoch": 0.8903365906623235,
+      "grad_norm": 0.4771921932697296,
+      "learning_rate": 0.0002,
+      "loss": 1.4186,
+      "step": 1230
+    },
+    {
+      "epoch": 0.8975750995294969,
+      "grad_norm": 0.5191764235496521,
+      "learning_rate": 0.0002,
+      "loss": 1.6323,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9048136083966702,
+      "grad_norm": 0.756222128868103,
+      "learning_rate": 0.0002,
+      "loss": 1.6105,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9120521172638436,
+      "grad_norm": 0.623823881149292,
+      "learning_rate": 0.0002,
+      "loss": 1.4396,
+      "step": 1260
+    },
+    {
+      "epoch": 0.919290626131017,
+      "grad_norm": 0.8166571259498596,
+      "learning_rate": 0.0002,
+      "loss": 1.3097,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9265291349981903,
+      "grad_norm": 0.6059346795082092,
+      "learning_rate": 0.0002,
+      "loss": 1.4625,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9337676438653637,
+      "grad_norm": 0.5842690467834473,
+      "learning_rate": 0.0002,
+      "loss": 1.3555,
+      "step": 1290
+    },
+    {
+      "epoch": 0.9410061527325371,
+      "grad_norm": 0.7649800777435303,
+      "learning_rate": 0.0002,
+      "loss": 1.5859,
+      "step": 1300
+    },
+    {
+      "epoch": 0.9482446615997104,
+      "grad_norm": 0.6420919895172119,
+      "learning_rate": 0.0002,
+      "loss": 1.5915,
+      "step": 1310
+    },
+    {
+      "epoch": 0.9554831704668838,
+      "grad_norm": 0.7011452913284302,
+      "learning_rate": 0.0002,
+      "loss": 1.453,
+      "step": 1320
+    },
+    {
+      "epoch": 0.9627216793340572,
+      "grad_norm": 0.5783746242523193,
+      "learning_rate": 0.0002,
+      "loss": 1.6766,
+      "step": 1330
+    },
+    {
+      "epoch": 0.9699601882012305,
+      "grad_norm": 0.5973192453384399,
+      "learning_rate": 0.0002,
+      "loss": 1.6308,
+      "step": 1340
+    },
+    {
+      "epoch": 0.9771986970684039,
+      "grad_norm": 0.6181833744049072,
+      "learning_rate": 0.0002,
+      "loss": 1.5901,
+      "step": 1350
+    },
+    {
+      "epoch": 0.9844372059355773,
+      "grad_norm": 0.5563396215438843,
+      "learning_rate": 0.0002,
+      "loss": 1.5258,
+      "step": 1360
+    },
+    {
+      "epoch": 0.9916757148027506,
+      "grad_norm": 0.45723360776901245,
+      "learning_rate": 0.0002,
+      "loss": 1.4508,
+      "step": 1370
+    },
+    {
+      "epoch": 0.998914223669924,
+      "grad_norm": 0.5947498679161072,
+      "learning_rate": 0.0002,
+      "loss": 1.3291,
+      "step": 1380
+    },
+    {
+      "epoch": 0.9996380745566413,
+      "eval_loss": 1.480796456336975,
+      "eval_runtime": 27.3103,
+      "eval_samples_per_second": 15.965,
+      "eval_steps_per_second": 2.014,
+      "step": 1381
+    },
+    {
+      "epoch": 1.0061527325370974,
+      "grad_norm": 0.5599952936172485,
+      "learning_rate": 0.0002,
+      "loss": 1.3057,
+      "step": 1390
+    },
+    {
+      "epoch": 1.0133912414042707,
+      "grad_norm": 0.5932008028030396,
+      "learning_rate": 0.0002,
+      "loss": 1.4991,
+      "step": 1400
+    },
+    {
+      "epoch": 1.020629750271444,
+      "grad_norm": 0.6194121837615967,
+      "learning_rate": 0.0002,
+      "loss": 1.4506,
+      "step": 1410
+    },
+    {
+      "epoch": 1.0278682591386175,
+      "grad_norm": 0.6995621919631958,
+      "learning_rate": 0.0002,
+      "loss": 1.5966,
+      "step": 1420
+    },
+    {
+      "epoch": 1.0351067680057908,
+      "grad_norm": 0.7905810475349426,
+      "learning_rate": 0.0002,
+      "loss": 1.4153,
+      "step": 1430
+    },
+    {
+      "epoch": 1.0423452768729642,
+      "grad_norm": 0.7221615314483643,
+      "learning_rate": 0.0002,
+      "loss": 1.4414,
+      "step": 1440
+    },
+    {
+      "epoch": 1.0495837857401376,
+      "grad_norm": 0.6170642375946045,
+      "learning_rate": 0.0002,
+      "loss": 1.3859,
+      "step": 1450
+    },
+    {
+      "epoch": 1.056822294607311,
+      "grad_norm": 0.5844094753265381,
+      "learning_rate": 0.0002,
+      "loss": 1.3806,
+      "step": 1460
+    },
+    {
+      "epoch": 1.0640608034744843,
+      "grad_norm": 0.7731822729110718,
+      "learning_rate": 0.0002,
+      "loss": 1.4871,
+      "step": 1470
+    },
+    {
+      "epoch": 1.0712993123416577,
+      "grad_norm": 0.4554748237133026,
+      "learning_rate": 0.0002,
+      "loss": 1.4286,
+      "step": 1480
+    },
+    {
+      "epoch": 1.078537821208831,
+      "grad_norm": 0.6923259496688843,
+      "learning_rate": 0.0002,
+      "loss": 1.3977,
+      "step": 1490
+    },
+    {
+      "epoch": 1.0857763300760044,
+      "grad_norm": 0.6008219122886658,
+      "learning_rate": 0.0002,
+      "loss": 1.3936,
+      "step": 1500
+    },
+    {
+      "epoch": 1.0930148389431777,
+      "grad_norm": 0.6450045704841614,
+      "learning_rate": 0.0002,
+      "loss": 1.4821,
+      "step": 1510
+    },
+    {
+      "epoch": 1.1002533478103511,
+      "grad_norm": 0.7833753824234009,
+      "learning_rate": 0.0002,
+      "loss": 1.3295,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1074918566775245,
+      "grad_norm": 0.5076758861541748,
+      "learning_rate": 0.0002,
+      "loss": 1.3424,
+      "step": 1530
+    },
+    {
+      "epoch": 1.1147303655446978,
+      "grad_norm": 0.5661332011222839,
+      "learning_rate": 0.0002,
+      "loss": 1.4043,
+      "step": 1540
+    },
+    {
+      "epoch": 1.1219688744118712,
+      "grad_norm": 0.6526919603347778,
+      "learning_rate": 0.0002,
+      "loss": 1.4963,
+      "step": 1550
+    },
+    {
+      "epoch": 1.1292073832790446,
+      "grad_norm": 0.5613082647323608,
+      "learning_rate": 0.0002,
+      "loss": 1.3671,
+      "step": 1560
+    },
+    {
+      "epoch": 1.136445892146218,
+      "grad_norm": 0.6113885641098022,
+      "learning_rate": 0.0002,
+      "loss": 1.4458,
+      "step": 1570
+    },
+    {
+      "epoch": 1.1436844010133913,
+      "grad_norm": 0.6732510328292847,
+      "learning_rate": 0.0002,
+      "loss": 1.3552,
+      "step": 1580
+    },
+    {
+      "epoch": 1.1509229098805647,
+      "grad_norm": 0.6146392226219177,
+      "learning_rate": 0.0002,
+      "loss": 1.3114,
+      "step": 1590
+    },
+    {
+      "epoch": 1.158161418747738,
+      "grad_norm": 0.6766974329948425,
+      "learning_rate": 0.0002,
+      "loss": 1.411,
+      "step": 1600
+    },
+    {
+      "epoch": 1.1653999276149114,
+      "grad_norm": 0.7621957659721375,
+      "learning_rate": 0.0002,
+      "loss": 1.2401,
+      "step": 1610
+    },
+    {
+      "epoch": 1.1726384364820848,
+      "grad_norm": 0.6959581971168518,
+      "learning_rate": 0.0002,
+      "loss": 1.3758,
+      "step": 1620
+    },
+    {
+      "epoch": 1.1798769453492581,
+      "grad_norm": 0.6691278219223022,
+      "learning_rate": 0.0002,
+      "loss": 1.382,
+      "step": 1630
+    },
+    {
+      "epoch": 1.1871154542164315,
+      "grad_norm": 0.4927774965763092,
+      "learning_rate": 0.0002,
+      "loss": 1.4147,
+      "step": 1640
+    },
+    {
+      "epoch": 1.1943539630836049,
+      "grad_norm": 0.7724234461784363,
+      "learning_rate": 0.0002,
+      "loss": 1.449,
+      "step": 1650
+    },
+    {
+      "epoch": 1.2015924719507782,
+      "grad_norm": 0.6817787885665894,
+      "learning_rate": 0.0002,
+      "loss": 1.4778,
+      "step": 1660
+    },
+    {
+      "epoch": 1.2088309808179516,
+      "grad_norm": 0.6500699520111084,
+      "learning_rate": 0.0002,
+      "loss": 1.3776,
+      "step": 1670
+    },
+    {
+      "epoch": 1.216069489685125,
+      "grad_norm": 0.5703568458557129,
+      "learning_rate": 0.0002,
+      "loss": 1.3875,
+      "step": 1680
+    },
+    {
+      "epoch": 1.2233079985522983,
+      "grad_norm": 0.6261579990386963,
+      "learning_rate": 0.0002,
+      "loss": 1.4735,
+      "step": 1690
+    },
+    {
+      "epoch": 1.2305465074194717,
+      "grad_norm": 0.651713490486145,
+      "learning_rate": 0.0002,
+      "loss": 1.3898,
+      "step": 1700
+    },
+    {
+      "epoch": 1.237785016286645,
+      "grad_norm": 0.684399425983429,
+      "learning_rate": 0.0002,
+      "loss": 1.4002,
+      "step": 1710
+    },
+    {
+      "epoch": 1.2450235251538184,
+      "grad_norm": 0.6996857523918152,
+      "learning_rate": 0.0002,
+      "loss": 1.5027,
+      "step": 1720
+    },
+    {
+      "epoch": 1.2522620340209918,
+      "grad_norm": 0.7102537751197815,
+      "learning_rate": 0.0002,
+      "loss": 1.3326,
+      "step": 1730
+    },
+    {
+      "epoch": 1.2595005428881652,
+      "grad_norm": 0.45809897780418396,
+      "learning_rate": 0.0002,
+      "loss": 1.3675,
+      "step": 1740
+    },
+    {
+      "epoch": 1.2667390517553385,
+      "grad_norm": 0.6377046704292297,
+      "learning_rate": 0.0002,
+      "loss": 1.4175,
+      "step": 1750
+    },
+    {
+      "epoch": 1.2739775606225119,
+      "grad_norm": 0.6965704560279846,
+      "learning_rate": 0.0002,
+      "loss": 1.3479,
+      "step": 1760
+    },
+    {
+      "epoch": 1.2812160694896852,
+      "grad_norm": 0.5688214302062988,
+      "learning_rate": 0.0002,
+      "loss": 1.5647,
+      "step": 1770
+    },
+    {
+      "epoch": 1.2884545783568586,
+      "grad_norm": 0.6384190320968628,
+      "learning_rate": 0.0002,
+      "loss": 1.3967,
+      "step": 1780
+    },
+    {
+      "epoch": 1.295693087224032,
+      "grad_norm": 0.5629363656044006,
+      "learning_rate": 0.0002,
+      "loss": 1.3671,
+      "step": 1790
+    },
+    {
+      "epoch": 1.3029315960912053,
+      "grad_norm": 0.6148255467414856,
+      "learning_rate": 0.0002,
+      "loss": 1.2292,
+      "step": 1800
+    },
+    {
+      "epoch": 1.3101701049583787,
+      "grad_norm": 0.655580997467041,
+      "learning_rate": 0.0002,
+      "loss": 1.5806,
+      "step": 1810
+    },
+    {
+      "epoch": 1.3174086138255519,
+      "grad_norm": 0.5642657279968262,
+      "learning_rate": 0.0002,
+      "loss": 1.2398,
+      "step": 1820
+    },
+    {
+      "epoch": 1.3246471226927252,
+      "grad_norm": 0.59607994556427,
+      "learning_rate": 0.0002,
+      "loss": 1.3246,
+      "step": 1830
+    },
+    {
+      "epoch": 1.3318856315598986,
+      "grad_norm": 0.5564199090003967,
+      "learning_rate": 0.0002,
+      "loss": 1.3274,
+      "step": 1840
+    },
+    {
+      "epoch": 1.339124140427072,
+      "grad_norm": 0.6949955821037292,
+      "learning_rate": 0.0002,
+      "loss": 1.5834,
+      "step": 1850
+    },
+    {
+      "epoch": 1.3463626492942453,
+      "grad_norm": 0.7036856412887573,
+      "learning_rate": 0.0002,
+      "loss": 1.4722,
+      "step": 1860
+    },
+    {
+      "epoch": 1.3536011581614187,
+      "grad_norm": 0.722062885761261,
+      "learning_rate": 0.0002,
+      "loss": 1.333,
+      "step": 1870
+    },
+    {
+      "epoch": 1.360839667028592,
+      "grad_norm": 0.6098677515983582,
+      "learning_rate": 0.0002,
+      "loss": 1.4044,
+      "step": 1880
+    },
+    {
+      "epoch": 1.3680781758957654,
+      "grad_norm": 0.5376402735710144,
+      "learning_rate": 0.0002,
+      "loss": 1.6217,
+      "step": 1890
+    },
+    {
+      "epoch": 1.3753166847629388,
+      "grad_norm": 0.6974610090255737,
+      "learning_rate": 0.0002,
+      "loss": 1.5071,
+      "step": 1900
+    },
+    {
+      "epoch": 1.3825551936301121,
+      "grad_norm": 0.6520763635635376,
+      "learning_rate": 0.0002,
+      "loss": 1.5854,
+      "step": 1910
+    },
+    {
+      "epoch": 1.3897937024972855,
+      "grad_norm": 0.6604374647140503,
+      "learning_rate": 0.0002,
+      "loss": 1.4271,
+      "step": 1920
+    },
+    {
+      "epoch": 1.3970322113644589,
+      "grad_norm": 0.7364398241043091,
+      "learning_rate": 0.0002,
+      "loss": 1.419,
+      "step": 1930
+    },
+    {
+      "epoch": 1.4042707202316322,
+      "grad_norm": 0.6849475502967834,
+      "learning_rate": 0.0002,
+      "loss": 1.4585,
+      "step": 1940
+    },
+    {
+      "epoch": 1.4115092290988056,
+      "grad_norm": 0.6562670469284058,
+      "learning_rate": 0.0002,
+      "loss": 1.5577,
+      "step": 1950
+    },
+    {
+      "epoch": 1.418747737965979,
+      "grad_norm": 0.5695616006851196,
+      "learning_rate": 0.0002,
+      "loss": 1.4725,
+      "step": 1960
+    },
+    {
+      "epoch": 1.4259862468331523,
+      "grad_norm": 0.5244464874267578,
+      "learning_rate": 0.0002,
+      "loss": 1.3088,
+      "step": 1970
+    },
+    {
+      "epoch": 1.4332247557003257,
+      "grad_norm": 0.6347293257713318,
+      "learning_rate": 0.0002,
+      "loss": 1.5069,
+      "step": 1980
+    },
+    {
+      "epoch": 1.440463264567499,
+      "grad_norm": 0.5528361201286316,
+      "learning_rate": 0.0002,
+      "loss": 1.3502,
+      "step": 1990
+    },
+    {
+      "epoch": 1.4477017734346724,
+      "grad_norm": 0.6987585425376892,
+      "learning_rate": 0.0002,
+      "loss": 1.3978,
+      "step": 2000
+    },
+    {
+      "epoch": 1.4549402823018458,
+      "grad_norm": 0.6568987369537354,
+      "learning_rate": 0.0002,
+      "loss": 1.4262,
+      "step": 2010
+    },
+    {
+      "epoch": 1.4621787911690192,
+      "grad_norm": 0.7665994763374329,
+      "learning_rate": 0.0002,
+      "loss": 1.4175,
+      "step": 2020
+    },
+    {
+      "epoch": 1.4694173000361925,
+      "grad_norm": 0.5127707123756409,
+      "learning_rate": 0.0002,
+      "loss": 1.244,
+      "step": 2030
+    },
+    {
+      "epoch": 1.476655808903366,
+      "grad_norm": 0.5406824946403503,
+      "learning_rate": 0.0002,
+      "loss": 1.3699,
+      "step": 2040
+    },
+    {
+      "epoch": 1.4838943177705393,
+      "grad_norm": 0.5990166664123535,
+      "learning_rate": 0.0002,
+      "loss": 1.3353,
+      "step": 2050
+    },
+    {
+      "epoch": 1.4911328266377126,
+      "grad_norm": 0.6186193823814392,
+      "learning_rate": 0.0002,
+      "loss": 1.2454,
+      "step": 2060
+    },
+    {
+      "epoch": 1.498371335504886,
+      "grad_norm": 0.6154307126998901,
+      "learning_rate": 0.0002,
+      "loss": 1.428,
+      "step": 2070
+    },
+    {
+      "epoch": 1.5056098443720594,
+      "grad_norm": 0.5606056451797485,
+      "learning_rate": 0.0002,
+      "loss": 1.4528,
+      "step": 2080
+    },
+    {
+      "epoch": 1.5128483532392327,
+      "grad_norm": 0.5006417036056519,
+      "learning_rate": 0.0002,
+      "loss": 1.2405,
+      "step": 2090
+    },
+    {
+      "epoch": 1.520086862106406,
+      "grad_norm": 0.5968486070632935,
+      "learning_rate": 0.0002,
+      "loss": 1.4258,
+      "step": 2100
+    },
+    {
+      "epoch": 1.5273253709735795,
+      "grad_norm": 0.5835496187210083,
+      "learning_rate": 0.0002,
+      "loss": 1.2752,
+      "step": 2110
+    },
+    {
+      "epoch": 1.5345638798407528,
+      "grad_norm": 0.6753535270690918,
+      "learning_rate": 0.0002,
+      "loss": 1.5443,
+      "step": 2120
+    },
+    {
+      "epoch": 1.5418023887079262,
+      "grad_norm": 0.7299720644950867,
+      "learning_rate": 0.0002,
+      "loss": 1.2139,
+      "step": 2130
+    },
+    {
+      "epoch": 1.5490408975750996,
+      "grad_norm": 0.5105988383293152,
+      "learning_rate": 0.0002,
+      "loss": 1.2364,
+      "step": 2140
+    },
+    {
+      "epoch": 1.556279406442273,
+      "grad_norm": 0.5675431489944458,
+      "learning_rate": 0.0002,
+      "loss": 1.4528,
+      "step": 2150
+    },
+    {
+      "epoch": 1.5635179153094463,
+      "grad_norm": 0.6246723532676697,
+      "learning_rate": 0.0002,
+      "loss": 1.4563,
+      "step": 2160
+    },
+    {
+      "epoch": 1.5707564241766196,
+      "grad_norm": 0.7291720509529114,
+      "learning_rate": 0.0002,
+      "loss": 1.5255,
+      "step": 2170
+    },
+    {
+      "epoch": 1.577994933043793,
+      "grad_norm": 0.678114116191864,
+      "learning_rate": 0.0002,
+      "loss": 1.5432,
+      "step": 2180
+    },
+    {
+      "epoch": 1.5852334419109664,
+      "grad_norm": 0.5136260986328125,
+      "learning_rate": 0.0002,
+      "loss": 1.5212,
+      "step": 2190
+    },
+    {
+      "epoch": 1.5924719507781397,
+      "grad_norm": 0.6359935998916626,
+      "learning_rate": 0.0002,
+      "loss": 1.3271,
+      "step": 2200
+    },
+    {
+      "epoch": 1.599710459645313,
+      "grad_norm": 0.7650278806686401,
+      "learning_rate": 0.0002,
+      "loss": 1.4038,
+      "step": 2210
+    },
+    {
+      "epoch": 1.6069489685124865,
+      "grad_norm": 0.7256110906600952,
+      "learning_rate": 0.0002,
+      "loss": 1.5478,
+      "step": 2220
+    },
+    {
+      "epoch": 1.6141874773796598,
+      "grad_norm": 0.688689649105072,
+      "learning_rate": 0.0002,
+      "loss": 1.4387,
+      "step": 2230
+    },
+    {
+      "epoch": 1.6214259862468332,
+      "grad_norm": 0.6045311093330383,
+      "learning_rate": 0.0002,
+      "loss": 1.4096,
+      "step": 2240
+    },
+    {
+      "epoch": 1.6286644951140063,
+      "grad_norm": 0.7064604163169861,
+      "learning_rate": 0.0002,
+      "loss": 1.4097,
+      "step": 2250
+    },
+    {
+      "epoch": 1.6359030039811797,
+      "grad_norm": 0.5309562087059021,
+      "learning_rate": 0.0002,
+      "loss": 1.3477,
+      "step": 2260
+    },
+    {
+      "epoch": 1.643141512848353,
+      "grad_norm": 0.5687053203582764,
+      "learning_rate": 0.0002,
+      "loss": 1.4022,
+      "step": 2270
+    },
+    {
+      "epoch": 1.6503800217155264,
+      "grad_norm": 0.535872757434845,
+      "learning_rate": 0.0002,
+      "loss": 1.2977,
+      "step": 2280
+    },
+    {
+      "epoch": 1.6576185305826998,
+      "grad_norm": 0.5502381920814514,
+      "learning_rate": 0.0002,
+      "loss": 1.3844,
+      "step": 2290
+    },
+    {
+      "epoch": 1.6648570394498732,
+      "grad_norm": 0.6158602237701416,
+      "learning_rate": 0.0002,
+      "loss": 1.3764,
+      "step": 2300
+    },
+    {
+      "epoch": 1.6720955483170465,
+      "grad_norm": 0.5804675817489624,
+      "learning_rate": 0.0002,
+      "loss": 1.3515,
+      "step": 2310
+    },
+    {
+      "epoch": 1.67933405718422,
+      "grad_norm": 0.600742757320404,
+      "learning_rate": 0.0002,
+      "loss": 1.2532,
+      "step": 2320
+    },
+    {
+      "epoch": 1.6865725660513933,
+      "grad_norm": 0.7101941108703613,
+      "learning_rate": 0.0002,
+      "loss": 1.477,
+      "step": 2330
+    },
+    {
+      "epoch": 1.6938110749185666,
+      "grad_norm": 0.7507809996604919,
+      "learning_rate": 0.0002,
+      "loss": 1.4849,
+      "step": 2340
+    },
+    {
+      "epoch": 1.70104958378574,
+      "grad_norm": 0.768502414226532,
+      "learning_rate": 0.0002,
+      "loss": 1.2703,
+      "step": 2350
+    },
+    {
+      "epoch": 1.7082880926529134,
+      "grad_norm": 0.4801851212978363,
+      "learning_rate": 0.0002,
+      "loss": 1.3332,
+      "step": 2360
+    },
+    {
+      "epoch": 1.7155266015200867,
+      "grad_norm": 0.5322122573852539,
+      "learning_rate": 0.0002,
+      "loss": 1.4158,
+      "step": 2370
+    },
+    {
+      "epoch": 1.72276511038726,
+      "grad_norm": 0.587661862373352,
+      "learning_rate": 0.0002,
+      "loss": 1.4136,
+      "step": 2380
+    },
+    {
+      "epoch": 1.7300036192544335,
+      "grad_norm": 0.6073525547981262,
+      "learning_rate": 0.0002,
+      "loss": 1.3771,
+      "step": 2390
+    },
+    {
+      "epoch": 1.7372421281216068,
+      "grad_norm": 0.6950460076332092,
+      "learning_rate": 0.0002,
+      "loss": 1.2754,
+      "step": 2400
+    },
+    {
+      "epoch": 1.7444806369887802,
+      "grad_norm": 0.5981102585792542,
+      "learning_rate": 0.0002,
+      "loss": 1.3858,
+      "step": 2410
+    },
+    {
+      "epoch": 1.7517191458559536,
+      "grad_norm": 0.544570803642273,
+      "learning_rate": 0.0002,
+      "loss": 1.4075,
+      "step": 2420
+    },
+    {
+      "epoch": 1.758957654723127,
+      "grad_norm": 0.5304399728775024,
+      "learning_rate": 0.0002,
+      "loss": 1.3861,
+      "step": 2430
+    },
+    {
+      "epoch": 1.7661961635903003,
+      "grad_norm": 0.7921594977378845,
+      "learning_rate": 0.0002,
+      "loss": 1.4244,
+      "step": 2440
+    },
+    {
+      "epoch": 1.7734346724574737,
+      "grad_norm": 0.6084808707237244,
+      "learning_rate": 0.0002,
+      "loss": 1.3053,
+      "step": 2450
+    },
+    {
+      "epoch": 1.780673181324647,
+      "grad_norm": 0.8844701051712036,
+      "learning_rate": 0.0002,
+      "loss": 1.3781,
+      "step": 2460
+    },
+    {
+      "epoch": 1.7879116901918204,
+      "grad_norm": 0.5729258060455322,
+      "learning_rate": 0.0002,
+      "loss": 1.3227,
+      "step": 2470
+    },
+    {
+      "epoch": 1.7951501990589938,
+      "grad_norm": 0.6303611993789673,
+      "learning_rate": 0.0002,
+      "loss": 1.3422,
+      "step": 2480
+    },
+    {
+      "epoch": 1.8023887079261671,
+      "grad_norm": 0.5627942085266113,
+      "learning_rate": 0.0002,
+      "loss": 1.3926,
+      "step": 2490
+    },
+    {
+      "epoch": 1.8096272167933405,
+      "grad_norm": 0.6724274158477783,
+      "learning_rate": 0.0002,
+      "loss": 1.3816,
+      "step": 2500
+    },
+    {
+      "epoch": 1.8168657256605139,
+      "grad_norm": 0.5030826330184937,
+      "learning_rate": 0.0002,
+      "loss": 1.2951,
+      "step": 2510
+    },
+    {
+      "epoch": 1.8241042345276872,
+      "grad_norm": 0.5504099130630493,
+      "learning_rate": 0.0002,
+      "loss": 1.2839,
+      "step": 2520
+    },
+    {
+      "epoch": 1.8313427433948606,
+      "grad_norm": 0.6338945627212524,
+      "learning_rate": 0.0002,
+      "loss": 1.4264,
+      "step": 2530
+    },
+    {
+      "epoch": 1.838581252262034,
+      "grad_norm": 0.5902037620544434,
+      "learning_rate": 0.0002,
+      "loss": 1.563,
+      "step": 2540
+    },
+    {
+      "epoch": 1.8458197611292073,
+      "grad_norm": 0.48814457654953003,
+      "learning_rate": 0.0002,
+      "loss": 1.2961,
+      "step": 2550
+    },
+    {
+      "epoch": 1.8530582699963807,
+      "grad_norm": 0.6216312646865845,
+      "learning_rate": 0.0002,
+      "loss": 1.466,
+      "step": 2560
+    },
+    {
+      "epoch": 1.860296778863554,
+      "grad_norm": 0.635603666305542,
+      "learning_rate": 0.0002,
+      "loss": 1.5123,
+      "step": 2570
+    },
+    {
+      "epoch": 1.8675352877307274,
+      "grad_norm": 0.6938216090202332,
+      "learning_rate": 0.0002,
+      "loss": 1.372,
+      "step": 2580
+    },
+    {
+      "epoch": 1.8747737965979008,
+      "grad_norm": 0.599557638168335,
+      "learning_rate": 0.0002,
+      "loss": 1.5011,
+      "step": 2590
+    },
+    {
+      "epoch": 1.8820123054650741,
+      "grad_norm": 0.564424455165863,
+      "learning_rate": 0.0002,
+      "loss": 1.2714,
+      "step": 2600
+    },
+    {
+      "epoch": 1.8892508143322475,
+      "grad_norm": 0.5430700182914734,
+      "learning_rate": 0.0002,
+      "loss": 1.3403,
+      "step": 2610
+    },
+    {
+      "epoch": 1.8964893231994209,
+      "grad_norm": 0.6150169372558594,
+      "learning_rate": 0.0002,
+      "loss": 1.4347,
+      "step": 2620
+    },
+    {
+      "epoch": 1.9037278320665942,
+      "grad_norm": 0.48159119486808777,
+      "learning_rate": 0.0002,
+      "loss": 1.2474,
+      "step": 2630
+    },
+    {
+      "epoch": 1.9109663409337676,
+      "grad_norm": 0.5608997941017151,
+      "learning_rate": 0.0002,
+      "loss": 1.3716,
+      "step": 2640
+    },
+    {
+      "epoch": 1.918204849800941,
+      "grad_norm": 0.6454501748085022,
+      "learning_rate": 0.0002,
+      "loss": 1.5787,
+      "step": 2650
+    },
+    {
+      "epoch": 1.9254433586681143,
+      "grad_norm": 0.5458073616027832,
+      "learning_rate": 0.0002,
+      "loss": 1.3238,
+      "step": 2660
+    },
+    {
+      "epoch": 1.9326818675352877,
+      "grad_norm": 0.5328490734100342,
+      "learning_rate": 0.0002,
+      "loss": 1.3208,
+      "step": 2670
+    },
+    {
+      "epoch": 1.939920376402461,
+      "grad_norm": 0.6444696187973022,
+      "learning_rate": 0.0002,
+      "loss": 1.4971,
+      "step": 2680
+    },
+    {
+      "epoch": 1.9471588852696344,
+      "grad_norm": 0.7126023769378662,
+      "learning_rate": 0.0002,
+      "loss": 1.5387,
+      "step": 2690
+    },
+    {
+      "epoch": 1.9543973941368078,
+      "grad_norm": 0.5164045095443726,
+      "learning_rate": 0.0002,
+      "loss": 1.3637,
+      "step": 2700
+    },
+    {
+      "epoch": 1.9616359030039812,
+      "grad_norm": 0.5347061157226562,
+      "learning_rate": 0.0002,
+      "loss": 1.5303,
+      "step": 2710
+    },
+    {
+      "epoch": 1.9688744118711545,
+      "grad_norm": 0.5297950506210327,
+      "learning_rate": 0.0002,
+      "loss": 1.2815,
+      "step": 2720
+    },
+    {
+      "epoch": 1.976112920738328,
+      "grad_norm": 0.6537790298461914,
+      "learning_rate": 0.0002,
+      "loss": 1.3566,
+      "step": 2730
+    },
+    {
+      "epoch": 1.9833514296055013,
+      "grad_norm": 0.5536222457885742,
+      "learning_rate": 0.0002,
+      "loss": 1.332,
+      "step": 2740
+    },
+    {
+      "epoch": 1.9905899384726746,
+      "grad_norm": 0.4856105446815491,
+      "learning_rate": 0.0002,
+      "loss": 1.3333,
+      "step": 2750
+    },
+    {
+      "epoch": 1.997828447339848,
+      "grad_norm": 0.6642730832099915,
+      "learning_rate": 0.0002,
+      "loss": 1.3521,
+      "step": 2760
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.4366681575775146,
+      "eval_runtime": 27.3729,
+      "eval_samples_per_second": 15.928,
+      "eval_steps_per_second": 2.009,
+      "step": 2763
+    },
+    {
+      "epoch": 2.0050669562070214,
+      "grad_norm": 0.740253210067749,
+      "learning_rate": 0.0002,
+      "loss": 1.4322,
+      "step": 2770
+    },
+    {
+      "epoch": 2.0123054650741947,
+      "grad_norm": 0.5826276540756226,
+      "learning_rate": 0.0002,
+      "loss": 1.277,
+      "step": 2780
+    },
+    {
+      "epoch": 2.019543973941368,
+      "grad_norm": 0.607356071472168,
+      "learning_rate": 0.0002,
+      "loss": 1.2424,
+      "step": 2790
+    },
+    {
+      "epoch": 2.0267824828085415,
+      "grad_norm": 0.5918063521385193,
+      "learning_rate": 0.0002,
+      "loss": 1.2601,
+      "step": 2800
+    },
+    {
+      "epoch": 2.034020991675715,
+      "grad_norm": 0.5610089898109436,
+      "learning_rate": 0.0002,
+      "loss": 1.3715,
+      "step": 2810
+    },
+    {
+      "epoch": 2.041259500542888,
+      "grad_norm": 0.5869926810264587,
+      "learning_rate": 0.0002,
+      "loss": 1.2092,
+      "step": 2820
+    },
+    {
+      "epoch": 2.0484980094100615,
+      "grad_norm": 0.5753467679023743,
+      "learning_rate": 0.0002,
+      "loss": 1.1929,
+      "step": 2830
+    },
+    {
+      "epoch": 2.055736518277235,
+      "grad_norm": 0.7096508145332336,
+      "learning_rate": 0.0002,
+      "loss": 1.333,
+      "step": 2840
+    },
+    {
+      "epoch": 2.0629750271444083,
+      "grad_norm": 0.7653635144233704,
+      "learning_rate": 0.0002,
+      "loss": 1.1766,
+      "step": 2850
+    },
+    {
+      "epoch": 2.0702135360115816,
+      "grad_norm": 0.6202841997146606,
+      "learning_rate": 0.0002,
+      "loss": 1.2331,
+      "step": 2860
+    },
+    {
+      "epoch": 2.077452044878755,
+      "grad_norm": 0.6810227632522583,
+      "learning_rate": 0.0002,
+      "loss": 1.3298,
+      "step": 2870
+    },
+    {
+      "epoch": 2.0846905537459284,
+      "grad_norm": 0.7481493353843689,
+      "learning_rate": 0.0002,
+      "loss": 1.2505,
+      "step": 2880
+    },
+    {
+      "epoch": 2.0919290626131017,
+      "grad_norm": 0.7089637517929077,
+      "learning_rate": 0.0002,
+      "loss": 1.2484,
+      "step": 2890
+    },
+    {
+      "epoch": 2.099167571480275,
+      "grad_norm": 0.7472923398017883,
+      "learning_rate": 0.0002,
+      "loss": 1.3095,
+      "step": 2900
+    },
+    {
+      "epoch": 2.1064060803474485,
+      "grad_norm": 0.8135465979576111,
+      "learning_rate": 0.0002,
+      "loss": 1.304,
+      "step": 2910
+    },
+    {
+      "epoch": 2.113644589214622,
+      "grad_norm": 0.6097133159637451,
+      "learning_rate": 0.0002,
+      "loss": 1.273,
+      "step": 2920
+    },
+    {
+      "epoch": 2.120883098081795,
+      "grad_norm": 0.5970117449760437,
+      "learning_rate": 0.0002,
+      "loss": 1.3384,
+      "step": 2930
+    },
+    {
+      "epoch": 2.1281216069489686,
+      "grad_norm": 0.6169309616088867,
+      "learning_rate": 0.0002,
+      "loss": 1.3233,
+      "step": 2940
+    },
+    {
+      "epoch": 2.135360115816142,
+      "grad_norm": 0.9428738355636597,
+      "learning_rate": 0.0002,
+      "loss": 1.4246,
+      "step": 2950
+    },
+    {
+      "epoch": 2.1425986246833153,
+      "grad_norm": 0.5671679973602295,
+      "learning_rate": 0.0002,
+      "loss": 1.3527,
+      "step": 2960
+    },
+    {
+      "epoch": 2.1498371335504887,
+      "grad_norm": 0.7007262110710144,
+      "learning_rate": 0.0002,
+      "loss": 1.1375,
+      "step": 2970
+    },
+    {
+      "epoch": 2.157075642417662,
+      "grad_norm": 0.6294044256210327,
+      "learning_rate": 0.0002,
+      "loss": 1.2015,
+      "step": 2980
+    },
+    {
+      "epoch": 2.1643141512848354,
+      "grad_norm": 0.6105241775512695,
+      "learning_rate": 0.0002,
+      "loss": 1.2167,
+      "step": 2990
+    },
+    {
+      "epoch": 2.1715526601520088,
+      "grad_norm": 0.557124137878418,
+      "learning_rate": 0.0002,
+      "loss": 1.2065,
+      "step": 3000
+    },
+    {
+      "epoch": 2.178791169019182,
+      "grad_norm": 0.6250392198562622,
+      "learning_rate": 0.0002,
+      "loss": 1.2515,
+      "step": 3010
+    },
+    {
+      "epoch": 2.1860296778863555,
+      "grad_norm": 0.645218551158905,
+      "learning_rate": 0.0002,
+      "loss": 1.385,
+      "step": 3020
+    },
+    {
+      "epoch": 2.193268186753529,
+      "grad_norm": 0.9033605456352234,
+      "learning_rate": 0.0002,
+      "loss": 1.3928,
+      "step": 3030
+    },
+    {
+      "epoch": 2.2005066956207022,
+      "grad_norm": 0.5325747132301331,
+      "learning_rate": 0.0002,
+      "loss": 1.2458,
+      "step": 3040
+    },
+    {
+      "epoch": 2.2077452044878756,
+      "grad_norm": 0.6334700584411621,
+      "learning_rate": 0.0002,
+      "loss": 1.261,
+      "step": 3050
+    },
+    {
+      "epoch": 2.214983713355049,
+      "grad_norm": 0.5206325054168701,
+      "learning_rate": 0.0002,
+      "loss": 1.2385,
+      "step": 3060
+    },
+    {
+      "epoch": 2.2222222222222223,
+      "grad_norm": 0.5987200140953064,
+      "learning_rate": 0.0002,
+      "loss": 1.3103,
+      "step": 3070
+    },
+    {
+      "epoch": 2.2294607310893957,
+      "grad_norm": 0.5893264412879944,
+      "learning_rate": 0.0002,
+      "loss": 1.1756,
+      "step": 3080
+    },
+    {
+      "epoch": 2.236699239956569,
+      "grad_norm": 0.6869237422943115,
+      "learning_rate": 0.0002,
+      "loss": 1.235,
+      "step": 3090
+    },
+    {
+      "epoch": 2.2439377488237424,
+      "grad_norm": 0.5040048360824585,
+      "learning_rate": 0.0002,
+      "loss": 1.3285,
+      "step": 3100
+    },
+    {
+      "epoch": 2.251176257690916,
+      "grad_norm": 0.6660613417625427,
+      "learning_rate": 0.0002,
+      "loss": 1.3316,
+      "step": 3110
+    },
+    {
+      "epoch": 2.258414766558089,
+      "grad_norm": 0.5890918970108032,
+      "learning_rate": 0.0002,
+      "loss": 1.3108,
+      "step": 3120
+    },
+    {
+      "epoch": 2.2656532754252625,
+      "grad_norm": 0.6458896994590759,
+      "learning_rate": 0.0002,
+      "loss": 1.248,
+      "step": 3130
+    },
+    {
+      "epoch": 2.272891784292436,
+      "grad_norm": 0.6832690834999084,
+      "learning_rate": 0.0002,
+      "loss": 1.4151,
+      "step": 3140
+    },
+    {
+      "epoch": 2.2801302931596092,
+      "grad_norm": 0.833908200263977,
+      "learning_rate": 0.0002,
+      "loss": 1.4458,
+      "step": 3150
+    },
+    {
+      "epoch": 2.2873688020267826,
+      "grad_norm": 0.4596034586429596,
+      "learning_rate": 0.0002,
+      "loss": 1.2931,
+      "step": 3160
+    },
+    {
+      "epoch": 2.294607310893956,
+      "grad_norm": 0.9130966067314148,
+      "learning_rate": 0.0002,
+      "loss": 1.449,
+      "step": 3170
+    },
+    {
+      "epoch": 2.3018458197611293,
+      "grad_norm": 0.7143292427062988,
+      "learning_rate": 0.0002,
+      "loss": 1.3806,
+      "step": 3180
+    },
+    {
+      "epoch": 2.3090843286283027,
+      "grad_norm": 0.5388900637626648,
+      "learning_rate": 0.0002,
+      "loss": 1.2692,
+      "step": 3190
+    },
+    {
+      "epoch": 2.316322837495476,
+      "grad_norm": 0.5607513189315796,
+      "learning_rate": 0.0002,
+      "loss": 1.2402,
+      "step": 3200
+    },
+    {
+      "epoch": 2.3235613463626494,
+      "grad_norm": 0.6795142292976379,
+      "learning_rate": 0.0002,
+      "loss": 1.3874,
+      "step": 3210
+    },
+    {
+      "epoch": 2.330799855229823,
+      "grad_norm": 0.6561070680618286,
+      "learning_rate": 0.0002,
+      "loss": 1.3042,
+      "step": 3220
+    },
+    {
+      "epoch": 2.338038364096996,
+      "grad_norm": 0.8858118057250977,
+      "learning_rate": 0.0002,
+      "loss": 1.4636,
+      "step": 3230
+    },
+    {
+      "epoch": 2.3452768729641695,
+      "grad_norm": 0.6604151725769043,
+      "learning_rate": 0.0002,
+      "loss": 1.3214,
+      "step": 3240
+    },
+    {
+      "epoch": 2.352515381831343,
+      "grad_norm": 0.6755785346031189,
+      "learning_rate": 0.0002,
+      "loss": 1.4004,
+      "step": 3250
+    },
+    {
+      "epoch": 2.3597538906985163,
+      "grad_norm": 0.6981677412986755,
+      "learning_rate": 0.0002,
+      "loss": 1.2503,
+      "step": 3260
+    },
+    {
+      "epoch": 2.3669923995656896,
+      "grad_norm": 0.6338568329811096,
+      "learning_rate": 0.0002,
+      "loss": 1.3078,
+      "step": 3270
+    },
+    {
+      "epoch": 2.374230908432863,
+      "grad_norm": 0.5754265785217285,
+      "learning_rate": 0.0002,
+      "loss": 1.285,
+      "step": 3280
+    },
+    {
+      "epoch": 2.3814694173000364,
+      "grad_norm": 0.7533153295516968,
+      "learning_rate": 0.0002,
+      "loss": 1.2924,
+      "step": 3290
+    },
+    {
+      "epoch": 2.3887079261672097,
+      "grad_norm": 0.675065279006958,
+      "learning_rate": 0.0002,
+      "loss": 1.3711,
+      "step": 3300
+    },
+    {
+      "epoch": 2.395946435034383,
+      "grad_norm": 0.5686452984809875,
+      "learning_rate": 0.0002,
+      "loss": 1.3548,
+      "step": 3310
+    },
+    {
+      "epoch": 2.4031849439015565,
+      "grad_norm": 0.8129481673240662,
+      "learning_rate": 0.0002,
+      "loss": 1.1998,
+      "step": 3320
+    },
+    {
+      "epoch": 2.41042345276873,
+      "grad_norm": 0.6615934371948242,
+      "learning_rate": 0.0002,
+      "loss": 1.2584,
+      "step": 3330
+    },
+    {
+      "epoch": 2.417661961635903,
+      "grad_norm": 0.6678834557533264,
+      "learning_rate": 0.0002,
+      "loss": 1.3691,
+      "step": 3340
+    },
+    {
+      "epoch": 2.4249004705030766,
+      "grad_norm": 0.5581308007240295,
+      "learning_rate": 0.0002,
+      "loss": 1.2381,
+      "step": 3350
+    },
+    {
+      "epoch": 2.43213897937025,
+      "grad_norm": 0.6098920106887817,
+      "learning_rate": 0.0002,
+      "loss": 1.3853,
+      "step": 3360
+    },
+    {
+      "epoch": 2.4393774882374233,
+      "grad_norm": 0.8101736903190613,
+      "learning_rate": 0.0002,
+      "loss": 1.3692,
+      "step": 3370
+    },
+    {
+      "epoch": 2.4466159971045967,
+      "grad_norm": 0.6621488928794861,
+      "learning_rate": 0.0002,
+      "loss": 1.4418,
+      "step": 3380
+    },
+    {
+      "epoch": 2.45385450597177,
+      "grad_norm": 0.8693289160728455,
+      "learning_rate": 0.0002,
+      "loss": 1.4579,
+      "step": 3390
+    },
+    {
+      "epoch": 2.4610930148389434,
+      "grad_norm": 0.6724580526351929,
+      "learning_rate": 0.0002,
+      "loss": 1.3644,
+      "step": 3400
+    },
+    {
+      "epoch": 2.4683315237061167,
+      "grad_norm": 0.6776891946792603,
+      "learning_rate": 0.0002,
+      "loss": 1.2006,
+      "step": 3410
+    },
+    {
+      "epoch": 2.47557003257329,
+      "grad_norm": 0.7214453816413879,
+      "learning_rate": 0.0002,
+      "loss": 1.2937,
+      "step": 3420
+    },
+    {
+      "epoch": 2.4828085414404635,
+      "grad_norm": 0.8390451073646545,
+      "learning_rate": 0.0002,
+      "loss": 1.4051,
+      "step": 3430
+    },
+    {
+      "epoch": 2.490047050307637,
+      "grad_norm": 0.7130982279777527,
+      "learning_rate": 0.0002,
+      "loss": 1.25,
+      "step": 3440
+    },
+    {
+      "epoch": 2.49728555917481,
+      "grad_norm": 0.8873937129974365,
+      "learning_rate": 0.0002,
+      "loss": 1.2231,
+      "step": 3450
+    },
+    {
+      "epoch": 2.5045240680419836,
+      "grad_norm": 0.725185751914978,
+      "learning_rate": 0.0002,
+      "loss": 1.1429,
+      "step": 3460
+    },
+    {
+      "epoch": 2.511762576909157,
+      "grad_norm": 0.6120352149009705,
+      "learning_rate": 0.0002,
+      "loss": 1.2699,
+      "step": 3470
+    },
+    {
+      "epoch": 2.5190010857763303,
+      "grad_norm": 0.7713613510131836,
+      "learning_rate": 0.0002,
+      "loss": 1.2552,
+      "step": 3480
+    },
+    {
+      "epoch": 2.5262395946435037,
+      "grad_norm": 0.895309567451477,
+      "learning_rate": 0.0002,
+      "loss": 1.4648,
+      "step": 3490
+    },
+    {
+      "epoch": 2.533478103510677,
+      "grad_norm": 0.9631021022796631,
+      "learning_rate": 0.0002,
+      "loss": 1.3043,
+      "step": 3500
+    },
+    {
+      "epoch": 2.5407166123778504,
+      "grad_norm": 0.7475683093070984,
+      "learning_rate": 0.0002,
+      "loss": 1.3492,
+      "step": 3510
+    },
+    {
+      "epoch": 2.5479551212450238,
+      "grad_norm": 0.7271341681480408,
+      "learning_rate": 0.0002,
+      "loss": 1.3637,
+      "step": 3520
+    },
+    {
+      "epoch": 2.555193630112197,
+      "grad_norm": 0.6979510188102722,
+      "learning_rate": 0.0002,
+      "loss": 1.304,
+      "step": 3530
+    },
+    {
+      "epoch": 2.5624321389793705,
+      "grad_norm": 0.6504196524620056,
+      "learning_rate": 0.0002,
+      "loss": 1.2353,
+      "step": 3540
+    },
+    {
+      "epoch": 2.569670647846544,
+      "grad_norm": 0.7226675748825073,
+      "learning_rate": 0.0002,
+      "loss": 1.2699,
+      "step": 3550
+    },
+    {
+      "epoch": 2.5769091567137172,
+      "grad_norm": 0.6143222451210022,
+      "learning_rate": 0.0002,
+      "loss": 1.3002,
+      "step": 3560
+    },
+    {
+      "epoch": 2.5841476655808906,
+      "grad_norm": 0.7245154976844788,
+      "learning_rate": 0.0002,
+      "loss": 1.1585,
+      "step": 3570
+    },
+    {
+      "epoch": 2.591386174448064,
+      "grad_norm": 0.943540632724762,
+      "learning_rate": 0.0002,
+      "loss": 1.3651,
+      "step": 3580
+    },
+    {
+      "epoch": 2.5986246833152373,
+      "grad_norm": 0.7707241773605347,
+      "learning_rate": 0.0002,
+      "loss": 1.3034,
+      "step": 3590
+    },
+    {
+      "epoch": 2.6058631921824107,
+      "grad_norm": 0.6705001592636108,
+      "learning_rate": 0.0002,
+      "loss": 1.3063,
+      "step": 3600
+    },
+    {
+      "epoch": 2.613101701049584,
+      "grad_norm": 0.6360933780670166,
+      "learning_rate": 0.0002,
+      "loss": 1.2437,
+      "step": 3610
+    },
+    {
+      "epoch": 2.6203402099167574,
+      "grad_norm": 0.5846424698829651,
+      "learning_rate": 0.0002,
+      "loss": 1.1844,
+      "step": 3620
+    },
+    {
+      "epoch": 2.6275787187839303,
+      "grad_norm": 0.5958625674247742,
+      "learning_rate": 0.0002,
+      "loss": 1.3674,
+      "step": 3630
+    },
+    {
+      "epoch": 2.6348172276511037,
+      "grad_norm": 0.6819243431091309,
+      "learning_rate": 0.0002,
+      "loss": 1.3599,
+      "step": 3640
+    },
+    {
+      "epoch": 2.642055736518277,
+      "grad_norm": 0.7033445835113525,
+      "learning_rate": 0.0002,
+      "loss": 1.3884,
+      "step": 3650
+    },
+    {
+      "epoch": 2.6492942453854504,
+      "grad_norm": 0.6134849786758423,
+      "learning_rate": 0.0002,
+      "loss": 1.3392,
+      "step": 3660
+    },
+    {
+      "epoch": 2.656532754252624,
+      "grad_norm": 0.658009946346283,
+      "learning_rate": 0.0002,
+      "loss": 1.2661,
+      "step": 3670
+    },
+    {
+      "epoch": 2.663771263119797,
+      "grad_norm": 0.6280999779701233,
+      "learning_rate": 0.0002,
+      "loss": 1.3987,
+      "step": 3680
+    },
+    {
+      "epoch": 2.6710097719869705,
+      "grad_norm": 0.5536085963249207,
+      "learning_rate": 0.0002,
+      "loss": 1.2995,
+      "step": 3690
+    },
+    {
+      "epoch": 2.678248280854144,
+      "grad_norm": 0.8603981733322144,
+      "learning_rate": 0.0002,
+      "loss": 1.2044,
+      "step": 3700
+    },
+    {
+      "epoch": 2.6854867897213173,
+      "grad_norm": 0.5509994626045227,
+      "learning_rate": 0.0002,
+      "loss": 1.3879,
+      "step": 3710
+    },
+    {
+      "epoch": 2.6927252985884906,
+      "grad_norm": 0.9093621969223022,
+      "learning_rate": 0.0002,
+      "loss": 1.3253,
+      "step": 3720
+    },
+    {
+      "epoch": 2.699963807455664,
+      "grad_norm": 0.7525952458381653,
+      "learning_rate": 0.0002,
+      "loss": 1.2668,
+      "step": 3730
+    },
+    {
+      "epoch": 2.7072023163228374,
+      "grad_norm": 0.6737023591995239,
+      "learning_rate": 0.0002,
+      "loss": 1.248,
+      "step": 3740
+    },
+    {
+      "epoch": 2.7144408251900107,
+      "grad_norm": 0.8656924962997437,
+      "learning_rate": 0.0002,
+      "loss": 1.2981,
+      "step": 3750
+    },
+    {
+      "epoch": 2.721679334057184,
+      "grad_norm": 0.7494133114814758,
+      "learning_rate": 0.0002,
+      "loss": 1.2342,
+      "step": 3760
+    },
+    {
+      "epoch": 2.7289178429243575,
+      "grad_norm": 0.5725520849227905,
+      "learning_rate": 0.0002,
+      "loss": 1.2417,
+      "step": 3770
+    },
+    {
+      "epoch": 2.736156351791531,
+      "grad_norm": 0.836412787437439,
+      "learning_rate": 0.0002,
+      "loss": 1.28,
+      "step": 3780
+    },
+    {
+      "epoch": 2.743394860658704,
+      "grad_norm": 0.6893242597579956,
+      "learning_rate": 0.0002,
+      "loss": 1.3784,
+      "step": 3790
+    },
+    {
+      "epoch": 2.7506333695258776,
+      "grad_norm": 0.6696223020553589,
+      "learning_rate": 0.0002,
+      "loss": 1.2929,
+      "step": 3800
+    },
+    {
+      "epoch": 2.757871878393051,
+      "grad_norm": 0.6483015418052673,
+      "learning_rate": 0.0002,
+      "loss": 1.2449,
+      "step": 3810
+    },
+    {
+      "epoch": 2.7651103872602243,
+      "grad_norm": 0.8084456920623779,
+      "learning_rate": 0.0002,
+      "loss": 1.3282,
+      "step": 3820
+    },
+    {
+      "epoch": 2.7723488961273977,
+      "grad_norm": 0.6601949334144592,
+      "learning_rate": 0.0002,
+      "loss": 1.3694,
+      "step": 3830
+    },
+    {
+      "epoch": 2.779587404994571,
+      "grad_norm": 0.6905533671379089,
+      "learning_rate": 0.0002,
+      "loss": 1.3568,
+      "step": 3840
+    },
+    {
+      "epoch": 2.7868259138617444,
+      "grad_norm": 0.619318425655365,
+      "learning_rate": 0.0002,
+      "loss": 1.3854,
+      "step": 3850
+    },
+    {
+      "epoch": 2.7940644227289178,
+      "grad_norm": 0.5994023084640503,
+      "learning_rate": 0.0002,
+      "loss": 1.2551,
+      "step": 3860
+    },
+    {
+      "epoch": 2.801302931596091,
+      "grad_norm": 0.5627168416976929,
+      "learning_rate": 0.0002,
+      "loss": 1.2022,
+      "step": 3870
+    },
+    {
+      "epoch": 2.8085414404632645,
+      "grad_norm": 0.6001605987548828,
+      "learning_rate": 0.0002,
+      "loss": 1.3921,
+      "step": 3880
+    },
+    {
+      "epoch": 2.815779949330438,
+      "grad_norm": 0.6022412776947021,
+      "learning_rate": 0.0002,
+      "loss": 1.3026,
+      "step": 3890
+    },
+    {
+      "epoch": 2.823018458197611,
+      "grad_norm": 0.6832426190376282,
+      "learning_rate": 0.0002,
+      "loss": 1.2765,
+      "step": 3900
+    },
+    {
+      "epoch": 2.8302569670647846,
+      "grad_norm": 0.5936811566352844,
+      "learning_rate": 0.0002,
+      "loss": 1.1363,
+      "step": 3910
+    },
+    {
+      "epoch": 2.837495475931958,
+      "grad_norm": 0.6960572600364685,
+      "learning_rate": 0.0002,
+      "loss": 1.1707,
+      "step": 3920
+    },
+    {
+      "epoch": 2.8447339847991313,
+      "grad_norm": 0.5913406610488892,
+      "learning_rate": 0.0002,
+      "loss": 1.4063,
+      "step": 3930
+    },
+    {
+      "epoch": 2.8519724936663047,
+      "grad_norm": 0.678154706954956,
+      "learning_rate": 0.0002,
+      "loss": 1.3245,
+      "step": 3940
+    },
+    {
+      "epoch": 2.859211002533478,
+      "grad_norm": 0.7898936867713928,
+      "learning_rate": 0.0002,
+      "loss": 1.366,
+      "step": 3950
+    },
+    {
+      "epoch": 2.8664495114006514,
+      "grad_norm": 0.9234195351600647,
+      "learning_rate": 0.0002,
+      "loss": 1.3948,
+      "step": 3960
+    },
+    {
+      "epoch": 2.8736880202678248,
+      "grad_norm": 0.5960825085639954,
+      "learning_rate": 0.0002,
+      "loss": 1.2773,
+      "step": 3970
+    },
+    {
+      "epoch": 2.880926529134998,
+      "grad_norm": 0.677118182182312,
+      "learning_rate": 0.0002,
+      "loss": 1.3127,
+      "step": 3980
+    },
+    {
+      "epoch": 2.8881650380021715,
+      "grad_norm": 0.6505142450332642,
+      "learning_rate": 0.0002,
+      "loss": 1.2652,
+      "step": 3990
+    },
+    {
+      "epoch": 2.895403546869345,
+      "grad_norm": 0.550826907157898,
+      "learning_rate": 0.0002,
+      "loss": 1.2078,
+      "step": 4000
+    },
+    {
+      "epoch": 2.9026420557365182,
+      "grad_norm": 0.6209215521812439,
+      "learning_rate": 0.0002,
+      "loss": 1.1811,
+      "step": 4010
+    },
+    {
+      "epoch": 2.9098805646036916,
+      "grad_norm": 0.6549018025398254,
+      "learning_rate": 0.0002,
+      "loss": 1.4001,
+      "step": 4020
+    },
+    {
+      "epoch": 2.917119073470865,
+      "grad_norm": 0.570682168006897,
+      "learning_rate": 0.0002,
+      "loss": 1.2285,
+      "step": 4030
+    },
+    {
+      "epoch": 2.9243575823380383,
+      "grad_norm": 1.1807632446289062,
+      "learning_rate": 0.0002,
+      "loss": 1.0832,
+      "step": 4040
+    },
+    {
+      "epoch": 2.9315960912052117,
+      "grad_norm": 0.7058857679367065,
+      "learning_rate": 0.0002,
+      "loss": 1.2693,
+      "step": 4050
+    },
+    {
+      "epoch": 2.938834600072385,
+      "grad_norm": 0.5542812943458557,
+      "learning_rate": 0.0002,
+      "loss": 1.2905,
+      "step": 4060
+    },
+    {
+      "epoch": 2.9460731089395584,
+      "grad_norm": 0.63167804479599,
+      "learning_rate": 0.0002,
+      "loss": 1.33,
+      "step": 4070
+    },
+    {
+      "epoch": 2.953311617806732,
+      "grad_norm": 0.5702962279319763,
+      "learning_rate": 0.0002,
+      "loss": 1.3075,
+      "step": 4080
+    },
+    {
+      "epoch": 2.960550126673905,
+      "grad_norm": 0.620944082736969,
+      "learning_rate": 0.0002,
+      "loss": 1.2007,
+      "step": 4090
+    },
+    {
+      "epoch": 2.9677886355410785,
+      "grad_norm": 0.5866289734840393,
+      "learning_rate": 0.0002,
+      "loss": 1.2864,
+      "step": 4100
+    },
+    {
+      "epoch": 2.975027144408252,
+      "grad_norm": 0.560170590877533,
+      "learning_rate": 0.0002,
+      "loss": 1.3293,
+      "step": 4110
+    },
+    {
+      "epoch": 2.9822656532754253,
+      "grad_norm": 0.675082802772522,
+      "learning_rate": 0.0002,
+      "loss": 1.2071,
+      "step": 4120
+    },
+    {
+      "epoch": 2.9895041621425986,
+      "grad_norm": 0.62708580493927,
+      "learning_rate": 0.0002,
+      "loss": 1.2981,
+      "step": 4130
+    },
+    {
+      "epoch": 2.996742671009772,
+      "grad_norm": 0.7893929481506348,
+      "learning_rate": 0.0002,
+      "loss": 1.2758,
+      "step": 4140
+    },
+    {
+      "epoch": 2.9996380745566413,
+      "eval_loss": 1.4217946529388428,
+      "eval_runtime": 27.1596,
+      "eval_samples_per_second": 16.053,
+      "eval_steps_per_second": 2.025,
+      "step": 4144
+    },
+    {
+      "epoch": 3.0039811798769454,
+      "grad_norm": 0.7043836116790771,
+      "learning_rate": 0.0002,
+      "loss": 1.2152,
+      "step": 4150
+    },
+    {
+      "epoch": 3.0112196887441187,
+      "grad_norm": 0.6806283593177795,
+      "learning_rate": 0.0002,
+      "loss": 1.1664,
+      "step": 4160
+    },
+    {
+      "epoch": 3.018458197611292,
+      "grad_norm": 0.7684550285339355,
+      "learning_rate": 0.0002,
+      "loss": 1.292,
+      "step": 4170
+    },
+    {
+      "epoch": 3.0256967064784654,
+      "grad_norm": 0.7895237803459167,
+      "learning_rate": 0.0002,
+      "loss": 1.3467,
+      "step": 4180
+    },
+    {
+      "epoch": 3.032935215345639,
+      "grad_norm": 0.7464531064033508,
+      "learning_rate": 0.0002,
+      "loss": 1.1324,
+      "step": 4190
+    },
+    {
+      "epoch": 3.040173724212812,
+      "grad_norm": 0.9358500838279724,
+      "learning_rate": 0.0002,
+      "loss": 1.1614,
+      "step": 4200
+    },
+    {
+      "epoch": 3.0474122330799855,
+      "grad_norm": 1.1066628694534302,
+      "learning_rate": 0.0002,
+      "loss": 1.1834,
+      "step": 4210
+    },
+    {
+      "epoch": 3.054650741947159,
+      "grad_norm": 0.6663267612457275,
+      "learning_rate": 0.0002,
+      "loss": 1.1557,
+      "step": 4220
+    },
+    {
+      "epoch": 3.0618892508143323,
+      "grad_norm": 0.6669464707374573,
+      "learning_rate": 0.0002,
+      "loss": 1.1707,
+      "step": 4230
+    },
+    {
+      "epoch": 3.0691277596815056,
+      "grad_norm": 0.7052164077758789,
+      "learning_rate": 0.0002,
+      "loss": 1.1841,
+      "step": 4240
+    },
+    {
+      "epoch": 3.076366268548679,
+      "grad_norm": 0.6118432879447937,
+      "learning_rate": 0.0002,
+      "loss": 1.2913,
+      "step": 4250
+    },
+    {
+      "epoch": 3.0836047774158524,
+      "grad_norm": 0.6915903687477112,
+      "learning_rate": 0.0002,
+      "loss": 1.1526,
+      "step": 4260
+    },
+    {
+      "epoch": 3.0908432862830257,
+      "grad_norm": 0.7441644668579102,
+      "learning_rate": 0.0002,
+      "loss": 1.1348,
+      "step": 4270
+    },
+    {
+      "epoch": 3.098081795150199,
+      "grad_norm": 0.823850691318512,
+      "learning_rate": 0.0002,
+      "loss": 1.1672,
+      "step": 4280
+    },
+    {
+      "epoch": 3.1053203040173725,
+      "grad_norm": 0.9677883386611938,
+      "learning_rate": 0.0002,
+      "loss": 1.2655,
+      "step": 4290
+    },
+    {
+      "epoch": 3.112558812884546,
+      "grad_norm": 0.7002579569816589,
+      "learning_rate": 0.0002,
+      "loss": 1.1794,
+      "step": 4300
+    },
+    {
+      "epoch": 3.119797321751719,
+      "grad_norm": 0.778789758682251,
+      "learning_rate": 0.0002,
+      "loss": 1.135,
+      "step": 4310
+    },
+    {
+      "epoch": 3.1270358306188926,
+      "grad_norm": 0.7236007452011108,
+      "learning_rate": 0.0002,
+      "loss": 1.0818,
+      "step": 4320
+    },
+    {
+      "epoch": 3.134274339486066,
+      "grad_norm": 0.8809133768081665,
+      "learning_rate": 0.0002,
+      "loss": 1.1803,
+      "step": 4330
+    },
+    {
+      "epoch": 3.1415128483532393,
+      "grad_norm": 0.7924913167953491,
+      "learning_rate": 0.0002,
+      "loss": 1.2571,
+      "step": 4340
+    },
+    {
+      "epoch": 3.1487513572204127,
+      "grad_norm": 0.7437422275543213,
+      "learning_rate": 0.0002,
+      "loss": 1.1413,
+      "step": 4350
+    },
+    {
+      "epoch": 3.155989866087586,
+      "grad_norm": 0.6428450345993042,
+      "learning_rate": 0.0002,
+      "loss": 1.2088,
+      "step": 4360
+    },
+    {
+      "epoch": 3.1632283749547594,
+      "grad_norm": 0.7922873497009277,
+      "learning_rate": 0.0002,
+      "loss": 1.3032,
+      "step": 4370
+    },
+    {
+      "epoch": 3.1704668838219328,
+      "grad_norm": 0.5252506732940674,
+      "learning_rate": 0.0002,
+      "loss": 1.216,
+      "step": 4380
+    },
+    {
+      "epoch": 3.177705392689106,
+      "grad_norm": 0.8570457696914673,
+      "learning_rate": 0.0002,
+      "loss": 1.1297,
+      "step": 4390
+    },
+    {
+      "epoch": 3.1849439015562795,
+      "grad_norm": 0.7218987345695496,
+      "learning_rate": 0.0002,
+      "loss": 1.0994,
+      "step": 4400
+    },
+    {
+      "epoch": 3.192182410423453,
+      "grad_norm": 0.6921393275260925,
+      "learning_rate": 0.0002,
+      "loss": 1.2891,
+      "step": 4410
+    },
+    {
+      "epoch": 3.199420919290626,
+      "grad_norm": 0.7386137843132019,
+      "learning_rate": 0.0002,
+      "loss": 1.2668,
+      "step": 4420
+    },
+    {
+      "epoch": 3.2066594281577996,
+      "grad_norm": 0.6227759122848511,
+      "learning_rate": 0.0002,
+      "loss": 1.1654,
+      "step": 4430
+    },
+    {
+      "epoch": 3.213897937024973,
+      "grad_norm": 0.7180278897285461,
+      "learning_rate": 0.0002,
+      "loss": 1.1752,
+      "step": 4440
+    },
+    {
+      "epoch": 3.2211364458921463,
+      "grad_norm": 0.745830774307251,
+      "learning_rate": 0.0002,
+      "loss": 1.1757,
+      "step": 4450
+    },
+    {
+      "epoch": 3.2283749547593197,
+      "grad_norm": 0.6766072511672974,
+      "learning_rate": 0.0002,
+      "loss": 1.234,
+      "step": 4460
+    },
+    {
+      "epoch": 3.235613463626493,
+      "grad_norm": 0.8325067162513733,
+      "learning_rate": 0.0002,
+      "loss": 1.1999,
+      "step": 4470
+    },
+    {
+      "epoch": 3.2428519724936664,
+      "grad_norm": 0.7148305177688599,
+      "learning_rate": 0.0002,
+      "loss": 1.1606,
+      "step": 4480
+    },
+    {
+      "epoch": 3.25009048136084,
+      "grad_norm": 0.7752676010131836,
+      "learning_rate": 0.0002,
+      "loss": 1.1383,
+      "step": 4490
+    },
+    {
+      "epoch": 3.257328990228013,
+      "grad_norm": 0.6776860952377319,
+      "learning_rate": 0.0002,
+      "loss": 1.3006,
+      "step": 4500
+    },
+    {
+      "epoch": 3.2645674990951865,
+      "grad_norm": 0.704359769821167,
+      "learning_rate": 0.0002,
+      "loss": 1.0796,
+      "step": 4510
+    },
+    {
+      "epoch": 3.27180600796236,
+      "grad_norm": 0.6880282163619995,
+      "learning_rate": 0.0002,
+      "loss": 1.2496,
+      "step": 4520
+    },
+    {
+      "epoch": 3.2790445168295332,
+      "grad_norm": 0.8179270029067993,
+      "learning_rate": 0.0002,
+      "loss": 1.0947,
+      "step": 4530
+    },
+    {
+      "epoch": 3.2862830256967066,
+      "grad_norm": 0.6718448996543884,
+      "learning_rate": 0.0002,
+      "loss": 1.1909,
+      "step": 4540
+    },
+    {
+      "epoch": 3.29352153456388,
+      "grad_norm": 0.8300657868385315,
+      "learning_rate": 0.0002,
+      "loss": 1.2708,
+      "step": 4550
+    },
+    {
+      "epoch": 3.3007600434310533,
+      "grad_norm": 0.6433690786361694,
+      "learning_rate": 0.0002,
+      "loss": 1.2594,
+      "step": 4560
+    },
+    {
+      "epoch": 3.3079985522982267,
+      "grad_norm": 0.690262496471405,
+      "learning_rate": 0.0002,
+      "loss": 1.2479,
+      "step": 4570
+    },
+    {
+      "epoch": 3.3152370611654,
+      "grad_norm": 0.7022852301597595,
+      "learning_rate": 0.0002,
+      "loss": 1.1342,
+      "step": 4580
+    },
+    {
+      "epoch": 3.3224755700325734,
+      "grad_norm": 0.6438387632369995,
+      "learning_rate": 0.0002,
+      "loss": 1.0844,
+      "step": 4590
+    },
+    {
+      "epoch": 3.329714078899747,
+      "grad_norm": 0.6866899132728577,
+      "learning_rate": 0.0002,
+      "loss": 1.17,
+      "step": 4600
+    },
+    {
+      "epoch": 3.33695258776692,
+      "grad_norm": 0.8233968019485474,
+      "learning_rate": 0.0002,
+      "loss": 1.1289,
+      "step": 4610
+    },
+    {
+      "epoch": 3.3441910966340935,
+      "grad_norm": 0.7251574993133545,
+      "learning_rate": 0.0002,
+      "loss": 1.1855,
+      "step": 4620
+    },
+    {
+      "epoch": 3.351429605501267,
+      "grad_norm": 0.7855110168457031,
+      "learning_rate": 0.0002,
+      "loss": 1.3403,
+      "step": 4630
+    },
+    {
+      "epoch": 3.3586681143684403,
+      "grad_norm": 0.8487356305122375,
+      "learning_rate": 0.0002,
+      "loss": 1.2922,
+      "step": 4640
+    },
+    {
+      "epoch": 3.3659066232356136,
+      "grad_norm": 0.6429011225700378,
+      "learning_rate": 0.0002,
+      "loss": 1.2462,
+      "step": 4650
+    },
+    {
+      "epoch": 3.373145132102787,
+      "grad_norm": 0.7095270156860352,
+      "learning_rate": 0.0002,
+      "loss": 1.129,
+      "step": 4660
+    },
+    {
+      "epoch": 3.3803836409699604,
+      "grad_norm": 0.6792303323745728,
+      "learning_rate": 0.0002,
+      "loss": 1.262,
+      "step": 4670
+    },
+    {
+      "epoch": 3.3876221498371337,
+      "grad_norm": 0.6784825921058655,
+      "learning_rate": 0.0002,
+      "loss": 1.256,
+      "step": 4680
+    },
+    {
+      "epoch": 3.394860658704307,
+      "grad_norm": 0.6362888216972351,
+      "learning_rate": 0.0002,
+      "loss": 1.0838,
+      "step": 4690
+    },
+    {
+      "epoch": 3.4020991675714805,
+      "grad_norm": 0.7794778943061829,
+      "learning_rate": 0.0002,
+      "loss": 1.2165,
+      "step": 4700
+    },
+    {
+      "epoch": 3.409337676438654,
+      "grad_norm": 0.7287485003471375,
+      "learning_rate": 0.0002,
+      "loss": 1.0644,
+      "step": 4710
+    },
+    {
+      "epoch": 3.416576185305827,
+      "grad_norm": 0.6481451392173767,
+      "learning_rate": 0.0002,
+      "loss": 1.2925,
+      "step": 4720
+    },
+    {
+      "epoch": 3.4238146941730006,
+      "grad_norm": 0.9200371503829956,
+      "learning_rate": 0.0002,
+      "loss": 1.2121,
+      "step": 4730
+    },
+    {
+      "epoch": 3.431053203040174,
+      "grad_norm": 1.074180245399475,
+      "learning_rate": 0.0002,
+      "loss": 1.072,
+      "step": 4740
+    },
+    {
+      "epoch": 3.438291711907347,
+      "grad_norm": 0.6722986698150635,
+      "learning_rate": 0.0002,
+      "loss": 1.0421,
+      "step": 4750
+    },
+    {
+      "epoch": 3.44553022077452,
+      "grad_norm": 0.7945933938026428,
+      "learning_rate": 0.0002,
+      "loss": 1.2258,
+      "step": 4760
+    },
+    {
+      "epoch": 3.4527687296416936,
+      "grad_norm": 0.7624640464782715,
+      "learning_rate": 0.0002,
+      "loss": 1.0927,
+      "step": 4770
+    },
+    {
+      "epoch": 3.460007238508867,
+      "grad_norm": 0.7763656377792358,
+      "learning_rate": 0.0002,
+      "loss": 1.2428,
+      "step": 4780
+    },
+    {
+      "epoch": 3.4672457473760403,
+      "grad_norm": 0.7736947536468506,
+      "learning_rate": 0.0002,
+      "loss": 1.2584,
+      "step": 4790
+    },
+    {
+      "epoch": 3.4744842562432137,
+      "grad_norm": 0.8450354933738708,
+      "learning_rate": 0.0002,
+      "loss": 1.1953,
+      "step": 4800
+    },
+    {
+      "epoch": 3.481722765110387,
+      "grad_norm": 0.6480133533477783,
+      "learning_rate": 0.0002,
+      "loss": 1.1362,
+      "step": 4810
+    },
+    {
+      "epoch": 3.4889612739775604,
+      "grad_norm": 0.8437445759773254,
+      "learning_rate": 0.0002,
+      "loss": 1.1882,
+      "step": 4820
+    },
+    {
+      "epoch": 3.4961997828447338,
+      "grad_norm": 0.7781730890274048,
+      "learning_rate": 0.0002,
+      "loss": 1.1519,
+      "step": 4830
+    },
+    {
+      "epoch": 3.503438291711907,
+      "grad_norm": 0.8523228168487549,
+      "learning_rate": 0.0002,
+      "loss": 1.1836,
+      "step": 4840
+    },
+    {
+      "epoch": 3.5106768005790805,
+      "grad_norm": 0.6236732006072998,
+      "learning_rate": 0.0002,
+      "loss": 1.1672,
+      "step": 4850
+    },
+    {
+      "epoch": 3.517915309446254,
+      "grad_norm": 0.7500787377357483,
+      "learning_rate": 0.0002,
+      "loss": 1.1926,
+      "step": 4860
+    },
+    {
+      "epoch": 3.5251538183134272,
+      "grad_norm": 0.7665374875068665,
+      "learning_rate": 0.0002,
+      "loss": 1.1998,
+      "step": 4870
+    },
+    {
+      "epoch": 3.5323923271806006,
+      "grad_norm": 0.787857711315155,
+      "learning_rate": 0.0002,
+      "loss": 1.1551,
+      "step": 4880
+    },
+    {
+      "epoch": 3.539630836047774,
+      "grad_norm": 0.970595121383667,
+      "learning_rate": 0.0002,
+      "loss": 1.2758,
+      "step": 4890
+    },
+    {
+      "epoch": 3.5468693449149473,
+      "grad_norm": 0.6409347057342529,
+      "learning_rate": 0.0002,
+      "loss": 1.1274,
+      "step": 4900
+    },
+    {
+      "epoch": 3.5541078537821207,
+      "grad_norm": 0.888551652431488,
+      "learning_rate": 0.0002,
+      "loss": 1.1596,
+      "step": 4910
+    },
+    {
+      "epoch": 3.561346362649294,
+      "grad_norm": 1.0808377265930176,
+      "learning_rate": 0.0002,
+      "loss": 1.1644,
+      "step": 4920
+    },
+    {
+      "epoch": 3.5685848715164674,
+      "grad_norm": 0.7501053214073181,
+      "learning_rate": 0.0002,
+      "loss": 1.2564,
+      "step": 4930
+    },
+    {
+      "epoch": 3.575823380383641,
+      "grad_norm": 0.7375240325927734,
+      "learning_rate": 0.0002,
+      "loss": 1.2351,
+      "step": 4940
+    },
+    {
+      "epoch": 3.583061889250814,
+      "grad_norm": 0.7075039744377136,
+      "learning_rate": 0.0002,
+      "loss": 1.3568,
+      "step": 4950
+    },
+    {
+      "epoch": 3.5903003981179875,
+      "grad_norm": 0.939337432384491,
+      "learning_rate": 0.0002,
+      "loss": 1.3355,
+      "step": 4960
+    },
+    {
+      "epoch": 3.597538906985161,
+      "grad_norm": 0.6717396974563599,
+      "learning_rate": 0.0002,
+      "loss": 1.1722,
+      "step": 4970
+    },
+    {
+      "epoch": 3.6047774158523342,
+      "grad_norm": 0.7141643762588501,
+      "learning_rate": 0.0002,
+      "loss": 1.1186,
+      "step": 4980
+    },
+    {
+      "epoch": 3.6120159247195076,
+      "grad_norm": 0.7109216451644897,
+      "learning_rate": 0.0002,
+      "loss": 1.1011,
+      "step": 4990
+    },
+    {
+      "epoch": 3.619254433586681,
+      "grad_norm": 0.7020776867866516,
+      "learning_rate": 0.0002,
+      "loss": 1.2178,
+      "step": 5000
+    },
+    {
+      "epoch": 3.6264929424538543,
+      "grad_norm": 0.7158873677253723,
+      "learning_rate": 0.0002,
+      "loss": 1.1939,
+      "step": 5010
+    },
+    {
+      "epoch": 3.6337314513210277,
+      "grad_norm": 0.7062035202980042,
+      "learning_rate": 0.0002,
+      "loss": 1.2624,
+      "step": 5020
+    },
+    {
+      "epoch": 3.640969960188201,
+      "grad_norm": 0.7081155776977539,
+      "learning_rate": 0.0002,
+      "loss": 1.0224,
+      "step": 5030
+    },
+    {
+      "epoch": 3.6482084690553744,
+      "grad_norm": 1.2210607528686523,
+      "learning_rate": 0.0002,
+      "loss": 1.2195,
+      "step": 5040
+    },
+    {
+      "epoch": 3.655446977922548,
+      "grad_norm": 0.6650236248970032,
+      "learning_rate": 0.0002,
+      "loss": 1.2596,
+      "step": 5050
+    },
+    {
+      "epoch": 3.662685486789721,
+      "grad_norm": 0.6884829998016357,
+      "learning_rate": 0.0002,
+      "loss": 1.1072,
+      "step": 5060
+    },
+    {
+      "epoch": 3.6699239956568945,
+      "grad_norm": 0.7317819595336914,
+      "learning_rate": 0.0002,
+      "loss": 1.2292,
+      "step": 5070
+    },
+    {
+      "epoch": 3.677162504524068,
+      "grad_norm": 0.7406691908836365,
+      "learning_rate": 0.0002,
+      "loss": 1.1917,
+      "step": 5080
+    },
+    {
+      "epoch": 3.6844010133912413,
+      "grad_norm": 0.9009454250335693,
+      "learning_rate": 0.0002,
+      "loss": 1.2949,
+      "step": 5090
+    },
+    {
+      "epoch": 3.6916395222584146,
+      "grad_norm": 0.8189385533332825,
+      "learning_rate": 0.0002,
+      "loss": 1.1528,
+      "step": 5100
+    },
+    {
+      "epoch": 3.698878031125588,
+      "grad_norm": 1.0793628692626953,
+      "learning_rate": 0.0002,
+      "loss": 1.3408,
+      "step": 5110
+    },
+    {
+      "epoch": 3.7061165399927614,
+      "grad_norm": 0.8593027591705322,
+      "learning_rate": 0.0002,
+      "loss": 1.2417,
+      "step": 5120
+    },
+    {
+      "epoch": 3.7133550488599347,
+      "grad_norm": 0.8481812477111816,
+      "learning_rate": 0.0002,
+      "loss": 1.2141,
+      "step": 5130
+    },
+    {
+      "epoch": 3.720593557727108,
+      "grad_norm": 0.6527451276779175,
+      "learning_rate": 0.0002,
+      "loss": 1.125,
+      "step": 5140
+    },
+    {
+      "epoch": 3.7278320665942815,
+      "grad_norm": 0.9220114350318909,
+      "learning_rate": 0.0002,
+      "loss": 1.1584,
+      "step": 5150
+    },
+    {
+      "epoch": 3.735070575461455,
+      "grad_norm": 1.0842019319534302,
+      "learning_rate": 0.0002,
+      "loss": 1.2267,
+      "step": 5160
+    },
+    {
+      "epoch": 3.742309084328628,
+      "grad_norm": 0.965453565120697,
+      "learning_rate": 0.0002,
+      "loss": 1.3083,
+      "step": 5170
+    },
+    {
+      "epoch": 3.7495475931958016,
+      "grad_norm": 0.9903319478034973,
+      "learning_rate": 0.0002,
+      "loss": 1.1772,
+      "step": 5180
+    },
+    {
+      "epoch": 3.756786102062975,
+      "grad_norm": 0.7434818148612976,
+      "learning_rate": 0.0002,
+      "loss": 1.2515,
+      "step": 5190
+    },
+    {
+      "epoch": 3.7640246109301483,
+      "grad_norm": 0.6717280745506287,
+      "learning_rate": 0.0002,
+      "loss": 1.2631,
+      "step": 5200
+    },
+    {
+      "epoch": 3.7712631197973217,
+      "grad_norm": 0.7754665613174438,
+      "learning_rate": 0.0002,
+      "loss": 1.2012,
+      "step": 5210
+    },
+    {
+      "epoch": 3.778501628664495,
+      "grad_norm": 1.028374433517456,
+      "learning_rate": 0.0002,
+      "loss": 1.305,
+      "step": 5220
+    },
+    {
+      "epoch": 3.7857401375316684,
+      "grad_norm": 0.6026996374130249,
+      "learning_rate": 0.0002,
+      "loss": 1.1866,
+      "step": 5230
+    },
+    {
+      "epoch": 3.7929786463988417,
+      "grad_norm": 0.6978490948677063,
+      "learning_rate": 0.0002,
+      "loss": 1.1901,
+      "step": 5240
+    },
+    {
+      "epoch": 3.800217155266015,
+      "grad_norm": 0.7303446531295776,
+      "learning_rate": 0.0002,
+      "loss": 1.2576,
+      "step": 5250
+    },
+    {
+      "epoch": 3.8074556641331885,
+      "grad_norm": 1.0734210014343262,
+      "learning_rate": 0.0002,
+      "loss": 1.3173,
+      "step": 5260
+    },
+    {
+      "epoch": 3.814694173000362,
+      "grad_norm": 0.6383201479911804,
+      "learning_rate": 0.0002,
+      "loss": 1.1137,
+      "step": 5270
+    },
+    {
+      "epoch": 3.821932681867535,
+      "grad_norm": 0.7742630243301392,
+      "learning_rate": 0.0002,
+      "loss": 1.0904,
+      "step": 5280
+    },
+    {
+      "epoch": 3.8291711907347086,
+      "grad_norm": 0.8477074503898621,
+      "learning_rate": 0.0002,
+      "loss": 1.2232,
+      "step": 5290
+    },
+    {
+      "epoch": 3.836409699601882,
+      "grad_norm": 0.6675317883491516,
+      "learning_rate": 0.0002,
+      "loss": 1.2047,
+      "step": 5300
+    },
+    {
+      "epoch": 3.8436482084690553,
+      "grad_norm": 0.7515445351600647,
+      "learning_rate": 0.0002,
+      "loss": 1.2275,
+      "step": 5310
+    },
+    {
+      "epoch": 3.8508867173362287,
+      "grad_norm": 1.1441220045089722,
+      "learning_rate": 0.0002,
+      "loss": 1.2569,
+      "step": 5320
+    },
+    {
+      "epoch": 3.858125226203402,
+      "grad_norm": 0.7968795895576477,
+      "learning_rate": 0.0002,
+      "loss": 1.1512,
+      "step": 5330
+    },
+    {
+      "epoch": 3.8653637350705754,
+      "grad_norm": 0.7842824459075928,
+      "learning_rate": 0.0002,
+      "loss": 1.232,
+      "step": 5340
+    },
+    {
+      "epoch": 3.8726022439377488,
+      "grad_norm": 0.8272225260734558,
+      "learning_rate": 0.0002,
+      "loss": 1.1847,
+      "step": 5350
+    },
+    {
+      "epoch": 3.879840752804922,
+      "grad_norm": 0.8413397669792175,
+      "learning_rate": 0.0002,
+      "loss": 1.1381,
+      "step": 5360
+    },
+    {
+      "epoch": 3.8870792616720955,
+      "grad_norm": 1.141764760017395,
+      "learning_rate": 0.0002,
+      "loss": 1.2349,
+      "step": 5370
+    },
+    {
+      "epoch": 3.894317770539269,
+      "grad_norm": 0.9826975464820862,
+      "learning_rate": 0.0002,
+      "loss": 1.212,
+      "step": 5380
+    },
+    {
+      "epoch": 3.9015562794064422,
+      "grad_norm": 0.8598255515098572,
+      "learning_rate": 0.0002,
+      "loss": 1.1833,
+      "step": 5390
+    },
+    {
+      "epoch": 3.9087947882736156,
+      "grad_norm": 0.6271058320999146,
+      "learning_rate": 0.0002,
+      "loss": 1.1247,
+      "step": 5400
+    },
+    {
+      "epoch": 3.916033297140789,
+      "grad_norm": 0.6379870772361755,
+      "learning_rate": 0.0002,
+      "loss": 1.2212,
+      "step": 5410
+    },
+    {
+      "epoch": 3.9232718060079623,
+      "grad_norm": 1.0313376188278198,
+      "learning_rate": 0.0002,
+      "loss": 1.2481,
+      "step": 5420
+    },
+    {
+      "epoch": 3.9305103148751357,
+      "grad_norm": 0.8220619559288025,
+      "learning_rate": 0.0002,
+      "loss": 1.1872,
+      "step": 5430
+    },
+    {
+      "epoch": 3.937748823742309,
+      "grad_norm": 0.7576116919517517,
+      "learning_rate": 0.0002,
+      "loss": 1.2006,
+      "step": 5440
+    },
+    {
+      "epoch": 3.9449873326094824,
+      "grad_norm": 1.226235032081604,
+      "learning_rate": 0.0002,
+      "loss": 1.1969,
+      "step": 5450
+    },
+    {
+      "epoch": 3.952225841476656,
+      "grad_norm": 0.7979229688644409,
+      "learning_rate": 0.0002,
+      "loss": 1.2945,
+      "step": 5460
+    },
+    {
+      "epoch": 3.959464350343829,
+      "grad_norm": 0.9911929965019226,
+      "learning_rate": 0.0002,
+      "loss": 1.1922,
+      "step": 5470
+    },
+    {
+      "epoch": 3.9667028592110025,
+      "grad_norm": 0.643738865852356,
+      "learning_rate": 0.0002,
+      "loss": 1.0924,
+      "step": 5480
+    },
+    {
+      "epoch": 3.973941368078176,
+      "grad_norm": 0.682305634021759,
+      "learning_rate": 0.0002,
+      "loss": 1.0607,
+      "step": 5490
+    },
+    {
+      "epoch": 3.9811798769453492,
+      "grad_norm": 1.18373441696167,
+      "learning_rate": 0.0002,
+      "loss": 1.2908,
+      "step": 5500
+    },
+    {
+      "epoch": 3.9884183858125226,
+      "grad_norm": 0.7190203070640564,
+      "learning_rate": 0.0002,
+      "loss": 1.0889,
+      "step": 5510
+    },
+    {
+      "epoch": 3.995656894679696,
+      "grad_norm": 0.7516948580741882,
+      "learning_rate": 0.0002,
+      "loss": 1.2745,
+      "step": 5520
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.4252897500991821,
+      "eval_runtime": 27.235,
+      "eval_samples_per_second": 16.009,
+      "eval_steps_per_second": 2.019,
+      "step": 5526
+    },
+    {
+      "epoch": 4.002895403546869,
+      "grad_norm": 0.6353074312210083,
+      "learning_rate": 0.0002,
+      "loss": 1.0088,
+      "step": 5530
+    },
+    {
+      "epoch": 4.010133912414043,
+      "grad_norm": 0.7424906492233276,
+      "learning_rate": 0.0002,
+      "loss": 1.0326,
+      "step": 5540
+    },
+    {
+      "epoch": 4.017372421281216,
+      "grad_norm": 0.8856638073921204,
+      "learning_rate": 0.0002,
+      "loss": 1.0667,
+      "step": 5550
+    },
+    {
+      "epoch": 4.024610930148389,
+      "grad_norm": 0.9627974033355713,
+      "learning_rate": 0.0002,
+      "loss": 1.0905,
+      "step": 5560
+    },
+    {
+      "epoch": 4.031849439015563,
+      "grad_norm": 0.9048978686332703,
+      "learning_rate": 0.0002,
+      "loss": 1.0965,
+      "step": 5570
+    },
+    {
+      "epoch": 4.039087947882736,
+      "grad_norm": 0.921119213104248,
+      "learning_rate": 0.0002,
+      "loss": 1.1108,
+      "step": 5580
+    },
+    {
+      "epoch": 4.0463264567499095,
+      "grad_norm": 0.8654361963272095,
+      "learning_rate": 0.0002,
+      "loss": 1.1235,
+      "step": 5590
+    },
+    {
+      "epoch": 4.053564965617083,
+      "grad_norm": 0.7947945594787598,
+      "learning_rate": 0.0002,
+      "loss": 1.0794,
+      "step": 5600
+    },
+    {
+      "epoch": 4.060803474484256,
+      "grad_norm": 0.8307326436042786,
+      "learning_rate": 0.0002,
+      "loss": 1.0674,
+      "step": 5610
+    },
+    {
+      "epoch": 4.06804198335143,
+      "grad_norm": 0.793273389339447,
+      "learning_rate": 0.0002,
+      "loss": 1.0076,
+      "step": 5620
+    },
+    {
+      "epoch": 4.075280492218603,
+      "grad_norm": 0.8748673796653748,
+      "learning_rate": 0.0002,
+      "loss": 1.0651,
+      "step": 5630
+    },
+    {
+      "epoch": 4.082519001085776,
+      "grad_norm": 0.7926856279373169,
+      "learning_rate": 0.0002,
+      "loss": 1.111,
+      "step": 5640
+    },
+    {
+      "epoch": 4.08975750995295,
+      "grad_norm": 0.922645092010498,
+      "learning_rate": 0.0002,
+      "loss": 1.044,
+      "step": 5650
+    },
+    {
+      "epoch": 4.096996018820123,
+      "grad_norm": 0.9539641737937927,
+      "learning_rate": 0.0002,
+      "loss": 1.109,
+      "step": 5660
+    },
+    {
+      "epoch": 4.1042345276872965,
+      "grad_norm": 0.8674443364143372,
+      "learning_rate": 0.0002,
+      "loss": 1.0788,
+      "step": 5670
+    },
+    {
+      "epoch": 4.11147303655447,
+      "grad_norm": 0.7097609043121338,
+      "learning_rate": 0.0002,
+      "loss": 0.9867,
+      "step": 5680
+    },
+    {
+      "epoch": 4.118711545421643,
+      "grad_norm": 0.8875522613525391,
+      "learning_rate": 0.0002,
+      "loss": 1.1154,
+      "step": 5690
+    },
+    {
+      "epoch": 4.125950054288817,
+      "grad_norm": 0.8583634495735168,
+      "learning_rate": 0.0002,
+      "loss": 1.1217,
+      "step": 5700
+    },
+    {
+      "epoch": 4.13318856315599,
+      "grad_norm": 0.6736377477645874,
+      "learning_rate": 0.0002,
+      "loss": 1.0973,
+      "step": 5710
+    },
+    {
+      "epoch": 4.140427072023163,
+      "grad_norm": 0.9349062442779541,
+      "learning_rate": 0.0002,
+      "loss": 1.1199,
+      "step": 5720
+    },
+    {
+      "epoch": 4.147665580890337,
+      "grad_norm": 1.0610365867614746,
+      "learning_rate": 0.0002,
+      "loss": 1.0508,
+      "step": 5730
+    },
+    {
+      "epoch": 4.15490408975751,
+      "grad_norm": 1.5838189125061035,
+      "learning_rate": 0.0002,
+      "loss": 1.1146,
+      "step": 5740
+    },
+    {
+      "epoch": 4.162142598624683,
+      "grad_norm": 0.747522234916687,
+      "learning_rate": 0.0002,
+      "loss": 1.0222,
+      "step": 5750
+    },
+    {
+      "epoch": 4.169381107491857,
+      "grad_norm": 1.3247915506362915,
+      "learning_rate": 0.0002,
+      "loss": 1.1328,
+      "step": 5760
+    },
+    {
+      "epoch": 4.17661961635903,
+      "grad_norm": 0.8750247955322266,
+      "learning_rate": 0.0002,
+      "loss": 1.1655,
+      "step": 5770
+    },
+    {
+      "epoch": 4.1838581252262035,
+      "grad_norm": 0.7914144992828369,
+      "learning_rate": 0.0002,
+      "loss": 1.199,
+      "step": 5780
+    },
+    {
+      "epoch": 4.191096634093377,
+      "grad_norm": 0.9493299126625061,
+      "learning_rate": 0.0002,
+      "loss": 1.1213,
+      "step": 5790
+    },
+    {
+      "epoch": 4.19833514296055,
+      "grad_norm": 0.7802295088768005,
+      "learning_rate": 0.0002,
+      "loss": 1.1515,
+      "step": 5800
+    },
+    {
+      "epoch": 4.205573651827724,
+      "grad_norm": 0.6987314820289612,
+      "learning_rate": 0.0002,
+      "loss": 1.0704,
+      "step": 5810
+    },
+    {
+      "epoch": 4.212812160694897,
+      "grad_norm": 0.9220341444015503,
+      "learning_rate": 0.0002,
+      "loss": 1.1699,
+      "step": 5820
+    },
+    {
+      "epoch": 4.22005066956207,
+      "grad_norm": 0.8932939767837524,
+      "learning_rate": 0.0002,
+      "loss": 1.1394,
+      "step": 5830
+    },
+    {
+      "epoch": 4.227289178429244,
+      "grad_norm": 0.920002818107605,
+      "learning_rate": 0.0002,
+      "loss": 1.0048,
+      "step": 5840
+    },
+    {
+      "epoch": 4.234527687296417,
+      "grad_norm": 0.6662752032279968,
+      "learning_rate": 0.0002,
+      "loss": 0.964,
+      "step": 5850
+    },
+    {
+      "epoch": 4.24176619616359,
+      "grad_norm": 0.8679718971252441,
+      "learning_rate": 0.0002,
+      "loss": 0.986,
+      "step": 5860
+    },
+    {
+      "epoch": 4.249004705030764,
+      "grad_norm": 0.7020887732505798,
+      "learning_rate": 0.0002,
+      "loss": 0.8991,
+      "step": 5870
+    },
+    {
+      "epoch": 4.256243213897937,
+      "grad_norm": 0.869611382484436,
+      "learning_rate": 0.0002,
+      "loss": 1.1132,
+      "step": 5880
+    },
+    {
+      "epoch": 4.2634817227651105,
+      "grad_norm": 0.7796585559844971,
+      "learning_rate": 0.0002,
+      "loss": 1.1026,
+      "step": 5890
+    },
+    {
+      "epoch": 4.270720231632284,
+      "grad_norm": 0.8978819251060486,
+      "learning_rate": 0.0002,
+      "loss": 1.0957,
+      "step": 5900
+    },
+    {
+      "epoch": 4.277958740499457,
+      "grad_norm": 1.0837205648422241,
+      "learning_rate": 0.0002,
+      "loss": 1.1325,
+      "step": 5910
+    },
+    {
+      "epoch": 4.285197249366631,
+      "grad_norm": 0.7584353089332581,
+      "learning_rate": 0.0002,
+      "loss": 1.1279,
+      "step": 5920
+    },
+    {
+      "epoch": 4.292435758233804,
+      "grad_norm": 0.7313185334205627,
+      "learning_rate": 0.0002,
+      "loss": 1.0513,
+      "step": 5930
+    },
+    {
+      "epoch": 4.299674267100977,
+      "grad_norm": 0.8004671335220337,
+      "learning_rate": 0.0002,
+      "loss": 1.1101,
+      "step": 5940
+    },
+    {
+      "epoch": 4.306912775968151,
+      "grad_norm": 2.154958724975586,
+      "learning_rate": 0.0002,
+      "loss": 1.14,
+      "step": 5950
+    },
+    {
+      "epoch": 4.314151284835324,
+      "grad_norm": 0.9163479804992676,
+      "learning_rate": 0.0002,
+      "loss": 1.1206,
+      "step": 5960
+    },
+    {
+      "epoch": 4.321389793702497,
+      "grad_norm": 0.9151589274406433,
+      "learning_rate": 0.0002,
+      "loss": 0.9941,
+      "step": 5970
+    },
+    {
+      "epoch": 4.328628302569671,
+      "grad_norm": 0.8624112010002136,
+      "learning_rate": 0.0002,
+      "loss": 1.0606,
+      "step": 5980
+    },
+    {
+      "epoch": 4.335866811436844,
+      "grad_norm": 0.9357741475105286,
+      "learning_rate": 0.0002,
+      "loss": 1.1625,
+      "step": 5990
+    },
+    {
+      "epoch": 4.3431053203040175,
+      "grad_norm": 1.3482335805892944,
+      "learning_rate": 0.0002,
+      "loss": 1.0712,
+      "step": 6000
+    },
+    {
+      "epoch": 4.350343829171191,
+      "grad_norm": 0.7156149744987488,
+      "learning_rate": 0.0002,
+      "loss": 1.1224,
+      "step": 6010
+    },
+    {
+      "epoch": 4.357582338038364,
+      "grad_norm": 0.8480049967765808,
+      "learning_rate": 0.0002,
+      "loss": 1.0753,
+      "step": 6020
+    },
+    {
+      "epoch": 4.364820846905538,
+      "grad_norm": 0.8262244462966919,
+      "learning_rate": 0.0002,
+      "loss": 1.051,
+      "step": 6030
+    },
+    {
+      "epoch": 4.372059355772711,
+      "grad_norm": 0.7733905911445618,
+      "learning_rate": 0.0002,
+      "loss": 0.9966,
+      "step": 6040
+    },
+    {
+      "epoch": 4.379297864639884,
+      "grad_norm": 0.8553919792175293,
+      "learning_rate": 0.0002,
+      "loss": 1.1008,
+      "step": 6050
+    },
+    {
+      "epoch": 4.386536373507058,
+      "grad_norm": 0.8666832447052002,
+      "learning_rate": 0.0002,
+      "loss": 1.1777,
+      "step": 6060
+    },
+    {
+      "epoch": 4.393774882374231,
+      "grad_norm": 0.9168295860290527,
+      "learning_rate": 0.0002,
+      "loss": 1.1934,
+      "step": 6070
+    },
+    {
+      "epoch": 4.4010133912414044,
+      "grad_norm": 0.7315238118171692,
+      "learning_rate": 0.0002,
+      "loss": 1.0988,
+      "step": 6080
+    },
+    {
+      "epoch": 4.408251900108578,
+      "grad_norm": 1.020263433456421,
+      "learning_rate": 0.0002,
+      "loss": 1.1599,
+      "step": 6090
+    },
+    {
+      "epoch": 4.415490408975751,
+      "grad_norm": 0.9978243708610535,
+      "learning_rate": 0.0002,
+      "loss": 1.133,
+      "step": 6100
+    },
+    {
+      "epoch": 4.4227289178429245,
+      "grad_norm": 0.995453953742981,
+      "learning_rate": 0.0002,
+      "loss": 1.1324,
+      "step": 6110
+    },
+    {
+      "epoch": 4.429967426710098,
+      "grad_norm": 0.9360884428024292,
+      "learning_rate": 0.0002,
+      "loss": 1.0957,
+      "step": 6120
+    },
+    {
+      "epoch": 4.437205935577271,
+      "grad_norm": 0.8099448084831238,
+      "learning_rate": 0.0002,
+      "loss": 0.9506,
+      "step": 6130
+    },
+    {
+      "epoch": 4.444444444444445,
+      "grad_norm": 0.8173841238021851,
+      "learning_rate": 0.0002,
+      "loss": 1.0887,
+      "step": 6140
+    },
+    {
+      "epoch": 4.451682953311618,
+      "grad_norm": 0.7972666025161743,
+      "learning_rate": 0.0002,
+      "loss": 1.1219,
+      "step": 6150
+    },
+    {
+      "epoch": 4.458921462178791,
+      "grad_norm": 0.7685779333114624,
+      "learning_rate": 0.0002,
+      "loss": 1.0226,
+      "step": 6160
+    },
+    {
+      "epoch": 4.466159971045965,
+      "grad_norm": 0.7872623801231384,
+      "learning_rate": 0.0002,
+      "loss": 1.0732,
+      "step": 6170
+    },
+    {
+      "epoch": 4.473398479913138,
+      "grad_norm": 0.7677070498466492,
+      "learning_rate": 0.0002,
+      "loss": 0.9911,
+      "step": 6180
+    },
+    {
+      "epoch": 4.4806369887803115,
+      "grad_norm": 0.7878316044807434,
+      "learning_rate": 0.0002,
+      "loss": 1.0919,
+      "step": 6190
+    },
+    {
+      "epoch": 4.487875497647485,
+      "grad_norm": 0.8178079724311829,
+      "learning_rate": 0.0002,
+      "loss": 1.018,
+      "step": 6200
+    },
+    {
+      "epoch": 4.495114006514658,
+      "grad_norm": 1.2820082902908325,
+      "learning_rate": 0.0002,
+      "loss": 1.0517,
+      "step": 6210
+    },
+    {
+      "epoch": 4.502352515381832,
+      "grad_norm": 0.9380832314491272,
+      "learning_rate": 0.0002,
+      "loss": 1.3101,
+      "step": 6220
+    },
+    {
+      "epoch": 4.509591024249005,
+      "grad_norm": 0.7810422778129578,
+      "learning_rate": 0.0002,
+      "loss": 0.9818,
+      "step": 6230
+    },
+    {
+      "epoch": 4.516829533116178,
+      "grad_norm": 1.1022917032241821,
+      "learning_rate": 0.0002,
+      "loss": 1.1677,
+      "step": 6240
+    },
+    {
+      "epoch": 4.524068041983352,
+      "grad_norm": 1.4275553226470947,
+      "learning_rate": 0.0002,
+      "loss": 1.1579,
+      "step": 6250
+    },
+    {
+      "epoch": 4.531306550850525,
+      "grad_norm": 0.7597777247428894,
+      "learning_rate": 0.0002,
+      "loss": 1.3237,
+      "step": 6260
+    },
+    {
+      "epoch": 4.538545059717698,
+      "grad_norm": 1.10992431640625,
+      "learning_rate": 0.0002,
+      "loss": 1.1529,
+      "step": 6270
+    },
+    {
+      "epoch": 4.545783568584872,
+      "grad_norm": 0.8981178998947144,
+      "learning_rate": 0.0002,
+      "loss": 1.0732,
+      "step": 6280
+    },
+    {
+      "epoch": 4.553022077452045,
+      "grad_norm": 0.7863979339599609,
+      "learning_rate": 0.0002,
+      "loss": 1.086,
+      "step": 6290
+    },
+    {
+      "epoch": 4.5602605863192185,
+      "grad_norm": 0.9071474671363831,
+      "learning_rate": 0.0002,
+      "loss": 1.2008,
+      "step": 6300
+    },
+    {
+      "epoch": 4.567499095186392,
+      "grad_norm": 0.7429424524307251,
+      "learning_rate": 0.0002,
+      "loss": 1.0916,
+      "step": 6310
+    },
+    {
+      "epoch": 4.574737604053565,
+      "grad_norm": 1.0767850875854492,
+      "learning_rate": 0.0002,
+      "loss": 1.095,
+      "step": 6320
+    },
+    {
+      "epoch": 4.581976112920739,
+      "grad_norm": 0.7885915637016296,
+      "learning_rate": 0.0002,
+      "loss": 1.1023,
+      "step": 6330
+    },
+    {
+      "epoch": 4.589214621787912,
+      "grad_norm": 0.8350457549095154,
+      "learning_rate": 0.0002,
+      "loss": 1.1131,
+      "step": 6340
+    },
+    {
+      "epoch": 4.596453130655085,
+      "grad_norm": 0.7853530645370483,
+      "learning_rate": 0.0002,
+      "loss": 1.0743,
+      "step": 6350
+    },
+    {
+      "epoch": 4.603691639522259,
+      "grad_norm": 1.1220661401748657,
+      "learning_rate": 0.0002,
+      "loss": 1.1912,
+      "step": 6360
+    },
+    {
+      "epoch": 4.610930148389432,
+      "grad_norm": 0.7959423065185547,
+      "learning_rate": 0.0002,
+      "loss": 1.0927,
+      "step": 6370
+    },
+    {
+      "epoch": 4.618168657256605,
+      "grad_norm": 0.7782652378082275,
+      "learning_rate": 0.0002,
+      "loss": 1.1542,
+      "step": 6380
+    },
+    {
+      "epoch": 4.625407166123779,
+      "grad_norm": 0.7882203459739685,
+      "learning_rate": 0.0002,
+      "loss": 1.0753,
+      "step": 6390
+    },
+    {
+      "epoch": 4.632645674990952,
+      "grad_norm": 0.8841899037361145,
+      "learning_rate": 0.0002,
+      "loss": 1.0676,
+      "step": 6400
+    },
+    {
+      "epoch": 4.6398841838581255,
+      "grad_norm": 0.7936127781867981,
+      "learning_rate": 0.0002,
+      "loss": 1.0815,
+      "step": 6410
+    },
+    {
+      "epoch": 4.647122692725299,
+      "grad_norm": 0.9213966131210327,
+      "learning_rate": 0.0002,
+      "loss": 1.0198,
+      "step": 6420
+    },
+    {
+      "epoch": 4.654361201592472,
+      "grad_norm": 0.9246473908424377,
+      "learning_rate": 0.0002,
+      "loss": 0.9872,
+      "step": 6430
+    },
+    {
+      "epoch": 4.661599710459646,
+      "grad_norm": 0.766572892665863,
+      "learning_rate": 0.0002,
+      "loss": 1.1309,
+      "step": 6440
+    },
+    {
+      "epoch": 4.668838219326819,
+      "grad_norm": 0.8596171736717224,
+      "learning_rate": 0.0002,
+      "loss": 1.1095,
+      "step": 6450
+    },
+    {
+      "epoch": 4.676076728193992,
+      "grad_norm": 0.8482751846313477,
+      "learning_rate": 0.0002,
+      "loss": 1.1869,
+      "step": 6460
+    },
+    {
+      "epoch": 4.683315237061166,
+      "grad_norm": 1.0826905965805054,
+      "learning_rate": 0.0002,
+      "loss": 1.0622,
+      "step": 6470
+    },
+    {
+      "epoch": 4.690553745928339,
+      "grad_norm": 1.1048457622528076,
+      "learning_rate": 0.0002,
+      "loss": 1.0256,
+      "step": 6480
+    },
+    {
+      "epoch": 4.697792254795512,
+      "grad_norm": 0.9429134726524353,
+      "learning_rate": 0.0002,
+      "loss": 1.0514,
+      "step": 6490
+    },
+    {
+      "epoch": 4.705030763662686,
+      "grad_norm": 0.8587502837181091,
+      "learning_rate": 0.0002,
+      "loss": 1.1351,
+      "step": 6500
+    },
+    {
+      "epoch": 4.712269272529859,
+      "grad_norm": 1.0387083292007446,
+      "learning_rate": 0.0002,
+      "loss": 1.0969,
+      "step": 6510
+    },
+    {
+      "epoch": 4.7195077813970325,
+      "grad_norm": 0.7471951842308044,
+      "learning_rate": 0.0002,
+      "loss": 1.0493,
+      "step": 6520
+    },
+    {
+      "epoch": 4.726746290264206,
+      "grad_norm": 0.8800424933433533,
+      "learning_rate": 0.0002,
+      "loss": 1.2632,
+      "step": 6530
+    },
+    {
+      "epoch": 4.733984799131379,
+      "grad_norm": 0.8136811852455139,
+      "learning_rate": 0.0002,
+      "loss": 1.2126,
+      "step": 6540
+    },
+    {
+      "epoch": 4.741223307998553,
+      "grad_norm": 0.9910339713096619,
+      "learning_rate": 0.0002,
+      "loss": 1.195,
+      "step": 6550
+    },
+    {
+      "epoch": 4.748461816865726,
+      "grad_norm": 1.0679163932800293,
+      "learning_rate": 0.0002,
+      "loss": 1.1201,
+      "step": 6560
+    },
+    {
+      "epoch": 4.755700325732899,
+      "grad_norm": 0.8468248248100281,
+      "learning_rate": 0.0002,
+      "loss": 1.0297,
+      "step": 6570
+    },
+    {
+      "epoch": 4.762938834600073,
+      "grad_norm": 0.8771235942840576,
+      "learning_rate": 0.0002,
+      "loss": 1.0858,
+      "step": 6580
+    },
+    {
+      "epoch": 4.770177343467246,
+      "grad_norm": 0.7024846076965332,
+      "learning_rate": 0.0002,
+      "loss": 1.077,
+      "step": 6590
+    },
+    {
+      "epoch": 4.7774158523344195,
+      "grad_norm": 0.7836683392524719,
+      "learning_rate": 0.0002,
+      "loss": 1.0876,
+      "step": 6600
+    },
+    {
+      "epoch": 4.784654361201593,
+      "grad_norm": 0.7717288136482239,
+      "learning_rate": 0.0002,
+      "loss": 1.1006,
+      "step": 6610
+    },
+    {
+      "epoch": 4.791892870068766,
+      "grad_norm": 0.884183943271637,
+      "learning_rate": 0.0002,
+      "loss": 1.0376,
+      "step": 6620
+    },
+    {
+      "epoch": 4.7991313789359396,
+      "grad_norm": 1.383867621421814,
+      "learning_rate": 0.0002,
+      "loss": 1.1757,
+      "step": 6630
+    },
+    {
+      "epoch": 4.806369887803113,
+      "grad_norm": 0.9741523861885071,
+      "learning_rate": 0.0002,
+      "loss": 1.0861,
+      "step": 6640
+    },
+    {
+      "epoch": 4.813608396670286,
+      "grad_norm": 0.9723693132400513,
+      "learning_rate": 0.0002,
+      "loss": 1.0884,
+      "step": 6650
+    },
+    {
+      "epoch": 4.82084690553746,
+      "grad_norm": 1.8324809074401855,
+      "learning_rate": 0.0002,
+      "loss": 1.2203,
+      "step": 6660
+    },
+    {
+      "epoch": 4.828085414404633,
+      "grad_norm": 0.904909074306488,
+      "learning_rate": 0.0002,
+      "loss": 1.0292,
+      "step": 6670
+    },
+    {
+      "epoch": 4.835323923271806,
+      "grad_norm": 0.7355411648750305,
+      "learning_rate": 0.0002,
+      "loss": 1.0349,
+      "step": 6680
+    },
+    {
+      "epoch": 4.84256243213898,
+      "grad_norm": 0.8934960961341858,
+      "learning_rate": 0.0002,
+      "loss": 1.0793,
+      "step": 6690
+    },
+    {
+      "epoch": 4.849800941006153,
+      "grad_norm": 1.4596954584121704,
+      "learning_rate": 0.0002,
+      "loss": 1.0375,
+      "step": 6700
+    },
+    {
+      "epoch": 4.8570394498733265,
+      "grad_norm": 0.8310341238975525,
+      "learning_rate": 0.0002,
+      "loss": 1.1065,
+      "step": 6710
+    },
+    {
+      "epoch": 4.8642779587405,
+      "grad_norm": 0.9709894061088562,
+      "learning_rate": 0.0002,
+      "loss": 1.1089,
+      "step": 6720
+    },
+    {
+      "epoch": 4.871516467607673,
+      "grad_norm": 0.852142333984375,
+      "learning_rate": 0.0002,
+      "loss": 1.0069,
+      "step": 6730
+    },
+    {
+      "epoch": 4.878754976474847,
+      "grad_norm": 1.0643625259399414,
+      "learning_rate": 0.0002,
+      "loss": 1.0507,
+      "step": 6740
+    },
+    {
+      "epoch": 4.88599348534202,
+      "grad_norm": 0.9419508576393127,
+      "learning_rate": 0.0002,
+      "loss": 1.056,
+      "step": 6750
+    },
+    {
+      "epoch": 4.893231994209193,
+      "grad_norm": 1.1818498373031616,
+      "learning_rate": 0.0002,
+      "loss": 1.1995,
+      "step": 6760
+    },
+    {
+      "epoch": 4.900470503076367,
+      "grad_norm": 0.9369569420814514,
+      "learning_rate": 0.0002,
+      "loss": 1.0925,
+      "step": 6770
+    },
+    {
+      "epoch": 4.90770901194354,
+      "grad_norm": 0.7012579441070557,
+      "learning_rate": 0.0002,
+      "loss": 1.1648,
+      "step": 6780
+    },
+    {
+      "epoch": 4.914947520810713,
+      "grad_norm": 0.9109319448471069,
+      "learning_rate": 0.0002,
+      "loss": 1.0926,
+      "step": 6790
+    },
+    {
+      "epoch": 4.922186029677887,
+      "grad_norm": 0.8077534437179565,
+      "learning_rate": 0.0002,
+      "loss": 1.0358,
+      "step": 6800
+    },
+    {
+      "epoch": 4.92942453854506,
+      "grad_norm": 0.7571148872375488,
+      "learning_rate": 0.0002,
+      "loss": 1.2549,
+      "step": 6810
+    },
+    {
+      "epoch": 4.9366630474122335,
+      "grad_norm": 0.7325633764266968,
+      "learning_rate": 0.0002,
+      "loss": 0.9638,
+      "step": 6820
+    },
+    {
+      "epoch": 4.943901556279407,
+      "grad_norm": 0.8465084433555603,
+      "learning_rate": 0.0002,
+      "loss": 1.0128,
+      "step": 6830
+    },
+    {
+      "epoch": 4.95114006514658,
+      "grad_norm": 0.8753737807273865,
+      "learning_rate": 0.0002,
+      "loss": 1.153,
+      "step": 6840
+    },
+    {
+      "epoch": 4.958378574013754,
+      "grad_norm": 0.9421748518943787,
+      "learning_rate": 0.0002,
+      "loss": 1.0247,
+      "step": 6850
+    },
+    {
+      "epoch": 4.965617082880927,
+      "grad_norm": 0.8245896697044373,
+      "learning_rate": 0.0002,
+      "loss": 1.1483,
+      "step": 6860
+    },
+    {
+      "epoch": 4.9728555917481,
+      "grad_norm": 0.8823089599609375,
+      "learning_rate": 0.0002,
+      "loss": 0.9905,
+      "step": 6870
+    },
+    {
+      "epoch": 4.980094100615274,
+      "grad_norm": 0.8406389355659485,
+      "learning_rate": 0.0002,
+      "loss": 1.1664,
+      "step": 6880
+    },
+    {
+      "epoch": 4.987332609482447,
+      "grad_norm": 0.9732868075370789,
+      "learning_rate": 0.0002,
+      "loss": 1.0944,
+      "step": 6890
+    },
+    {
+      "epoch": 4.99457111834962,
+      "grad_norm": 2.125141143798828,
+      "learning_rate": 0.0002,
+      "loss": 1.1776,
+      "step": 6900
+    },
+    {
+      "epoch": 4.999638074556641,
+      "eval_loss": 1.445176601409912,
+      "eval_runtime": 27.2351,
+      "eval_samples_per_second": 16.009,
+      "eval_steps_per_second": 2.019,
+      "step": 6907
+    },
+    {
+      "epoch": 5.001809627216793,
+      "grad_norm": 0.9465792775154114,
+      "learning_rate": 0.0002,
+      "loss": 1.1362,
+      "step": 6910
+    },
+    {
+      "epoch": 5.009048136083966,
+      "grad_norm": 1.2834891080856323,
+      "learning_rate": 0.0002,
+      "loss": 0.982,
+      "step": 6920
+    },
+    {
+      "epoch": 5.01628664495114,
+      "grad_norm": 1.0297378301620483,
+      "learning_rate": 0.0002,
+      "loss": 0.9803,
+      "step": 6930
+    },
+    {
+      "epoch": 5.023525153818313,
+      "grad_norm": 1.1705161333084106,
+      "learning_rate": 0.0002,
+      "loss": 1.0447,
+      "step": 6940
+    },
+    {
+      "epoch": 5.030763662685486,
+      "grad_norm": 0.8293961882591248,
+      "learning_rate": 0.0002,
+      "loss": 1.0113,
+      "step": 6950
+    },
+    {
+      "epoch": 5.03800217155266,
+      "grad_norm": 1.0422210693359375,
+      "learning_rate": 0.0002,
+      "loss": 0.9203,
+      "step": 6960
+    },
+    {
+      "epoch": 5.045240680419833,
+      "grad_norm": 1.116104245185852,
+      "learning_rate": 0.0002,
+      "loss": 1.0553,
+      "step": 6970
+    },
+    {
+      "epoch": 5.0524791892870065,
+      "grad_norm": 1.5118416547775269,
+      "learning_rate": 0.0002,
+      "loss": 0.9011,
+      "step": 6980
+    },
+    {
+      "epoch": 5.05971769815418,
+      "grad_norm": 0.8383979797363281,
+      "learning_rate": 0.0002,
+      "loss": 0.9969,
+      "step": 6990
+    },
+    {
+      "epoch": 5.066956207021353,
+      "grad_norm": 1.3378649950027466,
+      "learning_rate": 0.0002,
+      "loss": 0.9659,
+      "step": 7000
+    },
+    {
+      "epoch": 5.0741947158885266,
+      "grad_norm": 1.1840510368347168,
+      "learning_rate": 0.0002,
+      "loss": 1.0212,
+      "step": 7010
+    },
+    {
+      "epoch": 5.0814332247557,
+      "grad_norm": 1.2354751825332642,
+      "learning_rate": 0.0002,
+      "loss": 0.9939,
+      "step": 7020
+    },
+    {
+      "epoch": 5.088671733622873,
+      "grad_norm": 1.3830451965332031,
+      "learning_rate": 0.0002,
+      "loss": 0.9831,
+      "step": 7030
+    },
+    {
+      "epoch": 5.095910242490047,
+      "grad_norm": 0.8101674318313599,
+      "learning_rate": 0.0002,
+      "loss": 1.1827,
+      "step": 7040
+    },
+    {
+      "epoch": 5.10314875135722,
+      "grad_norm": 0.897982656955719,
+      "learning_rate": 0.0002,
+      "loss": 0.9255,
+      "step": 7050
+    },
+    {
+      "epoch": 5.110387260224393,
+      "grad_norm": 1.2049678564071655,
+      "learning_rate": 0.0002,
+      "loss": 0.8784,
+      "step": 7060
+    },
+    {
+      "epoch": 5.117625769091567,
+      "grad_norm": 1.5912116765975952,
+      "learning_rate": 0.0002,
+      "loss": 1.0182,
+      "step": 7070
+    },
+    {
+      "epoch": 5.12486427795874,
+      "grad_norm": 0.9261530041694641,
+      "learning_rate": 0.0002,
+      "loss": 1.0909,
+      "step": 7080
+    },
+    {
+      "epoch": 5.1321027868259135,
+      "grad_norm": 1.1454812288284302,
+      "learning_rate": 0.0002,
+      "loss": 0.9603,
+      "step": 7090
+    },
+    {
+      "epoch": 5.139341295693087,
+      "grad_norm": 1.0049978494644165,
+      "learning_rate": 0.0002,
+      "loss": 0.9149,
+      "step": 7100
+    },
+    {
+      "epoch": 5.14657980456026,
+      "grad_norm": 1.4513251781463623,
+      "learning_rate": 0.0002,
+      "loss": 0.9463,
+      "step": 7110
+    },
+    {
+      "epoch": 5.153818313427434,
+      "grad_norm": 0.9800849556922913,
+      "learning_rate": 0.0002,
+      "loss": 0.8995,
+      "step": 7120
+    },
+    {
+      "epoch": 5.161056822294607,
+      "grad_norm": 0.9698708653450012,
+      "learning_rate": 0.0002,
+      "loss": 0.9835,
+      "step": 7130
+    },
+    {
+      "epoch": 5.16829533116178,
+      "grad_norm": 1.1126646995544434,
+      "learning_rate": 0.0002,
+      "loss": 0.9672,
+      "step": 7140
+    },
+    {
+      "epoch": 5.175533840028954,
+      "grad_norm": 0.9248330593109131,
+      "learning_rate": 0.0002,
+      "loss": 0.9384,
+      "step": 7150
+    },
+    {
+      "epoch": 5.182772348896127,
+      "grad_norm": 0.7967255711555481,
+      "learning_rate": 0.0002,
+      "loss": 0.826,
+      "step": 7160
+    },
+    {
+      "epoch": 5.1900108577633,
+      "grad_norm": 0.9933333992958069,
+      "learning_rate": 0.0002,
+      "loss": 1.0078,
+      "step": 7170
+    },
+    {
+      "epoch": 5.197249366630474,
+      "grad_norm": 1.0080649852752686,
+      "learning_rate": 0.0002,
+      "loss": 1.0276,
+      "step": 7180
+    },
+    {
+      "epoch": 5.204487875497647,
+      "grad_norm": 1.3954921960830688,
+      "learning_rate": 0.0002,
+      "loss": 1.0201,
+      "step": 7190
+    },
+    {
+      "epoch": 5.2117263843648205,
+      "grad_norm": 1.2386271953582764,
+      "learning_rate": 0.0002,
+      "loss": 1.0863,
+      "step": 7200
+    },
+    {
+      "epoch": 5.218964893231994,
+      "grad_norm": 1.2379488945007324,
+      "learning_rate": 0.0002,
+      "loss": 0.8863,
+      "step": 7210
+    },
+    {
+      "epoch": 5.226203402099167,
+      "grad_norm": 0.9882503747940063,
+      "learning_rate": 0.0002,
+      "loss": 1.0518,
+      "step": 7220
+    },
+    {
+      "epoch": 5.233441910966341,
+      "grad_norm": 1.1728729009628296,
+      "learning_rate": 0.0002,
+      "loss": 0.9834,
+      "step": 7230
+    },
+    {
+      "epoch": 5.240680419833514,
+      "grad_norm": 0.9849673509597778,
+      "learning_rate": 0.0002,
+      "loss": 0.9269,
+      "step": 7240
+    },
+    {
+      "epoch": 5.247918928700687,
+      "grad_norm": 1.177639365196228,
+      "learning_rate": 0.0002,
+      "loss": 0.9935,
+      "step": 7250
+    },
+    {
+      "epoch": 5.255157437567861,
+      "grad_norm": 1.2395055294036865,
+      "learning_rate": 0.0002,
+      "loss": 1.0639,
+      "step": 7260
+    },
+    {
+      "epoch": 5.262395946435034,
+      "grad_norm": 1.3999171257019043,
+      "learning_rate": 0.0002,
+      "loss": 1.0138,
+      "step": 7270
+    },
+    {
+      "epoch": 5.269634455302207,
+      "grad_norm": 0.7698732018470764,
+      "learning_rate": 0.0002,
+      "loss": 0.9745,
+      "step": 7280
+    },
+    {
+      "epoch": 5.276872964169381,
+      "grad_norm": 0.9167453646659851,
+      "learning_rate": 0.0002,
+      "loss": 1.0389,
+      "step": 7290
+    },
+    {
+      "epoch": 5.284111473036554,
+      "grad_norm": 1.113830804824829,
+      "learning_rate": 0.0002,
+      "loss": 0.9858,
+      "step": 7300
+    },
+    {
+      "epoch": 5.2913499819037275,
+      "grad_norm": 0.9644396901130676,
+      "learning_rate": 0.0002,
+      "loss": 0.9577,
+      "step": 7310
+    },
+    {
+      "epoch": 5.298588490770901,
+      "grad_norm": 1.462435007095337,
+      "learning_rate": 0.0002,
+      "loss": 1.0556,
+      "step": 7320
+    },
+    {
+      "epoch": 5.305826999638074,
+      "grad_norm": 0.9406287670135498,
+      "learning_rate": 0.0002,
+      "loss": 0.871,
+      "step": 7330
+    },
+    {
+      "epoch": 5.313065508505248,
+      "grad_norm": 0.9698247909545898,
+      "learning_rate": 0.0002,
+      "loss": 1.0022,
+      "step": 7340
+    },
+    {
+      "epoch": 5.320304017372421,
+      "grad_norm": 1.12003755569458,
+      "learning_rate": 0.0002,
+      "loss": 0.915,
+      "step": 7350
+    },
+    {
+      "epoch": 5.327542526239594,
+      "grad_norm": 1.598681926727295,
+      "learning_rate": 0.0002,
+      "loss": 0.9838,
+      "step": 7360
+    },
+    {
+      "epoch": 5.334781035106768,
+      "grad_norm": 1.0450010299682617,
+      "learning_rate": 0.0002,
+      "loss": 1.0,
+      "step": 7370
+    },
+    {
+      "epoch": 5.342019543973941,
+      "grad_norm": 0.8680008053779602,
+      "learning_rate": 0.0002,
+      "loss": 0.9983,
+      "step": 7380
+    },
+    {
+      "epoch": 5.349258052841114,
+      "grad_norm": 1.0115476846694946,
+      "learning_rate": 0.0002,
+      "loss": 0.9851,
+      "step": 7390
+    },
+    {
+      "epoch": 5.356496561708288,
+      "grad_norm": 0.9589748382568359,
+      "learning_rate": 0.0002,
+      "loss": 1.0702,
+      "step": 7400
+    },
+    {
+      "epoch": 5.363735070575461,
+      "grad_norm": 0.6729998588562012,
+      "learning_rate": 0.0002,
+      "loss": 0.9366,
+      "step": 7410
+    },
+    {
+      "epoch": 5.3709735794426345,
+      "grad_norm": 0.9246699213981628,
+      "learning_rate": 0.0002,
+      "loss": 1.0126,
+      "step": 7420
+    },
+    {
+      "epoch": 5.378212088309808,
+      "grad_norm": 1.1266791820526123,
+      "learning_rate": 0.0002,
+      "loss": 0.9815,
+      "step": 7430
+    },
+    {
+      "epoch": 5.385450597176981,
+      "grad_norm": 1.8056942224502563,
+      "learning_rate": 0.0002,
+      "loss": 1.1166,
+      "step": 7440
+    },
+    {
+      "epoch": 5.392689106044155,
+      "grad_norm": 0.9802932739257812,
+      "learning_rate": 0.0002,
+      "loss": 0.9604,
+      "step": 7450
+    },
+    {
+      "epoch": 5.399927614911328,
+      "grad_norm": 1.0504707098007202,
+      "learning_rate": 0.0002,
+      "loss": 0.9656,
+      "step": 7460
+    },
+    {
+      "epoch": 5.407166123778501,
+      "grad_norm": 1.1915022134780884,
+      "learning_rate": 0.0002,
+      "loss": 1.0132,
+      "step": 7470
+    },
+    {
+      "epoch": 5.414404632645675,
+      "grad_norm": 1.1856611967086792,
+      "learning_rate": 0.0002,
+      "loss": 1.0041,
+      "step": 7480
+    },
+    {
+      "epoch": 5.421643141512848,
+      "grad_norm": 1.292152762413025,
+      "learning_rate": 0.0002,
+      "loss": 0.9747,
+      "step": 7490
+    },
+    {
+      "epoch": 5.4288816503800215,
+      "grad_norm": 1.2675740718841553,
+      "learning_rate": 0.0002,
+      "loss": 0.9659,
+      "step": 7500
+    },
+    {
+      "epoch": 5.436120159247195,
+      "grad_norm": 1.4034695625305176,
+      "learning_rate": 0.0002,
+      "loss": 1.0271,
+      "step": 7510
+    },
+    {
+      "epoch": 5.443358668114368,
+      "grad_norm": 0.984588623046875,
+      "learning_rate": 0.0002,
+      "loss": 1.0318,
+      "step": 7520
+    },
+    {
+      "epoch": 5.450597176981542,
+      "grad_norm": 0.8419108390808105,
+      "learning_rate": 0.0002,
+      "loss": 1.0726,
+      "step": 7530
+    },
+    {
+      "epoch": 5.457835685848715,
+      "grad_norm": 1.0270143747329712,
+      "learning_rate": 0.0002,
+      "loss": 1.0499,
+      "step": 7540
+    },
+    {
+      "epoch": 5.465074194715888,
+      "grad_norm": 2.2158689498901367,
+      "learning_rate": 0.0002,
+      "loss": 0.9804,
+      "step": 7550
+    },
+    {
+      "epoch": 5.472312703583062,
+      "grad_norm": 1.0740524530410767,
+      "learning_rate": 0.0002,
+      "loss": 0.9856,
+      "step": 7560
+    },
+    {
+      "epoch": 5.479551212450235,
+      "grad_norm": 1.3804482221603394,
+      "learning_rate": 0.0002,
+      "loss": 1.0522,
+      "step": 7570
+    },
+    {
+      "epoch": 5.486789721317408,
+      "grad_norm": 0.9428979754447937,
+      "learning_rate": 0.0002,
+      "loss": 1.0297,
+      "step": 7580
+    },
+    {
+      "epoch": 5.494028230184582,
+      "grad_norm": 0.9548295736312866,
+      "learning_rate": 0.0002,
+      "loss": 1.0906,
+      "step": 7590
+    },
+    {
+      "epoch": 5.501266739051755,
+      "grad_norm": 1.0691065788269043,
+      "learning_rate": 0.0002,
+      "loss": 0.8853,
+      "step": 7600
+    },
+    {
+      "epoch": 5.5085052479189285,
+      "grad_norm": 1.0987380743026733,
+      "learning_rate": 0.0002,
+      "loss": 1.0375,
+      "step": 7610
+    },
+    {
+      "epoch": 5.515743756786102,
+      "grad_norm": 0.9483979344367981,
+      "learning_rate": 0.0002,
+      "loss": 1.0162,
+      "step": 7620
+    },
+    {
+      "epoch": 5.522982265653275,
+      "grad_norm": 1.16624915599823,
+      "learning_rate": 0.0002,
+      "loss": 1.105,
+      "step": 7630
+    },
+    {
+      "epoch": 5.530220774520449,
+      "grad_norm": 0.8563777208328247,
+      "learning_rate": 0.0002,
+      "loss": 0.8695,
+      "step": 7640
+    },
+    {
+      "epoch": 5.537459283387622,
+      "grad_norm": 1.268186092376709,
+      "learning_rate": 0.0002,
+      "loss": 0.9297,
+      "step": 7650
+    },
+    {
+      "epoch": 5.544697792254795,
+      "grad_norm": 1.0752092599868774,
+      "learning_rate": 0.0002,
+      "loss": 1.1152,
+      "step": 7660
+    },
+    {
+      "epoch": 5.551936301121969,
+      "grad_norm": 1.210389256477356,
+      "learning_rate": 0.0002,
+      "loss": 0.9344,
+      "step": 7670
+    },
+    {
+      "epoch": 5.559174809989142,
+      "grad_norm": 1.669063925743103,
+      "learning_rate": 0.0002,
+      "loss": 1.0349,
+      "step": 7680
+    },
+    {
+      "epoch": 5.566413318856315,
+      "grad_norm": 1.038020133972168,
+      "learning_rate": 0.0002,
+      "loss": 0.9833,
+      "step": 7690
+    },
+    {
+      "epoch": 5.573651827723489,
+      "grad_norm": 1.316673994064331,
+      "learning_rate": 0.0002,
+      "loss": 0.8907,
+      "step": 7700
+    },
+    {
+      "epoch": 5.580890336590662,
+      "grad_norm": 1.029935359954834,
+      "learning_rate": 0.0002,
+      "loss": 0.9614,
+      "step": 7710
+    },
+    {
+      "epoch": 5.5881288454578355,
+      "grad_norm": 0.9401940703392029,
+      "learning_rate": 0.0002,
+      "loss": 1.0409,
+      "step": 7720
+    },
+    {
+      "epoch": 5.595367354325009,
+      "grad_norm": 2.4811816215515137,
+      "learning_rate": 0.0002,
+      "loss": 0.9272,
+      "step": 7730
+    },
+    {
+      "epoch": 5.602605863192182,
+      "grad_norm": 1.0329105854034424,
+      "learning_rate": 0.0002,
+      "loss": 0.992,
+      "step": 7740
+    },
+    {
+      "epoch": 5.609844372059356,
+      "grad_norm": 1.479629635810852,
+      "learning_rate": 0.0002,
+      "loss": 0.9493,
+      "step": 7750
+    },
+    {
+      "epoch": 5.617082880926529,
+      "grad_norm": 1.9232319593429565,
+      "learning_rate": 0.0002,
+      "loss": 1.0727,
+      "step": 7760
+    },
+    {
+      "epoch": 5.624321389793702,
+      "grad_norm": 1.0055509805679321,
+      "learning_rate": 0.0002,
+      "loss": 1.0741,
+      "step": 7770
+    },
+    {
+      "epoch": 5.631559898660876,
+      "grad_norm": 1.0037437677383423,
+      "learning_rate": 0.0002,
+      "loss": 1.0731,
+      "step": 7780
+    },
+    {
+      "epoch": 5.638798407528049,
+      "grad_norm": 1.4245030879974365,
+      "learning_rate": 0.0002,
+      "loss": 1.0913,
+      "step": 7790
+    },
+    {
+      "epoch": 5.646036916395222,
+      "grad_norm": 1.080687403678894,
+      "learning_rate": 0.0002,
+      "loss": 0.9711,
+      "step": 7800
+    },
+    {
+      "epoch": 5.653275425262396,
+      "grad_norm": 1.354953408241272,
+      "learning_rate": 0.0002,
+      "loss": 1.0276,
+      "step": 7810
+    },
+    {
+      "epoch": 5.660513934129569,
+      "grad_norm": 0.8966761231422424,
+      "learning_rate": 0.0002,
+      "loss": 1.0534,
+      "step": 7820
+    },
+    {
+      "epoch": 5.6677524429967425,
+      "grad_norm": 1.0675480365753174,
+      "learning_rate": 0.0002,
+      "loss": 1.0662,
+      "step": 7830
+    },
+    {
+      "epoch": 5.674990951863916,
+      "grad_norm": 1.2104216814041138,
+      "learning_rate": 0.0002,
+      "loss": 1.1077,
+      "step": 7840
+    },
+    {
+      "epoch": 5.682229460731089,
+      "grad_norm": 1.105790376663208,
+      "learning_rate": 0.0002,
+      "loss": 0.9627,
+      "step": 7850
+    },
+    {
+      "epoch": 5.689467969598263,
+      "grad_norm": 1.0915391445159912,
+      "learning_rate": 0.0002,
+      "loss": 1.0483,
+      "step": 7860
+    },
+    {
+      "epoch": 5.696706478465436,
+      "grad_norm": 0.8957812786102295,
+      "learning_rate": 0.0002,
+      "loss": 1.0291,
+      "step": 7870
+    },
+    {
+      "epoch": 5.703944987332609,
+      "grad_norm": 1.9189311265945435,
+      "learning_rate": 0.0002,
+      "loss": 0.9785,
+      "step": 7880
+    },
+    {
+      "epoch": 5.711183496199783,
+      "grad_norm": 1.0867321491241455,
+      "learning_rate": 0.0002,
+      "loss": 1.0076,
+      "step": 7890
+    },
+    {
+      "epoch": 5.718422005066956,
+      "grad_norm": 1.0233147144317627,
+      "learning_rate": 0.0002,
+      "loss": 1.0236,
+      "step": 7900
+    },
+    {
+      "epoch": 5.7256605139341294,
+      "grad_norm": 1.16460382938385,
+      "learning_rate": 0.0002,
+      "loss": 0.9872,
+      "step": 7910
+    },
+    {
+      "epoch": 5.732899022801303,
+      "grad_norm": 1.1098358631134033,
+      "learning_rate": 0.0002,
+      "loss": 1.0762,
+      "step": 7920
+    },
+    {
+      "epoch": 5.740137531668476,
+      "grad_norm": 0.8555701375007629,
+      "learning_rate": 0.0002,
+      "loss": 0.9937,
+      "step": 7930
+    },
+    {
+      "epoch": 5.7473760405356495,
+      "grad_norm": 0.9885705709457397,
+      "learning_rate": 0.0002,
+      "loss": 1.0081,
+      "step": 7940
+    },
+    {
+      "epoch": 5.754614549402823,
+      "grad_norm": 0.9184203147888184,
+      "learning_rate": 0.0002,
+      "loss": 0.9909,
+      "step": 7950
+    },
+    {
+      "epoch": 5.761853058269996,
+      "grad_norm": 0.9653698205947876,
+      "learning_rate": 0.0002,
+      "loss": 1.0767,
+      "step": 7960
+    },
+    {
+      "epoch": 5.76909156713717,
+      "grad_norm": 1.0014251470565796,
+      "learning_rate": 0.0002,
+      "loss": 0.9317,
+      "step": 7970
+    },
+    {
+      "epoch": 5.776330076004343,
+      "grad_norm": 1.004701018333435,
+      "learning_rate": 0.0002,
+      "loss": 1.0271,
+      "step": 7980
+    },
+    {
+      "epoch": 5.783568584871516,
+      "grad_norm": 0.950577974319458,
+      "learning_rate": 0.0002,
+      "loss": 1.0397,
+      "step": 7990
+    },
+    {
+      "epoch": 5.79080709373869,
+      "grad_norm": 1.2986834049224854,
+      "learning_rate": 0.0002,
+      "loss": 0.9725,
+      "step": 8000
+    },
+    {
+      "epoch": 5.798045602605863,
+      "grad_norm": 1.3353424072265625,
+      "learning_rate": 0.0002,
+      "loss": 1.039,
+      "step": 8010
+    },
+    {
+      "epoch": 5.8052841114730365,
+      "grad_norm": 0.7650562524795532,
+      "learning_rate": 0.0002,
+      "loss": 1.0626,
+      "step": 8020
+    },
+    {
+      "epoch": 5.81252262034021,
+      "grad_norm": 1.0156235694885254,
+      "learning_rate": 0.0002,
+      "loss": 1.0802,
+      "step": 8030
+    },
+    {
+      "epoch": 5.819761129207383,
+      "grad_norm": 1.3092900514602661,
+      "learning_rate": 0.0002,
+      "loss": 1.0185,
+      "step": 8040
+    },
+    {
+      "epoch": 5.826999638074557,
+      "grad_norm": 1.184428095817566,
+      "learning_rate": 0.0002,
+      "loss": 0.9905,
+      "step": 8050
+    },
+    {
+      "epoch": 5.83423814694173,
+      "grad_norm": 0.979401707649231,
+      "learning_rate": 0.0002,
+      "loss": 1.0548,
+      "step": 8060
+    },
+    {
+      "epoch": 5.841476655808903,
+      "grad_norm": 1.3557400703430176,
+      "learning_rate": 0.0002,
+      "loss": 0.9721,
+      "step": 8070
+    },
+    {
+      "epoch": 5.848715164676077,
+      "grad_norm": 0.8429333567619324,
+      "learning_rate": 0.0002,
+      "loss": 1.0235,
+      "step": 8080
+    },
+    {
+      "epoch": 5.85595367354325,
+      "grad_norm": 1.3167692422866821,
+      "learning_rate": 0.0002,
+      "loss": 0.952,
+      "step": 8090
+    },
+    {
+      "epoch": 5.863192182410423,
+      "grad_norm": 0.9750998020172119,
+      "learning_rate": 0.0002,
+      "loss": 0.9609,
+      "step": 8100
+    },
+    {
+      "epoch": 5.870430691277597,
+      "grad_norm": 1.1869813203811646,
+      "learning_rate": 0.0002,
+      "loss": 1.0789,
+      "step": 8110
+    },
+    {
+      "epoch": 5.87766920014477,
+      "grad_norm": 1.508615255355835,
+      "learning_rate": 0.0002,
+      "loss": 1.0331,
+      "step": 8120
+    },
+    {
+      "epoch": 5.8849077090119435,
+      "grad_norm": 0.9439908266067505,
+      "learning_rate": 0.0002,
+      "loss": 1.0171,
+      "step": 8130
+    },
+    {
+      "epoch": 5.892146217879117,
+      "grad_norm": 0.910508930683136,
+      "learning_rate": 0.0002,
+      "loss": 0.9682,
+      "step": 8140
+    },
+    {
+      "epoch": 5.89938472674629,
+      "grad_norm": 1.111501932144165,
+      "learning_rate": 0.0002,
+      "loss": 1.0032,
+      "step": 8150
+    },
+    {
+      "epoch": 5.906623235613464,
+      "grad_norm": 0.726554274559021,
+      "learning_rate": 0.0002,
+      "loss": 1.0266,
+      "step": 8160
+    },
+    {
+      "epoch": 5.913861744480637,
+      "grad_norm": 1.1084556579589844,
+      "learning_rate": 0.0002,
+      "loss": 1.0681,
+      "step": 8170
+    },
+    {
+      "epoch": 5.92110025334781,
+      "grad_norm": 0.9695167541503906,
+      "learning_rate": 0.0002,
+      "loss": 0.969,
+      "step": 8180
+    },
+    {
+      "epoch": 5.928338762214984,
+      "grad_norm": 1.1169592142105103,
+      "learning_rate": 0.0002,
+      "loss": 0.9858,
+      "step": 8190
+    },
+    {
+      "epoch": 5.935577271082157,
+      "grad_norm": 1.5116780996322632,
+      "learning_rate": 0.0002,
+      "loss": 1.0924,
+      "step": 8200
+    },
+    {
+      "epoch": 5.94281577994933,
+      "grad_norm": 1.0073388814926147,
+      "learning_rate": 0.0002,
+      "loss": 0.878,
+      "step": 8210
+    },
+    {
+      "epoch": 5.950054288816504,
+      "grad_norm": 0.9323263168334961,
+      "learning_rate": 0.0002,
+      "loss": 1.0462,
+      "step": 8220
+    },
+    {
+      "epoch": 5.957292797683677,
+      "grad_norm": 0.9422887563705444,
+      "learning_rate": 0.0002,
+      "loss": 1.0291,
+      "step": 8230
+    },
+    {
+      "epoch": 5.9645313065508505,
+      "grad_norm": 0.9691047668457031,
+      "learning_rate": 0.0002,
+      "loss": 0.953,
+      "step": 8240
+    },
+    {
+      "epoch": 5.971769815418024,
+      "grad_norm": 0.9650622606277466,
+      "learning_rate": 0.0002,
+      "loss": 0.9842,
+      "step": 8250
+    },
+    {
+      "epoch": 5.979008324285197,
+      "grad_norm": 1.077958345413208,
+      "learning_rate": 0.0002,
+      "loss": 0.907,
+      "step": 8260
+    },
+    {
+      "epoch": 5.986246833152371,
+      "grad_norm": 0.8946306109428406,
+      "learning_rate": 0.0002,
+      "loss": 0.9162,
+      "step": 8270
+    },
+    {
+      "epoch": 5.993485342019544,
+      "grad_norm": 1.34098219871521,
+      "learning_rate": 0.0002,
+      "loss": 1.0439,
+      "step": 8280
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 1.4714229106903076,
+      "eval_runtime": 26.301,
+      "eval_samples_per_second": 16.577,
+      "eval_steps_per_second": 2.091,
+      "step": 8289
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 11048,
+  "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": 1.0130812762167706e+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-0.9-num-8828-sd-1/checkpoint-8289/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-0.9-num-8828-sd-1/checkpoint-8289/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..4d7b7431bbbe8c9bf29b925bca391a558af5ff8c
--- /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-0.9-num-8828-sd-1/checkpoint-8289/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ad613885e4f267fc04125f1a836d42cfa796bbe12e536f9ee60c955de02cdb5a
+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-0.9-num-8828-sd-1/checkpoint-9670/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-0.9-num-8828-sd-1/checkpoint-9670/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-0.9-num-8828-sd-1/checkpoint-9670/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-0.9-num-8828-sd-1/checkpoint-9670/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-0.9-num-8828-sd-1/checkpoint-9670/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e99bbcd43df1c19d98706c7e3be95c93844c5349
--- /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-0.9-num-8828-sd-1/checkpoint-9670/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-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-8828-sd-1/checkpoint-9670/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-0.9-num-8828-sd-1/checkpoint-9670/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..5927413cd6210d26978b09f0d6db928597d370cb
--- /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-0.9-num-8828-sd-1/checkpoint-9670/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e7d325e62cb89159e996e41924b13e86460baf25ad8e6335045fc107f85fb39b
+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-0.9-num-8828-sd-1/checkpoint-9670/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-0.9-num-8828-sd-1/checkpoint-9670/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..966f2f94ca868ec2c430c250f47949da6960a17a
--- /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-0.9-num-8828-sd-1/checkpoint-9670/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:03eb9b89691ccce742b81163a460675912dbecdb45aac3bcb0b7f01e7487bc8e
+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-0.9-num-8828-sd-1/checkpoint-9670/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-0.9-num-8828-sd-1/checkpoint-9670/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..dd4137551d3fee184cc9b694c303a190f5778edd
--- /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-0.9-num-8828-sd-1/checkpoint-9670/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:86980a9d744b0b83618f85562e4bb02bf4e224ee1e6dd81b109fdc19b8619627
+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-0.9-num-8828-sd-1/checkpoint-9670/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-0.9-num-8828-sd-1/checkpoint-9670/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..9c74709cc1d4b3eefc8991b311b4299b91d52f6b
--- /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-0.9-num-8828-sd-1/checkpoint-9670/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eef3501f6ad2d312c19551b58b1762e6f84df60507cfb6d4e18ac4e3ca5b0721
+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-0.9-num-8828-sd-1/checkpoint-9670/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-0.9-num-8828-sd-1/checkpoint-9670/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-0.9-num-8828-sd-1/checkpoint-9670/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-0.9-num-8828-sd-1/checkpoint-9670/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-0.9-num-8828-sd-1/checkpoint-9670/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-0.9-num-8828-sd-1/checkpoint-9670/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-0.9-num-8828-sd-1/checkpoint-9670/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-0.9-num-8828-sd-1/checkpoint-9670/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-0.9-num-8828-sd-1/checkpoint-9670/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-0.9-num-8828-sd-1/checkpoint-9670/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-0.9-num-8828-sd-1/checkpoint-9670/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-0.9-num-8828-sd-1/checkpoint-9670/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-0.9-num-8828-sd-1/checkpoint-9670/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-0.9-num-8828-sd-1/checkpoint-9670/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..28a8418758ac0c73331f559afa1f8718048c16fe
--- /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-0.9-num-8828-sd-1/checkpoint-9670/trainer_state.json
@@ -0,0 +1,6858 @@
+{
+  "best_metric": 1.4217946529388428,
+  "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-0.9-num-8828-sd-1/checkpoint-4144",
+  "epoch": 6.999638074556641,
+  "eval_steps": 10,
+  "global_step": 9670,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007238508867173362,
+      "grad_norm": 1.2523442506790161,
+      "learning_rate": 0.0002,
+      "loss": 4.7061,
+      "step": 10
+    },
+    {
+      "epoch": 0.014477017734346724,
+      "grad_norm": 1.8887330293655396,
+      "learning_rate": 0.0002,
+      "loss": 3.3493,
+      "step": 20
+    },
+    {
+      "epoch": 0.021715526601520086,
+      "grad_norm": 0.9668035507202148,
+      "learning_rate": 0.0002,
+      "loss": 2.7585,
+      "step": 30
+    },
+    {
+      "epoch": 0.028954035468693448,
+      "grad_norm": 2.9167306423187256,
+      "learning_rate": 0.0002,
+      "loss": 2.3699,
+      "step": 40
+    },
+    {
+      "epoch": 0.036192544335866814,
+      "grad_norm": 2.649867296218872,
+      "learning_rate": 0.0002,
+      "loss": 2.2679,
+      "step": 50
+    },
+    {
+      "epoch": 0.04343105320304017,
+      "grad_norm": 1.5120655298233032,
+      "learning_rate": 0.0002,
+      "loss": 2.2202,
+      "step": 60
+    },
+    {
+      "epoch": 0.05066956207021354,
+      "grad_norm": 0.7879868149757385,
+      "learning_rate": 0.0002,
+      "loss": 2.2026,
+      "step": 70
+    },
+    {
+      "epoch": 0.057908070937386896,
+      "grad_norm": 0.7616953253746033,
+      "learning_rate": 0.0002,
+      "loss": 1.9447,
+      "step": 80
+    },
+    {
+      "epoch": 0.06514657980456026,
+      "grad_norm": 1.8809149265289307,
+      "learning_rate": 0.0002,
+      "loss": 2.0112,
+      "step": 90
+    },
+    {
+      "epoch": 0.07238508867173363,
+      "grad_norm": 0.9294016361236572,
+      "learning_rate": 0.0002,
+      "loss": 1.8337,
+      "step": 100
+    },
+    {
+      "epoch": 0.07962359753890698,
+      "grad_norm": 0.7145281434059143,
+      "learning_rate": 0.0002,
+      "loss": 1.8419,
+      "step": 110
+    },
+    {
+      "epoch": 0.08686210640608034,
+      "grad_norm": 0.7564446330070496,
+      "learning_rate": 0.0002,
+      "loss": 2.0036,
+      "step": 120
+    },
+    {
+      "epoch": 0.09410061527325371,
+      "grad_norm": 1.1681925058364868,
+      "learning_rate": 0.0002,
+      "loss": 1.9306,
+      "step": 130
+    },
+    {
+      "epoch": 0.10133912414042708,
+      "grad_norm": 0.6708641648292542,
+      "learning_rate": 0.0002,
+      "loss": 1.7875,
+      "step": 140
+    },
+    {
+      "epoch": 0.10857763300760044,
+      "grad_norm": 0.7625647783279419,
+      "learning_rate": 0.0002,
+      "loss": 1.786,
+      "step": 150
+    },
+    {
+      "epoch": 0.11581614187477379,
+      "grad_norm": 0.8463464975357056,
+      "learning_rate": 0.0002,
+      "loss": 1.6687,
+      "step": 160
+    },
+    {
+      "epoch": 0.12305465074194716,
+      "grad_norm": 0.7502335906028748,
+      "learning_rate": 0.0002,
+      "loss": 1.6214,
+      "step": 170
+    },
+    {
+      "epoch": 0.13029315960912052,
+      "grad_norm": 0.6929958462715149,
+      "learning_rate": 0.0002,
+      "loss": 1.7433,
+      "step": 180
+    },
+    {
+      "epoch": 0.1375316684762939,
+      "grad_norm": 0.6798707842826843,
+      "learning_rate": 0.0002,
+      "loss": 1.6009,
+      "step": 190
+    },
+    {
+      "epoch": 0.14477017734346725,
+      "grad_norm": 0.7566508650779724,
+      "learning_rate": 0.0002,
+      "loss": 1.6208,
+      "step": 200
+    },
+    {
+      "epoch": 0.15200868621064062,
+      "grad_norm": 0.7196869850158691,
+      "learning_rate": 0.0002,
+      "loss": 1.5823,
+      "step": 210
+    },
+    {
+      "epoch": 0.15924719507781396,
+      "grad_norm": 0.8401045799255371,
+      "learning_rate": 0.0002,
+      "loss": 1.738,
+      "step": 220
+    },
+    {
+      "epoch": 0.16648570394498732,
+      "grad_norm": 0.8503773212432861,
+      "learning_rate": 0.0002,
+      "loss": 1.7574,
+      "step": 230
+    },
+    {
+      "epoch": 0.1737242128121607,
+      "grad_norm": 0.7183733582496643,
+      "learning_rate": 0.0002,
+      "loss": 1.7861,
+      "step": 240
+    },
+    {
+      "epoch": 0.18096272167933405,
+      "grad_norm": 0.7082605957984924,
+      "learning_rate": 0.0002,
+      "loss": 1.6693,
+      "step": 250
+    },
+    {
+      "epoch": 0.18820123054650742,
+      "grad_norm": 0.9386326670646667,
+      "learning_rate": 0.0002,
+      "loss": 1.619,
+      "step": 260
+    },
+    {
+      "epoch": 0.19543973941368079,
+      "grad_norm": 0.7332451939582825,
+      "learning_rate": 0.0002,
+      "loss": 1.6511,
+      "step": 270
+    },
+    {
+      "epoch": 0.20267824828085415,
+      "grad_norm": 0.7092869877815247,
+      "learning_rate": 0.0002,
+      "loss": 1.6353,
+      "step": 280
+    },
+    {
+      "epoch": 0.20991675714802752,
+      "grad_norm": 0.7256413698196411,
+      "learning_rate": 0.0002,
+      "loss": 1.5996,
+      "step": 290
+    },
+    {
+      "epoch": 0.21715526601520088,
+      "grad_norm": 0.6398681402206421,
+      "learning_rate": 0.0002,
+      "loss": 1.6754,
+      "step": 300
+    },
+    {
+      "epoch": 0.22439377488237422,
+      "grad_norm": 0.6273287534713745,
+      "learning_rate": 0.0002,
+      "loss": 1.397,
+      "step": 310
+    },
+    {
+      "epoch": 0.23163228374954759,
+      "grad_norm": 0.511648416519165,
+      "learning_rate": 0.0002,
+      "loss": 1.5115,
+      "step": 320
+    },
+    {
+      "epoch": 0.23887079261672095,
+      "grad_norm": 0.8677352070808411,
+      "learning_rate": 0.0002,
+      "loss": 1.5424,
+      "step": 330
+    },
+    {
+      "epoch": 0.24610930148389432,
+      "grad_norm": 0.6270743012428284,
+      "learning_rate": 0.0002,
+      "loss": 1.6779,
+      "step": 340
+    },
+    {
+      "epoch": 0.2533478103510677,
+      "grad_norm": 0.7980281114578247,
+      "learning_rate": 0.0002,
+      "loss": 1.626,
+      "step": 350
+    },
+    {
+      "epoch": 0.26058631921824105,
+      "grad_norm": 0.632486879825592,
+      "learning_rate": 0.0002,
+      "loss": 1.5238,
+      "step": 360
+    },
+    {
+      "epoch": 0.2678248280854144,
+      "grad_norm": 0.6527034640312195,
+      "learning_rate": 0.0002,
+      "loss": 1.5175,
+      "step": 370
+    },
+    {
+      "epoch": 0.2750633369525878,
+      "grad_norm": 0.7672118544578552,
+      "learning_rate": 0.0002,
+      "loss": 1.627,
+      "step": 380
+    },
+    {
+      "epoch": 0.28230184581976114,
+      "grad_norm": 0.6035117506980896,
+      "learning_rate": 0.0002,
+      "loss": 1.5605,
+      "step": 390
+    },
+    {
+      "epoch": 0.2895403546869345,
+      "grad_norm": 0.5955103039741516,
+      "learning_rate": 0.0002,
+      "loss": 1.4603,
+      "step": 400
+    },
+    {
+      "epoch": 0.2967788635541079,
+      "grad_norm": 0.6015191674232483,
+      "learning_rate": 0.0002,
+      "loss": 1.558,
+      "step": 410
+    },
+    {
+      "epoch": 0.30401737242128124,
+      "grad_norm": 0.6380982398986816,
+      "learning_rate": 0.0002,
+      "loss": 1.6091,
+      "step": 420
+    },
+    {
+      "epoch": 0.3112558812884546,
+      "grad_norm": 0.6707863211631775,
+      "learning_rate": 0.0002,
+      "loss": 1.5292,
+      "step": 430
+    },
+    {
+      "epoch": 0.3184943901556279,
+      "grad_norm": 0.7010176777839661,
+      "learning_rate": 0.0002,
+      "loss": 1.4426,
+      "step": 440
+    },
+    {
+      "epoch": 0.3257328990228013,
+      "grad_norm": 0.8263739943504333,
+      "learning_rate": 0.0002,
+      "loss": 1.5572,
+      "step": 450
+    },
+    {
+      "epoch": 0.33297140788997465,
+      "grad_norm": 0.7253276109695435,
+      "learning_rate": 0.0002,
+      "loss": 1.5188,
+      "step": 460
+    },
+    {
+      "epoch": 0.340209916757148,
+      "grad_norm": 0.5238934755325317,
+      "learning_rate": 0.0002,
+      "loss": 1.584,
+      "step": 470
+    },
+    {
+      "epoch": 0.3474484256243214,
+      "grad_norm": 0.7869495749473572,
+      "learning_rate": 0.0002,
+      "loss": 1.7035,
+      "step": 480
+    },
+    {
+      "epoch": 0.35468693449149474,
+      "grad_norm": 0.7485215663909912,
+      "learning_rate": 0.0002,
+      "loss": 1.5776,
+      "step": 490
+    },
+    {
+      "epoch": 0.3619254433586681,
+      "grad_norm": 0.5413193106651306,
+      "learning_rate": 0.0002,
+      "loss": 1.6274,
+      "step": 500
+    },
+    {
+      "epoch": 0.3691639522258415,
+      "grad_norm": 0.7615048885345459,
+      "learning_rate": 0.0002,
+      "loss": 1.7323,
+      "step": 510
+    },
+    {
+      "epoch": 0.37640246109301484,
+      "grad_norm": 0.7685340046882629,
+      "learning_rate": 0.0002,
+      "loss": 1.532,
+      "step": 520
+    },
+    {
+      "epoch": 0.3836409699601882,
+      "grad_norm": 0.6379081010818481,
+      "learning_rate": 0.0002,
+      "loss": 1.6312,
+      "step": 530
+    },
+    {
+      "epoch": 0.39087947882736157,
+      "grad_norm": 0.7946939468383789,
+      "learning_rate": 0.0002,
+      "loss": 1.5645,
+      "step": 540
+    },
+    {
+      "epoch": 0.39811798769453494,
+      "grad_norm": 0.6287278532981873,
+      "learning_rate": 0.0002,
+      "loss": 1.4001,
+      "step": 550
+    },
+    {
+      "epoch": 0.4053564965617083,
+      "grad_norm": 0.6811642646789551,
+      "learning_rate": 0.0002,
+      "loss": 1.5982,
+      "step": 560
+    },
+    {
+      "epoch": 0.41259500542888167,
+      "grad_norm": 0.671073317527771,
+      "learning_rate": 0.0002,
+      "loss": 1.4953,
+      "step": 570
+    },
+    {
+      "epoch": 0.41983351429605503,
+      "grad_norm": 0.6313900351524353,
+      "learning_rate": 0.0002,
+      "loss": 1.6753,
+      "step": 580
+    },
+    {
+      "epoch": 0.4270720231632284,
+      "grad_norm": 0.5291772484779358,
+      "learning_rate": 0.0002,
+      "loss": 1.546,
+      "step": 590
+    },
+    {
+      "epoch": 0.43431053203040176,
+      "grad_norm": 0.62503582239151,
+      "learning_rate": 0.0002,
+      "loss": 1.5441,
+      "step": 600
+    },
+    {
+      "epoch": 0.4415490408975751,
+      "grad_norm": 0.5777305364608765,
+      "learning_rate": 0.0002,
+      "loss": 1.6276,
+      "step": 610
+    },
+    {
+      "epoch": 0.44878754976474844,
+      "grad_norm": 0.7013497352600098,
+      "learning_rate": 0.0002,
+      "loss": 1.4758,
+      "step": 620
+    },
+    {
+      "epoch": 0.4560260586319218,
+      "grad_norm": 0.8044822216033936,
+      "learning_rate": 0.0002,
+      "loss": 1.4029,
+      "step": 630
+    },
+    {
+      "epoch": 0.46326456749909517,
+      "grad_norm": 0.672531247138977,
+      "learning_rate": 0.0002,
+      "loss": 1.7195,
+      "step": 640
+    },
+    {
+      "epoch": 0.47050307636626854,
+      "grad_norm": 0.6233910322189331,
+      "learning_rate": 0.0002,
+      "loss": 1.614,
+      "step": 650
+    },
+    {
+      "epoch": 0.4777415852334419,
+      "grad_norm": 0.651524543762207,
+      "learning_rate": 0.0002,
+      "loss": 1.6041,
+      "step": 660
+    },
+    {
+      "epoch": 0.48498009410061527,
+      "grad_norm": 0.7213939428329468,
+      "learning_rate": 0.0002,
+      "loss": 1.5842,
+      "step": 670
+    },
+    {
+      "epoch": 0.49221860296778863,
+      "grad_norm": 0.6541454792022705,
+      "learning_rate": 0.0002,
+      "loss": 1.5453,
+      "step": 680
+    },
+    {
+      "epoch": 0.499457111834962,
+      "grad_norm": 0.6568936109542847,
+      "learning_rate": 0.0002,
+      "loss": 1.662,
+      "step": 690
+    },
+    {
+      "epoch": 0.5066956207021354,
+      "grad_norm": 0.7176415324211121,
+      "learning_rate": 0.0002,
+      "loss": 1.624,
+      "step": 700
+    },
+    {
+      "epoch": 0.5139341295693087,
+      "grad_norm": 0.6553855538368225,
+      "learning_rate": 0.0002,
+      "loss": 1.6099,
+      "step": 710
+    },
+    {
+      "epoch": 0.5211726384364821,
+      "grad_norm": 0.5654335618019104,
+      "learning_rate": 0.0002,
+      "loss": 1.5508,
+      "step": 720
+    },
+    {
+      "epoch": 0.5284111473036555,
+      "grad_norm": 0.5671001672744751,
+      "learning_rate": 0.0002,
+      "loss": 1.392,
+      "step": 730
+    },
+    {
+      "epoch": 0.5356496561708288,
+      "grad_norm": 0.7914412021636963,
+      "learning_rate": 0.0002,
+      "loss": 1.388,
+      "step": 740
+    },
+    {
+      "epoch": 0.5428881650380022,
+      "grad_norm": 0.6172138452529907,
+      "learning_rate": 0.0002,
+      "loss": 1.5931,
+      "step": 750
+    },
+    {
+      "epoch": 0.5501266739051756,
+      "grad_norm": 0.6132623553276062,
+      "learning_rate": 0.0002,
+      "loss": 1.4018,
+      "step": 760
+    },
+    {
+      "epoch": 0.5573651827723489,
+      "grad_norm": 0.654000461101532,
+      "learning_rate": 0.0002,
+      "loss": 1.513,
+      "step": 770
+    },
+    {
+      "epoch": 0.5646036916395223,
+      "grad_norm": 0.5691370964050293,
+      "learning_rate": 0.0002,
+      "loss": 1.5035,
+      "step": 780
+    },
+    {
+      "epoch": 0.5718422005066957,
+      "grad_norm": 0.7922580242156982,
+      "learning_rate": 0.0002,
+      "loss": 1.65,
+      "step": 790
+    },
+    {
+      "epoch": 0.579080709373869,
+      "grad_norm": 0.6831880211830139,
+      "learning_rate": 0.0002,
+      "loss": 1.4521,
+      "step": 800
+    },
+    {
+      "epoch": 0.5863192182410424,
+      "grad_norm": 0.6740124821662903,
+      "learning_rate": 0.0002,
+      "loss": 1.4734,
+      "step": 810
+    },
+    {
+      "epoch": 0.5935577271082157,
+      "grad_norm": 1.380016803741455,
+      "learning_rate": 0.0002,
+      "loss": 1.6498,
+      "step": 820
+    },
+    {
+      "epoch": 0.6007962359753891,
+      "grad_norm": 0.6552878022193909,
+      "learning_rate": 0.0002,
+      "loss": 1.4642,
+      "step": 830
+    },
+    {
+      "epoch": 0.6080347448425625,
+      "grad_norm": 0.6649535298347473,
+      "learning_rate": 0.0002,
+      "loss": 1.6271,
+      "step": 840
+    },
+    {
+      "epoch": 0.6152732537097358,
+      "grad_norm": 0.561738133430481,
+      "learning_rate": 0.0002,
+      "loss": 1.5886,
+      "step": 850
+    },
+    {
+      "epoch": 0.6225117625769092,
+      "grad_norm": 0.6133047938346863,
+      "learning_rate": 0.0002,
+      "loss": 1.5364,
+      "step": 860
+    },
+    {
+      "epoch": 0.6297502714440825,
+      "grad_norm": 0.559843122959137,
+      "learning_rate": 0.0002,
+      "loss": 1.3489,
+      "step": 870
+    },
+    {
+      "epoch": 0.6369887803112558,
+      "grad_norm": 0.6117811799049377,
+      "learning_rate": 0.0002,
+      "loss": 1.4878,
+      "step": 880
+    },
+    {
+      "epoch": 0.6442272891784292,
+      "grad_norm": 0.6209776401519775,
+      "learning_rate": 0.0002,
+      "loss": 1.56,
+      "step": 890
+    },
+    {
+      "epoch": 0.6514657980456026,
+      "grad_norm": 0.6234082579612732,
+      "learning_rate": 0.0002,
+      "loss": 1.6747,
+      "step": 900
+    },
+    {
+      "epoch": 0.6587043069127759,
+      "grad_norm": 0.7623258233070374,
+      "learning_rate": 0.0002,
+      "loss": 1.6963,
+      "step": 910
+    },
+    {
+      "epoch": 0.6659428157799493,
+      "grad_norm": 0.6148061752319336,
+      "learning_rate": 0.0002,
+      "loss": 1.2424,
+      "step": 920
+    },
+    {
+      "epoch": 0.6731813246471227,
+      "grad_norm": 0.6682973504066467,
+      "learning_rate": 0.0002,
+      "loss": 1.4319,
+      "step": 930
+    },
+    {
+      "epoch": 0.680419833514296,
+      "grad_norm": 0.5513041615486145,
+      "learning_rate": 0.0002,
+      "loss": 1.5377,
+      "step": 940
+    },
+    {
+      "epoch": 0.6876583423814694,
+      "grad_norm": 0.5197525024414062,
+      "learning_rate": 0.0002,
+      "loss": 1.3991,
+      "step": 950
+    },
+    {
+      "epoch": 0.6948968512486428,
+      "grad_norm": 0.6490758061408997,
+      "learning_rate": 0.0002,
+      "loss": 1.4398,
+      "step": 960
+    },
+    {
+      "epoch": 0.7021353601158161,
+      "grad_norm": 0.6450682878494263,
+      "learning_rate": 0.0002,
+      "loss": 1.5251,
+      "step": 970
+    },
+    {
+      "epoch": 0.7093738689829895,
+      "grad_norm": 0.6203766465187073,
+      "learning_rate": 0.0002,
+      "loss": 1.5417,
+      "step": 980
+    },
+    {
+      "epoch": 0.7166123778501629,
+      "grad_norm": 0.6023609638214111,
+      "learning_rate": 0.0002,
+      "loss": 1.4575,
+      "step": 990
+    },
+    {
+      "epoch": 0.7238508867173362,
+      "grad_norm": 0.5765255093574524,
+      "learning_rate": 0.0002,
+      "loss": 1.4973,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7310893955845096,
+      "grad_norm": 0.6650075316429138,
+      "learning_rate": 0.0002,
+      "loss": 1.483,
+      "step": 1010
+    },
+    {
+      "epoch": 0.738327904451683,
+      "grad_norm": 0.5610854029655457,
+      "learning_rate": 0.0002,
+      "loss": 1.5959,
+      "step": 1020
+    },
+    {
+      "epoch": 0.7455664133188563,
+      "grad_norm": 0.7072813510894775,
+      "learning_rate": 0.0002,
+      "loss": 1.5248,
+      "step": 1030
+    },
+    {
+      "epoch": 0.7528049221860297,
+      "grad_norm": 0.6815407872200012,
+      "learning_rate": 0.0002,
+      "loss": 1.5776,
+      "step": 1040
+    },
+    {
+      "epoch": 0.760043431053203,
+      "grad_norm": 0.7932390570640564,
+      "learning_rate": 0.0002,
+      "loss": 1.4577,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7672819399203764,
+      "grad_norm": 0.5798183083534241,
+      "learning_rate": 0.0002,
+      "loss": 1.4515,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7745204487875498,
+      "grad_norm": 0.7898504137992859,
+      "learning_rate": 0.0002,
+      "loss": 1.5053,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7817589576547231,
+      "grad_norm": 0.4983280301094055,
+      "learning_rate": 0.0002,
+      "loss": 1.4776,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7889974665218965,
+      "grad_norm": 0.691403329372406,
+      "learning_rate": 0.0002,
+      "loss": 1.5007,
+      "step": 1090
+    },
+    {
+      "epoch": 0.7962359753890699,
+      "grad_norm": 0.5394481420516968,
+      "learning_rate": 0.0002,
+      "loss": 1.5153,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8034744842562432,
+      "grad_norm": 0.5136822462081909,
+      "learning_rate": 0.0002,
+      "loss": 1.6892,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8107129931234166,
+      "grad_norm": 0.6828126907348633,
+      "learning_rate": 0.0002,
+      "loss": 1.4902,
+      "step": 1120
+    },
+    {
+      "epoch": 0.81795150199059,
+      "grad_norm": 0.6799656748771667,
+      "learning_rate": 0.0002,
+      "loss": 1.4346,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8251900108577633,
+      "grad_norm": 0.5428406000137329,
+      "learning_rate": 0.0002,
+      "loss": 1.2678,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8324285197249367,
+      "grad_norm": 0.4811290502548218,
+      "learning_rate": 0.0002,
+      "loss": 1.4072,
+      "step": 1150
+    },
+    {
+      "epoch": 0.8396670285921101,
+      "grad_norm": 0.5519434809684753,
+      "learning_rate": 0.0002,
+      "loss": 1.4512,
+      "step": 1160
+    },
+    {
+      "epoch": 0.8469055374592834,
+      "grad_norm": 0.9748060703277588,
+      "learning_rate": 0.0002,
+      "loss": 1.4072,
+      "step": 1170
+    },
+    {
+      "epoch": 0.8541440463264568,
+      "grad_norm": 0.712609589099884,
+      "learning_rate": 0.0002,
+      "loss": 1.4309,
+      "step": 1180
+    },
+    {
+      "epoch": 0.8613825551936302,
+      "grad_norm": 0.6866157054901123,
+      "learning_rate": 0.0002,
+      "loss": 1.434,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8686210640608035,
+      "grad_norm": 0.5068854093551636,
+      "learning_rate": 0.0002,
+      "loss": 1.3704,
+      "step": 1200
+    },
+    {
+      "epoch": 0.8758595729279768,
+      "grad_norm": 0.6333245038986206,
+      "learning_rate": 0.0002,
+      "loss": 1.5601,
+      "step": 1210
+    },
+    {
+      "epoch": 0.8830980817951501,
+      "grad_norm": 0.6424421072006226,
+      "learning_rate": 0.0002,
+      "loss": 1.4636,
+      "step": 1220
+    },
+    {
+      "epoch": 0.8903365906623235,
+      "grad_norm": 0.4771921932697296,
+      "learning_rate": 0.0002,
+      "loss": 1.4186,
+      "step": 1230
+    },
+    {
+      "epoch": 0.8975750995294969,
+      "grad_norm": 0.5191764235496521,
+      "learning_rate": 0.0002,
+      "loss": 1.6323,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9048136083966702,
+      "grad_norm": 0.756222128868103,
+      "learning_rate": 0.0002,
+      "loss": 1.6105,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9120521172638436,
+      "grad_norm": 0.623823881149292,
+      "learning_rate": 0.0002,
+      "loss": 1.4396,
+      "step": 1260
+    },
+    {
+      "epoch": 0.919290626131017,
+      "grad_norm": 0.8166571259498596,
+      "learning_rate": 0.0002,
+      "loss": 1.3097,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9265291349981903,
+      "grad_norm": 0.6059346795082092,
+      "learning_rate": 0.0002,
+      "loss": 1.4625,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9337676438653637,
+      "grad_norm": 0.5842690467834473,
+      "learning_rate": 0.0002,
+      "loss": 1.3555,
+      "step": 1290
+    },
+    {
+      "epoch": 0.9410061527325371,
+      "grad_norm": 0.7649800777435303,
+      "learning_rate": 0.0002,
+      "loss": 1.5859,
+      "step": 1300
+    },
+    {
+      "epoch": 0.9482446615997104,
+      "grad_norm": 0.6420919895172119,
+      "learning_rate": 0.0002,
+      "loss": 1.5915,
+      "step": 1310
+    },
+    {
+      "epoch": 0.9554831704668838,
+      "grad_norm": 0.7011452913284302,
+      "learning_rate": 0.0002,
+      "loss": 1.453,
+      "step": 1320
+    },
+    {
+      "epoch": 0.9627216793340572,
+      "grad_norm": 0.5783746242523193,
+      "learning_rate": 0.0002,
+      "loss": 1.6766,
+      "step": 1330
+    },
+    {
+      "epoch": 0.9699601882012305,
+      "grad_norm": 0.5973192453384399,
+      "learning_rate": 0.0002,
+      "loss": 1.6308,
+      "step": 1340
+    },
+    {
+      "epoch": 0.9771986970684039,
+      "grad_norm": 0.6181833744049072,
+      "learning_rate": 0.0002,
+      "loss": 1.5901,
+      "step": 1350
+    },
+    {
+      "epoch": 0.9844372059355773,
+      "grad_norm": 0.5563396215438843,
+      "learning_rate": 0.0002,
+      "loss": 1.5258,
+      "step": 1360
+    },
+    {
+      "epoch": 0.9916757148027506,
+      "grad_norm": 0.45723360776901245,
+      "learning_rate": 0.0002,
+      "loss": 1.4508,
+      "step": 1370
+    },
+    {
+      "epoch": 0.998914223669924,
+      "grad_norm": 0.5947498679161072,
+      "learning_rate": 0.0002,
+      "loss": 1.3291,
+      "step": 1380
+    },
+    {
+      "epoch": 0.9996380745566413,
+      "eval_loss": 1.480796456336975,
+      "eval_runtime": 27.3103,
+      "eval_samples_per_second": 15.965,
+      "eval_steps_per_second": 2.014,
+      "step": 1381
+    },
+    {
+      "epoch": 1.0061527325370974,
+      "grad_norm": 0.5599952936172485,
+      "learning_rate": 0.0002,
+      "loss": 1.3057,
+      "step": 1390
+    },
+    {
+      "epoch": 1.0133912414042707,
+      "grad_norm": 0.5932008028030396,
+      "learning_rate": 0.0002,
+      "loss": 1.4991,
+      "step": 1400
+    },
+    {
+      "epoch": 1.020629750271444,
+      "grad_norm": 0.6194121837615967,
+      "learning_rate": 0.0002,
+      "loss": 1.4506,
+      "step": 1410
+    },
+    {
+      "epoch": 1.0278682591386175,
+      "grad_norm": 0.6995621919631958,
+      "learning_rate": 0.0002,
+      "loss": 1.5966,
+      "step": 1420
+    },
+    {
+      "epoch": 1.0351067680057908,
+      "grad_norm": 0.7905810475349426,
+      "learning_rate": 0.0002,
+      "loss": 1.4153,
+      "step": 1430
+    },
+    {
+      "epoch": 1.0423452768729642,
+      "grad_norm": 0.7221615314483643,
+      "learning_rate": 0.0002,
+      "loss": 1.4414,
+      "step": 1440
+    },
+    {
+      "epoch": 1.0495837857401376,
+      "grad_norm": 0.6170642375946045,
+      "learning_rate": 0.0002,
+      "loss": 1.3859,
+      "step": 1450
+    },
+    {
+      "epoch": 1.056822294607311,
+      "grad_norm": 0.5844094753265381,
+      "learning_rate": 0.0002,
+      "loss": 1.3806,
+      "step": 1460
+    },
+    {
+      "epoch": 1.0640608034744843,
+      "grad_norm": 0.7731822729110718,
+      "learning_rate": 0.0002,
+      "loss": 1.4871,
+      "step": 1470
+    },
+    {
+      "epoch": 1.0712993123416577,
+      "grad_norm": 0.4554748237133026,
+      "learning_rate": 0.0002,
+      "loss": 1.4286,
+      "step": 1480
+    },
+    {
+      "epoch": 1.078537821208831,
+      "grad_norm": 0.6923259496688843,
+      "learning_rate": 0.0002,
+      "loss": 1.3977,
+      "step": 1490
+    },
+    {
+      "epoch": 1.0857763300760044,
+      "grad_norm": 0.6008219122886658,
+      "learning_rate": 0.0002,
+      "loss": 1.3936,
+      "step": 1500
+    },
+    {
+      "epoch": 1.0930148389431777,
+      "grad_norm": 0.6450045704841614,
+      "learning_rate": 0.0002,
+      "loss": 1.4821,
+      "step": 1510
+    },
+    {
+      "epoch": 1.1002533478103511,
+      "grad_norm": 0.7833753824234009,
+      "learning_rate": 0.0002,
+      "loss": 1.3295,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1074918566775245,
+      "grad_norm": 0.5076758861541748,
+      "learning_rate": 0.0002,
+      "loss": 1.3424,
+      "step": 1530
+    },
+    {
+      "epoch": 1.1147303655446978,
+      "grad_norm": 0.5661332011222839,
+      "learning_rate": 0.0002,
+      "loss": 1.4043,
+      "step": 1540
+    },
+    {
+      "epoch": 1.1219688744118712,
+      "grad_norm": 0.6526919603347778,
+      "learning_rate": 0.0002,
+      "loss": 1.4963,
+      "step": 1550
+    },
+    {
+      "epoch": 1.1292073832790446,
+      "grad_norm": 0.5613082647323608,
+      "learning_rate": 0.0002,
+      "loss": 1.3671,
+      "step": 1560
+    },
+    {
+      "epoch": 1.136445892146218,
+      "grad_norm": 0.6113885641098022,
+      "learning_rate": 0.0002,
+      "loss": 1.4458,
+      "step": 1570
+    },
+    {
+      "epoch": 1.1436844010133913,
+      "grad_norm": 0.6732510328292847,
+      "learning_rate": 0.0002,
+      "loss": 1.3552,
+      "step": 1580
+    },
+    {
+      "epoch": 1.1509229098805647,
+      "grad_norm": 0.6146392226219177,
+      "learning_rate": 0.0002,
+      "loss": 1.3114,
+      "step": 1590
+    },
+    {
+      "epoch": 1.158161418747738,
+      "grad_norm": 0.6766974329948425,
+      "learning_rate": 0.0002,
+      "loss": 1.411,
+      "step": 1600
+    },
+    {
+      "epoch": 1.1653999276149114,
+      "grad_norm": 0.7621957659721375,
+      "learning_rate": 0.0002,
+      "loss": 1.2401,
+      "step": 1610
+    },
+    {
+      "epoch": 1.1726384364820848,
+      "grad_norm": 0.6959581971168518,
+      "learning_rate": 0.0002,
+      "loss": 1.3758,
+      "step": 1620
+    },
+    {
+      "epoch": 1.1798769453492581,
+      "grad_norm": 0.6691278219223022,
+      "learning_rate": 0.0002,
+      "loss": 1.382,
+      "step": 1630
+    },
+    {
+      "epoch": 1.1871154542164315,
+      "grad_norm": 0.4927774965763092,
+      "learning_rate": 0.0002,
+      "loss": 1.4147,
+      "step": 1640
+    },
+    {
+      "epoch": 1.1943539630836049,
+      "grad_norm": 0.7724234461784363,
+      "learning_rate": 0.0002,
+      "loss": 1.449,
+      "step": 1650
+    },
+    {
+      "epoch": 1.2015924719507782,
+      "grad_norm": 0.6817787885665894,
+      "learning_rate": 0.0002,
+      "loss": 1.4778,
+      "step": 1660
+    },
+    {
+      "epoch": 1.2088309808179516,
+      "grad_norm": 0.6500699520111084,
+      "learning_rate": 0.0002,
+      "loss": 1.3776,
+      "step": 1670
+    },
+    {
+      "epoch": 1.216069489685125,
+      "grad_norm": 0.5703568458557129,
+      "learning_rate": 0.0002,
+      "loss": 1.3875,
+      "step": 1680
+    },
+    {
+      "epoch": 1.2233079985522983,
+      "grad_norm": 0.6261579990386963,
+      "learning_rate": 0.0002,
+      "loss": 1.4735,
+      "step": 1690
+    },
+    {
+      "epoch": 1.2305465074194717,
+      "grad_norm": 0.651713490486145,
+      "learning_rate": 0.0002,
+      "loss": 1.3898,
+      "step": 1700
+    },
+    {
+      "epoch": 1.237785016286645,
+      "grad_norm": 0.684399425983429,
+      "learning_rate": 0.0002,
+      "loss": 1.4002,
+      "step": 1710
+    },
+    {
+      "epoch": 1.2450235251538184,
+      "grad_norm": 0.6996857523918152,
+      "learning_rate": 0.0002,
+      "loss": 1.5027,
+      "step": 1720
+    },
+    {
+      "epoch": 1.2522620340209918,
+      "grad_norm": 0.7102537751197815,
+      "learning_rate": 0.0002,
+      "loss": 1.3326,
+      "step": 1730
+    },
+    {
+      "epoch": 1.2595005428881652,
+      "grad_norm": 0.45809897780418396,
+      "learning_rate": 0.0002,
+      "loss": 1.3675,
+      "step": 1740
+    },
+    {
+      "epoch": 1.2667390517553385,
+      "grad_norm": 0.6377046704292297,
+      "learning_rate": 0.0002,
+      "loss": 1.4175,
+      "step": 1750
+    },
+    {
+      "epoch": 1.2739775606225119,
+      "grad_norm": 0.6965704560279846,
+      "learning_rate": 0.0002,
+      "loss": 1.3479,
+      "step": 1760
+    },
+    {
+      "epoch": 1.2812160694896852,
+      "grad_norm": 0.5688214302062988,
+      "learning_rate": 0.0002,
+      "loss": 1.5647,
+      "step": 1770
+    },
+    {
+      "epoch": 1.2884545783568586,
+      "grad_norm": 0.6384190320968628,
+      "learning_rate": 0.0002,
+      "loss": 1.3967,
+      "step": 1780
+    },
+    {
+      "epoch": 1.295693087224032,
+      "grad_norm": 0.5629363656044006,
+      "learning_rate": 0.0002,
+      "loss": 1.3671,
+      "step": 1790
+    },
+    {
+      "epoch": 1.3029315960912053,
+      "grad_norm": 0.6148255467414856,
+      "learning_rate": 0.0002,
+      "loss": 1.2292,
+      "step": 1800
+    },
+    {
+      "epoch": 1.3101701049583787,
+      "grad_norm": 0.655580997467041,
+      "learning_rate": 0.0002,
+      "loss": 1.5806,
+      "step": 1810
+    },
+    {
+      "epoch": 1.3174086138255519,
+      "grad_norm": 0.5642657279968262,
+      "learning_rate": 0.0002,
+      "loss": 1.2398,
+      "step": 1820
+    },
+    {
+      "epoch": 1.3246471226927252,
+      "grad_norm": 0.59607994556427,
+      "learning_rate": 0.0002,
+      "loss": 1.3246,
+      "step": 1830
+    },
+    {
+      "epoch": 1.3318856315598986,
+      "grad_norm": 0.5564199090003967,
+      "learning_rate": 0.0002,
+      "loss": 1.3274,
+      "step": 1840
+    },
+    {
+      "epoch": 1.339124140427072,
+      "grad_norm": 0.6949955821037292,
+      "learning_rate": 0.0002,
+      "loss": 1.5834,
+      "step": 1850
+    },
+    {
+      "epoch": 1.3463626492942453,
+      "grad_norm": 0.7036856412887573,
+      "learning_rate": 0.0002,
+      "loss": 1.4722,
+      "step": 1860
+    },
+    {
+      "epoch": 1.3536011581614187,
+      "grad_norm": 0.722062885761261,
+      "learning_rate": 0.0002,
+      "loss": 1.333,
+      "step": 1870
+    },
+    {
+      "epoch": 1.360839667028592,
+      "grad_norm": 0.6098677515983582,
+      "learning_rate": 0.0002,
+      "loss": 1.4044,
+      "step": 1880
+    },
+    {
+      "epoch": 1.3680781758957654,
+      "grad_norm": 0.5376402735710144,
+      "learning_rate": 0.0002,
+      "loss": 1.6217,
+      "step": 1890
+    },
+    {
+      "epoch": 1.3753166847629388,
+      "grad_norm": 0.6974610090255737,
+      "learning_rate": 0.0002,
+      "loss": 1.5071,
+      "step": 1900
+    },
+    {
+      "epoch": 1.3825551936301121,
+      "grad_norm": 0.6520763635635376,
+      "learning_rate": 0.0002,
+      "loss": 1.5854,
+      "step": 1910
+    },
+    {
+      "epoch": 1.3897937024972855,
+      "grad_norm": 0.6604374647140503,
+      "learning_rate": 0.0002,
+      "loss": 1.4271,
+      "step": 1920
+    },
+    {
+      "epoch": 1.3970322113644589,
+      "grad_norm": 0.7364398241043091,
+      "learning_rate": 0.0002,
+      "loss": 1.419,
+      "step": 1930
+    },
+    {
+      "epoch": 1.4042707202316322,
+      "grad_norm": 0.6849475502967834,
+      "learning_rate": 0.0002,
+      "loss": 1.4585,
+      "step": 1940
+    },
+    {
+      "epoch": 1.4115092290988056,
+      "grad_norm": 0.6562670469284058,
+      "learning_rate": 0.0002,
+      "loss": 1.5577,
+      "step": 1950
+    },
+    {
+      "epoch": 1.418747737965979,
+      "grad_norm": 0.5695616006851196,
+      "learning_rate": 0.0002,
+      "loss": 1.4725,
+      "step": 1960
+    },
+    {
+      "epoch": 1.4259862468331523,
+      "grad_norm": 0.5244464874267578,
+      "learning_rate": 0.0002,
+      "loss": 1.3088,
+      "step": 1970
+    },
+    {
+      "epoch": 1.4332247557003257,
+      "grad_norm": 0.6347293257713318,
+      "learning_rate": 0.0002,
+      "loss": 1.5069,
+      "step": 1980
+    },
+    {
+      "epoch": 1.440463264567499,
+      "grad_norm": 0.5528361201286316,
+      "learning_rate": 0.0002,
+      "loss": 1.3502,
+      "step": 1990
+    },
+    {
+      "epoch": 1.4477017734346724,
+      "grad_norm": 0.6987585425376892,
+      "learning_rate": 0.0002,
+      "loss": 1.3978,
+      "step": 2000
+    },
+    {
+      "epoch": 1.4549402823018458,
+      "grad_norm": 0.6568987369537354,
+      "learning_rate": 0.0002,
+      "loss": 1.4262,
+      "step": 2010
+    },
+    {
+      "epoch": 1.4621787911690192,
+      "grad_norm": 0.7665994763374329,
+      "learning_rate": 0.0002,
+      "loss": 1.4175,
+      "step": 2020
+    },
+    {
+      "epoch": 1.4694173000361925,
+      "grad_norm": 0.5127707123756409,
+      "learning_rate": 0.0002,
+      "loss": 1.244,
+      "step": 2030
+    },
+    {
+      "epoch": 1.476655808903366,
+      "grad_norm": 0.5406824946403503,
+      "learning_rate": 0.0002,
+      "loss": 1.3699,
+      "step": 2040
+    },
+    {
+      "epoch": 1.4838943177705393,
+      "grad_norm": 0.5990166664123535,
+      "learning_rate": 0.0002,
+      "loss": 1.3353,
+      "step": 2050
+    },
+    {
+      "epoch": 1.4911328266377126,
+      "grad_norm": 0.6186193823814392,
+      "learning_rate": 0.0002,
+      "loss": 1.2454,
+      "step": 2060
+    },
+    {
+      "epoch": 1.498371335504886,
+      "grad_norm": 0.6154307126998901,
+      "learning_rate": 0.0002,
+      "loss": 1.428,
+      "step": 2070
+    },
+    {
+      "epoch": 1.5056098443720594,
+      "grad_norm": 0.5606056451797485,
+      "learning_rate": 0.0002,
+      "loss": 1.4528,
+      "step": 2080
+    },
+    {
+      "epoch": 1.5128483532392327,
+      "grad_norm": 0.5006417036056519,
+      "learning_rate": 0.0002,
+      "loss": 1.2405,
+      "step": 2090
+    },
+    {
+      "epoch": 1.520086862106406,
+      "grad_norm": 0.5968486070632935,
+      "learning_rate": 0.0002,
+      "loss": 1.4258,
+      "step": 2100
+    },
+    {
+      "epoch": 1.5273253709735795,
+      "grad_norm": 0.5835496187210083,
+      "learning_rate": 0.0002,
+      "loss": 1.2752,
+      "step": 2110
+    },
+    {
+      "epoch": 1.5345638798407528,
+      "grad_norm": 0.6753535270690918,
+      "learning_rate": 0.0002,
+      "loss": 1.5443,
+      "step": 2120
+    },
+    {
+      "epoch": 1.5418023887079262,
+      "grad_norm": 0.7299720644950867,
+      "learning_rate": 0.0002,
+      "loss": 1.2139,
+      "step": 2130
+    },
+    {
+      "epoch": 1.5490408975750996,
+      "grad_norm": 0.5105988383293152,
+      "learning_rate": 0.0002,
+      "loss": 1.2364,
+      "step": 2140
+    },
+    {
+      "epoch": 1.556279406442273,
+      "grad_norm": 0.5675431489944458,
+      "learning_rate": 0.0002,
+      "loss": 1.4528,
+      "step": 2150
+    },
+    {
+      "epoch": 1.5635179153094463,
+      "grad_norm": 0.6246723532676697,
+      "learning_rate": 0.0002,
+      "loss": 1.4563,
+      "step": 2160
+    },
+    {
+      "epoch": 1.5707564241766196,
+      "grad_norm": 0.7291720509529114,
+      "learning_rate": 0.0002,
+      "loss": 1.5255,
+      "step": 2170
+    },
+    {
+      "epoch": 1.577994933043793,
+      "grad_norm": 0.678114116191864,
+      "learning_rate": 0.0002,
+      "loss": 1.5432,
+      "step": 2180
+    },
+    {
+      "epoch": 1.5852334419109664,
+      "grad_norm": 0.5136260986328125,
+      "learning_rate": 0.0002,
+      "loss": 1.5212,
+      "step": 2190
+    },
+    {
+      "epoch": 1.5924719507781397,
+      "grad_norm": 0.6359935998916626,
+      "learning_rate": 0.0002,
+      "loss": 1.3271,
+      "step": 2200
+    },
+    {
+      "epoch": 1.599710459645313,
+      "grad_norm": 0.7650278806686401,
+      "learning_rate": 0.0002,
+      "loss": 1.4038,
+      "step": 2210
+    },
+    {
+      "epoch": 1.6069489685124865,
+      "grad_norm": 0.7256110906600952,
+      "learning_rate": 0.0002,
+      "loss": 1.5478,
+      "step": 2220
+    },
+    {
+      "epoch": 1.6141874773796598,
+      "grad_norm": 0.688689649105072,
+      "learning_rate": 0.0002,
+      "loss": 1.4387,
+      "step": 2230
+    },
+    {
+      "epoch": 1.6214259862468332,
+      "grad_norm": 0.6045311093330383,
+      "learning_rate": 0.0002,
+      "loss": 1.4096,
+      "step": 2240
+    },
+    {
+      "epoch": 1.6286644951140063,
+      "grad_norm": 0.7064604163169861,
+      "learning_rate": 0.0002,
+      "loss": 1.4097,
+      "step": 2250
+    },
+    {
+      "epoch": 1.6359030039811797,
+      "grad_norm": 0.5309562087059021,
+      "learning_rate": 0.0002,
+      "loss": 1.3477,
+      "step": 2260
+    },
+    {
+      "epoch": 1.643141512848353,
+      "grad_norm": 0.5687053203582764,
+      "learning_rate": 0.0002,
+      "loss": 1.4022,
+      "step": 2270
+    },
+    {
+      "epoch": 1.6503800217155264,
+      "grad_norm": 0.535872757434845,
+      "learning_rate": 0.0002,
+      "loss": 1.2977,
+      "step": 2280
+    },
+    {
+      "epoch": 1.6576185305826998,
+      "grad_norm": 0.5502381920814514,
+      "learning_rate": 0.0002,
+      "loss": 1.3844,
+      "step": 2290
+    },
+    {
+      "epoch": 1.6648570394498732,
+      "grad_norm": 0.6158602237701416,
+      "learning_rate": 0.0002,
+      "loss": 1.3764,
+      "step": 2300
+    },
+    {
+      "epoch": 1.6720955483170465,
+      "grad_norm": 0.5804675817489624,
+      "learning_rate": 0.0002,
+      "loss": 1.3515,
+      "step": 2310
+    },
+    {
+      "epoch": 1.67933405718422,
+      "grad_norm": 0.600742757320404,
+      "learning_rate": 0.0002,
+      "loss": 1.2532,
+      "step": 2320
+    },
+    {
+      "epoch": 1.6865725660513933,
+      "grad_norm": 0.7101941108703613,
+      "learning_rate": 0.0002,
+      "loss": 1.477,
+      "step": 2330
+    },
+    {
+      "epoch": 1.6938110749185666,
+      "grad_norm": 0.7507809996604919,
+      "learning_rate": 0.0002,
+      "loss": 1.4849,
+      "step": 2340
+    },
+    {
+      "epoch": 1.70104958378574,
+      "grad_norm": 0.768502414226532,
+      "learning_rate": 0.0002,
+      "loss": 1.2703,
+      "step": 2350
+    },
+    {
+      "epoch": 1.7082880926529134,
+      "grad_norm": 0.4801851212978363,
+      "learning_rate": 0.0002,
+      "loss": 1.3332,
+      "step": 2360
+    },
+    {
+      "epoch": 1.7155266015200867,
+      "grad_norm": 0.5322122573852539,
+      "learning_rate": 0.0002,
+      "loss": 1.4158,
+      "step": 2370
+    },
+    {
+      "epoch": 1.72276511038726,
+      "grad_norm": 0.587661862373352,
+      "learning_rate": 0.0002,
+      "loss": 1.4136,
+      "step": 2380
+    },
+    {
+      "epoch": 1.7300036192544335,
+      "grad_norm": 0.6073525547981262,
+      "learning_rate": 0.0002,
+      "loss": 1.3771,
+      "step": 2390
+    },
+    {
+      "epoch": 1.7372421281216068,
+      "grad_norm": 0.6950460076332092,
+      "learning_rate": 0.0002,
+      "loss": 1.2754,
+      "step": 2400
+    },
+    {
+      "epoch": 1.7444806369887802,
+      "grad_norm": 0.5981102585792542,
+      "learning_rate": 0.0002,
+      "loss": 1.3858,
+      "step": 2410
+    },
+    {
+      "epoch": 1.7517191458559536,
+      "grad_norm": 0.544570803642273,
+      "learning_rate": 0.0002,
+      "loss": 1.4075,
+      "step": 2420
+    },
+    {
+      "epoch": 1.758957654723127,
+      "grad_norm": 0.5304399728775024,
+      "learning_rate": 0.0002,
+      "loss": 1.3861,
+      "step": 2430
+    },
+    {
+      "epoch": 1.7661961635903003,
+      "grad_norm": 0.7921594977378845,
+      "learning_rate": 0.0002,
+      "loss": 1.4244,
+      "step": 2440
+    },
+    {
+      "epoch": 1.7734346724574737,
+      "grad_norm": 0.6084808707237244,
+      "learning_rate": 0.0002,
+      "loss": 1.3053,
+      "step": 2450
+    },
+    {
+      "epoch": 1.780673181324647,
+      "grad_norm": 0.8844701051712036,
+      "learning_rate": 0.0002,
+      "loss": 1.3781,
+      "step": 2460
+    },
+    {
+      "epoch": 1.7879116901918204,
+      "grad_norm": 0.5729258060455322,
+      "learning_rate": 0.0002,
+      "loss": 1.3227,
+      "step": 2470
+    },
+    {
+      "epoch": 1.7951501990589938,
+      "grad_norm": 0.6303611993789673,
+      "learning_rate": 0.0002,
+      "loss": 1.3422,
+      "step": 2480
+    },
+    {
+      "epoch": 1.8023887079261671,
+      "grad_norm": 0.5627942085266113,
+      "learning_rate": 0.0002,
+      "loss": 1.3926,
+      "step": 2490
+    },
+    {
+      "epoch": 1.8096272167933405,
+      "grad_norm": 0.6724274158477783,
+      "learning_rate": 0.0002,
+      "loss": 1.3816,
+      "step": 2500
+    },
+    {
+      "epoch": 1.8168657256605139,
+      "grad_norm": 0.5030826330184937,
+      "learning_rate": 0.0002,
+      "loss": 1.2951,
+      "step": 2510
+    },
+    {
+      "epoch": 1.8241042345276872,
+      "grad_norm": 0.5504099130630493,
+      "learning_rate": 0.0002,
+      "loss": 1.2839,
+      "step": 2520
+    },
+    {
+      "epoch": 1.8313427433948606,
+      "grad_norm": 0.6338945627212524,
+      "learning_rate": 0.0002,
+      "loss": 1.4264,
+      "step": 2530
+    },
+    {
+      "epoch": 1.838581252262034,
+      "grad_norm": 0.5902037620544434,
+      "learning_rate": 0.0002,
+      "loss": 1.563,
+      "step": 2540
+    },
+    {
+      "epoch": 1.8458197611292073,
+      "grad_norm": 0.48814457654953003,
+      "learning_rate": 0.0002,
+      "loss": 1.2961,
+      "step": 2550
+    },
+    {
+      "epoch": 1.8530582699963807,
+      "grad_norm": 0.6216312646865845,
+      "learning_rate": 0.0002,
+      "loss": 1.466,
+      "step": 2560
+    },
+    {
+      "epoch": 1.860296778863554,
+      "grad_norm": 0.635603666305542,
+      "learning_rate": 0.0002,
+      "loss": 1.5123,
+      "step": 2570
+    },
+    {
+      "epoch": 1.8675352877307274,
+      "grad_norm": 0.6938216090202332,
+      "learning_rate": 0.0002,
+      "loss": 1.372,
+      "step": 2580
+    },
+    {
+      "epoch": 1.8747737965979008,
+      "grad_norm": 0.599557638168335,
+      "learning_rate": 0.0002,
+      "loss": 1.5011,
+      "step": 2590
+    },
+    {
+      "epoch": 1.8820123054650741,
+      "grad_norm": 0.564424455165863,
+      "learning_rate": 0.0002,
+      "loss": 1.2714,
+      "step": 2600
+    },
+    {
+      "epoch": 1.8892508143322475,
+      "grad_norm": 0.5430700182914734,
+      "learning_rate": 0.0002,
+      "loss": 1.3403,
+      "step": 2610
+    },
+    {
+      "epoch": 1.8964893231994209,
+      "grad_norm": 0.6150169372558594,
+      "learning_rate": 0.0002,
+      "loss": 1.4347,
+      "step": 2620
+    },
+    {
+      "epoch": 1.9037278320665942,
+      "grad_norm": 0.48159119486808777,
+      "learning_rate": 0.0002,
+      "loss": 1.2474,
+      "step": 2630
+    },
+    {
+      "epoch": 1.9109663409337676,
+      "grad_norm": 0.5608997941017151,
+      "learning_rate": 0.0002,
+      "loss": 1.3716,
+      "step": 2640
+    },
+    {
+      "epoch": 1.918204849800941,
+      "grad_norm": 0.6454501748085022,
+      "learning_rate": 0.0002,
+      "loss": 1.5787,
+      "step": 2650
+    },
+    {
+      "epoch": 1.9254433586681143,
+      "grad_norm": 0.5458073616027832,
+      "learning_rate": 0.0002,
+      "loss": 1.3238,
+      "step": 2660
+    },
+    {
+      "epoch": 1.9326818675352877,
+      "grad_norm": 0.5328490734100342,
+      "learning_rate": 0.0002,
+      "loss": 1.3208,
+      "step": 2670
+    },
+    {
+      "epoch": 1.939920376402461,
+      "grad_norm": 0.6444696187973022,
+      "learning_rate": 0.0002,
+      "loss": 1.4971,
+      "step": 2680
+    },
+    {
+      "epoch": 1.9471588852696344,
+      "grad_norm": 0.7126023769378662,
+      "learning_rate": 0.0002,
+      "loss": 1.5387,
+      "step": 2690
+    },
+    {
+      "epoch": 1.9543973941368078,
+      "grad_norm": 0.5164045095443726,
+      "learning_rate": 0.0002,
+      "loss": 1.3637,
+      "step": 2700
+    },
+    {
+      "epoch": 1.9616359030039812,
+      "grad_norm": 0.5347061157226562,
+      "learning_rate": 0.0002,
+      "loss": 1.5303,
+      "step": 2710
+    },
+    {
+      "epoch": 1.9688744118711545,
+      "grad_norm": 0.5297950506210327,
+      "learning_rate": 0.0002,
+      "loss": 1.2815,
+      "step": 2720
+    },
+    {
+      "epoch": 1.976112920738328,
+      "grad_norm": 0.6537790298461914,
+      "learning_rate": 0.0002,
+      "loss": 1.3566,
+      "step": 2730
+    },
+    {
+      "epoch": 1.9833514296055013,
+      "grad_norm": 0.5536222457885742,
+      "learning_rate": 0.0002,
+      "loss": 1.332,
+      "step": 2740
+    },
+    {
+      "epoch": 1.9905899384726746,
+      "grad_norm": 0.4856105446815491,
+      "learning_rate": 0.0002,
+      "loss": 1.3333,
+      "step": 2750
+    },
+    {
+      "epoch": 1.997828447339848,
+      "grad_norm": 0.6642730832099915,
+      "learning_rate": 0.0002,
+      "loss": 1.3521,
+      "step": 2760
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.4366681575775146,
+      "eval_runtime": 27.3729,
+      "eval_samples_per_second": 15.928,
+      "eval_steps_per_second": 2.009,
+      "step": 2763
+    },
+    {
+      "epoch": 2.0050669562070214,
+      "grad_norm": 0.740253210067749,
+      "learning_rate": 0.0002,
+      "loss": 1.4322,
+      "step": 2770
+    },
+    {
+      "epoch": 2.0123054650741947,
+      "grad_norm": 0.5826276540756226,
+      "learning_rate": 0.0002,
+      "loss": 1.277,
+      "step": 2780
+    },
+    {
+      "epoch": 2.019543973941368,
+      "grad_norm": 0.607356071472168,
+      "learning_rate": 0.0002,
+      "loss": 1.2424,
+      "step": 2790
+    },
+    {
+      "epoch": 2.0267824828085415,
+      "grad_norm": 0.5918063521385193,
+      "learning_rate": 0.0002,
+      "loss": 1.2601,
+      "step": 2800
+    },
+    {
+      "epoch": 2.034020991675715,
+      "grad_norm": 0.5610089898109436,
+      "learning_rate": 0.0002,
+      "loss": 1.3715,
+      "step": 2810
+    },
+    {
+      "epoch": 2.041259500542888,
+      "grad_norm": 0.5869926810264587,
+      "learning_rate": 0.0002,
+      "loss": 1.2092,
+      "step": 2820
+    },
+    {
+      "epoch": 2.0484980094100615,
+      "grad_norm": 0.5753467679023743,
+      "learning_rate": 0.0002,
+      "loss": 1.1929,
+      "step": 2830
+    },
+    {
+      "epoch": 2.055736518277235,
+      "grad_norm": 0.7096508145332336,
+      "learning_rate": 0.0002,
+      "loss": 1.333,
+      "step": 2840
+    },
+    {
+      "epoch": 2.0629750271444083,
+      "grad_norm": 0.7653635144233704,
+      "learning_rate": 0.0002,
+      "loss": 1.1766,
+      "step": 2850
+    },
+    {
+      "epoch": 2.0702135360115816,
+      "grad_norm": 0.6202841997146606,
+      "learning_rate": 0.0002,
+      "loss": 1.2331,
+      "step": 2860
+    },
+    {
+      "epoch": 2.077452044878755,
+      "grad_norm": 0.6810227632522583,
+      "learning_rate": 0.0002,
+      "loss": 1.3298,
+      "step": 2870
+    },
+    {
+      "epoch": 2.0846905537459284,
+      "grad_norm": 0.7481493353843689,
+      "learning_rate": 0.0002,
+      "loss": 1.2505,
+      "step": 2880
+    },
+    {
+      "epoch": 2.0919290626131017,
+      "grad_norm": 0.7089637517929077,
+      "learning_rate": 0.0002,
+      "loss": 1.2484,
+      "step": 2890
+    },
+    {
+      "epoch": 2.099167571480275,
+      "grad_norm": 0.7472923398017883,
+      "learning_rate": 0.0002,
+      "loss": 1.3095,
+      "step": 2900
+    },
+    {
+      "epoch": 2.1064060803474485,
+      "grad_norm": 0.8135465979576111,
+      "learning_rate": 0.0002,
+      "loss": 1.304,
+      "step": 2910
+    },
+    {
+      "epoch": 2.113644589214622,
+      "grad_norm": 0.6097133159637451,
+      "learning_rate": 0.0002,
+      "loss": 1.273,
+      "step": 2920
+    },
+    {
+      "epoch": 2.120883098081795,
+      "grad_norm": 0.5970117449760437,
+      "learning_rate": 0.0002,
+      "loss": 1.3384,
+      "step": 2930
+    },
+    {
+      "epoch": 2.1281216069489686,
+      "grad_norm": 0.6169309616088867,
+      "learning_rate": 0.0002,
+      "loss": 1.3233,
+      "step": 2940
+    },
+    {
+      "epoch": 2.135360115816142,
+      "grad_norm": 0.9428738355636597,
+      "learning_rate": 0.0002,
+      "loss": 1.4246,
+      "step": 2950
+    },
+    {
+      "epoch": 2.1425986246833153,
+      "grad_norm": 0.5671679973602295,
+      "learning_rate": 0.0002,
+      "loss": 1.3527,
+      "step": 2960
+    },
+    {
+      "epoch": 2.1498371335504887,
+      "grad_norm": 0.7007262110710144,
+      "learning_rate": 0.0002,
+      "loss": 1.1375,
+      "step": 2970
+    },
+    {
+      "epoch": 2.157075642417662,
+      "grad_norm": 0.6294044256210327,
+      "learning_rate": 0.0002,
+      "loss": 1.2015,
+      "step": 2980
+    },
+    {
+      "epoch": 2.1643141512848354,
+      "grad_norm": 0.6105241775512695,
+      "learning_rate": 0.0002,
+      "loss": 1.2167,
+      "step": 2990
+    },
+    {
+      "epoch": 2.1715526601520088,
+      "grad_norm": 0.557124137878418,
+      "learning_rate": 0.0002,
+      "loss": 1.2065,
+      "step": 3000
+    },
+    {
+      "epoch": 2.178791169019182,
+      "grad_norm": 0.6250392198562622,
+      "learning_rate": 0.0002,
+      "loss": 1.2515,
+      "step": 3010
+    },
+    {
+      "epoch": 2.1860296778863555,
+      "grad_norm": 0.645218551158905,
+      "learning_rate": 0.0002,
+      "loss": 1.385,
+      "step": 3020
+    },
+    {
+      "epoch": 2.193268186753529,
+      "grad_norm": 0.9033605456352234,
+      "learning_rate": 0.0002,
+      "loss": 1.3928,
+      "step": 3030
+    },
+    {
+      "epoch": 2.2005066956207022,
+      "grad_norm": 0.5325747132301331,
+      "learning_rate": 0.0002,
+      "loss": 1.2458,
+      "step": 3040
+    },
+    {
+      "epoch": 2.2077452044878756,
+      "grad_norm": 0.6334700584411621,
+      "learning_rate": 0.0002,
+      "loss": 1.261,
+      "step": 3050
+    },
+    {
+      "epoch": 2.214983713355049,
+      "grad_norm": 0.5206325054168701,
+      "learning_rate": 0.0002,
+      "loss": 1.2385,
+      "step": 3060
+    },
+    {
+      "epoch": 2.2222222222222223,
+      "grad_norm": 0.5987200140953064,
+      "learning_rate": 0.0002,
+      "loss": 1.3103,
+      "step": 3070
+    },
+    {
+      "epoch": 2.2294607310893957,
+      "grad_norm": 0.5893264412879944,
+      "learning_rate": 0.0002,
+      "loss": 1.1756,
+      "step": 3080
+    },
+    {
+      "epoch": 2.236699239956569,
+      "grad_norm": 0.6869237422943115,
+      "learning_rate": 0.0002,
+      "loss": 1.235,
+      "step": 3090
+    },
+    {
+      "epoch": 2.2439377488237424,
+      "grad_norm": 0.5040048360824585,
+      "learning_rate": 0.0002,
+      "loss": 1.3285,
+      "step": 3100
+    },
+    {
+      "epoch": 2.251176257690916,
+      "grad_norm": 0.6660613417625427,
+      "learning_rate": 0.0002,
+      "loss": 1.3316,
+      "step": 3110
+    },
+    {
+      "epoch": 2.258414766558089,
+      "grad_norm": 0.5890918970108032,
+      "learning_rate": 0.0002,
+      "loss": 1.3108,
+      "step": 3120
+    },
+    {
+      "epoch": 2.2656532754252625,
+      "grad_norm": 0.6458896994590759,
+      "learning_rate": 0.0002,
+      "loss": 1.248,
+      "step": 3130
+    },
+    {
+      "epoch": 2.272891784292436,
+      "grad_norm": 0.6832690834999084,
+      "learning_rate": 0.0002,
+      "loss": 1.4151,
+      "step": 3140
+    },
+    {
+      "epoch": 2.2801302931596092,
+      "grad_norm": 0.833908200263977,
+      "learning_rate": 0.0002,
+      "loss": 1.4458,
+      "step": 3150
+    },
+    {
+      "epoch": 2.2873688020267826,
+      "grad_norm": 0.4596034586429596,
+      "learning_rate": 0.0002,
+      "loss": 1.2931,
+      "step": 3160
+    },
+    {
+      "epoch": 2.294607310893956,
+      "grad_norm": 0.9130966067314148,
+      "learning_rate": 0.0002,
+      "loss": 1.449,
+      "step": 3170
+    },
+    {
+      "epoch": 2.3018458197611293,
+      "grad_norm": 0.7143292427062988,
+      "learning_rate": 0.0002,
+      "loss": 1.3806,
+      "step": 3180
+    },
+    {
+      "epoch": 2.3090843286283027,
+      "grad_norm": 0.5388900637626648,
+      "learning_rate": 0.0002,
+      "loss": 1.2692,
+      "step": 3190
+    },
+    {
+      "epoch": 2.316322837495476,
+      "grad_norm": 0.5607513189315796,
+      "learning_rate": 0.0002,
+      "loss": 1.2402,
+      "step": 3200
+    },
+    {
+      "epoch": 2.3235613463626494,
+      "grad_norm": 0.6795142292976379,
+      "learning_rate": 0.0002,
+      "loss": 1.3874,
+      "step": 3210
+    },
+    {
+      "epoch": 2.330799855229823,
+      "grad_norm": 0.6561070680618286,
+      "learning_rate": 0.0002,
+      "loss": 1.3042,
+      "step": 3220
+    },
+    {
+      "epoch": 2.338038364096996,
+      "grad_norm": 0.8858118057250977,
+      "learning_rate": 0.0002,
+      "loss": 1.4636,
+      "step": 3230
+    },
+    {
+      "epoch": 2.3452768729641695,
+      "grad_norm": 0.6604151725769043,
+      "learning_rate": 0.0002,
+      "loss": 1.3214,
+      "step": 3240
+    },
+    {
+      "epoch": 2.352515381831343,
+      "grad_norm": 0.6755785346031189,
+      "learning_rate": 0.0002,
+      "loss": 1.4004,
+      "step": 3250
+    },
+    {
+      "epoch": 2.3597538906985163,
+      "grad_norm": 0.6981677412986755,
+      "learning_rate": 0.0002,
+      "loss": 1.2503,
+      "step": 3260
+    },
+    {
+      "epoch": 2.3669923995656896,
+      "grad_norm": 0.6338568329811096,
+      "learning_rate": 0.0002,
+      "loss": 1.3078,
+      "step": 3270
+    },
+    {
+      "epoch": 2.374230908432863,
+      "grad_norm": 0.5754265785217285,
+      "learning_rate": 0.0002,
+      "loss": 1.285,
+      "step": 3280
+    },
+    {
+      "epoch": 2.3814694173000364,
+      "grad_norm": 0.7533153295516968,
+      "learning_rate": 0.0002,
+      "loss": 1.2924,
+      "step": 3290
+    },
+    {
+      "epoch": 2.3887079261672097,
+      "grad_norm": 0.675065279006958,
+      "learning_rate": 0.0002,
+      "loss": 1.3711,
+      "step": 3300
+    },
+    {
+      "epoch": 2.395946435034383,
+      "grad_norm": 0.5686452984809875,
+      "learning_rate": 0.0002,
+      "loss": 1.3548,
+      "step": 3310
+    },
+    {
+      "epoch": 2.4031849439015565,
+      "grad_norm": 0.8129481673240662,
+      "learning_rate": 0.0002,
+      "loss": 1.1998,
+      "step": 3320
+    },
+    {
+      "epoch": 2.41042345276873,
+      "grad_norm": 0.6615934371948242,
+      "learning_rate": 0.0002,
+      "loss": 1.2584,
+      "step": 3330
+    },
+    {
+      "epoch": 2.417661961635903,
+      "grad_norm": 0.6678834557533264,
+      "learning_rate": 0.0002,
+      "loss": 1.3691,
+      "step": 3340
+    },
+    {
+      "epoch": 2.4249004705030766,
+      "grad_norm": 0.5581308007240295,
+      "learning_rate": 0.0002,
+      "loss": 1.2381,
+      "step": 3350
+    },
+    {
+      "epoch": 2.43213897937025,
+      "grad_norm": 0.6098920106887817,
+      "learning_rate": 0.0002,
+      "loss": 1.3853,
+      "step": 3360
+    },
+    {
+      "epoch": 2.4393774882374233,
+      "grad_norm": 0.8101736903190613,
+      "learning_rate": 0.0002,
+      "loss": 1.3692,
+      "step": 3370
+    },
+    {
+      "epoch": 2.4466159971045967,
+      "grad_norm": 0.6621488928794861,
+      "learning_rate": 0.0002,
+      "loss": 1.4418,
+      "step": 3380
+    },
+    {
+      "epoch": 2.45385450597177,
+      "grad_norm": 0.8693289160728455,
+      "learning_rate": 0.0002,
+      "loss": 1.4579,
+      "step": 3390
+    },
+    {
+      "epoch": 2.4610930148389434,
+      "grad_norm": 0.6724580526351929,
+      "learning_rate": 0.0002,
+      "loss": 1.3644,
+      "step": 3400
+    },
+    {
+      "epoch": 2.4683315237061167,
+      "grad_norm": 0.6776891946792603,
+      "learning_rate": 0.0002,
+      "loss": 1.2006,
+      "step": 3410
+    },
+    {
+      "epoch": 2.47557003257329,
+      "grad_norm": 0.7214453816413879,
+      "learning_rate": 0.0002,
+      "loss": 1.2937,
+      "step": 3420
+    },
+    {
+      "epoch": 2.4828085414404635,
+      "grad_norm": 0.8390451073646545,
+      "learning_rate": 0.0002,
+      "loss": 1.4051,
+      "step": 3430
+    },
+    {
+      "epoch": 2.490047050307637,
+      "grad_norm": 0.7130982279777527,
+      "learning_rate": 0.0002,
+      "loss": 1.25,
+      "step": 3440
+    },
+    {
+      "epoch": 2.49728555917481,
+      "grad_norm": 0.8873937129974365,
+      "learning_rate": 0.0002,
+      "loss": 1.2231,
+      "step": 3450
+    },
+    {
+      "epoch": 2.5045240680419836,
+      "grad_norm": 0.725185751914978,
+      "learning_rate": 0.0002,
+      "loss": 1.1429,
+      "step": 3460
+    },
+    {
+      "epoch": 2.511762576909157,
+      "grad_norm": 0.6120352149009705,
+      "learning_rate": 0.0002,
+      "loss": 1.2699,
+      "step": 3470
+    },
+    {
+      "epoch": 2.5190010857763303,
+      "grad_norm": 0.7713613510131836,
+      "learning_rate": 0.0002,
+      "loss": 1.2552,
+      "step": 3480
+    },
+    {
+      "epoch": 2.5262395946435037,
+      "grad_norm": 0.895309567451477,
+      "learning_rate": 0.0002,
+      "loss": 1.4648,
+      "step": 3490
+    },
+    {
+      "epoch": 2.533478103510677,
+      "grad_norm": 0.9631021022796631,
+      "learning_rate": 0.0002,
+      "loss": 1.3043,
+      "step": 3500
+    },
+    {
+      "epoch": 2.5407166123778504,
+      "grad_norm": 0.7475683093070984,
+      "learning_rate": 0.0002,
+      "loss": 1.3492,
+      "step": 3510
+    },
+    {
+      "epoch": 2.5479551212450238,
+      "grad_norm": 0.7271341681480408,
+      "learning_rate": 0.0002,
+      "loss": 1.3637,
+      "step": 3520
+    },
+    {
+      "epoch": 2.555193630112197,
+      "grad_norm": 0.6979510188102722,
+      "learning_rate": 0.0002,
+      "loss": 1.304,
+      "step": 3530
+    },
+    {
+      "epoch": 2.5624321389793705,
+      "grad_norm": 0.6504196524620056,
+      "learning_rate": 0.0002,
+      "loss": 1.2353,
+      "step": 3540
+    },
+    {
+      "epoch": 2.569670647846544,
+      "grad_norm": 0.7226675748825073,
+      "learning_rate": 0.0002,
+      "loss": 1.2699,
+      "step": 3550
+    },
+    {
+      "epoch": 2.5769091567137172,
+      "grad_norm": 0.6143222451210022,
+      "learning_rate": 0.0002,
+      "loss": 1.3002,
+      "step": 3560
+    },
+    {
+      "epoch": 2.5841476655808906,
+      "grad_norm": 0.7245154976844788,
+      "learning_rate": 0.0002,
+      "loss": 1.1585,
+      "step": 3570
+    },
+    {
+      "epoch": 2.591386174448064,
+      "grad_norm": 0.943540632724762,
+      "learning_rate": 0.0002,
+      "loss": 1.3651,
+      "step": 3580
+    },
+    {
+      "epoch": 2.5986246833152373,
+      "grad_norm": 0.7707241773605347,
+      "learning_rate": 0.0002,
+      "loss": 1.3034,
+      "step": 3590
+    },
+    {
+      "epoch": 2.6058631921824107,
+      "grad_norm": 0.6705001592636108,
+      "learning_rate": 0.0002,
+      "loss": 1.3063,
+      "step": 3600
+    },
+    {
+      "epoch": 2.613101701049584,
+      "grad_norm": 0.6360933780670166,
+      "learning_rate": 0.0002,
+      "loss": 1.2437,
+      "step": 3610
+    },
+    {
+      "epoch": 2.6203402099167574,
+      "grad_norm": 0.5846424698829651,
+      "learning_rate": 0.0002,
+      "loss": 1.1844,
+      "step": 3620
+    },
+    {
+      "epoch": 2.6275787187839303,
+      "grad_norm": 0.5958625674247742,
+      "learning_rate": 0.0002,
+      "loss": 1.3674,
+      "step": 3630
+    },
+    {
+      "epoch": 2.6348172276511037,
+      "grad_norm": 0.6819243431091309,
+      "learning_rate": 0.0002,
+      "loss": 1.3599,
+      "step": 3640
+    },
+    {
+      "epoch": 2.642055736518277,
+      "grad_norm": 0.7033445835113525,
+      "learning_rate": 0.0002,
+      "loss": 1.3884,
+      "step": 3650
+    },
+    {
+      "epoch": 2.6492942453854504,
+      "grad_norm": 0.6134849786758423,
+      "learning_rate": 0.0002,
+      "loss": 1.3392,
+      "step": 3660
+    },
+    {
+      "epoch": 2.656532754252624,
+      "grad_norm": 0.658009946346283,
+      "learning_rate": 0.0002,
+      "loss": 1.2661,
+      "step": 3670
+    },
+    {
+      "epoch": 2.663771263119797,
+      "grad_norm": 0.6280999779701233,
+      "learning_rate": 0.0002,
+      "loss": 1.3987,
+      "step": 3680
+    },
+    {
+      "epoch": 2.6710097719869705,
+      "grad_norm": 0.5536085963249207,
+      "learning_rate": 0.0002,
+      "loss": 1.2995,
+      "step": 3690
+    },
+    {
+      "epoch": 2.678248280854144,
+      "grad_norm": 0.8603981733322144,
+      "learning_rate": 0.0002,
+      "loss": 1.2044,
+      "step": 3700
+    },
+    {
+      "epoch": 2.6854867897213173,
+      "grad_norm": 0.5509994626045227,
+      "learning_rate": 0.0002,
+      "loss": 1.3879,
+      "step": 3710
+    },
+    {
+      "epoch": 2.6927252985884906,
+      "grad_norm": 0.9093621969223022,
+      "learning_rate": 0.0002,
+      "loss": 1.3253,
+      "step": 3720
+    },
+    {
+      "epoch": 2.699963807455664,
+      "grad_norm": 0.7525952458381653,
+      "learning_rate": 0.0002,
+      "loss": 1.2668,
+      "step": 3730
+    },
+    {
+      "epoch": 2.7072023163228374,
+      "grad_norm": 0.6737023591995239,
+      "learning_rate": 0.0002,
+      "loss": 1.248,
+      "step": 3740
+    },
+    {
+      "epoch": 2.7144408251900107,
+      "grad_norm": 0.8656924962997437,
+      "learning_rate": 0.0002,
+      "loss": 1.2981,
+      "step": 3750
+    },
+    {
+      "epoch": 2.721679334057184,
+      "grad_norm": 0.7494133114814758,
+      "learning_rate": 0.0002,
+      "loss": 1.2342,
+      "step": 3760
+    },
+    {
+      "epoch": 2.7289178429243575,
+      "grad_norm": 0.5725520849227905,
+      "learning_rate": 0.0002,
+      "loss": 1.2417,
+      "step": 3770
+    },
+    {
+      "epoch": 2.736156351791531,
+      "grad_norm": 0.836412787437439,
+      "learning_rate": 0.0002,
+      "loss": 1.28,
+      "step": 3780
+    },
+    {
+      "epoch": 2.743394860658704,
+      "grad_norm": 0.6893242597579956,
+      "learning_rate": 0.0002,
+      "loss": 1.3784,
+      "step": 3790
+    },
+    {
+      "epoch": 2.7506333695258776,
+      "grad_norm": 0.6696223020553589,
+      "learning_rate": 0.0002,
+      "loss": 1.2929,
+      "step": 3800
+    },
+    {
+      "epoch": 2.757871878393051,
+      "grad_norm": 0.6483015418052673,
+      "learning_rate": 0.0002,
+      "loss": 1.2449,
+      "step": 3810
+    },
+    {
+      "epoch": 2.7651103872602243,
+      "grad_norm": 0.8084456920623779,
+      "learning_rate": 0.0002,
+      "loss": 1.3282,
+      "step": 3820
+    },
+    {
+      "epoch": 2.7723488961273977,
+      "grad_norm": 0.6601949334144592,
+      "learning_rate": 0.0002,
+      "loss": 1.3694,
+      "step": 3830
+    },
+    {
+      "epoch": 2.779587404994571,
+      "grad_norm": 0.6905533671379089,
+      "learning_rate": 0.0002,
+      "loss": 1.3568,
+      "step": 3840
+    },
+    {
+      "epoch": 2.7868259138617444,
+      "grad_norm": 0.619318425655365,
+      "learning_rate": 0.0002,
+      "loss": 1.3854,
+      "step": 3850
+    },
+    {
+      "epoch": 2.7940644227289178,
+      "grad_norm": 0.5994023084640503,
+      "learning_rate": 0.0002,
+      "loss": 1.2551,
+      "step": 3860
+    },
+    {
+      "epoch": 2.801302931596091,
+      "grad_norm": 0.5627168416976929,
+      "learning_rate": 0.0002,
+      "loss": 1.2022,
+      "step": 3870
+    },
+    {
+      "epoch": 2.8085414404632645,
+      "grad_norm": 0.6001605987548828,
+      "learning_rate": 0.0002,
+      "loss": 1.3921,
+      "step": 3880
+    },
+    {
+      "epoch": 2.815779949330438,
+      "grad_norm": 0.6022412776947021,
+      "learning_rate": 0.0002,
+      "loss": 1.3026,
+      "step": 3890
+    },
+    {
+      "epoch": 2.823018458197611,
+      "grad_norm": 0.6832426190376282,
+      "learning_rate": 0.0002,
+      "loss": 1.2765,
+      "step": 3900
+    },
+    {
+      "epoch": 2.8302569670647846,
+      "grad_norm": 0.5936811566352844,
+      "learning_rate": 0.0002,
+      "loss": 1.1363,
+      "step": 3910
+    },
+    {
+      "epoch": 2.837495475931958,
+      "grad_norm": 0.6960572600364685,
+      "learning_rate": 0.0002,
+      "loss": 1.1707,
+      "step": 3920
+    },
+    {
+      "epoch": 2.8447339847991313,
+      "grad_norm": 0.5913406610488892,
+      "learning_rate": 0.0002,
+      "loss": 1.4063,
+      "step": 3930
+    },
+    {
+      "epoch": 2.8519724936663047,
+      "grad_norm": 0.678154706954956,
+      "learning_rate": 0.0002,
+      "loss": 1.3245,
+      "step": 3940
+    },
+    {
+      "epoch": 2.859211002533478,
+      "grad_norm": 0.7898936867713928,
+      "learning_rate": 0.0002,
+      "loss": 1.366,
+      "step": 3950
+    },
+    {
+      "epoch": 2.8664495114006514,
+      "grad_norm": 0.9234195351600647,
+      "learning_rate": 0.0002,
+      "loss": 1.3948,
+      "step": 3960
+    },
+    {
+      "epoch": 2.8736880202678248,
+      "grad_norm": 0.5960825085639954,
+      "learning_rate": 0.0002,
+      "loss": 1.2773,
+      "step": 3970
+    },
+    {
+      "epoch": 2.880926529134998,
+      "grad_norm": 0.677118182182312,
+      "learning_rate": 0.0002,
+      "loss": 1.3127,
+      "step": 3980
+    },
+    {
+      "epoch": 2.8881650380021715,
+      "grad_norm": 0.6505142450332642,
+      "learning_rate": 0.0002,
+      "loss": 1.2652,
+      "step": 3990
+    },
+    {
+      "epoch": 2.895403546869345,
+      "grad_norm": 0.550826907157898,
+      "learning_rate": 0.0002,
+      "loss": 1.2078,
+      "step": 4000
+    },
+    {
+      "epoch": 2.9026420557365182,
+      "grad_norm": 0.6209215521812439,
+      "learning_rate": 0.0002,
+      "loss": 1.1811,
+      "step": 4010
+    },
+    {
+      "epoch": 2.9098805646036916,
+      "grad_norm": 0.6549018025398254,
+      "learning_rate": 0.0002,
+      "loss": 1.4001,
+      "step": 4020
+    },
+    {
+      "epoch": 2.917119073470865,
+      "grad_norm": 0.570682168006897,
+      "learning_rate": 0.0002,
+      "loss": 1.2285,
+      "step": 4030
+    },
+    {
+      "epoch": 2.9243575823380383,
+      "grad_norm": 1.1807632446289062,
+      "learning_rate": 0.0002,
+      "loss": 1.0832,
+      "step": 4040
+    },
+    {
+      "epoch": 2.9315960912052117,
+      "grad_norm": 0.7058857679367065,
+      "learning_rate": 0.0002,
+      "loss": 1.2693,
+      "step": 4050
+    },
+    {
+      "epoch": 2.938834600072385,
+      "grad_norm": 0.5542812943458557,
+      "learning_rate": 0.0002,
+      "loss": 1.2905,
+      "step": 4060
+    },
+    {
+      "epoch": 2.9460731089395584,
+      "grad_norm": 0.63167804479599,
+      "learning_rate": 0.0002,
+      "loss": 1.33,
+      "step": 4070
+    },
+    {
+      "epoch": 2.953311617806732,
+      "grad_norm": 0.5702962279319763,
+      "learning_rate": 0.0002,
+      "loss": 1.3075,
+      "step": 4080
+    },
+    {
+      "epoch": 2.960550126673905,
+      "grad_norm": 0.620944082736969,
+      "learning_rate": 0.0002,
+      "loss": 1.2007,
+      "step": 4090
+    },
+    {
+      "epoch": 2.9677886355410785,
+      "grad_norm": 0.5866289734840393,
+      "learning_rate": 0.0002,
+      "loss": 1.2864,
+      "step": 4100
+    },
+    {
+      "epoch": 2.975027144408252,
+      "grad_norm": 0.560170590877533,
+      "learning_rate": 0.0002,
+      "loss": 1.3293,
+      "step": 4110
+    },
+    {
+      "epoch": 2.9822656532754253,
+      "grad_norm": 0.675082802772522,
+      "learning_rate": 0.0002,
+      "loss": 1.2071,
+      "step": 4120
+    },
+    {
+      "epoch": 2.9895041621425986,
+      "grad_norm": 0.62708580493927,
+      "learning_rate": 0.0002,
+      "loss": 1.2981,
+      "step": 4130
+    },
+    {
+      "epoch": 2.996742671009772,
+      "grad_norm": 0.7893929481506348,
+      "learning_rate": 0.0002,
+      "loss": 1.2758,
+      "step": 4140
+    },
+    {
+      "epoch": 2.9996380745566413,
+      "eval_loss": 1.4217946529388428,
+      "eval_runtime": 27.1596,
+      "eval_samples_per_second": 16.053,
+      "eval_steps_per_second": 2.025,
+      "step": 4144
+    },
+    {
+      "epoch": 3.0039811798769454,
+      "grad_norm": 0.7043836116790771,
+      "learning_rate": 0.0002,
+      "loss": 1.2152,
+      "step": 4150
+    },
+    {
+      "epoch": 3.0112196887441187,
+      "grad_norm": 0.6806283593177795,
+      "learning_rate": 0.0002,
+      "loss": 1.1664,
+      "step": 4160
+    },
+    {
+      "epoch": 3.018458197611292,
+      "grad_norm": 0.7684550285339355,
+      "learning_rate": 0.0002,
+      "loss": 1.292,
+      "step": 4170
+    },
+    {
+      "epoch": 3.0256967064784654,
+      "grad_norm": 0.7895237803459167,
+      "learning_rate": 0.0002,
+      "loss": 1.3467,
+      "step": 4180
+    },
+    {
+      "epoch": 3.032935215345639,
+      "grad_norm": 0.7464531064033508,
+      "learning_rate": 0.0002,
+      "loss": 1.1324,
+      "step": 4190
+    },
+    {
+      "epoch": 3.040173724212812,
+      "grad_norm": 0.9358500838279724,
+      "learning_rate": 0.0002,
+      "loss": 1.1614,
+      "step": 4200
+    },
+    {
+      "epoch": 3.0474122330799855,
+      "grad_norm": 1.1066628694534302,
+      "learning_rate": 0.0002,
+      "loss": 1.1834,
+      "step": 4210
+    },
+    {
+      "epoch": 3.054650741947159,
+      "grad_norm": 0.6663267612457275,
+      "learning_rate": 0.0002,
+      "loss": 1.1557,
+      "step": 4220
+    },
+    {
+      "epoch": 3.0618892508143323,
+      "grad_norm": 0.6669464707374573,
+      "learning_rate": 0.0002,
+      "loss": 1.1707,
+      "step": 4230
+    },
+    {
+      "epoch": 3.0691277596815056,
+      "grad_norm": 0.7052164077758789,
+      "learning_rate": 0.0002,
+      "loss": 1.1841,
+      "step": 4240
+    },
+    {
+      "epoch": 3.076366268548679,
+      "grad_norm": 0.6118432879447937,
+      "learning_rate": 0.0002,
+      "loss": 1.2913,
+      "step": 4250
+    },
+    {
+      "epoch": 3.0836047774158524,
+      "grad_norm": 0.6915903687477112,
+      "learning_rate": 0.0002,
+      "loss": 1.1526,
+      "step": 4260
+    },
+    {
+      "epoch": 3.0908432862830257,
+      "grad_norm": 0.7441644668579102,
+      "learning_rate": 0.0002,
+      "loss": 1.1348,
+      "step": 4270
+    },
+    {
+      "epoch": 3.098081795150199,
+      "grad_norm": 0.823850691318512,
+      "learning_rate": 0.0002,
+      "loss": 1.1672,
+      "step": 4280
+    },
+    {
+      "epoch": 3.1053203040173725,
+      "grad_norm": 0.9677883386611938,
+      "learning_rate": 0.0002,
+      "loss": 1.2655,
+      "step": 4290
+    },
+    {
+      "epoch": 3.112558812884546,
+      "grad_norm": 0.7002579569816589,
+      "learning_rate": 0.0002,
+      "loss": 1.1794,
+      "step": 4300
+    },
+    {
+      "epoch": 3.119797321751719,
+      "grad_norm": 0.778789758682251,
+      "learning_rate": 0.0002,
+      "loss": 1.135,
+      "step": 4310
+    },
+    {
+      "epoch": 3.1270358306188926,
+      "grad_norm": 0.7236007452011108,
+      "learning_rate": 0.0002,
+      "loss": 1.0818,
+      "step": 4320
+    },
+    {
+      "epoch": 3.134274339486066,
+      "grad_norm": 0.8809133768081665,
+      "learning_rate": 0.0002,
+      "loss": 1.1803,
+      "step": 4330
+    },
+    {
+      "epoch": 3.1415128483532393,
+      "grad_norm": 0.7924913167953491,
+      "learning_rate": 0.0002,
+      "loss": 1.2571,
+      "step": 4340
+    },
+    {
+      "epoch": 3.1487513572204127,
+      "grad_norm": 0.7437422275543213,
+      "learning_rate": 0.0002,
+      "loss": 1.1413,
+      "step": 4350
+    },
+    {
+      "epoch": 3.155989866087586,
+      "grad_norm": 0.6428450345993042,
+      "learning_rate": 0.0002,
+      "loss": 1.2088,
+      "step": 4360
+    },
+    {
+      "epoch": 3.1632283749547594,
+      "grad_norm": 0.7922873497009277,
+      "learning_rate": 0.0002,
+      "loss": 1.3032,
+      "step": 4370
+    },
+    {
+      "epoch": 3.1704668838219328,
+      "grad_norm": 0.5252506732940674,
+      "learning_rate": 0.0002,
+      "loss": 1.216,
+      "step": 4380
+    },
+    {
+      "epoch": 3.177705392689106,
+      "grad_norm": 0.8570457696914673,
+      "learning_rate": 0.0002,
+      "loss": 1.1297,
+      "step": 4390
+    },
+    {
+      "epoch": 3.1849439015562795,
+      "grad_norm": 0.7218987345695496,
+      "learning_rate": 0.0002,
+      "loss": 1.0994,
+      "step": 4400
+    },
+    {
+      "epoch": 3.192182410423453,
+      "grad_norm": 0.6921393275260925,
+      "learning_rate": 0.0002,
+      "loss": 1.2891,
+      "step": 4410
+    },
+    {
+      "epoch": 3.199420919290626,
+      "grad_norm": 0.7386137843132019,
+      "learning_rate": 0.0002,
+      "loss": 1.2668,
+      "step": 4420
+    },
+    {
+      "epoch": 3.2066594281577996,
+      "grad_norm": 0.6227759122848511,
+      "learning_rate": 0.0002,
+      "loss": 1.1654,
+      "step": 4430
+    },
+    {
+      "epoch": 3.213897937024973,
+      "grad_norm": 0.7180278897285461,
+      "learning_rate": 0.0002,
+      "loss": 1.1752,
+      "step": 4440
+    },
+    {
+      "epoch": 3.2211364458921463,
+      "grad_norm": 0.745830774307251,
+      "learning_rate": 0.0002,
+      "loss": 1.1757,
+      "step": 4450
+    },
+    {
+      "epoch": 3.2283749547593197,
+      "grad_norm": 0.6766072511672974,
+      "learning_rate": 0.0002,
+      "loss": 1.234,
+      "step": 4460
+    },
+    {
+      "epoch": 3.235613463626493,
+      "grad_norm": 0.8325067162513733,
+      "learning_rate": 0.0002,
+      "loss": 1.1999,
+      "step": 4470
+    },
+    {
+      "epoch": 3.2428519724936664,
+      "grad_norm": 0.7148305177688599,
+      "learning_rate": 0.0002,
+      "loss": 1.1606,
+      "step": 4480
+    },
+    {
+      "epoch": 3.25009048136084,
+      "grad_norm": 0.7752676010131836,
+      "learning_rate": 0.0002,
+      "loss": 1.1383,
+      "step": 4490
+    },
+    {
+      "epoch": 3.257328990228013,
+      "grad_norm": 0.6776860952377319,
+      "learning_rate": 0.0002,
+      "loss": 1.3006,
+      "step": 4500
+    },
+    {
+      "epoch": 3.2645674990951865,
+      "grad_norm": 0.704359769821167,
+      "learning_rate": 0.0002,
+      "loss": 1.0796,
+      "step": 4510
+    },
+    {
+      "epoch": 3.27180600796236,
+      "grad_norm": 0.6880282163619995,
+      "learning_rate": 0.0002,
+      "loss": 1.2496,
+      "step": 4520
+    },
+    {
+      "epoch": 3.2790445168295332,
+      "grad_norm": 0.8179270029067993,
+      "learning_rate": 0.0002,
+      "loss": 1.0947,
+      "step": 4530
+    },
+    {
+      "epoch": 3.2862830256967066,
+      "grad_norm": 0.6718448996543884,
+      "learning_rate": 0.0002,
+      "loss": 1.1909,
+      "step": 4540
+    },
+    {
+      "epoch": 3.29352153456388,
+      "grad_norm": 0.8300657868385315,
+      "learning_rate": 0.0002,
+      "loss": 1.2708,
+      "step": 4550
+    },
+    {
+      "epoch": 3.3007600434310533,
+      "grad_norm": 0.6433690786361694,
+      "learning_rate": 0.0002,
+      "loss": 1.2594,
+      "step": 4560
+    },
+    {
+      "epoch": 3.3079985522982267,
+      "grad_norm": 0.690262496471405,
+      "learning_rate": 0.0002,
+      "loss": 1.2479,
+      "step": 4570
+    },
+    {
+      "epoch": 3.3152370611654,
+      "grad_norm": 0.7022852301597595,
+      "learning_rate": 0.0002,
+      "loss": 1.1342,
+      "step": 4580
+    },
+    {
+      "epoch": 3.3224755700325734,
+      "grad_norm": 0.6438387632369995,
+      "learning_rate": 0.0002,
+      "loss": 1.0844,
+      "step": 4590
+    },
+    {
+      "epoch": 3.329714078899747,
+      "grad_norm": 0.6866899132728577,
+      "learning_rate": 0.0002,
+      "loss": 1.17,
+      "step": 4600
+    },
+    {
+      "epoch": 3.33695258776692,
+      "grad_norm": 0.8233968019485474,
+      "learning_rate": 0.0002,
+      "loss": 1.1289,
+      "step": 4610
+    },
+    {
+      "epoch": 3.3441910966340935,
+      "grad_norm": 0.7251574993133545,
+      "learning_rate": 0.0002,
+      "loss": 1.1855,
+      "step": 4620
+    },
+    {
+      "epoch": 3.351429605501267,
+      "grad_norm": 0.7855110168457031,
+      "learning_rate": 0.0002,
+      "loss": 1.3403,
+      "step": 4630
+    },
+    {
+      "epoch": 3.3586681143684403,
+      "grad_norm": 0.8487356305122375,
+      "learning_rate": 0.0002,
+      "loss": 1.2922,
+      "step": 4640
+    },
+    {
+      "epoch": 3.3659066232356136,
+      "grad_norm": 0.6429011225700378,
+      "learning_rate": 0.0002,
+      "loss": 1.2462,
+      "step": 4650
+    },
+    {
+      "epoch": 3.373145132102787,
+      "grad_norm": 0.7095270156860352,
+      "learning_rate": 0.0002,
+      "loss": 1.129,
+      "step": 4660
+    },
+    {
+      "epoch": 3.3803836409699604,
+      "grad_norm": 0.6792303323745728,
+      "learning_rate": 0.0002,
+      "loss": 1.262,
+      "step": 4670
+    },
+    {
+      "epoch": 3.3876221498371337,
+      "grad_norm": 0.6784825921058655,
+      "learning_rate": 0.0002,
+      "loss": 1.256,
+      "step": 4680
+    },
+    {
+      "epoch": 3.394860658704307,
+      "grad_norm": 0.6362888216972351,
+      "learning_rate": 0.0002,
+      "loss": 1.0838,
+      "step": 4690
+    },
+    {
+      "epoch": 3.4020991675714805,
+      "grad_norm": 0.7794778943061829,
+      "learning_rate": 0.0002,
+      "loss": 1.2165,
+      "step": 4700
+    },
+    {
+      "epoch": 3.409337676438654,
+      "grad_norm": 0.7287485003471375,
+      "learning_rate": 0.0002,
+      "loss": 1.0644,
+      "step": 4710
+    },
+    {
+      "epoch": 3.416576185305827,
+      "grad_norm": 0.6481451392173767,
+      "learning_rate": 0.0002,
+      "loss": 1.2925,
+      "step": 4720
+    },
+    {
+      "epoch": 3.4238146941730006,
+      "grad_norm": 0.9200371503829956,
+      "learning_rate": 0.0002,
+      "loss": 1.2121,
+      "step": 4730
+    },
+    {
+      "epoch": 3.431053203040174,
+      "grad_norm": 1.074180245399475,
+      "learning_rate": 0.0002,
+      "loss": 1.072,
+      "step": 4740
+    },
+    {
+      "epoch": 3.438291711907347,
+      "grad_norm": 0.6722986698150635,
+      "learning_rate": 0.0002,
+      "loss": 1.0421,
+      "step": 4750
+    },
+    {
+      "epoch": 3.44553022077452,
+      "grad_norm": 0.7945933938026428,
+      "learning_rate": 0.0002,
+      "loss": 1.2258,
+      "step": 4760
+    },
+    {
+      "epoch": 3.4527687296416936,
+      "grad_norm": 0.7624640464782715,
+      "learning_rate": 0.0002,
+      "loss": 1.0927,
+      "step": 4770
+    },
+    {
+      "epoch": 3.460007238508867,
+      "grad_norm": 0.7763656377792358,
+      "learning_rate": 0.0002,
+      "loss": 1.2428,
+      "step": 4780
+    },
+    {
+      "epoch": 3.4672457473760403,
+      "grad_norm": 0.7736947536468506,
+      "learning_rate": 0.0002,
+      "loss": 1.2584,
+      "step": 4790
+    },
+    {
+      "epoch": 3.4744842562432137,
+      "grad_norm": 0.8450354933738708,
+      "learning_rate": 0.0002,
+      "loss": 1.1953,
+      "step": 4800
+    },
+    {
+      "epoch": 3.481722765110387,
+      "grad_norm": 0.6480133533477783,
+      "learning_rate": 0.0002,
+      "loss": 1.1362,
+      "step": 4810
+    },
+    {
+      "epoch": 3.4889612739775604,
+      "grad_norm": 0.8437445759773254,
+      "learning_rate": 0.0002,
+      "loss": 1.1882,
+      "step": 4820
+    },
+    {
+      "epoch": 3.4961997828447338,
+      "grad_norm": 0.7781730890274048,
+      "learning_rate": 0.0002,
+      "loss": 1.1519,
+      "step": 4830
+    },
+    {
+      "epoch": 3.503438291711907,
+      "grad_norm": 0.8523228168487549,
+      "learning_rate": 0.0002,
+      "loss": 1.1836,
+      "step": 4840
+    },
+    {
+      "epoch": 3.5106768005790805,
+      "grad_norm": 0.6236732006072998,
+      "learning_rate": 0.0002,
+      "loss": 1.1672,
+      "step": 4850
+    },
+    {
+      "epoch": 3.517915309446254,
+      "grad_norm": 0.7500787377357483,
+      "learning_rate": 0.0002,
+      "loss": 1.1926,
+      "step": 4860
+    },
+    {
+      "epoch": 3.5251538183134272,
+      "grad_norm": 0.7665374875068665,
+      "learning_rate": 0.0002,
+      "loss": 1.1998,
+      "step": 4870
+    },
+    {
+      "epoch": 3.5323923271806006,
+      "grad_norm": 0.787857711315155,
+      "learning_rate": 0.0002,
+      "loss": 1.1551,
+      "step": 4880
+    },
+    {
+      "epoch": 3.539630836047774,
+      "grad_norm": 0.970595121383667,
+      "learning_rate": 0.0002,
+      "loss": 1.2758,
+      "step": 4890
+    },
+    {
+      "epoch": 3.5468693449149473,
+      "grad_norm": 0.6409347057342529,
+      "learning_rate": 0.0002,
+      "loss": 1.1274,
+      "step": 4900
+    },
+    {
+      "epoch": 3.5541078537821207,
+      "grad_norm": 0.888551652431488,
+      "learning_rate": 0.0002,
+      "loss": 1.1596,
+      "step": 4910
+    },
+    {
+      "epoch": 3.561346362649294,
+      "grad_norm": 1.0808377265930176,
+      "learning_rate": 0.0002,
+      "loss": 1.1644,
+      "step": 4920
+    },
+    {
+      "epoch": 3.5685848715164674,
+      "grad_norm": 0.7501053214073181,
+      "learning_rate": 0.0002,
+      "loss": 1.2564,
+      "step": 4930
+    },
+    {
+      "epoch": 3.575823380383641,
+      "grad_norm": 0.7375240325927734,
+      "learning_rate": 0.0002,
+      "loss": 1.2351,
+      "step": 4940
+    },
+    {
+      "epoch": 3.583061889250814,
+      "grad_norm": 0.7075039744377136,
+      "learning_rate": 0.0002,
+      "loss": 1.3568,
+      "step": 4950
+    },
+    {
+      "epoch": 3.5903003981179875,
+      "grad_norm": 0.939337432384491,
+      "learning_rate": 0.0002,
+      "loss": 1.3355,
+      "step": 4960
+    },
+    {
+      "epoch": 3.597538906985161,
+      "grad_norm": 0.6717396974563599,
+      "learning_rate": 0.0002,
+      "loss": 1.1722,
+      "step": 4970
+    },
+    {
+      "epoch": 3.6047774158523342,
+      "grad_norm": 0.7141643762588501,
+      "learning_rate": 0.0002,
+      "loss": 1.1186,
+      "step": 4980
+    },
+    {
+      "epoch": 3.6120159247195076,
+      "grad_norm": 0.7109216451644897,
+      "learning_rate": 0.0002,
+      "loss": 1.1011,
+      "step": 4990
+    },
+    {
+      "epoch": 3.619254433586681,
+      "grad_norm": 0.7020776867866516,
+      "learning_rate": 0.0002,
+      "loss": 1.2178,
+      "step": 5000
+    },
+    {
+      "epoch": 3.6264929424538543,
+      "grad_norm": 0.7158873677253723,
+      "learning_rate": 0.0002,
+      "loss": 1.1939,
+      "step": 5010
+    },
+    {
+      "epoch": 3.6337314513210277,
+      "grad_norm": 0.7062035202980042,
+      "learning_rate": 0.0002,
+      "loss": 1.2624,
+      "step": 5020
+    },
+    {
+      "epoch": 3.640969960188201,
+      "grad_norm": 0.7081155776977539,
+      "learning_rate": 0.0002,
+      "loss": 1.0224,
+      "step": 5030
+    },
+    {
+      "epoch": 3.6482084690553744,
+      "grad_norm": 1.2210607528686523,
+      "learning_rate": 0.0002,
+      "loss": 1.2195,
+      "step": 5040
+    },
+    {
+      "epoch": 3.655446977922548,
+      "grad_norm": 0.6650236248970032,
+      "learning_rate": 0.0002,
+      "loss": 1.2596,
+      "step": 5050
+    },
+    {
+      "epoch": 3.662685486789721,
+      "grad_norm": 0.6884829998016357,
+      "learning_rate": 0.0002,
+      "loss": 1.1072,
+      "step": 5060
+    },
+    {
+      "epoch": 3.6699239956568945,
+      "grad_norm": 0.7317819595336914,
+      "learning_rate": 0.0002,
+      "loss": 1.2292,
+      "step": 5070
+    },
+    {
+      "epoch": 3.677162504524068,
+      "grad_norm": 0.7406691908836365,
+      "learning_rate": 0.0002,
+      "loss": 1.1917,
+      "step": 5080
+    },
+    {
+      "epoch": 3.6844010133912413,
+      "grad_norm": 0.9009454250335693,
+      "learning_rate": 0.0002,
+      "loss": 1.2949,
+      "step": 5090
+    },
+    {
+      "epoch": 3.6916395222584146,
+      "grad_norm": 0.8189385533332825,
+      "learning_rate": 0.0002,
+      "loss": 1.1528,
+      "step": 5100
+    },
+    {
+      "epoch": 3.698878031125588,
+      "grad_norm": 1.0793628692626953,
+      "learning_rate": 0.0002,
+      "loss": 1.3408,
+      "step": 5110
+    },
+    {
+      "epoch": 3.7061165399927614,
+      "grad_norm": 0.8593027591705322,
+      "learning_rate": 0.0002,
+      "loss": 1.2417,
+      "step": 5120
+    },
+    {
+      "epoch": 3.7133550488599347,
+      "grad_norm": 0.8481812477111816,
+      "learning_rate": 0.0002,
+      "loss": 1.2141,
+      "step": 5130
+    },
+    {
+      "epoch": 3.720593557727108,
+      "grad_norm": 0.6527451276779175,
+      "learning_rate": 0.0002,
+      "loss": 1.125,
+      "step": 5140
+    },
+    {
+      "epoch": 3.7278320665942815,
+      "grad_norm": 0.9220114350318909,
+      "learning_rate": 0.0002,
+      "loss": 1.1584,
+      "step": 5150
+    },
+    {
+      "epoch": 3.735070575461455,
+      "grad_norm": 1.0842019319534302,
+      "learning_rate": 0.0002,
+      "loss": 1.2267,
+      "step": 5160
+    },
+    {
+      "epoch": 3.742309084328628,
+      "grad_norm": 0.965453565120697,
+      "learning_rate": 0.0002,
+      "loss": 1.3083,
+      "step": 5170
+    },
+    {
+      "epoch": 3.7495475931958016,
+      "grad_norm": 0.9903319478034973,
+      "learning_rate": 0.0002,
+      "loss": 1.1772,
+      "step": 5180
+    },
+    {
+      "epoch": 3.756786102062975,
+      "grad_norm": 0.7434818148612976,
+      "learning_rate": 0.0002,
+      "loss": 1.2515,
+      "step": 5190
+    },
+    {
+      "epoch": 3.7640246109301483,
+      "grad_norm": 0.6717280745506287,
+      "learning_rate": 0.0002,
+      "loss": 1.2631,
+      "step": 5200
+    },
+    {
+      "epoch": 3.7712631197973217,
+      "grad_norm": 0.7754665613174438,
+      "learning_rate": 0.0002,
+      "loss": 1.2012,
+      "step": 5210
+    },
+    {
+      "epoch": 3.778501628664495,
+      "grad_norm": 1.028374433517456,
+      "learning_rate": 0.0002,
+      "loss": 1.305,
+      "step": 5220
+    },
+    {
+      "epoch": 3.7857401375316684,
+      "grad_norm": 0.6026996374130249,
+      "learning_rate": 0.0002,
+      "loss": 1.1866,
+      "step": 5230
+    },
+    {
+      "epoch": 3.7929786463988417,
+      "grad_norm": 0.6978490948677063,
+      "learning_rate": 0.0002,
+      "loss": 1.1901,
+      "step": 5240
+    },
+    {
+      "epoch": 3.800217155266015,
+      "grad_norm": 0.7303446531295776,
+      "learning_rate": 0.0002,
+      "loss": 1.2576,
+      "step": 5250
+    },
+    {
+      "epoch": 3.8074556641331885,
+      "grad_norm": 1.0734210014343262,
+      "learning_rate": 0.0002,
+      "loss": 1.3173,
+      "step": 5260
+    },
+    {
+      "epoch": 3.814694173000362,
+      "grad_norm": 0.6383201479911804,
+      "learning_rate": 0.0002,
+      "loss": 1.1137,
+      "step": 5270
+    },
+    {
+      "epoch": 3.821932681867535,
+      "grad_norm": 0.7742630243301392,
+      "learning_rate": 0.0002,
+      "loss": 1.0904,
+      "step": 5280
+    },
+    {
+      "epoch": 3.8291711907347086,
+      "grad_norm": 0.8477074503898621,
+      "learning_rate": 0.0002,
+      "loss": 1.2232,
+      "step": 5290
+    },
+    {
+      "epoch": 3.836409699601882,
+      "grad_norm": 0.6675317883491516,
+      "learning_rate": 0.0002,
+      "loss": 1.2047,
+      "step": 5300
+    },
+    {
+      "epoch": 3.8436482084690553,
+      "grad_norm": 0.7515445351600647,
+      "learning_rate": 0.0002,
+      "loss": 1.2275,
+      "step": 5310
+    },
+    {
+      "epoch": 3.8508867173362287,
+      "grad_norm": 1.1441220045089722,
+      "learning_rate": 0.0002,
+      "loss": 1.2569,
+      "step": 5320
+    },
+    {
+      "epoch": 3.858125226203402,
+      "grad_norm": 0.7968795895576477,
+      "learning_rate": 0.0002,
+      "loss": 1.1512,
+      "step": 5330
+    },
+    {
+      "epoch": 3.8653637350705754,
+      "grad_norm": 0.7842824459075928,
+      "learning_rate": 0.0002,
+      "loss": 1.232,
+      "step": 5340
+    },
+    {
+      "epoch": 3.8726022439377488,
+      "grad_norm": 0.8272225260734558,
+      "learning_rate": 0.0002,
+      "loss": 1.1847,
+      "step": 5350
+    },
+    {
+      "epoch": 3.879840752804922,
+      "grad_norm": 0.8413397669792175,
+      "learning_rate": 0.0002,
+      "loss": 1.1381,
+      "step": 5360
+    },
+    {
+      "epoch": 3.8870792616720955,
+      "grad_norm": 1.141764760017395,
+      "learning_rate": 0.0002,
+      "loss": 1.2349,
+      "step": 5370
+    },
+    {
+      "epoch": 3.894317770539269,
+      "grad_norm": 0.9826975464820862,
+      "learning_rate": 0.0002,
+      "loss": 1.212,
+      "step": 5380
+    },
+    {
+      "epoch": 3.9015562794064422,
+      "grad_norm": 0.8598255515098572,
+      "learning_rate": 0.0002,
+      "loss": 1.1833,
+      "step": 5390
+    },
+    {
+      "epoch": 3.9087947882736156,
+      "grad_norm": 0.6271058320999146,
+      "learning_rate": 0.0002,
+      "loss": 1.1247,
+      "step": 5400
+    },
+    {
+      "epoch": 3.916033297140789,
+      "grad_norm": 0.6379870772361755,
+      "learning_rate": 0.0002,
+      "loss": 1.2212,
+      "step": 5410
+    },
+    {
+      "epoch": 3.9232718060079623,
+      "grad_norm": 1.0313376188278198,
+      "learning_rate": 0.0002,
+      "loss": 1.2481,
+      "step": 5420
+    },
+    {
+      "epoch": 3.9305103148751357,
+      "grad_norm": 0.8220619559288025,
+      "learning_rate": 0.0002,
+      "loss": 1.1872,
+      "step": 5430
+    },
+    {
+      "epoch": 3.937748823742309,
+      "grad_norm": 0.7576116919517517,
+      "learning_rate": 0.0002,
+      "loss": 1.2006,
+      "step": 5440
+    },
+    {
+      "epoch": 3.9449873326094824,
+      "grad_norm": 1.226235032081604,
+      "learning_rate": 0.0002,
+      "loss": 1.1969,
+      "step": 5450
+    },
+    {
+      "epoch": 3.952225841476656,
+      "grad_norm": 0.7979229688644409,
+      "learning_rate": 0.0002,
+      "loss": 1.2945,
+      "step": 5460
+    },
+    {
+      "epoch": 3.959464350343829,
+      "grad_norm": 0.9911929965019226,
+      "learning_rate": 0.0002,
+      "loss": 1.1922,
+      "step": 5470
+    },
+    {
+      "epoch": 3.9667028592110025,
+      "grad_norm": 0.643738865852356,
+      "learning_rate": 0.0002,
+      "loss": 1.0924,
+      "step": 5480
+    },
+    {
+      "epoch": 3.973941368078176,
+      "grad_norm": 0.682305634021759,
+      "learning_rate": 0.0002,
+      "loss": 1.0607,
+      "step": 5490
+    },
+    {
+      "epoch": 3.9811798769453492,
+      "grad_norm": 1.18373441696167,
+      "learning_rate": 0.0002,
+      "loss": 1.2908,
+      "step": 5500
+    },
+    {
+      "epoch": 3.9884183858125226,
+      "grad_norm": 0.7190203070640564,
+      "learning_rate": 0.0002,
+      "loss": 1.0889,
+      "step": 5510
+    },
+    {
+      "epoch": 3.995656894679696,
+      "grad_norm": 0.7516948580741882,
+      "learning_rate": 0.0002,
+      "loss": 1.2745,
+      "step": 5520
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.4252897500991821,
+      "eval_runtime": 27.235,
+      "eval_samples_per_second": 16.009,
+      "eval_steps_per_second": 2.019,
+      "step": 5526
+    },
+    {
+      "epoch": 4.002895403546869,
+      "grad_norm": 0.6353074312210083,
+      "learning_rate": 0.0002,
+      "loss": 1.0088,
+      "step": 5530
+    },
+    {
+      "epoch": 4.010133912414043,
+      "grad_norm": 0.7424906492233276,
+      "learning_rate": 0.0002,
+      "loss": 1.0326,
+      "step": 5540
+    },
+    {
+      "epoch": 4.017372421281216,
+      "grad_norm": 0.8856638073921204,
+      "learning_rate": 0.0002,
+      "loss": 1.0667,
+      "step": 5550
+    },
+    {
+      "epoch": 4.024610930148389,
+      "grad_norm": 0.9627974033355713,
+      "learning_rate": 0.0002,
+      "loss": 1.0905,
+      "step": 5560
+    },
+    {
+      "epoch": 4.031849439015563,
+      "grad_norm": 0.9048978686332703,
+      "learning_rate": 0.0002,
+      "loss": 1.0965,
+      "step": 5570
+    },
+    {
+      "epoch": 4.039087947882736,
+      "grad_norm": 0.921119213104248,
+      "learning_rate": 0.0002,
+      "loss": 1.1108,
+      "step": 5580
+    },
+    {
+      "epoch": 4.0463264567499095,
+      "grad_norm": 0.8654361963272095,
+      "learning_rate": 0.0002,
+      "loss": 1.1235,
+      "step": 5590
+    },
+    {
+      "epoch": 4.053564965617083,
+      "grad_norm": 0.7947945594787598,
+      "learning_rate": 0.0002,
+      "loss": 1.0794,
+      "step": 5600
+    },
+    {
+      "epoch": 4.060803474484256,
+      "grad_norm": 0.8307326436042786,
+      "learning_rate": 0.0002,
+      "loss": 1.0674,
+      "step": 5610
+    },
+    {
+      "epoch": 4.06804198335143,
+      "grad_norm": 0.793273389339447,
+      "learning_rate": 0.0002,
+      "loss": 1.0076,
+      "step": 5620
+    },
+    {
+      "epoch": 4.075280492218603,
+      "grad_norm": 0.8748673796653748,
+      "learning_rate": 0.0002,
+      "loss": 1.0651,
+      "step": 5630
+    },
+    {
+      "epoch": 4.082519001085776,
+      "grad_norm": 0.7926856279373169,
+      "learning_rate": 0.0002,
+      "loss": 1.111,
+      "step": 5640
+    },
+    {
+      "epoch": 4.08975750995295,
+      "grad_norm": 0.922645092010498,
+      "learning_rate": 0.0002,
+      "loss": 1.044,
+      "step": 5650
+    },
+    {
+      "epoch": 4.096996018820123,
+      "grad_norm": 0.9539641737937927,
+      "learning_rate": 0.0002,
+      "loss": 1.109,
+      "step": 5660
+    },
+    {
+      "epoch": 4.1042345276872965,
+      "grad_norm": 0.8674443364143372,
+      "learning_rate": 0.0002,
+      "loss": 1.0788,
+      "step": 5670
+    },
+    {
+      "epoch": 4.11147303655447,
+      "grad_norm": 0.7097609043121338,
+      "learning_rate": 0.0002,
+      "loss": 0.9867,
+      "step": 5680
+    },
+    {
+      "epoch": 4.118711545421643,
+      "grad_norm": 0.8875522613525391,
+      "learning_rate": 0.0002,
+      "loss": 1.1154,
+      "step": 5690
+    },
+    {
+      "epoch": 4.125950054288817,
+      "grad_norm": 0.8583634495735168,
+      "learning_rate": 0.0002,
+      "loss": 1.1217,
+      "step": 5700
+    },
+    {
+      "epoch": 4.13318856315599,
+      "grad_norm": 0.6736377477645874,
+      "learning_rate": 0.0002,
+      "loss": 1.0973,
+      "step": 5710
+    },
+    {
+      "epoch": 4.140427072023163,
+      "grad_norm": 0.9349062442779541,
+      "learning_rate": 0.0002,
+      "loss": 1.1199,
+      "step": 5720
+    },
+    {
+      "epoch": 4.147665580890337,
+      "grad_norm": 1.0610365867614746,
+      "learning_rate": 0.0002,
+      "loss": 1.0508,
+      "step": 5730
+    },
+    {
+      "epoch": 4.15490408975751,
+      "grad_norm": 1.5838189125061035,
+      "learning_rate": 0.0002,
+      "loss": 1.1146,
+      "step": 5740
+    },
+    {
+      "epoch": 4.162142598624683,
+      "grad_norm": 0.747522234916687,
+      "learning_rate": 0.0002,
+      "loss": 1.0222,
+      "step": 5750
+    },
+    {
+      "epoch": 4.169381107491857,
+      "grad_norm": 1.3247915506362915,
+      "learning_rate": 0.0002,
+      "loss": 1.1328,
+      "step": 5760
+    },
+    {
+      "epoch": 4.17661961635903,
+      "grad_norm": 0.8750247955322266,
+      "learning_rate": 0.0002,
+      "loss": 1.1655,
+      "step": 5770
+    },
+    {
+      "epoch": 4.1838581252262035,
+      "grad_norm": 0.7914144992828369,
+      "learning_rate": 0.0002,
+      "loss": 1.199,
+      "step": 5780
+    },
+    {
+      "epoch": 4.191096634093377,
+      "grad_norm": 0.9493299126625061,
+      "learning_rate": 0.0002,
+      "loss": 1.1213,
+      "step": 5790
+    },
+    {
+      "epoch": 4.19833514296055,
+      "grad_norm": 0.7802295088768005,
+      "learning_rate": 0.0002,
+      "loss": 1.1515,
+      "step": 5800
+    },
+    {
+      "epoch": 4.205573651827724,
+      "grad_norm": 0.6987314820289612,
+      "learning_rate": 0.0002,
+      "loss": 1.0704,
+      "step": 5810
+    },
+    {
+      "epoch": 4.212812160694897,
+      "grad_norm": 0.9220341444015503,
+      "learning_rate": 0.0002,
+      "loss": 1.1699,
+      "step": 5820
+    },
+    {
+      "epoch": 4.22005066956207,
+      "grad_norm": 0.8932939767837524,
+      "learning_rate": 0.0002,
+      "loss": 1.1394,
+      "step": 5830
+    },
+    {
+      "epoch": 4.227289178429244,
+      "grad_norm": 0.920002818107605,
+      "learning_rate": 0.0002,
+      "loss": 1.0048,
+      "step": 5840
+    },
+    {
+      "epoch": 4.234527687296417,
+      "grad_norm": 0.6662752032279968,
+      "learning_rate": 0.0002,
+      "loss": 0.964,
+      "step": 5850
+    },
+    {
+      "epoch": 4.24176619616359,
+      "grad_norm": 0.8679718971252441,
+      "learning_rate": 0.0002,
+      "loss": 0.986,
+      "step": 5860
+    },
+    {
+      "epoch": 4.249004705030764,
+      "grad_norm": 0.7020887732505798,
+      "learning_rate": 0.0002,
+      "loss": 0.8991,
+      "step": 5870
+    },
+    {
+      "epoch": 4.256243213897937,
+      "grad_norm": 0.869611382484436,
+      "learning_rate": 0.0002,
+      "loss": 1.1132,
+      "step": 5880
+    },
+    {
+      "epoch": 4.2634817227651105,
+      "grad_norm": 0.7796585559844971,
+      "learning_rate": 0.0002,
+      "loss": 1.1026,
+      "step": 5890
+    },
+    {
+      "epoch": 4.270720231632284,
+      "grad_norm": 0.8978819251060486,
+      "learning_rate": 0.0002,
+      "loss": 1.0957,
+      "step": 5900
+    },
+    {
+      "epoch": 4.277958740499457,
+      "grad_norm": 1.0837205648422241,
+      "learning_rate": 0.0002,
+      "loss": 1.1325,
+      "step": 5910
+    },
+    {
+      "epoch": 4.285197249366631,
+      "grad_norm": 0.7584353089332581,
+      "learning_rate": 0.0002,
+      "loss": 1.1279,
+      "step": 5920
+    },
+    {
+      "epoch": 4.292435758233804,
+      "grad_norm": 0.7313185334205627,
+      "learning_rate": 0.0002,
+      "loss": 1.0513,
+      "step": 5930
+    },
+    {
+      "epoch": 4.299674267100977,
+      "grad_norm": 0.8004671335220337,
+      "learning_rate": 0.0002,
+      "loss": 1.1101,
+      "step": 5940
+    },
+    {
+      "epoch": 4.306912775968151,
+      "grad_norm": 2.154958724975586,
+      "learning_rate": 0.0002,
+      "loss": 1.14,
+      "step": 5950
+    },
+    {
+      "epoch": 4.314151284835324,
+      "grad_norm": 0.9163479804992676,
+      "learning_rate": 0.0002,
+      "loss": 1.1206,
+      "step": 5960
+    },
+    {
+      "epoch": 4.321389793702497,
+      "grad_norm": 0.9151589274406433,
+      "learning_rate": 0.0002,
+      "loss": 0.9941,
+      "step": 5970
+    },
+    {
+      "epoch": 4.328628302569671,
+      "grad_norm": 0.8624112010002136,
+      "learning_rate": 0.0002,
+      "loss": 1.0606,
+      "step": 5980
+    },
+    {
+      "epoch": 4.335866811436844,
+      "grad_norm": 0.9357741475105286,
+      "learning_rate": 0.0002,
+      "loss": 1.1625,
+      "step": 5990
+    },
+    {
+      "epoch": 4.3431053203040175,
+      "grad_norm": 1.3482335805892944,
+      "learning_rate": 0.0002,
+      "loss": 1.0712,
+      "step": 6000
+    },
+    {
+      "epoch": 4.350343829171191,
+      "grad_norm": 0.7156149744987488,
+      "learning_rate": 0.0002,
+      "loss": 1.1224,
+      "step": 6010
+    },
+    {
+      "epoch": 4.357582338038364,
+      "grad_norm": 0.8480049967765808,
+      "learning_rate": 0.0002,
+      "loss": 1.0753,
+      "step": 6020
+    },
+    {
+      "epoch": 4.364820846905538,
+      "grad_norm": 0.8262244462966919,
+      "learning_rate": 0.0002,
+      "loss": 1.051,
+      "step": 6030
+    },
+    {
+      "epoch": 4.372059355772711,
+      "grad_norm": 0.7733905911445618,
+      "learning_rate": 0.0002,
+      "loss": 0.9966,
+      "step": 6040
+    },
+    {
+      "epoch": 4.379297864639884,
+      "grad_norm": 0.8553919792175293,
+      "learning_rate": 0.0002,
+      "loss": 1.1008,
+      "step": 6050
+    },
+    {
+      "epoch": 4.386536373507058,
+      "grad_norm": 0.8666832447052002,
+      "learning_rate": 0.0002,
+      "loss": 1.1777,
+      "step": 6060
+    },
+    {
+      "epoch": 4.393774882374231,
+      "grad_norm": 0.9168295860290527,
+      "learning_rate": 0.0002,
+      "loss": 1.1934,
+      "step": 6070
+    },
+    {
+      "epoch": 4.4010133912414044,
+      "grad_norm": 0.7315238118171692,
+      "learning_rate": 0.0002,
+      "loss": 1.0988,
+      "step": 6080
+    },
+    {
+      "epoch": 4.408251900108578,
+      "grad_norm": 1.020263433456421,
+      "learning_rate": 0.0002,
+      "loss": 1.1599,
+      "step": 6090
+    },
+    {
+      "epoch": 4.415490408975751,
+      "grad_norm": 0.9978243708610535,
+      "learning_rate": 0.0002,
+      "loss": 1.133,
+      "step": 6100
+    },
+    {
+      "epoch": 4.4227289178429245,
+      "grad_norm": 0.995453953742981,
+      "learning_rate": 0.0002,
+      "loss": 1.1324,
+      "step": 6110
+    },
+    {
+      "epoch": 4.429967426710098,
+      "grad_norm": 0.9360884428024292,
+      "learning_rate": 0.0002,
+      "loss": 1.0957,
+      "step": 6120
+    },
+    {
+      "epoch": 4.437205935577271,
+      "grad_norm": 0.8099448084831238,
+      "learning_rate": 0.0002,
+      "loss": 0.9506,
+      "step": 6130
+    },
+    {
+      "epoch": 4.444444444444445,
+      "grad_norm": 0.8173841238021851,
+      "learning_rate": 0.0002,
+      "loss": 1.0887,
+      "step": 6140
+    },
+    {
+      "epoch": 4.451682953311618,
+      "grad_norm": 0.7972666025161743,
+      "learning_rate": 0.0002,
+      "loss": 1.1219,
+      "step": 6150
+    },
+    {
+      "epoch": 4.458921462178791,
+      "grad_norm": 0.7685779333114624,
+      "learning_rate": 0.0002,
+      "loss": 1.0226,
+      "step": 6160
+    },
+    {
+      "epoch": 4.466159971045965,
+      "grad_norm": 0.7872623801231384,
+      "learning_rate": 0.0002,
+      "loss": 1.0732,
+      "step": 6170
+    },
+    {
+      "epoch": 4.473398479913138,
+      "grad_norm": 0.7677070498466492,
+      "learning_rate": 0.0002,
+      "loss": 0.9911,
+      "step": 6180
+    },
+    {
+      "epoch": 4.4806369887803115,
+      "grad_norm": 0.7878316044807434,
+      "learning_rate": 0.0002,
+      "loss": 1.0919,
+      "step": 6190
+    },
+    {
+      "epoch": 4.487875497647485,
+      "grad_norm": 0.8178079724311829,
+      "learning_rate": 0.0002,
+      "loss": 1.018,
+      "step": 6200
+    },
+    {
+      "epoch": 4.495114006514658,
+      "grad_norm": 1.2820082902908325,
+      "learning_rate": 0.0002,
+      "loss": 1.0517,
+      "step": 6210
+    },
+    {
+      "epoch": 4.502352515381832,
+      "grad_norm": 0.9380832314491272,
+      "learning_rate": 0.0002,
+      "loss": 1.3101,
+      "step": 6220
+    },
+    {
+      "epoch": 4.509591024249005,
+      "grad_norm": 0.7810422778129578,
+      "learning_rate": 0.0002,
+      "loss": 0.9818,
+      "step": 6230
+    },
+    {
+      "epoch": 4.516829533116178,
+      "grad_norm": 1.1022917032241821,
+      "learning_rate": 0.0002,
+      "loss": 1.1677,
+      "step": 6240
+    },
+    {
+      "epoch": 4.524068041983352,
+      "grad_norm": 1.4275553226470947,
+      "learning_rate": 0.0002,
+      "loss": 1.1579,
+      "step": 6250
+    },
+    {
+      "epoch": 4.531306550850525,
+      "grad_norm": 0.7597777247428894,
+      "learning_rate": 0.0002,
+      "loss": 1.3237,
+      "step": 6260
+    },
+    {
+      "epoch": 4.538545059717698,
+      "grad_norm": 1.10992431640625,
+      "learning_rate": 0.0002,
+      "loss": 1.1529,
+      "step": 6270
+    },
+    {
+      "epoch": 4.545783568584872,
+      "grad_norm": 0.8981178998947144,
+      "learning_rate": 0.0002,
+      "loss": 1.0732,
+      "step": 6280
+    },
+    {
+      "epoch": 4.553022077452045,
+      "grad_norm": 0.7863979339599609,
+      "learning_rate": 0.0002,
+      "loss": 1.086,
+      "step": 6290
+    },
+    {
+      "epoch": 4.5602605863192185,
+      "grad_norm": 0.9071474671363831,
+      "learning_rate": 0.0002,
+      "loss": 1.2008,
+      "step": 6300
+    },
+    {
+      "epoch": 4.567499095186392,
+      "grad_norm": 0.7429424524307251,
+      "learning_rate": 0.0002,
+      "loss": 1.0916,
+      "step": 6310
+    },
+    {
+      "epoch": 4.574737604053565,
+      "grad_norm": 1.0767850875854492,
+      "learning_rate": 0.0002,
+      "loss": 1.095,
+      "step": 6320
+    },
+    {
+      "epoch": 4.581976112920739,
+      "grad_norm": 0.7885915637016296,
+      "learning_rate": 0.0002,
+      "loss": 1.1023,
+      "step": 6330
+    },
+    {
+      "epoch": 4.589214621787912,
+      "grad_norm": 0.8350457549095154,
+      "learning_rate": 0.0002,
+      "loss": 1.1131,
+      "step": 6340
+    },
+    {
+      "epoch": 4.596453130655085,
+      "grad_norm": 0.7853530645370483,
+      "learning_rate": 0.0002,
+      "loss": 1.0743,
+      "step": 6350
+    },
+    {
+      "epoch": 4.603691639522259,
+      "grad_norm": 1.1220661401748657,
+      "learning_rate": 0.0002,
+      "loss": 1.1912,
+      "step": 6360
+    },
+    {
+      "epoch": 4.610930148389432,
+      "grad_norm": 0.7959423065185547,
+      "learning_rate": 0.0002,
+      "loss": 1.0927,
+      "step": 6370
+    },
+    {
+      "epoch": 4.618168657256605,
+      "grad_norm": 0.7782652378082275,
+      "learning_rate": 0.0002,
+      "loss": 1.1542,
+      "step": 6380
+    },
+    {
+      "epoch": 4.625407166123779,
+      "grad_norm": 0.7882203459739685,
+      "learning_rate": 0.0002,
+      "loss": 1.0753,
+      "step": 6390
+    },
+    {
+      "epoch": 4.632645674990952,
+      "grad_norm": 0.8841899037361145,
+      "learning_rate": 0.0002,
+      "loss": 1.0676,
+      "step": 6400
+    },
+    {
+      "epoch": 4.6398841838581255,
+      "grad_norm": 0.7936127781867981,
+      "learning_rate": 0.0002,
+      "loss": 1.0815,
+      "step": 6410
+    },
+    {
+      "epoch": 4.647122692725299,
+      "grad_norm": 0.9213966131210327,
+      "learning_rate": 0.0002,
+      "loss": 1.0198,
+      "step": 6420
+    },
+    {
+      "epoch": 4.654361201592472,
+      "grad_norm": 0.9246473908424377,
+      "learning_rate": 0.0002,
+      "loss": 0.9872,
+      "step": 6430
+    },
+    {
+      "epoch": 4.661599710459646,
+      "grad_norm": 0.766572892665863,
+      "learning_rate": 0.0002,
+      "loss": 1.1309,
+      "step": 6440
+    },
+    {
+      "epoch": 4.668838219326819,
+      "grad_norm": 0.8596171736717224,
+      "learning_rate": 0.0002,
+      "loss": 1.1095,
+      "step": 6450
+    },
+    {
+      "epoch": 4.676076728193992,
+      "grad_norm": 0.8482751846313477,
+      "learning_rate": 0.0002,
+      "loss": 1.1869,
+      "step": 6460
+    },
+    {
+      "epoch": 4.683315237061166,
+      "grad_norm": 1.0826905965805054,
+      "learning_rate": 0.0002,
+      "loss": 1.0622,
+      "step": 6470
+    },
+    {
+      "epoch": 4.690553745928339,
+      "grad_norm": 1.1048457622528076,
+      "learning_rate": 0.0002,
+      "loss": 1.0256,
+      "step": 6480
+    },
+    {
+      "epoch": 4.697792254795512,
+      "grad_norm": 0.9429134726524353,
+      "learning_rate": 0.0002,
+      "loss": 1.0514,
+      "step": 6490
+    },
+    {
+      "epoch": 4.705030763662686,
+      "grad_norm": 0.8587502837181091,
+      "learning_rate": 0.0002,
+      "loss": 1.1351,
+      "step": 6500
+    },
+    {
+      "epoch": 4.712269272529859,
+      "grad_norm": 1.0387083292007446,
+      "learning_rate": 0.0002,
+      "loss": 1.0969,
+      "step": 6510
+    },
+    {
+      "epoch": 4.7195077813970325,
+      "grad_norm": 0.7471951842308044,
+      "learning_rate": 0.0002,
+      "loss": 1.0493,
+      "step": 6520
+    },
+    {
+      "epoch": 4.726746290264206,
+      "grad_norm": 0.8800424933433533,
+      "learning_rate": 0.0002,
+      "loss": 1.2632,
+      "step": 6530
+    },
+    {
+      "epoch": 4.733984799131379,
+      "grad_norm": 0.8136811852455139,
+      "learning_rate": 0.0002,
+      "loss": 1.2126,
+      "step": 6540
+    },
+    {
+      "epoch": 4.741223307998553,
+      "grad_norm": 0.9910339713096619,
+      "learning_rate": 0.0002,
+      "loss": 1.195,
+      "step": 6550
+    },
+    {
+      "epoch": 4.748461816865726,
+      "grad_norm": 1.0679163932800293,
+      "learning_rate": 0.0002,
+      "loss": 1.1201,
+      "step": 6560
+    },
+    {
+      "epoch": 4.755700325732899,
+      "grad_norm": 0.8468248248100281,
+      "learning_rate": 0.0002,
+      "loss": 1.0297,
+      "step": 6570
+    },
+    {
+      "epoch": 4.762938834600073,
+      "grad_norm": 0.8771235942840576,
+      "learning_rate": 0.0002,
+      "loss": 1.0858,
+      "step": 6580
+    },
+    {
+      "epoch": 4.770177343467246,
+      "grad_norm": 0.7024846076965332,
+      "learning_rate": 0.0002,
+      "loss": 1.077,
+      "step": 6590
+    },
+    {
+      "epoch": 4.7774158523344195,
+      "grad_norm": 0.7836683392524719,
+      "learning_rate": 0.0002,
+      "loss": 1.0876,
+      "step": 6600
+    },
+    {
+      "epoch": 4.784654361201593,
+      "grad_norm": 0.7717288136482239,
+      "learning_rate": 0.0002,
+      "loss": 1.1006,
+      "step": 6610
+    },
+    {
+      "epoch": 4.791892870068766,
+      "grad_norm": 0.884183943271637,
+      "learning_rate": 0.0002,
+      "loss": 1.0376,
+      "step": 6620
+    },
+    {
+      "epoch": 4.7991313789359396,
+      "grad_norm": 1.383867621421814,
+      "learning_rate": 0.0002,
+      "loss": 1.1757,
+      "step": 6630
+    },
+    {
+      "epoch": 4.806369887803113,
+      "grad_norm": 0.9741523861885071,
+      "learning_rate": 0.0002,
+      "loss": 1.0861,
+      "step": 6640
+    },
+    {
+      "epoch": 4.813608396670286,
+      "grad_norm": 0.9723693132400513,
+      "learning_rate": 0.0002,
+      "loss": 1.0884,
+      "step": 6650
+    },
+    {
+      "epoch": 4.82084690553746,
+      "grad_norm": 1.8324809074401855,
+      "learning_rate": 0.0002,
+      "loss": 1.2203,
+      "step": 6660
+    },
+    {
+      "epoch": 4.828085414404633,
+      "grad_norm": 0.904909074306488,
+      "learning_rate": 0.0002,
+      "loss": 1.0292,
+      "step": 6670
+    },
+    {
+      "epoch": 4.835323923271806,
+      "grad_norm": 0.7355411648750305,
+      "learning_rate": 0.0002,
+      "loss": 1.0349,
+      "step": 6680
+    },
+    {
+      "epoch": 4.84256243213898,
+      "grad_norm": 0.8934960961341858,
+      "learning_rate": 0.0002,
+      "loss": 1.0793,
+      "step": 6690
+    },
+    {
+      "epoch": 4.849800941006153,
+      "grad_norm": 1.4596954584121704,
+      "learning_rate": 0.0002,
+      "loss": 1.0375,
+      "step": 6700
+    },
+    {
+      "epoch": 4.8570394498733265,
+      "grad_norm": 0.8310341238975525,
+      "learning_rate": 0.0002,
+      "loss": 1.1065,
+      "step": 6710
+    },
+    {
+      "epoch": 4.8642779587405,
+      "grad_norm": 0.9709894061088562,
+      "learning_rate": 0.0002,
+      "loss": 1.1089,
+      "step": 6720
+    },
+    {
+      "epoch": 4.871516467607673,
+      "grad_norm": 0.852142333984375,
+      "learning_rate": 0.0002,
+      "loss": 1.0069,
+      "step": 6730
+    },
+    {
+      "epoch": 4.878754976474847,
+      "grad_norm": 1.0643625259399414,
+      "learning_rate": 0.0002,
+      "loss": 1.0507,
+      "step": 6740
+    },
+    {
+      "epoch": 4.88599348534202,
+      "grad_norm": 0.9419508576393127,
+      "learning_rate": 0.0002,
+      "loss": 1.056,
+      "step": 6750
+    },
+    {
+      "epoch": 4.893231994209193,
+      "grad_norm": 1.1818498373031616,
+      "learning_rate": 0.0002,
+      "loss": 1.1995,
+      "step": 6760
+    },
+    {
+      "epoch": 4.900470503076367,
+      "grad_norm": 0.9369569420814514,
+      "learning_rate": 0.0002,
+      "loss": 1.0925,
+      "step": 6770
+    },
+    {
+      "epoch": 4.90770901194354,
+      "grad_norm": 0.7012579441070557,
+      "learning_rate": 0.0002,
+      "loss": 1.1648,
+      "step": 6780
+    },
+    {
+      "epoch": 4.914947520810713,
+      "grad_norm": 0.9109319448471069,
+      "learning_rate": 0.0002,
+      "loss": 1.0926,
+      "step": 6790
+    },
+    {
+      "epoch": 4.922186029677887,
+      "grad_norm": 0.8077534437179565,
+      "learning_rate": 0.0002,
+      "loss": 1.0358,
+      "step": 6800
+    },
+    {
+      "epoch": 4.92942453854506,
+      "grad_norm": 0.7571148872375488,
+      "learning_rate": 0.0002,
+      "loss": 1.2549,
+      "step": 6810
+    },
+    {
+      "epoch": 4.9366630474122335,
+      "grad_norm": 0.7325633764266968,
+      "learning_rate": 0.0002,
+      "loss": 0.9638,
+      "step": 6820
+    },
+    {
+      "epoch": 4.943901556279407,
+      "grad_norm": 0.8465084433555603,
+      "learning_rate": 0.0002,
+      "loss": 1.0128,
+      "step": 6830
+    },
+    {
+      "epoch": 4.95114006514658,
+      "grad_norm": 0.8753737807273865,
+      "learning_rate": 0.0002,
+      "loss": 1.153,
+      "step": 6840
+    },
+    {
+      "epoch": 4.958378574013754,
+      "grad_norm": 0.9421748518943787,
+      "learning_rate": 0.0002,
+      "loss": 1.0247,
+      "step": 6850
+    },
+    {
+      "epoch": 4.965617082880927,
+      "grad_norm": 0.8245896697044373,
+      "learning_rate": 0.0002,
+      "loss": 1.1483,
+      "step": 6860
+    },
+    {
+      "epoch": 4.9728555917481,
+      "grad_norm": 0.8823089599609375,
+      "learning_rate": 0.0002,
+      "loss": 0.9905,
+      "step": 6870
+    },
+    {
+      "epoch": 4.980094100615274,
+      "grad_norm": 0.8406389355659485,
+      "learning_rate": 0.0002,
+      "loss": 1.1664,
+      "step": 6880
+    },
+    {
+      "epoch": 4.987332609482447,
+      "grad_norm": 0.9732868075370789,
+      "learning_rate": 0.0002,
+      "loss": 1.0944,
+      "step": 6890
+    },
+    {
+      "epoch": 4.99457111834962,
+      "grad_norm": 2.125141143798828,
+      "learning_rate": 0.0002,
+      "loss": 1.1776,
+      "step": 6900
+    },
+    {
+      "epoch": 4.999638074556641,
+      "eval_loss": 1.445176601409912,
+      "eval_runtime": 27.2351,
+      "eval_samples_per_second": 16.009,
+      "eval_steps_per_second": 2.019,
+      "step": 6907
+    },
+    {
+      "epoch": 5.001809627216793,
+      "grad_norm": 0.9465792775154114,
+      "learning_rate": 0.0002,
+      "loss": 1.1362,
+      "step": 6910
+    },
+    {
+      "epoch": 5.009048136083966,
+      "grad_norm": 1.2834891080856323,
+      "learning_rate": 0.0002,
+      "loss": 0.982,
+      "step": 6920
+    },
+    {
+      "epoch": 5.01628664495114,
+      "grad_norm": 1.0297378301620483,
+      "learning_rate": 0.0002,
+      "loss": 0.9803,
+      "step": 6930
+    },
+    {
+      "epoch": 5.023525153818313,
+      "grad_norm": 1.1705161333084106,
+      "learning_rate": 0.0002,
+      "loss": 1.0447,
+      "step": 6940
+    },
+    {
+      "epoch": 5.030763662685486,
+      "grad_norm": 0.8293961882591248,
+      "learning_rate": 0.0002,
+      "loss": 1.0113,
+      "step": 6950
+    },
+    {
+      "epoch": 5.03800217155266,
+      "grad_norm": 1.0422210693359375,
+      "learning_rate": 0.0002,
+      "loss": 0.9203,
+      "step": 6960
+    },
+    {
+      "epoch": 5.045240680419833,
+      "grad_norm": 1.116104245185852,
+      "learning_rate": 0.0002,
+      "loss": 1.0553,
+      "step": 6970
+    },
+    {
+      "epoch": 5.0524791892870065,
+      "grad_norm": 1.5118416547775269,
+      "learning_rate": 0.0002,
+      "loss": 0.9011,
+      "step": 6980
+    },
+    {
+      "epoch": 5.05971769815418,
+      "grad_norm": 0.8383979797363281,
+      "learning_rate": 0.0002,
+      "loss": 0.9969,
+      "step": 6990
+    },
+    {
+      "epoch": 5.066956207021353,
+      "grad_norm": 1.3378649950027466,
+      "learning_rate": 0.0002,
+      "loss": 0.9659,
+      "step": 7000
+    },
+    {
+      "epoch": 5.0741947158885266,
+      "grad_norm": 1.1840510368347168,
+      "learning_rate": 0.0002,
+      "loss": 1.0212,
+      "step": 7010
+    },
+    {
+      "epoch": 5.0814332247557,
+      "grad_norm": 1.2354751825332642,
+      "learning_rate": 0.0002,
+      "loss": 0.9939,
+      "step": 7020
+    },
+    {
+      "epoch": 5.088671733622873,
+      "grad_norm": 1.3830451965332031,
+      "learning_rate": 0.0002,
+      "loss": 0.9831,
+      "step": 7030
+    },
+    {
+      "epoch": 5.095910242490047,
+      "grad_norm": 0.8101674318313599,
+      "learning_rate": 0.0002,
+      "loss": 1.1827,
+      "step": 7040
+    },
+    {
+      "epoch": 5.10314875135722,
+      "grad_norm": 0.897982656955719,
+      "learning_rate": 0.0002,
+      "loss": 0.9255,
+      "step": 7050
+    },
+    {
+      "epoch": 5.110387260224393,
+      "grad_norm": 1.2049678564071655,
+      "learning_rate": 0.0002,
+      "loss": 0.8784,
+      "step": 7060
+    },
+    {
+      "epoch": 5.117625769091567,
+      "grad_norm": 1.5912116765975952,
+      "learning_rate": 0.0002,
+      "loss": 1.0182,
+      "step": 7070
+    },
+    {
+      "epoch": 5.12486427795874,
+      "grad_norm": 0.9261530041694641,
+      "learning_rate": 0.0002,
+      "loss": 1.0909,
+      "step": 7080
+    },
+    {
+      "epoch": 5.1321027868259135,
+      "grad_norm": 1.1454812288284302,
+      "learning_rate": 0.0002,
+      "loss": 0.9603,
+      "step": 7090
+    },
+    {
+      "epoch": 5.139341295693087,
+      "grad_norm": 1.0049978494644165,
+      "learning_rate": 0.0002,
+      "loss": 0.9149,
+      "step": 7100
+    },
+    {
+      "epoch": 5.14657980456026,
+      "grad_norm": 1.4513251781463623,
+      "learning_rate": 0.0002,
+      "loss": 0.9463,
+      "step": 7110
+    },
+    {
+      "epoch": 5.153818313427434,
+      "grad_norm": 0.9800849556922913,
+      "learning_rate": 0.0002,
+      "loss": 0.8995,
+      "step": 7120
+    },
+    {
+      "epoch": 5.161056822294607,
+      "grad_norm": 0.9698708653450012,
+      "learning_rate": 0.0002,
+      "loss": 0.9835,
+      "step": 7130
+    },
+    {
+      "epoch": 5.16829533116178,
+      "grad_norm": 1.1126646995544434,
+      "learning_rate": 0.0002,
+      "loss": 0.9672,
+      "step": 7140
+    },
+    {
+      "epoch": 5.175533840028954,
+      "grad_norm": 0.9248330593109131,
+      "learning_rate": 0.0002,
+      "loss": 0.9384,
+      "step": 7150
+    },
+    {
+      "epoch": 5.182772348896127,
+      "grad_norm": 0.7967255711555481,
+      "learning_rate": 0.0002,
+      "loss": 0.826,
+      "step": 7160
+    },
+    {
+      "epoch": 5.1900108577633,
+      "grad_norm": 0.9933333992958069,
+      "learning_rate": 0.0002,
+      "loss": 1.0078,
+      "step": 7170
+    },
+    {
+      "epoch": 5.197249366630474,
+      "grad_norm": 1.0080649852752686,
+      "learning_rate": 0.0002,
+      "loss": 1.0276,
+      "step": 7180
+    },
+    {
+      "epoch": 5.204487875497647,
+      "grad_norm": 1.3954921960830688,
+      "learning_rate": 0.0002,
+      "loss": 1.0201,
+      "step": 7190
+    },
+    {
+      "epoch": 5.2117263843648205,
+      "grad_norm": 1.2386271953582764,
+      "learning_rate": 0.0002,
+      "loss": 1.0863,
+      "step": 7200
+    },
+    {
+      "epoch": 5.218964893231994,
+      "grad_norm": 1.2379488945007324,
+      "learning_rate": 0.0002,
+      "loss": 0.8863,
+      "step": 7210
+    },
+    {
+      "epoch": 5.226203402099167,
+      "grad_norm": 0.9882503747940063,
+      "learning_rate": 0.0002,
+      "loss": 1.0518,
+      "step": 7220
+    },
+    {
+      "epoch": 5.233441910966341,
+      "grad_norm": 1.1728729009628296,
+      "learning_rate": 0.0002,
+      "loss": 0.9834,
+      "step": 7230
+    },
+    {
+      "epoch": 5.240680419833514,
+      "grad_norm": 0.9849673509597778,
+      "learning_rate": 0.0002,
+      "loss": 0.9269,
+      "step": 7240
+    },
+    {
+      "epoch": 5.247918928700687,
+      "grad_norm": 1.177639365196228,
+      "learning_rate": 0.0002,
+      "loss": 0.9935,
+      "step": 7250
+    },
+    {
+      "epoch": 5.255157437567861,
+      "grad_norm": 1.2395055294036865,
+      "learning_rate": 0.0002,
+      "loss": 1.0639,
+      "step": 7260
+    },
+    {
+      "epoch": 5.262395946435034,
+      "grad_norm": 1.3999171257019043,
+      "learning_rate": 0.0002,
+      "loss": 1.0138,
+      "step": 7270
+    },
+    {
+      "epoch": 5.269634455302207,
+      "grad_norm": 0.7698732018470764,
+      "learning_rate": 0.0002,
+      "loss": 0.9745,
+      "step": 7280
+    },
+    {
+      "epoch": 5.276872964169381,
+      "grad_norm": 0.9167453646659851,
+      "learning_rate": 0.0002,
+      "loss": 1.0389,
+      "step": 7290
+    },
+    {
+      "epoch": 5.284111473036554,
+      "grad_norm": 1.113830804824829,
+      "learning_rate": 0.0002,
+      "loss": 0.9858,
+      "step": 7300
+    },
+    {
+      "epoch": 5.2913499819037275,
+      "grad_norm": 0.9644396901130676,
+      "learning_rate": 0.0002,
+      "loss": 0.9577,
+      "step": 7310
+    },
+    {
+      "epoch": 5.298588490770901,
+      "grad_norm": 1.462435007095337,
+      "learning_rate": 0.0002,
+      "loss": 1.0556,
+      "step": 7320
+    },
+    {
+      "epoch": 5.305826999638074,
+      "grad_norm": 0.9406287670135498,
+      "learning_rate": 0.0002,
+      "loss": 0.871,
+      "step": 7330
+    },
+    {
+      "epoch": 5.313065508505248,
+      "grad_norm": 0.9698247909545898,
+      "learning_rate": 0.0002,
+      "loss": 1.0022,
+      "step": 7340
+    },
+    {
+      "epoch": 5.320304017372421,
+      "grad_norm": 1.12003755569458,
+      "learning_rate": 0.0002,
+      "loss": 0.915,
+      "step": 7350
+    },
+    {
+      "epoch": 5.327542526239594,
+      "grad_norm": 1.598681926727295,
+      "learning_rate": 0.0002,
+      "loss": 0.9838,
+      "step": 7360
+    },
+    {
+      "epoch": 5.334781035106768,
+      "grad_norm": 1.0450010299682617,
+      "learning_rate": 0.0002,
+      "loss": 1.0,
+      "step": 7370
+    },
+    {
+      "epoch": 5.342019543973941,
+      "grad_norm": 0.8680008053779602,
+      "learning_rate": 0.0002,
+      "loss": 0.9983,
+      "step": 7380
+    },
+    {
+      "epoch": 5.349258052841114,
+      "grad_norm": 1.0115476846694946,
+      "learning_rate": 0.0002,
+      "loss": 0.9851,
+      "step": 7390
+    },
+    {
+      "epoch": 5.356496561708288,
+      "grad_norm": 0.9589748382568359,
+      "learning_rate": 0.0002,
+      "loss": 1.0702,
+      "step": 7400
+    },
+    {
+      "epoch": 5.363735070575461,
+      "grad_norm": 0.6729998588562012,
+      "learning_rate": 0.0002,
+      "loss": 0.9366,
+      "step": 7410
+    },
+    {
+      "epoch": 5.3709735794426345,
+      "grad_norm": 0.9246699213981628,
+      "learning_rate": 0.0002,
+      "loss": 1.0126,
+      "step": 7420
+    },
+    {
+      "epoch": 5.378212088309808,
+      "grad_norm": 1.1266791820526123,
+      "learning_rate": 0.0002,
+      "loss": 0.9815,
+      "step": 7430
+    },
+    {
+      "epoch": 5.385450597176981,
+      "grad_norm": 1.8056942224502563,
+      "learning_rate": 0.0002,
+      "loss": 1.1166,
+      "step": 7440
+    },
+    {
+      "epoch": 5.392689106044155,
+      "grad_norm": 0.9802932739257812,
+      "learning_rate": 0.0002,
+      "loss": 0.9604,
+      "step": 7450
+    },
+    {
+      "epoch": 5.399927614911328,
+      "grad_norm": 1.0504707098007202,
+      "learning_rate": 0.0002,
+      "loss": 0.9656,
+      "step": 7460
+    },
+    {
+      "epoch": 5.407166123778501,
+      "grad_norm": 1.1915022134780884,
+      "learning_rate": 0.0002,
+      "loss": 1.0132,
+      "step": 7470
+    },
+    {
+      "epoch": 5.414404632645675,
+      "grad_norm": 1.1856611967086792,
+      "learning_rate": 0.0002,
+      "loss": 1.0041,
+      "step": 7480
+    },
+    {
+      "epoch": 5.421643141512848,
+      "grad_norm": 1.292152762413025,
+      "learning_rate": 0.0002,
+      "loss": 0.9747,
+      "step": 7490
+    },
+    {
+      "epoch": 5.4288816503800215,
+      "grad_norm": 1.2675740718841553,
+      "learning_rate": 0.0002,
+      "loss": 0.9659,
+      "step": 7500
+    },
+    {
+      "epoch": 5.436120159247195,
+      "grad_norm": 1.4034695625305176,
+      "learning_rate": 0.0002,
+      "loss": 1.0271,
+      "step": 7510
+    },
+    {
+      "epoch": 5.443358668114368,
+      "grad_norm": 0.984588623046875,
+      "learning_rate": 0.0002,
+      "loss": 1.0318,
+      "step": 7520
+    },
+    {
+      "epoch": 5.450597176981542,
+      "grad_norm": 0.8419108390808105,
+      "learning_rate": 0.0002,
+      "loss": 1.0726,
+      "step": 7530
+    },
+    {
+      "epoch": 5.457835685848715,
+      "grad_norm": 1.0270143747329712,
+      "learning_rate": 0.0002,
+      "loss": 1.0499,
+      "step": 7540
+    },
+    {
+      "epoch": 5.465074194715888,
+      "grad_norm": 2.2158689498901367,
+      "learning_rate": 0.0002,
+      "loss": 0.9804,
+      "step": 7550
+    },
+    {
+      "epoch": 5.472312703583062,
+      "grad_norm": 1.0740524530410767,
+      "learning_rate": 0.0002,
+      "loss": 0.9856,
+      "step": 7560
+    },
+    {
+      "epoch": 5.479551212450235,
+      "grad_norm": 1.3804482221603394,
+      "learning_rate": 0.0002,
+      "loss": 1.0522,
+      "step": 7570
+    },
+    {
+      "epoch": 5.486789721317408,
+      "grad_norm": 0.9428979754447937,
+      "learning_rate": 0.0002,
+      "loss": 1.0297,
+      "step": 7580
+    },
+    {
+      "epoch": 5.494028230184582,
+      "grad_norm": 0.9548295736312866,
+      "learning_rate": 0.0002,
+      "loss": 1.0906,
+      "step": 7590
+    },
+    {
+      "epoch": 5.501266739051755,
+      "grad_norm": 1.0691065788269043,
+      "learning_rate": 0.0002,
+      "loss": 0.8853,
+      "step": 7600
+    },
+    {
+      "epoch": 5.5085052479189285,
+      "grad_norm": 1.0987380743026733,
+      "learning_rate": 0.0002,
+      "loss": 1.0375,
+      "step": 7610
+    },
+    {
+      "epoch": 5.515743756786102,
+      "grad_norm": 0.9483979344367981,
+      "learning_rate": 0.0002,
+      "loss": 1.0162,
+      "step": 7620
+    },
+    {
+      "epoch": 5.522982265653275,
+      "grad_norm": 1.16624915599823,
+      "learning_rate": 0.0002,
+      "loss": 1.105,
+      "step": 7630
+    },
+    {
+      "epoch": 5.530220774520449,
+      "grad_norm": 0.8563777208328247,
+      "learning_rate": 0.0002,
+      "loss": 0.8695,
+      "step": 7640
+    },
+    {
+      "epoch": 5.537459283387622,
+      "grad_norm": 1.268186092376709,
+      "learning_rate": 0.0002,
+      "loss": 0.9297,
+      "step": 7650
+    },
+    {
+      "epoch": 5.544697792254795,
+      "grad_norm": 1.0752092599868774,
+      "learning_rate": 0.0002,
+      "loss": 1.1152,
+      "step": 7660
+    },
+    {
+      "epoch": 5.551936301121969,
+      "grad_norm": 1.210389256477356,
+      "learning_rate": 0.0002,
+      "loss": 0.9344,
+      "step": 7670
+    },
+    {
+      "epoch": 5.559174809989142,
+      "grad_norm": 1.669063925743103,
+      "learning_rate": 0.0002,
+      "loss": 1.0349,
+      "step": 7680
+    },
+    {
+      "epoch": 5.566413318856315,
+      "grad_norm": 1.038020133972168,
+      "learning_rate": 0.0002,
+      "loss": 0.9833,
+      "step": 7690
+    },
+    {
+      "epoch": 5.573651827723489,
+      "grad_norm": 1.316673994064331,
+      "learning_rate": 0.0002,
+      "loss": 0.8907,
+      "step": 7700
+    },
+    {
+      "epoch": 5.580890336590662,
+      "grad_norm": 1.029935359954834,
+      "learning_rate": 0.0002,
+      "loss": 0.9614,
+      "step": 7710
+    },
+    {
+      "epoch": 5.5881288454578355,
+      "grad_norm": 0.9401940703392029,
+      "learning_rate": 0.0002,
+      "loss": 1.0409,
+      "step": 7720
+    },
+    {
+      "epoch": 5.595367354325009,
+      "grad_norm": 2.4811816215515137,
+      "learning_rate": 0.0002,
+      "loss": 0.9272,
+      "step": 7730
+    },
+    {
+      "epoch": 5.602605863192182,
+      "grad_norm": 1.0329105854034424,
+      "learning_rate": 0.0002,
+      "loss": 0.992,
+      "step": 7740
+    },
+    {
+      "epoch": 5.609844372059356,
+      "grad_norm": 1.479629635810852,
+      "learning_rate": 0.0002,
+      "loss": 0.9493,
+      "step": 7750
+    },
+    {
+      "epoch": 5.617082880926529,
+      "grad_norm": 1.9232319593429565,
+      "learning_rate": 0.0002,
+      "loss": 1.0727,
+      "step": 7760
+    },
+    {
+      "epoch": 5.624321389793702,
+      "grad_norm": 1.0055509805679321,
+      "learning_rate": 0.0002,
+      "loss": 1.0741,
+      "step": 7770
+    },
+    {
+      "epoch": 5.631559898660876,
+      "grad_norm": 1.0037437677383423,
+      "learning_rate": 0.0002,
+      "loss": 1.0731,
+      "step": 7780
+    },
+    {
+      "epoch": 5.638798407528049,
+      "grad_norm": 1.4245030879974365,
+      "learning_rate": 0.0002,
+      "loss": 1.0913,
+      "step": 7790
+    },
+    {
+      "epoch": 5.646036916395222,
+      "grad_norm": 1.080687403678894,
+      "learning_rate": 0.0002,
+      "loss": 0.9711,
+      "step": 7800
+    },
+    {
+      "epoch": 5.653275425262396,
+      "grad_norm": 1.354953408241272,
+      "learning_rate": 0.0002,
+      "loss": 1.0276,
+      "step": 7810
+    },
+    {
+      "epoch": 5.660513934129569,
+      "grad_norm": 0.8966761231422424,
+      "learning_rate": 0.0002,
+      "loss": 1.0534,
+      "step": 7820
+    },
+    {
+      "epoch": 5.6677524429967425,
+      "grad_norm": 1.0675480365753174,
+      "learning_rate": 0.0002,
+      "loss": 1.0662,
+      "step": 7830
+    },
+    {
+      "epoch": 5.674990951863916,
+      "grad_norm": 1.2104216814041138,
+      "learning_rate": 0.0002,
+      "loss": 1.1077,
+      "step": 7840
+    },
+    {
+      "epoch": 5.682229460731089,
+      "grad_norm": 1.105790376663208,
+      "learning_rate": 0.0002,
+      "loss": 0.9627,
+      "step": 7850
+    },
+    {
+      "epoch": 5.689467969598263,
+      "grad_norm": 1.0915391445159912,
+      "learning_rate": 0.0002,
+      "loss": 1.0483,
+      "step": 7860
+    },
+    {
+      "epoch": 5.696706478465436,
+      "grad_norm": 0.8957812786102295,
+      "learning_rate": 0.0002,
+      "loss": 1.0291,
+      "step": 7870
+    },
+    {
+      "epoch": 5.703944987332609,
+      "grad_norm": 1.9189311265945435,
+      "learning_rate": 0.0002,
+      "loss": 0.9785,
+      "step": 7880
+    },
+    {
+      "epoch": 5.711183496199783,
+      "grad_norm": 1.0867321491241455,
+      "learning_rate": 0.0002,
+      "loss": 1.0076,
+      "step": 7890
+    },
+    {
+      "epoch": 5.718422005066956,
+      "grad_norm": 1.0233147144317627,
+      "learning_rate": 0.0002,
+      "loss": 1.0236,
+      "step": 7900
+    },
+    {
+      "epoch": 5.7256605139341294,
+      "grad_norm": 1.16460382938385,
+      "learning_rate": 0.0002,
+      "loss": 0.9872,
+      "step": 7910
+    },
+    {
+      "epoch": 5.732899022801303,
+      "grad_norm": 1.1098358631134033,
+      "learning_rate": 0.0002,
+      "loss": 1.0762,
+      "step": 7920
+    },
+    {
+      "epoch": 5.740137531668476,
+      "grad_norm": 0.8555701375007629,
+      "learning_rate": 0.0002,
+      "loss": 0.9937,
+      "step": 7930
+    },
+    {
+      "epoch": 5.7473760405356495,
+      "grad_norm": 0.9885705709457397,
+      "learning_rate": 0.0002,
+      "loss": 1.0081,
+      "step": 7940
+    },
+    {
+      "epoch": 5.754614549402823,
+      "grad_norm": 0.9184203147888184,
+      "learning_rate": 0.0002,
+      "loss": 0.9909,
+      "step": 7950
+    },
+    {
+      "epoch": 5.761853058269996,
+      "grad_norm": 0.9653698205947876,
+      "learning_rate": 0.0002,
+      "loss": 1.0767,
+      "step": 7960
+    },
+    {
+      "epoch": 5.76909156713717,
+      "grad_norm": 1.0014251470565796,
+      "learning_rate": 0.0002,
+      "loss": 0.9317,
+      "step": 7970
+    },
+    {
+      "epoch": 5.776330076004343,
+      "grad_norm": 1.004701018333435,
+      "learning_rate": 0.0002,
+      "loss": 1.0271,
+      "step": 7980
+    },
+    {
+      "epoch": 5.783568584871516,
+      "grad_norm": 0.950577974319458,
+      "learning_rate": 0.0002,
+      "loss": 1.0397,
+      "step": 7990
+    },
+    {
+      "epoch": 5.79080709373869,
+      "grad_norm": 1.2986834049224854,
+      "learning_rate": 0.0002,
+      "loss": 0.9725,
+      "step": 8000
+    },
+    {
+      "epoch": 5.798045602605863,
+      "grad_norm": 1.3353424072265625,
+      "learning_rate": 0.0002,
+      "loss": 1.039,
+      "step": 8010
+    },
+    {
+      "epoch": 5.8052841114730365,
+      "grad_norm": 0.7650562524795532,
+      "learning_rate": 0.0002,
+      "loss": 1.0626,
+      "step": 8020
+    },
+    {
+      "epoch": 5.81252262034021,
+      "grad_norm": 1.0156235694885254,
+      "learning_rate": 0.0002,
+      "loss": 1.0802,
+      "step": 8030
+    },
+    {
+      "epoch": 5.819761129207383,
+      "grad_norm": 1.3092900514602661,
+      "learning_rate": 0.0002,
+      "loss": 1.0185,
+      "step": 8040
+    },
+    {
+      "epoch": 5.826999638074557,
+      "grad_norm": 1.184428095817566,
+      "learning_rate": 0.0002,
+      "loss": 0.9905,
+      "step": 8050
+    },
+    {
+      "epoch": 5.83423814694173,
+      "grad_norm": 0.979401707649231,
+      "learning_rate": 0.0002,
+      "loss": 1.0548,
+      "step": 8060
+    },
+    {
+      "epoch": 5.841476655808903,
+      "grad_norm": 1.3557400703430176,
+      "learning_rate": 0.0002,
+      "loss": 0.9721,
+      "step": 8070
+    },
+    {
+      "epoch": 5.848715164676077,
+      "grad_norm": 0.8429333567619324,
+      "learning_rate": 0.0002,
+      "loss": 1.0235,
+      "step": 8080
+    },
+    {
+      "epoch": 5.85595367354325,
+      "grad_norm": 1.3167692422866821,
+      "learning_rate": 0.0002,
+      "loss": 0.952,
+      "step": 8090
+    },
+    {
+      "epoch": 5.863192182410423,
+      "grad_norm": 0.9750998020172119,
+      "learning_rate": 0.0002,
+      "loss": 0.9609,
+      "step": 8100
+    },
+    {
+      "epoch": 5.870430691277597,
+      "grad_norm": 1.1869813203811646,
+      "learning_rate": 0.0002,
+      "loss": 1.0789,
+      "step": 8110
+    },
+    {
+      "epoch": 5.87766920014477,
+      "grad_norm": 1.508615255355835,
+      "learning_rate": 0.0002,
+      "loss": 1.0331,
+      "step": 8120
+    },
+    {
+      "epoch": 5.8849077090119435,
+      "grad_norm": 0.9439908266067505,
+      "learning_rate": 0.0002,
+      "loss": 1.0171,
+      "step": 8130
+    },
+    {
+      "epoch": 5.892146217879117,
+      "grad_norm": 0.910508930683136,
+      "learning_rate": 0.0002,
+      "loss": 0.9682,
+      "step": 8140
+    },
+    {
+      "epoch": 5.89938472674629,
+      "grad_norm": 1.111501932144165,
+      "learning_rate": 0.0002,
+      "loss": 1.0032,
+      "step": 8150
+    },
+    {
+      "epoch": 5.906623235613464,
+      "grad_norm": 0.726554274559021,
+      "learning_rate": 0.0002,
+      "loss": 1.0266,
+      "step": 8160
+    },
+    {
+      "epoch": 5.913861744480637,
+      "grad_norm": 1.1084556579589844,
+      "learning_rate": 0.0002,
+      "loss": 1.0681,
+      "step": 8170
+    },
+    {
+      "epoch": 5.92110025334781,
+      "grad_norm": 0.9695167541503906,
+      "learning_rate": 0.0002,
+      "loss": 0.969,
+      "step": 8180
+    },
+    {
+      "epoch": 5.928338762214984,
+      "grad_norm": 1.1169592142105103,
+      "learning_rate": 0.0002,
+      "loss": 0.9858,
+      "step": 8190
+    },
+    {
+      "epoch": 5.935577271082157,
+      "grad_norm": 1.5116780996322632,
+      "learning_rate": 0.0002,
+      "loss": 1.0924,
+      "step": 8200
+    },
+    {
+      "epoch": 5.94281577994933,
+      "grad_norm": 1.0073388814926147,
+      "learning_rate": 0.0002,
+      "loss": 0.878,
+      "step": 8210
+    },
+    {
+      "epoch": 5.950054288816504,
+      "grad_norm": 0.9323263168334961,
+      "learning_rate": 0.0002,
+      "loss": 1.0462,
+      "step": 8220
+    },
+    {
+      "epoch": 5.957292797683677,
+      "grad_norm": 0.9422887563705444,
+      "learning_rate": 0.0002,
+      "loss": 1.0291,
+      "step": 8230
+    },
+    {
+      "epoch": 5.9645313065508505,
+      "grad_norm": 0.9691047668457031,
+      "learning_rate": 0.0002,
+      "loss": 0.953,
+      "step": 8240
+    },
+    {
+      "epoch": 5.971769815418024,
+      "grad_norm": 0.9650622606277466,
+      "learning_rate": 0.0002,
+      "loss": 0.9842,
+      "step": 8250
+    },
+    {
+      "epoch": 5.979008324285197,
+      "grad_norm": 1.077958345413208,
+      "learning_rate": 0.0002,
+      "loss": 0.907,
+      "step": 8260
+    },
+    {
+      "epoch": 5.986246833152371,
+      "grad_norm": 0.8946306109428406,
+      "learning_rate": 0.0002,
+      "loss": 0.9162,
+      "step": 8270
+    },
+    {
+      "epoch": 5.993485342019544,
+      "grad_norm": 1.34098219871521,
+      "learning_rate": 0.0002,
+      "loss": 1.0439,
+      "step": 8280
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 1.4714229106903076,
+      "eval_runtime": 26.301,
+      "eval_samples_per_second": 16.577,
+      "eval_steps_per_second": 2.091,
+      "step": 8289
+    },
+    {
+      "epoch": 6.000723850886717,
+      "grad_norm": 0.9737564325332642,
+      "learning_rate": 0.0002,
+      "loss": 1.1403,
+      "step": 8290
+    },
+    {
+      "epoch": 6.007962359753891,
+      "grad_norm": 1.2205945253372192,
+      "learning_rate": 0.0002,
+      "loss": 0.8875,
+      "step": 8300
+    },
+    {
+      "epoch": 6.015200868621064,
+      "grad_norm": 1.3529434204101562,
+      "learning_rate": 0.0002,
+      "loss": 0.8623,
+      "step": 8310
+    },
+    {
+      "epoch": 6.022439377488237,
+      "grad_norm": 1.2300174236297607,
+      "learning_rate": 0.0002,
+      "loss": 0.9427,
+      "step": 8320
+    },
+    {
+      "epoch": 6.029677886355411,
+      "grad_norm": 0.9248194098472595,
+      "learning_rate": 0.0002,
+      "loss": 0.9322,
+      "step": 8330
+    },
+    {
+      "epoch": 6.036916395222584,
+      "grad_norm": 1.1140035390853882,
+      "learning_rate": 0.0002,
+      "loss": 0.9302,
+      "step": 8340
+    },
+    {
+      "epoch": 6.0441549040897575,
+      "grad_norm": 1.2097352743148804,
+      "learning_rate": 0.0002,
+      "loss": 0.8255,
+      "step": 8350
+    },
+    {
+      "epoch": 6.051393412956931,
+      "grad_norm": 0.9472483396530151,
+      "learning_rate": 0.0002,
+      "loss": 0.8792,
+      "step": 8360
+    },
+    {
+      "epoch": 6.058631921824104,
+      "grad_norm": 1.0195368528366089,
+      "learning_rate": 0.0002,
+      "loss": 0.8865,
+      "step": 8370
+    },
+    {
+      "epoch": 6.065870430691278,
+      "grad_norm": 1.182735562324524,
+      "learning_rate": 0.0002,
+      "loss": 0.8858,
+      "step": 8380
+    },
+    {
+      "epoch": 6.073108939558451,
+      "grad_norm": 1.1042858362197876,
+      "learning_rate": 0.0002,
+      "loss": 0.9455,
+      "step": 8390
+    },
+    {
+      "epoch": 6.080347448425624,
+      "grad_norm": 0.8606401085853577,
+      "learning_rate": 0.0002,
+      "loss": 0.9723,
+      "step": 8400
+    },
+    {
+      "epoch": 6.087585957292798,
+      "grad_norm": 1.1015676259994507,
+      "learning_rate": 0.0002,
+      "loss": 0.8436,
+      "step": 8410
+    },
+    {
+      "epoch": 6.094824466159971,
+      "grad_norm": 1.690224289894104,
+      "learning_rate": 0.0002,
+      "loss": 0.8845,
+      "step": 8420
+    },
+    {
+      "epoch": 6.1020629750271445,
+      "grad_norm": 1.1928749084472656,
+      "learning_rate": 0.0002,
+      "loss": 0.8484,
+      "step": 8430
+    },
+    {
+      "epoch": 6.109301483894318,
+      "grad_norm": 1.0816864967346191,
+      "learning_rate": 0.0002,
+      "loss": 0.9546,
+      "step": 8440
+    },
+    {
+      "epoch": 6.116539992761491,
+      "grad_norm": 1.1638226509094238,
+      "learning_rate": 0.0002,
+      "loss": 0.8286,
+      "step": 8450
+    },
+    {
+      "epoch": 6.1237785016286646,
+      "grad_norm": 1.3782968521118164,
+      "learning_rate": 0.0002,
+      "loss": 0.8749,
+      "step": 8460
+    },
+    {
+      "epoch": 6.131017010495838,
+      "grad_norm": 1.2030094861984253,
+      "learning_rate": 0.0002,
+      "loss": 0.7956,
+      "step": 8470
+    },
+    {
+      "epoch": 6.138255519363011,
+      "grad_norm": 1.3227659463882446,
+      "learning_rate": 0.0002,
+      "loss": 0.8393,
+      "step": 8480
+    },
+    {
+      "epoch": 6.145494028230185,
+      "grad_norm": 1.104384422302246,
+      "learning_rate": 0.0002,
+      "loss": 0.9175,
+      "step": 8490
+    },
+    {
+      "epoch": 6.152732537097358,
+      "grad_norm": 1.518805980682373,
+      "learning_rate": 0.0002,
+      "loss": 0.861,
+      "step": 8500
+    },
+    {
+      "epoch": 6.159971045964531,
+      "grad_norm": 1.2029093503952026,
+      "learning_rate": 0.0002,
+      "loss": 0.9169,
+      "step": 8510
+    },
+    {
+      "epoch": 6.167209554831705,
+      "grad_norm": 1.2991217374801636,
+      "learning_rate": 0.0002,
+      "loss": 0.8701,
+      "step": 8520
+    },
+    {
+      "epoch": 6.174448063698878,
+      "grad_norm": 1.7002956867218018,
+      "learning_rate": 0.0002,
+      "loss": 0.9748,
+      "step": 8530
+    },
+    {
+      "epoch": 6.1816865725660515,
+      "grad_norm": 1.6653581857681274,
+      "learning_rate": 0.0002,
+      "loss": 0.8881,
+      "step": 8540
+    },
+    {
+      "epoch": 6.188925081433225,
+      "grad_norm": 1.0493303537368774,
+      "learning_rate": 0.0002,
+      "loss": 0.817,
+      "step": 8550
+    },
+    {
+      "epoch": 6.196163590300398,
+      "grad_norm": 1.539345622062683,
+      "learning_rate": 0.0002,
+      "loss": 0.8726,
+      "step": 8560
+    },
+    {
+      "epoch": 6.203402099167572,
+      "grad_norm": 1.2757070064544678,
+      "learning_rate": 0.0002,
+      "loss": 0.9452,
+      "step": 8570
+    },
+    {
+      "epoch": 6.210640608034745,
+      "grad_norm": 1.2416890859603882,
+      "learning_rate": 0.0002,
+      "loss": 0.8773,
+      "step": 8580
+    },
+    {
+      "epoch": 6.217879116901918,
+      "grad_norm": 1.617621898651123,
+      "learning_rate": 0.0002,
+      "loss": 0.815,
+      "step": 8590
+    },
+    {
+      "epoch": 6.225117625769092,
+      "grad_norm": 1.058962106704712,
+      "learning_rate": 0.0002,
+      "loss": 0.9137,
+      "step": 8600
+    },
+    {
+      "epoch": 6.232356134636265,
+      "grad_norm": 1.1489088535308838,
+      "learning_rate": 0.0002,
+      "loss": 0.8164,
+      "step": 8610
+    },
+    {
+      "epoch": 6.239594643503438,
+      "grad_norm": 0.9391577243804932,
+      "learning_rate": 0.0002,
+      "loss": 0.9476,
+      "step": 8620
+    },
+    {
+      "epoch": 6.246833152370612,
+      "grad_norm": 1.363706111907959,
+      "learning_rate": 0.0002,
+      "loss": 0.932,
+      "step": 8630
+    },
+    {
+      "epoch": 6.254071661237785,
+      "grad_norm": 0.779502809047699,
+      "learning_rate": 0.0002,
+      "loss": 0.8917,
+      "step": 8640
+    },
+    {
+      "epoch": 6.2613101701049585,
+      "grad_norm": 2.000821590423584,
+      "learning_rate": 0.0002,
+      "loss": 0.9196,
+      "step": 8650
+    },
+    {
+      "epoch": 6.268548678972132,
+      "grad_norm": 1.1521023511886597,
+      "learning_rate": 0.0002,
+      "loss": 0.9794,
+      "step": 8660
+    },
+    {
+      "epoch": 6.275787187839305,
+      "grad_norm": 1.3734570741653442,
+      "learning_rate": 0.0002,
+      "loss": 0.9147,
+      "step": 8670
+    },
+    {
+      "epoch": 6.283025696706479,
+      "grad_norm": 0.9550670385360718,
+      "learning_rate": 0.0002,
+      "loss": 0.795,
+      "step": 8680
+    },
+    {
+      "epoch": 6.290264205573652,
+      "grad_norm": 0.8937032222747803,
+      "learning_rate": 0.0002,
+      "loss": 0.9049,
+      "step": 8690
+    },
+    {
+      "epoch": 6.297502714440825,
+      "grad_norm": 1.3352779150009155,
+      "learning_rate": 0.0002,
+      "loss": 0.8526,
+      "step": 8700
+    },
+    {
+      "epoch": 6.304741223307999,
+      "grad_norm": 1.3057222366333008,
+      "learning_rate": 0.0002,
+      "loss": 0.8572,
+      "step": 8710
+    },
+    {
+      "epoch": 6.311979732175172,
+      "grad_norm": 0.9078314304351807,
+      "learning_rate": 0.0002,
+      "loss": 0.8825,
+      "step": 8720
+    },
+    {
+      "epoch": 6.319218241042345,
+      "grad_norm": 1.6663457155227661,
+      "learning_rate": 0.0002,
+      "loss": 0.8666,
+      "step": 8730
+    },
+    {
+      "epoch": 6.326456749909519,
+      "grad_norm": 1.2043739557266235,
+      "learning_rate": 0.0002,
+      "loss": 0.927,
+      "step": 8740
+    },
+    {
+      "epoch": 6.333695258776692,
+      "grad_norm": 0.9165967702865601,
+      "learning_rate": 0.0002,
+      "loss": 0.8014,
+      "step": 8750
+    },
+    {
+      "epoch": 6.3409337676438655,
+      "grad_norm": 1.016452670097351,
+      "learning_rate": 0.0002,
+      "loss": 0.9761,
+      "step": 8760
+    },
+    {
+      "epoch": 6.348172276511039,
+      "grad_norm": 1.2209261655807495,
+      "learning_rate": 0.0002,
+      "loss": 1.022,
+      "step": 8770
+    },
+    {
+      "epoch": 6.355410785378212,
+      "grad_norm": 1.3380663394927979,
+      "learning_rate": 0.0002,
+      "loss": 0.8012,
+      "step": 8780
+    },
+    {
+      "epoch": 6.362649294245386,
+      "grad_norm": 2.3311562538146973,
+      "learning_rate": 0.0002,
+      "loss": 0.9553,
+      "step": 8790
+    },
+    {
+      "epoch": 6.369887803112559,
+      "grad_norm": 1.0330604314804077,
+      "learning_rate": 0.0002,
+      "loss": 0.8676,
+      "step": 8800
+    },
+    {
+      "epoch": 6.377126311979732,
+      "grad_norm": 0.9655511975288391,
+      "learning_rate": 0.0002,
+      "loss": 0.98,
+      "step": 8810
+    },
+    {
+      "epoch": 6.384364820846906,
+      "grad_norm": 1.1065765619277954,
+      "learning_rate": 0.0002,
+      "loss": 1.0324,
+      "step": 8820
+    },
+    {
+      "epoch": 6.391603329714079,
+      "grad_norm": 1.2631285190582275,
+      "learning_rate": 0.0002,
+      "loss": 1.0078,
+      "step": 8830
+    },
+    {
+      "epoch": 6.398841838581252,
+      "grad_norm": 0.92459636926651,
+      "learning_rate": 0.0002,
+      "loss": 0.8989,
+      "step": 8840
+    },
+    {
+      "epoch": 6.406080347448426,
+      "grad_norm": 0.9982633590698242,
+      "learning_rate": 0.0002,
+      "loss": 0.8536,
+      "step": 8850
+    },
+    {
+      "epoch": 6.413318856315599,
+      "grad_norm": 1.0746768712997437,
+      "learning_rate": 0.0002,
+      "loss": 0.8949,
+      "step": 8860
+    },
+    {
+      "epoch": 6.4205573651827725,
+      "grad_norm": 1.3024073839187622,
+      "learning_rate": 0.0002,
+      "loss": 0.8547,
+      "step": 8870
+    },
+    {
+      "epoch": 6.427795874049946,
+      "grad_norm": 1.2764527797698975,
+      "learning_rate": 0.0002,
+      "loss": 0.9618,
+      "step": 8880
+    },
+    {
+      "epoch": 6.435034382917119,
+      "grad_norm": 0.8318809270858765,
+      "learning_rate": 0.0002,
+      "loss": 0.8905,
+      "step": 8890
+    },
+    {
+      "epoch": 6.442272891784293,
+      "grad_norm": 1.7350783348083496,
+      "learning_rate": 0.0002,
+      "loss": 0.917,
+      "step": 8900
+    },
+    {
+      "epoch": 6.449511400651466,
+      "grad_norm": 1.3430488109588623,
+      "learning_rate": 0.0002,
+      "loss": 1.0229,
+      "step": 8910
+    },
+    {
+      "epoch": 6.456749909518639,
+      "grad_norm": 1.5907495021820068,
+      "learning_rate": 0.0002,
+      "loss": 0.9678,
+      "step": 8920
+    },
+    {
+      "epoch": 6.463988418385813,
+      "grad_norm": 1.8579202890396118,
+      "learning_rate": 0.0002,
+      "loss": 0.9639,
+      "step": 8930
+    },
+    {
+      "epoch": 6.471226927252986,
+      "grad_norm": 1.2233413457870483,
+      "learning_rate": 0.0002,
+      "loss": 0.9302,
+      "step": 8940
+    },
+    {
+      "epoch": 6.4784654361201595,
+      "grad_norm": 1.009103775024414,
+      "learning_rate": 0.0002,
+      "loss": 0.9169,
+      "step": 8950
+    },
+    {
+      "epoch": 6.485703944987333,
+      "grad_norm": 1.1265181303024292,
+      "learning_rate": 0.0002,
+      "loss": 0.8969,
+      "step": 8960
+    },
+    {
+      "epoch": 6.492942453854506,
+      "grad_norm": 1.1733338832855225,
+      "learning_rate": 0.0002,
+      "loss": 0.8374,
+      "step": 8970
+    },
+    {
+      "epoch": 6.50018096272168,
+      "grad_norm": 1.0444518327713013,
+      "learning_rate": 0.0002,
+      "loss": 0.8764,
+      "step": 8980
+    },
+    {
+      "epoch": 6.507419471588853,
+      "grad_norm": 1.2296479940414429,
+      "learning_rate": 0.0002,
+      "loss": 0.9582,
+      "step": 8990
+    },
+    {
+      "epoch": 6.514657980456026,
+      "grad_norm": 1.370417833328247,
+      "learning_rate": 0.0002,
+      "loss": 0.8557,
+      "step": 9000
+    },
+    {
+      "epoch": 6.5218964893232,
+      "grad_norm": 1.4787620306015015,
+      "learning_rate": 0.0002,
+      "loss": 0.9787,
+      "step": 9010
+    },
+    {
+      "epoch": 6.529134998190373,
+      "grad_norm": 0.8550514578819275,
+      "learning_rate": 0.0002,
+      "loss": 0.967,
+      "step": 9020
+    },
+    {
+      "epoch": 6.536373507057546,
+      "grad_norm": 1.2327991724014282,
+      "learning_rate": 0.0002,
+      "loss": 0.9755,
+      "step": 9030
+    },
+    {
+      "epoch": 6.54361201592472,
+      "grad_norm": 1.0915621519088745,
+      "learning_rate": 0.0002,
+      "loss": 0.9248,
+      "step": 9040
+    },
+    {
+      "epoch": 6.550850524791893,
+      "grad_norm": 1.7243309020996094,
+      "learning_rate": 0.0002,
+      "loss": 1.0024,
+      "step": 9050
+    },
+    {
+      "epoch": 6.5580890336590665,
+      "grad_norm": 0.954359769821167,
+      "learning_rate": 0.0002,
+      "loss": 1.0123,
+      "step": 9060
+    },
+    {
+      "epoch": 6.56532754252624,
+      "grad_norm": 1.066051959991455,
+      "learning_rate": 0.0002,
+      "loss": 0.8261,
+      "step": 9070
+    },
+    {
+      "epoch": 6.572566051393413,
+      "grad_norm": 1.200271487236023,
+      "learning_rate": 0.0002,
+      "loss": 0.944,
+      "step": 9080
+    },
+    {
+      "epoch": 6.579804560260587,
+      "grad_norm": 1.4331457614898682,
+      "learning_rate": 0.0002,
+      "loss": 0.9788,
+      "step": 9090
+    },
+    {
+      "epoch": 6.58704306912776,
+      "grad_norm": 1.0892444849014282,
+      "learning_rate": 0.0002,
+      "loss": 1.0216,
+      "step": 9100
+    },
+    {
+      "epoch": 6.594281577994933,
+      "grad_norm": 1.849726915359497,
+      "learning_rate": 0.0002,
+      "loss": 0.8557,
+      "step": 9110
+    },
+    {
+      "epoch": 6.601520086862107,
+      "grad_norm": 1.1228708028793335,
+      "learning_rate": 0.0002,
+      "loss": 0.9495,
+      "step": 9120
+    },
+    {
+      "epoch": 6.60875859572928,
+      "grad_norm": 1.0928595066070557,
+      "learning_rate": 0.0002,
+      "loss": 1.0169,
+      "step": 9130
+    },
+    {
+      "epoch": 6.615997104596453,
+      "grad_norm": 1.2138155698776245,
+      "learning_rate": 0.0002,
+      "loss": 0.9342,
+      "step": 9140
+    },
+    {
+      "epoch": 6.623235613463627,
+      "grad_norm": 1.5155235528945923,
+      "learning_rate": 0.0002,
+      "loss": 0.8715,
+      "step": 9150
+    },
+    {
+      "epoch": 6.6304741223308,
+      "grad_norm": 1.3194212913513184,
+      "learning_rate": 0.0002,
+      "loss": 0.9806,
+      "step": 9160
+    },
+    {
+      "epoch": 6.6377126311979735,
+      "grad_norm": 1.045623779296875,
+      "learning_rate": 0.0002,
+      "loss": 0.8958,
+      "step": 9170
+    },
+    {
+      "epoch": 6.644951140065147,
+      "grad_norm": 0.9647570252418518,
+      "learning_rate": 0.0002,
+      "loss": 0.8698,
+      "step": 9180
+    },
+    {
+      "epoch": 6.65218964893232,
+      "grad_norm": 1.0818220376968384,
+      "learning_rate": 0.0002,
+      "loss": 0.8829,
+      "step": 9190
+    },
+    {
+      "epoch": 6.659428157799494,
+      "grad_norm": 1.2792822122573853,
+      "learning_rate": 0.0002,
+      "loss": 0.9745,
+      "step": 9200
+    },
+    {
+      "epoch": 6.666666666666667,
+      "grad_norm": 1.2764191627502441,
+      "learning_rate": 0.0002,
+      "loss": 0.8854,
+      "step": 9210
+    },
+    {
+      "epoch": 6.67390517553384,
+      "grad_norm": 1.0552066564559937,
+      "learning_rate": 0.0002,
+      "loss": 0.9709,
+      "step": 9220
+    },
+    {
+      "epoch": 6.681143684401014,
+      "grad_norm": 1.082476019859314,
+      "learning_rate": 0.0002,
+      "loss": 0.8855,
+      "step": 9230
+    },
+    {
+      "epoch": 6.688382193268187,
+      "grad_norm": 1.3313323259353638,
+      "learning_rate": 0.0002,
+      "loss": 0.9779,
+      "step": 9240
+    },
+    {
+      "epoch": 6.69562070213536,
+      "grad_norm": 1.130048394203186,
+      "learning_rate": 0.0002,
+      "loss": 1.005,
+      "step": 9250
+    },
+    {
+      "epoch": 6.702859211002534,
+      "grad_norm": 1.1997296810150146,
+      "learning_rate": 0.0002,
+      "loss": 0.9969,
+      "step": 9260
+    },
+    {
+      "epoch": 6.710097719869707,
+      "grad_norm": 1.0591834783554077,
+      "learning_rate": 0.0002,
+      "loss": 0.8691,
+      "step": 9270
+    },
+    {
+      "epoch": 6.7173362287368805,
+      "grad_norm": 1.2722901105880737,
+      "learning_rate": 0.0002,
+      "loss": 0.9603,
+      "step": 9280
+    },
+    {
+      "epoch": 6.724574737604054,
+      "grad_norm": 1.1150950193405151,
+      "learning_rate": 0.0002,
+      "loss": 0.9227,
+      "step": 9290
+    },
+    {
+      "epoch": 6.731813246471227,
+      "grad_norm": 1.1575992107391357,
+      "learning_rate": 0.0002,
+      "loss": 0.95,
+      "step": 9300
+    },
+    {
+      "epoch": 6.739051755338401,
+      "grad_norm": 0.9371691346168518,
+      "learning_rate": 0.0002,
+      "loss": 0.9822,
+      "step": 9310
+    },
+    {
+      "epoch": 6.746290264205574,
+      "grad_norm": 1.4924226999282837,
+      "learning_rate": 0.0002,
+      "loss": 0.9773,
+      "step": 9320
+    },
+    {
+      "epoch": 6.753528773072747,
+      "grad_norm": 1.1524218320846558,
+      "learning_rate": 0.0002,
+      "loss": 0.969,
+      "step": 9330
+    },
+    {
+      "epoch": 6.760767281939921,
+      "grad_norm": 0.9500471949577332,
+      "learning_rate": 0.0002,
+      "loss": 0.9271,
+      "step": 9340
+    },
+    {
+      "epoch": 6.768005790807094,
+      "grad_norm": 1.2062290906906128,
+      "learning_rate": 0.0002,
+      "loss": 0.9029,
+      "step": 9350
+    },
+    {
+      "epoch": 6.7752442996742674,
+      "grad_norm": 1.212631106376648,
+      "learning_rate": 0.0002,
+      "loss": 0.9121,
+      "step": 9360
+    },
+    {
+      "epoch": 6.782482808541441,
+      "grad_norm": 1.9135472774505615,
+      "learning_rate": 0.0002,
+      "loss": 0.8486,
+      "step": 9370
+    },
+    {
+      "epoch": 6.789721317408614,
+      "grad_norm": 0.9682775139808655,
+      "learning_rate": 0.0002,
+      "loss": 0.9332,
+      "step": 9380
+    },
+    {
+      "epoch": 6.7969598262757875,
+      "grad_norm": 1.1405237913131714,
+      "learning_rate": 0.0002,
+      "loss": 0.8548,
+      "step": 9390
+    },
+    {
+      "epoch": 6.804198335142961,
+      "grad_norm": 1.6855751276016235,
+      "learning_rate": 0.0002,
+      "loss": 0.8922,
+      "step": 9400
+    },
+    {
+      "epoch": 6.811436844010134,
+      "grad_norm": 1.6590169668197632,
+      "learning_rate": 0.0002,
+      "loss": 0.9417,
+      "step": 9410
+    },
+    {
+      "epoch": 6.818675352877308,
+      "grad_norm": 1.8795170783996582,
+      "learning_rate": 0.0002,
+      "loss": 0.868,
+      "step": 9420
+    },
+    {
+      "epoch": 6.825913861744481,
+      "grad_norm": 1.1087183952331543,
+      "learning_rate": 0.0002,
+      "loss": 0.9142,
+      "step": 9430
+    },
+    {
+      "epoch": 6.833152370611654,
+      "grad_norm": 1.4178446531295776,
+      "learning_rate": 0.0002,
+      "loss": 1.1427,
+      "step": 9440
+    },
+    {
+      "epoch": 6.840390879478828,
+      "grad_norm": 1.0792350769042969,
+      "learning_rate": 0.0002,
+      "loss": 0.8325,
+      "step": 9450
+    },
+    {
+      "epoch": 6.847629388346001,
+      "grad_norm": 1.2159196138381958,
+      "learning_rate": 0.0002,
+      "loss": 1.0078,
+      "step": 9460
+    },
+    {
+      "epoch": 6.8548678972131745,
+      "grad_norm": 0.9998821020126343,
+      "learning_rate": 0.0002,
+      "loss": 0.9536,
+      "step": 9470
+    },
+    {
+      "epoch": 6.862106406080348,
+      "grad_norm": 0.7940687537193298,
+      "learning_rate": 0.0002,
+      "loss": 0.9277,
+      "step": 9480
+    },
+    {
+      "epoch": 6.869344914947521,
+      "grad_norm": 0.9572826027870178,
+      "learning_rate": 0.0002,
+      "loss": 0.8612,
+      "step": 9490
+    },
+    {
+      "epoch": 6.876583423814694,
+      "grad_norm": 1.1086537837982178,
+      "learning_rate": 0.0002,
+      "loss": 0.9611,
+      "step": 9500
+    },
+    {
+      "epoch": 6.883821932681867,
+      "grad_norm": 1.1934887170791626,
+      "learning_rate": 0.0002,
+      "loss": 0.9276,
+      "step": 9510
+    },
+    {
+      "epoch": 6.89106044154904,
+      "grad_norm": 1.207324504852295,
+      "learning_rate": 0.0002,
+      "loss": 0.8416,
+      "step": 9520
+    },
+    {
+      "epoch": 6.898298950416214,
+      "grad_norm": 1.1303677558898926,
+      "learning_rate": 0.0002,
+      "loss": 0.9378,
+      "step": 9530
+    },
+    {
+      "epoch": 6.905537459283387,
+      "grad_norm": 1.4958926439285278,
+      "learning_rate": 0.0002,
+      "loss": 0.9599,
+      "step": 9540
+    },
+    {
+      "epoch": 6.9127759681505605,
+      "grad_norm": 1.2141553163528442,
+      "learning_rate": 0.0002,
+      "loss": 0.9365,
+      "step": 9550
+    },
+    {
+      "epoch": 6.920014477017734,
+      "grad_norm": 1.6544346809387207,
+      "learning_rate": 0.0002,
+      "loss": 1.0291,
+      "step": 9560
+    },
+    {
+      "epoch": 6.927252985884907,
+      "grad_norm": 1.0540320873260498,
+      "learning_rate": 0.0002,
+      "loss": 0.8439,
+      "step": 9570
+    },
+    {
+      "epoch": 6.934491494752081,
+      "grad_norm": 1.3095581531524658,
+      "learning_rate": 0.0002,
+      "loss": 0.9831,
+      "step": 9580
+    },
+    {
+      "epoch": 6.941730003619254,
+      "grad_norm": 1.4509341716766357,
+      "learning_rate": 0.0002,
+      "loss": 0.8694,
+      "step": 9590
+    },
+    {
+      "epoch": 6.948968512486427,
+      "grad_norm": 1.1091740131378174,
+      "learning_rate": 0.0002,
+      "loss": 0.983,
+      "step": 9600
+    },
+    {
+      "epoch": 6.956207021353601,
+      "grad_norm": 1.102929949760437,
+      "learning_rate": 0.0002,
+      "loss": 0.9126,
+      "step": 9610
+    },
+    {
+      "epoch": 6.963445530220774,
+      "grad_norm": 1.1377743482589722,
+      "learning_rate": 0.0002,
+      "loss": 0.9622,
+      "step": 9620
+    },
+    {
+      "epoch": 6.970684039087947,
+      "grad_norm": 1.2070361375808716,
+      "learning_rate": 0.0002,
+      "loss": 0.9045,
+      "step": 9630
+    },
+    {
+      "epoch": 6.977922547955121,
+      "grad_norm": 1.30153489112854,
+      "learning_rate": 0.0002,
+      "loss": 0.9714,
+      "step": 9640
+    },
+    {
+      "epoch": 6.985161056822294,
+      "grad_norm": 1.4641543626785278,
+      "learning_rate": 0.0002,
+      "loss": 0.9555,
+      "step": 9650
+    },
+    {
+      "epoch": 6.9923995656894675,
+      "grad_norm": 1.0497819185256958,
+      "learning_rate": 0.0002,
+      "loss": 0.9177,
+      "step": 9660
+    },
+    {
+      "epoch": 6.999638074556641,
+      "grad_norm": 1.2500354051589966,
+      "learning_rate": 0.0002,
+      "loss": 0.8369,
+      "step": 9670
+    },
+    {
+      "epoch": 6.999638074556641,
+      "eval_loss": 1.518465518951416,
+      "eval_runtime": 26.4525,
+      "eval_samples_per_second": 16.482,
+      "eval_steps_per_second": 2.079,
+      "step": 9670
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 11048,
+  "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": 1.1819281555862323e+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-0.9-num-8828-sd-1/checkpoint-9670/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-0.9-num-8828-sd-1/checkpoint-9670/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..4d7b7431bbbe8c9bf29b925bca391a558af5ff8c
--- /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-0.9-num-8828-sd-1/checkpoint-9670/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ad613885e4f267fc04125f1a836d42cfa796bbe12e536f9ee60c955de02cdb5a
+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-0.9-num-8828-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-0.9-num-8828-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-0.9-num-8828-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-0.9-num-8828-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-0.9-num-8828-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-0.9-num-8828-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-0.9-num-8828-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-0.9-num-8828-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-0.9-num-8828-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-0.9-num-8828-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-0.9-num-8828-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-0.9-num-8828-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-0.9-num-8828-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-0.9-num-8828-sd-1/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..4d7b7431bbbe8c9bf29b925bca391a558af5ff8c
--- /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-0.9-num-8828-sd-1/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ad613885e4f267fc04125f1a836d42cfa796bbe12e536f9ee60c955de02cdb5a
+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-0.9-num-8828-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-0.9-num-8828-sd-1/training_log.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..9adfb8801dcfa9f33ddcc9367233e4e892daecc9
--- /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-0.9-num-8828-sd-1/training_log.jsonl
@@ -0,0 +1,8 @@
+{"epoch": 0.9996380745566413, "step": 1381, "epoch_duration": 1362.1938304901123, "total_accumulated_duration": 1362.1938304901123, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 3048.73388671875}, "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": 4.7061, "grad_norm": 1.2523442506790161, "learning_rate": 0.0002, "epoch": 0.007238508867173362, "step": 10}, {"loss": 3.3493, "grad_norm": 1.8887330293655396, "learning_rate": 0.0002, "epoch": 0.014477017734346724, "step": 20}, {"loss": 2.7585, "grad_norm": 0.9668035507202148, "learning_rate": 0.0002, "epoch": 0.021715526601520086, "step": 30}, {"loss": 2.3699, "grad_norm": 2.9167306423187256, "learning_rate": 0.0002, "epoch": 0.028954035468693448, "step": 40}, {"loss": 2.2679, "grad_norm": 2.649867296218872, "learning_rate": 0.0002, "epoch": 0.036192544335866814, "step": 50}, {"loss": 2.2202, "grad_norm": 1.5120655298233032, "learning_rate": 0.0002, "epoch": 0.04343105320304017, "step": 60}, {"loss": 2.2026, "grad_norm": 0.7879868149757385, "learning_rate": 0.0002, "epoch": 0.05066956207021354, "step": 70}, {"loss": 1.9447, "grad_norm": 0.7616953253746033, "learning_rate": 0.0002, "epoch": 0.057908070937386896, "step": 80}, {"loss": 2.0112, "grad_norm": 1.8809149265289307, "learning_rate": 0.0002, "epoch": 0.06514657980456026, "step": 90}, {"loss": 1.8337, "grad_norm": 0.9294016361236572, "learning_rate": 0.0002, "epoch": 0.07238508867173363, "step": 100}, {"loss": 1.8419, "grad_norm": 0.7145281434059143, "learning_rate": 0.0002, "epoch": 0.07962359753890698, "step": 110}, {"loss": 2.0036, "grad_norm": 0.7564446330070496, "learning_rate": 0.0002, "epoch": 0.08686210640608034, "step": 120}, {"loss": 1.9306, "grad_norm": 1.1681925058364868, "learning_rate": 0.0002, "epoch": 0.09410061527325371, "step": 130}, {"loss": 1.7875, "grad_norm": 0.6708641648292542, "learning_rate": 0.0002, "epoch": 0.10133912414042708, "step": 140}, {"loss": 1.786, "grad_norm": 0.7625647783279419, "learning_rate": 0.0002, "epoch": 0.10857763300760044, "step": 150}, {"loss": 1.6687, "grad_norm": 0.8463464975357056, "learning_rate": 0.0002, "epoch": 0.11581614187477379, "step": 160}, {"loss": 1.6214, "grad_norm": 0.7502335906028748, "learning_rate": 0.0002, "epoch": 0.12305465074194716, "step": 170}, {"loss": 1.7433, "grad_norm": 0.6929958462715149, "learning_rate": 0.0002, "epoch": 0.13029315960912052, "step": 180}, {"loss": 1.6009, "grad_norm": 0.6798707842826843, "learning_rate": 0.0002, "epoch": 0.1375316684762939, "step": 190}, {"loss": 1.6208, "grad_norm": 0.7566508650779724, "learning_rate": 0.0002, "epoch": 0.14477017734346725, "step": 200}, {"loss": 1.5823, "grad_norm": 0.7196869850158691, "learning_rate": 0.0002, "epoch": 0.15200868621064062, "step": 210}, {"loss": 1.738, "grad_norm": 0.8401045799255371, "learning_rate": 0.0002, "epoch": 0.15924719507781396, "step": 220}, {"loss": 1.7574, "grad_norm": 0.8503773212432861, "learning_rate": 0.0002, "epoch": 0.16648570394498732, "step": 230}, {"loss": 1.7861, "grad_norm": 0.7183733582496643, "learning_rate": 0.0002, "epoch": 0.1737242128121607, "step": 240}, {"loss": 1.6693, "grad_norm": 0.7082605957984924, "learning_rate": 0.0002, "epoch": 0.18096272167933405, "step": 250}, {"loss": 1.619, "grad_norm": 0.9386326670646667, "learning_rate": 0.0002, "epoch": 0.18820123054650742, "step": 260}, {"loss": 1.6511, "grad_norm": 0.7332451939582825, "learning_rate": 0.0002, "epoch": 0.19543973941368079, "step": 270}, {"loss": 1.6353, "grad_norm": 0.7092869877815247, "learning_rate": 0.0002, "epoch": 0.20267824828085415, "step": 280}, {"loss": 1.5996, "grad_norm": 0.7256413698196411, "learning_rate": 0.0002, "epoch": 0.20991675714802752, "step": 290}, {"loss": 1.6754, "grad_norm": 0.6398681402206421, "learning_rate": 0.0002, "epoch": 0.21715526601520088, "step": 300}, {"loss": 1.397, "grad_norm": 0.6273287534713745, "learning_rate": 0.0002, "epoch": 0.22439377488237422, "step": 310}, {"loss": 1.5115, "grad_norm": 0.511648416519165, "learning_rate": 0.0002, "epoch": 0.23163228374954759, "step": 320}, {"loss": 1.5424, "grad_norm": 0.8677352070808411, "learning_rate": 0.0002, "epoch": 0.23887079261672095, "step": 330}, {"loss": 1.6779, "grad_norm": 0.6270743012428284, "learning_rate": 0.0002, "epoch": 0.24610930148389432, "step": 340}, {"loss": 1.626, "grad_norm": 0.7980281114578247, "learning_rate": 0.0002, "epoch": 0.2533478103510677, "step": 350}, {"loss": 1.5238, "grad_norm": 0.632486879825592, "learning_rate": 0.0002, "epoch": 0.26058631921824105, "step": 360}, {"loss": 1.5175, "grad_norm": 0.6527034640312195, "learning_rate": 0.0002, "epoch": 0.2678248280854144, "step": 370}, {"loss": 1.627, "grad_norm": 0.7672118544578552, "learning_rate": 0.0002, "epoch": 0.2750633369525878, "step": 380}, {"loss": 1.5605, "grad_norm": 0.6035117506980896, "learning_rate": 0.0002, "epoch": 0.28230184581976114, "step": 390}, {"loss": 1.4603, "grad_norm": 0.5955103039741516, "learning_rate": 0.0002, "epoch": 0.2895403546869345, "step": 400}, {"loss": 1.558, "grad_norm": 0.6015191674232483, "learning_rate": 0.0002, "epoch": 0.2967788635541079, "step": 410}, {"loss": 1.6091, "grad_norm": 0.6380982398986816, "learning_rate": 0.0002, "epoch": 0.30401737242128124, "step": 420}, {"loss": 1.5292, "grad_norm": 0.6707863211631775, "learning_rate": 0.0002, "epoch": 0.3112558812884546, "step": 430}, {"loss": 1.4426, "grad_norm": 0.7010176777839661, "learning_rate": 0.0002, "epoch": 0.3184943901556279, "step": 440}, {"loss": 1.5572, "grad_norm": 0.8263739943504333, "learning_rate": 0.0002, "epoch": 0.3257328990228013, "step": 450}, {"loss": 1.5188, "grad_norm": 0.7253276109695435, "learning_rate": 0.0002, "epoch": 0.33297140788997465, "step": 460}, {"loss": 1.584, "grad_norm": 0.5238934755325317, "learning_rate": 0.0002, "epoch": 0.340209916757148, "step": 470}, {"loss": 1.7035, "grad_norm": 0.7869495749473572, "learning_rate": 0.0002, "epoch": 0.3474484256243214, "step": 480}, {"loss": 1.5776, "grad_norm": 0.7485215663909912, "learning_rate": 0.0002, "epoch": 0.35468693449149474, "step": 490}, {"loss": 1.6274, "grad_norm": 0.5413193106651306, "learning_rate": 0.0002, "epoch": 0.3619254433586681, "step": 500}, {"loss": 1.7323, "grad_norm": 0.7615048885345459, "learning_rate": 0.0002, "epoch": 0.3691639522258415, "step": 510}, {"loss": 1.532, "grad_norm": 0.7685340046882629, "learning_rate": 0.0002, "epoch": 0.37640246109301484, "step": 520}, {"loss": 1.6312, "grad_norm": 0.6379081010818481, "learning_rate": 0.0002, "epoch": 0.3836409699601882, "step": 530}, {"loss": 1.5645, "grad_norm": 0.7946939468383789, "learning_rate": 0.0002, "epoch": 0.39087947882736157, "step": 540}, {"loss": 1.4001, "grad_norm": 0.6287278532981873, "learning_rate": 0.0002, "epoch": 0.39811798769453494, "step": 550}, {"loss": 1.5982, "grad_norm": 0.6811642646789551, "learning_rate": 0.0002, "epoch": 0.4053564965617083, "step": 560}, {"loss": 1.4953, "grad_norm": 0.671073317527771, "learning_rate": 0.0002, "epoch": 0.41259500542888167, "step": 570}, {"loss": 1.6753, "grad_norm": 0.6313900351524353, "learning_rate": 0.0002, "epoch": 0.41983351429605503, "step": 580}, {"loss": 1.546, "grad_norm": 0.5291772484779358, "learning_rate": 0.0002, "epoch": 0.4270720231632284, "step": 590}, {"loss": 1.5441, "grad_norm": 0.62503582239151, "learning_rate": 0.0002, "epoch": 0.43431053203040176, "step": 600}, {"loss": 1.6276, "grad_norm": 0.5777305364608765, "learning_rate": 0.0002, "epoch": 0.4415490408975751, "step": 610}, {"loss": 1.4758, "grad_norm": 0.7013497352600098, "learning_rate": 0.0002, "epoch": 0.44878754976474844, "step": 620}, {"loss": 1.4029, "grad_norm": 0.8044822216033936, "learning_rate": 0.0002, "epoch": 0.4560260586319218, "step": 630}, {"loss": 1.7195, "grad_norm": 0.672531247138977, "learning_rate": 0.0002, "epoch": 0.46326456749909517, "step": 640}, {"loss": 1.614, "grad_norm": 0.6233910322189331, "learning_rate": 0.0002, "epoch": 0.47050307636626854, "step": 650}, {"loss": 1.6041, "grad_norm": 0.651524543762207, "learning_rate": 0.0002, "epoch": 0.4777415852334419, "step": 660}, {"loss": 1.5842, "grad_norm": 0.7213939428329468, "learning_rate": 0.0002, "epoch": 0.48498009410061527, "step": 670}, {"loss": 1.5453, "grad_norm": 0.6541454792022705, "learning_rate": 0.0002, "epoch": 0.49221860296778863, "step": 680}, {"loss": 1.662, "grad_norm": 0.6568936109542847, "learning_rate": 0.0002, "epoch": 0.499457111834962, "step": 690}, {"loss": 1.624, "grad_norm": 0.7176415324211121, "learning_rate": 0.0002, "epoch": 0.5066956207021354, "step": 700}, {"loss": 1.6099, "grad_norm": 0.6553855538368225, "learning_rate": 0.0002, "epoch": 0.5139341295693087, "step": 710}, {"loss": 1.5508, "grad_norm": 0.5654335618019104, "learning_rate": 0.0002, "epoch": 0.5211726384364821, "step": 720}, {"loss": 1.392, "grad_norm": 0.5671001672744751, "learning_rate": 0.0002, "epoch": 0.5284111473036555, "step": 730}, {"loss": 1.388, "grad_norm": 0.7914412021636963, "learning_rate": 0.0002, "epoch": 0.5356496561708288, "step": 740}, {"loss": 1.5931, "grad_norm": 0.6172138452529907, "learning_rate": 0.0002, "epoch": 0.5428881650380022, "step": 750}, {"loss": 1.4018, "grad_norm": 0.6132623553276062, "learning_rate": 0.0002, "epoch": 0.5501266739051756, "step": 760}, {"loss": 1.513, "grad_norm": 0.654000461101532, "learning_rate": 0.0002, "epoch": 0.5573651827723489, "step": 770}, {"loss": 1.5035, "grad_norm": 0.5691370964050293, "learning_rate": 0.0002, "epoch": 0.5646036916395223, "step": 780}, {"loss": 1.65, "grad_norm": 0.7922580242156982, "learning_rate": 0.0002, "epoch": 0.5718422005066957, "step": 790}, {"loss": 1.4521, "grad_norm": 0.6831880211830139, "learning_rate": 0.0002, "epoch": 0.579080709373869, "step": 800}, {"loss": 1.4734, "grad_norm": 0.6740124821662903, "learning_rate": 0.0002, "epoch": 0.5863192182410424, "step": 810}, {"loss": 1.6498, "grad_norm": 1.380016803741455, "learning_rate": 0.0002, "epoch": 0.5935577271082157, "step": 820}, {"loss": 1.4642, "grad_norm": 0.6552878022193909, "learning_rate": 0.0002, "epoch": 0.6007962359753891, "step": 830}, {"loss": 1.6271, "grad_norm": 0.6649535298347473, "learning_rate": 0.0002, "epoch": 0.6080347448425625, "step": 840}, {"loss": 1.5886, "grad_norm": 0.561738133430481, "learning_rate": 0.0002, "epoch": 0.6152732537097358, "step": 850}, {"loss": 1.5364, "grad_norm": 0.6133047938346863, "learning_rate": 0.0002, "epoch": 0.6225117625769092, "step": 860}, {"loss": 1.3489, "grad_norm": 0.559843122959137, "learning_rate": 0.0002, "epoch": 0.6297502714440825, "step": 870}, {"loss": 1.4878, "grad_norm": 0.6117811799049377, "learning_rate": 0.0002, "epoch": 0.6369887803112558, "step": 880}, {"loss": 1.56, "grad_norm": 0.6209776401519775, "learning_rate": 0.0002, "epoch": 0.6442272891784292, "step": 890}, {"loss": 1.6747, "grad_norm": 0.6234082579612732, "learning_rate": 0.0002, "epoch": 0.6514657980456026, "step": 900}, {"loss": 1.6963, "grad_norm": 0.7623258233070374, "learning_rate": 0.0002, "epoch": 0.6587043069127759, "step": 910}, {"loss": 1.2424, "grad_norm": 0.6148061752319336, "learning_rate": 0.0002, "epoch": 0.6659428157799493, "step": 920}, {"loss": 1.4319, "grad_norm": 0.6682973504066467, "learning_rate": 0.0002, "epoch": 0.6731813246471227, "step": 930}, {"loss": 1.5377, "grad_norm": 0.5513041615486145, "learning_rate": 0.0002, "epoch": 0.680419833514296, "step": 940}, {"loss": 1.3991, "grad_norm": 0.5197525024414062, "learning_rate": 0.0002, "epoch": 0.6876583423814694, "step": 950}, {"loss": 1.4398, "grad_norm": 0.6490758061408997, "learning_rate": 0.0002, "epoch": 0.6948968512486428, "step": 960}, {"loss": 1.5251, "grad_norm": 0.6450682878494263, "learning_rate": 0.0002, "epoch": 0.7021353601158161, "step": 970}, {"loss": 1.5417, "grad_norm": 0.6203766465187073, "learning_rate": 0.0002, "epoch": 0.7093738689829895, "step": 980}, {"loss": 1.4575, "grad_norm": 0.6023609638214111, "learning_rate": 0.0002, "epoch": 0.7166123778501629, "step": 990}, {"loss": 1.4973, "grad_norm": 0.5765255093574524, "learning_rate": 0.0002, "epoch": 0.7238508867173362, "step": 1000}, {"loss": 1.483, "grad_norm": 0.6650075316429138, "learning_rate": 0.0002, "epoch": 0.7310893955845096, "step": 1010}, {"loss": 1.5959, "grad_norm": 0.5610854029655457, "learning_rate": 0.0002, "epoch": 0.738327904451683, "step": 1020}, {"loss": 1.5248, "grad_norm": 0.7072813510894775, "learning_rate": 0.0002, "epoch": 0.7455664133188563, "step": 1030}, {"loss": 1.5776, "grad_norm": 0.6815407872200012, "learning_rate": 0.0002, "epoch": 0.7528049221860297, "step": 1040}, {"loss": 1.4577, "grad_norm": 0.7932390570640564, "learning_rate": 0.0002, "epoch": 0.760043431053203, "step": 1050}, {"loss": 1.4515, "grad_norm": 0.5798183083534241, "learning_rate": 0.0002, "epoch": 0.7672819399203764, "step": 1060}, {"loss": 1.5053, "grad_norm": 0.7898504137992859, "learning_rate": 0.0002, "epoch": 0.7745204487875498, "step": 1070}, {"loss": 1.4776, "grad_norm": 0.4983280301094055, "learning_rate": 0.0002, "epoch": 0.7817589576547231, "step": 1080}, {"loss": 1.5007, "grad_norm": 0.691403329372406, "learning_rate": 0.0002, "epoch": 0.7889974665218965, "step": 1090}, {"loss": 1.5153, "grad_norm": 0.5394481420516968, "learning_rate": 0.0002, "epoch": 0.7962359753890699, "step": 1100}, {"loss": 1.6892, "grad_norm": 0.5136822462081909, "learning_rate": 0.0002, "epoch": 0.8034744842562432, "step": 1110}, {"loss": 1.4902, "grad_norm": 0.6828126907348633, "learning_rate": 0.0002, "epoch": 0.8107129931234166, "step": 1120}, {"loss": 1.4346, "grad_norm": 0.6799656748771667, "learning_rate": 0.0002, "epoch": 0.81795150199059, "step": 1130}, {"loss": 1.2678, "grad_norm": 0.5428406000137329, "learning_rate": 0.0002, "epoch": 0.8251900108577633, "step": 1140}, {"loss": 1.4072, "grad_norm": 0.4811290502548218, "learning_rate": 0.0002, "epoch": 0.8324285197249367, "step": 1150}, {"loss": 1.4512, "grad_norm": 0.5519434809684753, "learning_rate": 0.0002, "epoch": 0.8396670285921101, "step": 1160}, {"loss": 1.4072, "grad_norm": 0.9748060703277588, "learning_rate": 0.0002, "epoch": 0.8469055374592834, "step": 1170}, {"loss": 1.4309, "grad_norm": 0.712609589099884, "learning_rate": 0.0002, "epoch": 0.8541440463264568, "step": 1180}, {"loss": 1.434, "grad_norm": 0.6866157054901123, "learning_rate": 0.0002, "epoch": 0.8613825551936302, "step": 1190}, {"loss": 1.3704, "grad_norm": 0.5068854093551636, "learning_rate": 0.0002, "epoch": 0.8686210640608035, "step": 1200}, {"loss": 1.5601, "grad_norm": 0.6333245038986206, "learning_rate": 0.0002, "epoch": 0.8758595729279768, "step": 1210}, {"loss": 1.4636, "grad_norm": 0.6424421072006226, "learning_rate": 0.0002, "epoch": 0.8830980817951501, "step": 1220}, {"loss": 1.4186, "grad_norm": 0.4771921932697296, "learning_rate": 0.0002, "epoch": 0.8903365906623235, "step": 1230}, {"loss": 1.6323, "grad_norm": 0.5191764235496521, "learning_rate": 0.0002, "epoch": 0.8975750995294969, "step": 1240}, {"loss": 1.6105, "grad_norm": 0.756222128868103, "learning_rate": 0.0002, "epoch": 0.9048136083966702, "step": 1250}, {"loss": 1.4396, "grad_norm": 0.623823881149292, "learning_rate": 0.0002, "epoch": 0.9120521172638436, "step": 1260}, {"loss": 1.3097, "grad_norm": 0.8166571259498596, "learning_rate": 0.0002, "epoch": 0.919290626131017, "step": 1270}, {"loss": 1.4625, "grad_norm": 0.6059346795082092, "learning_rate": 0.0002, "epoch": 0.9265291349981903, "step": 1280}, {"loss": 1.3555, "grad_norm": 0.5842690467834473, "learning_rate": 0.0002, "epoch": 0.9337676438653637, "step": 1290}, {"loss": 1.5859, "grad_norm": 0.7649800777435303, "learning_rate": 0.0002, "epoch": 0.9410061527325371, "step": 1300}, {"loss": 1.5915, "grad_norm": 0.6420919895172119, "learning_rate": 0.0002, "epoch": 0.9482446615997104, "step": 1310}, {"loss": 1.453, "grad_norm": 0.7011452913284302, "learning_rate": 0.0002, "epoch": 0.9554831704668838, "step": 1320}, {"loss": 1.6766, "grad_norm": 0.5783746242523193, "learning_rate": 0.0002, "epoch": 0.9627216793340572, "step": 1330}, {"loss": 1.6308, "grad_norm": 0.5973192453384399, "learning_rate": 0.0002, "epoch": 0.9699601882012305, "step": 1340}, {"loss": 1.5901, "grad_norm": 0.6181833744049072, "learning_rate": 0.0002, "epoch": 0.9771986970684039, "step": 1350}, {"loss": 1.5258, "grad_norm": 0.5563396215438843, "learning_rate": 0.0002, "epoch": 0.9844372059355773, "step": 1360}, {"loss": 1.4508, "grad_norm": 0.45723360776901245, "learning_rate": 0.0002, "epoch": 0.9916757148027506, "step": 1370}, {"loss": 1.3291, "grad_norm": 0.5947498679161072, "learning_rate": 0.0002, "epoch": 0.998914223669924, "step": 1380}]}
+{"epoch": 2.0, "step": 2763, "epoch_duration": 1324.4045159816742, "total_accumulated_duration": 2686.5983464717865, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 3020.60888671875}, "peak_memory_usage": {"GPU_0": 15079.2998046875}, "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-0.9-num-8828-sd-1/checkpoint-1381", "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": 4.7061, "grad_norm": 1.2523442506790161, "learning_rate": 0.0002, "epoch": 0.007238508867173362, "step": 10}, {"loss": 3.3493, "grad_norm": 1.8887330293655396, "learning_rate": 0.0002, "epoch": 0.014477017734346724, "step": 20}, {"loss": 2.7585, "grad_norm": 0.9668035507202148, "learning_rate": 0.0002, "epoch": 0.021715526601520086, "step": 30}, {"loss": 2.3699, "grad_norm": 2.9167306423187256, "learning_rate": 0.0002, "epoch": 0.028954035468693448, "step": 40}, {"loss": 2.2679, "grad_norm": 2.649867296218872, "learning_rate": 0.0002, "epoch": 0.036192544335866814, "step": 50}, {"loss": 2.2202, "grad_norm": 1.5120655298233032, "learning_rate": 0.0002, "epoch": 0.04343105320304017, "step": 60}, {"loss": 2.2026, "grad_norm": 0.7879868149757385, "learning_rate": 0.0002, "epoch": 0.05066956207021354, "step": 70}, {"loss": 1.9447, "grad_norm": 0.7616953253746033, "learning_rate": 0.0002, "epoch": 0.057908070937386896, "step": 80}, {"loss": 2.0112, "grad_norm": 1.8809149265289307, "learning_rate": 0.0002, "epoch": 0.06514657980456026, "step": 90}, {"loss": 1.8337, "grad_norm": 0.9294016361236572, "learning_rate": 0.0002, "epoch": 0.07238508867173363, "step": 100}, {"loss": 1.8419, "grad_norm": 0.7145281434059143, "learning_rate": 0.0002, "epoch": 0.07962359753890698, "step": 110}, {"loss": 2.0036, "grad_norm": 0.7564446330070496, "learning_rate": 0.0002, "epoch": 0.08686210640608034, "step": 120}, {"loss": 1.9306, "grad_norm": 1.1681925058364868, "learning_rate": 0.0002, "epoch": 0.09410061527325371, "step": 130}, {"loss": 1.7875, "grad_norm": 0.6708641648292542, "learning_rate": 0.0002, "epoch": 0.10133912414042708, "step": 140}, {"loss": 1.786, "grad_norm": 0.7625647783279419, "learning_rate": 0.0002, "epoch": 0.10857763300760044, "step": 150}, {"loss": 1.6687, "grad_norm": 0.8463464975357056, "learning_rate": 0.0002, "epoch": 0.11581614187477379, "step": 160}, {"loss": 1.6214, "grad_norm": 0.7502335906028748, "learning_rate": 0.0002, "epoch": 0.12305465074194716, "step": 170}, {"loss": 1.7433, "grad_norm": 0.6929958462715149, "learning_rate": 0.0002, "epoch": 0.13029315960912052, "step": 180}, {"loss": 1.6009, "grad_norm": 0.6798707842826843, "learning_rate": 0.0002, "epoch": 0.1375316684762939, "step": 190}, {"loss": 1.6208, "grad_norm": 0.7566508650779724, "learning_rate": 0.0002, "epoch": 0.14477017734346725, "step": 200}, {"loss": 1.5823, "grad_norm": 0.7196869850158691, "learning_rate": 0.0002, "epoch": 0.15200868621064062, "step": 210}, {"loss": 1.738, "grad_norm": 0.8401045799255371, "learning_rate": 0.0002, "epoch": 0.15924719507781396, "step": 220}, {"loss": 1.7574, "grad_norm": 0.8503773212432861, "learning_rate": 0.0002, "epoch": 0.16648570394498732, "step": 230}, {"loss": 1.7861, "grad_norm": 0.7183733582496643, "learning_rate": 0.0002, "epoch": 0.1737242128121607, "step": 240}, {"loss": 1.6693, "grad_norm": 0.7082605957984924, "learning_rate": 0.0002, "epoch": 0.18096272167933405, "step": 250}, {"loss": 1.619, "grad_norm": 0.9386326670646667, "learning_rate": 0.0002, "epoch": 0.18820123054650742, "step": 260}, {"loss": 1.6511, "grad_norm": 0.7332451939582825, "learning_rate": 0.0002, "epoch": 0.19543973941368079, "step": 270}, {"loss": 1.6353, "grad_norm": 0.7092869877815247, "learning_rate": 0.0002, "epoch": 0.20267824828085415, "step": 280}, {"loss": 1.5996, "grad_norm": 0.7256413698196411, "learning_rate": 0.0002, "epoch": 0.20991675714802752, "step": 290}, {"loss": 1.6754, "grad_norm": 0.6398681402206421, "learning_rate": 0.0002, "epoch": 0.21715526601520088, "step": 300}, {"loss": 1.397, "grad_norm": 0.6273287534713745, "learning_rate": 0.0002, "epoch": 0.22439377488237422, "step": 310}, {"loss": 1.5115, "grad_norm": 0.511648416519165, "learning_rate": 0.0002, "epoch": 0.23163228374954759, "step": 320}, {"loss": 1.5424, "grad_norm": 0.8677352070808411, "learning_rate": 0.0002, "epoch": 0.23887079261672095, "step": 330}, {"loss": 1.6779, "grad_norm": 0.6270743012428284, "learning_rate": 0.0002, "epoch": 0.24610930148389432, "step": 340}, {"loss": 1.626, "grad_norm": 0.7980281114578247, "learning_rate": 0.0002, "epoch": 0.2533478103510677, "step": 350}, {"loss": 1.5238, "grad_norm": 0.632486879825592, "learning_rate": 0.0002, "epoch": 0.26058631921824105, "step": 360}, {"loss": 1.5175, "grad_norm": 0.6527034640312195, "learning_rate": 0.0002, "epoch": 0.2678248280854144, "step": 370}, {"loss": 1.627, "grad_norm": 0.7672118544578552, "learning_rate": 0.0002, "epoch": 0.2750633369525878, "step": 380}, {"loss": 1.5605, "grad_norm": 0.6035117506980896, "learning_rate": 0.0002, "epoch": 0.28230184581976114, "step": 390}, {"loss": 1.4603, "grad_norm": 0.5955103039741516, "learning_rate": 0.0002, "epoch": 0.2895403546869345, "step": 400}, {"loss": 1.558, "grad_norm": 0.6015191674232483, "learning_rate": 0.0002, "epoch": 0.2967788635541079, "step": 410}, {"loss": 1.6091, "grad_norm": 0.6380982398986816, "learning_rate": 0.0002, "epoch": 0.30401737242128124, "step": 420}, {"loss": 1.5292, "grad_norm": 0.6707863211631775, "learning_rate": 0.0002, "epoch": 0.3112558812884546, "step": 430}, {"loss": 1.4426, "grad_norm": 0.7010176777839661, "learning_rate": 0.0002, "epoch": 0.3184943901556279, "step": 440}, {"loss": 1.5572, "grad_norm": 0.8263739943504333, "learning_rate": 0.0002, "epoch": 0.3257328990228013, "step": 450}, {"loss": 1.5188, "grad_norm": 0.7253276109695435, "learning_rate": 0.0002, "epoch": 0.33297140788997465, "step": 460}, {"loss": 1.584, "grad_norm": 0.5238934755325317, "learning_rate": 0.0002, "epoch": 0.340209916757148, "step": 470}, {"loss": 1.7035, "grad_norm": 0.7869495749473572, "learning_rate": 0.0002, "epoch": 0.3474484256243214, "step": 480}, {"loss": 1.5776, "grad_norm": 0.7485215663909912, "learning_rate": 0.0002, "epoch": 0.35468693449149474, "step": 490}, {"loss": 1.6274, "grad_norm": 0.5413193106651306, "learning_rate": 0.0002, "epoch": 0.3619254433586681, "step": 500}, {"loss": 1.7323, "grad_norm": 0.7615048885345459, "learning_rate": 0.0002, "epoch": 0.3691639522258415, "step": 510}, {"loss": 1.532, "grad_norm": 0.7685340046882629, "learning_rate": 0.0002, "epoch": 0.37640246109301484, "step": 520}, {"loss": 1.6312, "grad_norm": 0.6379081010818481, "learning_rate": 0.0002, "epoch": 0.3836409699601882, "step": 530}, {"loss": 1.5645, "grad_norm": 0.7946939468383789, "learning_rate": 0.0002, "epoch": 0.39087947882736157, "step": 540}, {"loss": 1.4001, "grad_norm": 0.6287278532981873, "learning_rate": 0.0002, "epoch": 0.39811798769453494, "step": 550}, {"loss": 1.5982, "grad_norm": 0.6811642646789551, "learning_rate": 0.0002, "epoch": 0.4053564965617083, "step": 560}, {"loss": 1.4953, "grad_norm": 0.671073317527771, "learning_rate": 0.0002, "epoch": 0.41259500542888167, "step": 570}, {"loss": 1.6753, "grad_norm": 0.6313900351524353, "learning_rate": 0.0002, "epoch": 0.41983351429605503, "step": 580}, {"loss": 1.546, "grad_norm": 0.5291772484779358, "learning_rate": 0.0002, "epoch": 0.4270720231632284, "step": 590}, {"loss": 1.5441, "grad_norm": 0.62503582239151, "learning_rate": 0.0002, "epoch": 0.43431053203040176, "step": 600}, {"loss": 1.6276, "grad_norm": 0.5777305364608765, "learning_rate": 0.0002, "epoch": 0.4415490408975751, "step": 610}, {"loss": 1.4758, "grad_norm": 0.7013497352600098, "learning_rate": 0.0002, "epoch": 0.44878754976474844, "step": 620}, {"loss": 1.4029, "grad_norm": 0.8044822216033936, "learning_rate": 0.0002, "epoch": 0.4560260586319218, "step": 630}, {"loss": 1.7195, "grad_norm": 0.672531247138977, "learning_rate": 0.0002, "epoch": 0.46326456749909517, "step": 640}, {"loss": 1.614, "grad_norm": 0.6233910322189331, "learning_rate": 0.0002, "epoch": 0.47050307636626854, "step": 650}, {"loss": 1.6041, "grad_norm": 0.651524543762207, "learning_rate": 0.0002, "epoch": 0.4777415852334419, "step": 660}, {"loss": 1.5842, "grad_norm": 0.7213939428329468, "learning_rate": 0.0002, "epoch": 0.48498009410061527, "step": 670}, {"loss": 1.5453, "grad_norm": 0.6541454792022705, "learning_rate": 0.0002, "epoch": 0.49221860296778863, "step": 680}, {"loss": 1.662, "grad_norm": 0.6568936109542847, "learning_rate": 0.0002, "epoch": 0.499457111834962, "step": 690}, {"loss": 1.624, "grad_norm": 0.7176415324211121, "learning_rate": 0.0002, "epoch": 0.5066956207021354, "step": 700}, {"loss": 1.6099, "grad_norm": 0.6553855538368225, "learning_rate": 0.0002, "epoch": 0.5139341295693087, "step": 710}, {"loss": 1.5508, "grad_norm": 0.5654335618019104, "learning_rate": 0.0002, "epoch": 0.5211726384364821, "step": 720}, {"loss": 1.392, "grad_norm": 0.5671001672744751, "learning_rate": 0.0002, "epoch": 0.5284111473036555, "step": 730}, {"loss": 1.388, "grad_norm": 0.7914412021636963, "learning_rate": 0.0002, "epoch": 0.5356496561708288, "step": 740}, {"loss": 1.5931, "grad_norm": 0.6172138452529907, "learning_rate": 0.0002, "epoch": 0.5428881650380022, "step": 750}, {"loss": 1.4018, "grad_norm": 0.6132623553276062, "learning_rate": 0.0002, "epoch": 0.5501266739051756, "step": 760}, {"loss": 1.513, "grad_norm": 0.654000461101532, "learning_rate": 0.0002, "epoch": 0.5573651827723489, "step": 770}, {"loss": 1.5035, "grad_norm": 0.5691370964050293, "learning_rate": 0.0002, "epoch": 0.5646036916395223, "step": 780}, {"loss": 1.65, "grad_norm": 0.7922580242156982, "learning_rate": 0.0002, "epoch": 0.5718422005066957, "step": 790}, {"loss": 1.4521, "grad_norm": 0.6831880211830139, "learning_rate": 0.0002, "epoch": 0.579080709373869, "step": 800}, {"loss": 1.4734, "grad_norm": 0.6740124821662903, "learning_rate": 0.0002, "epoch": 0.5863192182410424, "step": 810}, {"loss": 1.6498, "grad_norm": 1.380016803741455, "learning_rate": 0.0002, "epoch": 0.5935577271082157, "step": 820}, {"loss": 1.4642, "grad_norm": 0.6552878022193909, "learning_rate": 0.0002, "epoch": 0.6007962359753891, "step": 830}, {"loss": 1.6271, "grad_norm": 0.6649535298347473, "learning_rate": 0.0002, "epoch": 0.6080347448425625, "step": 840}, {"loss": 1.5886, "grad_norm": 0.561738133430481, "learning_rate": 0.0002, "epoch": 0.6152732537097358, "step": 850}, {"loss": 1.5364, "grad_norm": 0.6133047938346863, "learning_rate": 0.0002, "epoch": 0.6225117625769092, "step": 860}, {"loss": 1.3489, "grad_norm": 0.559843122959137, "learning_rate": 0.0002, "epoch": 0.6297502714440825, "step": 870}, {"loss": 1.4878, "grad_norm": 0.6117811799049377, "learning_rate": 0.0002, "epoch": 0.6369887803112558, "step": 880}, {"loss": 1.56, "grad_norm": 0.6209776401519775, "learning_rate": 0.0002, "epoch": 0.6442272891784292, "step": 890}, {"loss": 1.6747, "grad_norm": 0.6234082579612732, "learning_rate": 0.0002, "epoch": 0.6514657980456026, "step": 900}, {"loss": 1.6963, "grad_norm": 0.7623258233070374, "learning_rate": 0.0002, "epoch": 0.6587043069127759, "step": 910}, {"loss": 1.2424, "grad_norm": 0.6148061752319336, "learning_rate": 0.0002, "epoch": 0.6659428157799493, "step": 920}, {"loss": 1.4319, "grad_norm": 0.6682973504066467, "learning_rate": 0.0002, "epoch": 0.6731813246471227, "step": 930}, {"loss": 1.5377, "grad_norm": 0.5513041615486145, "learning_rate": 0.0002, "epoch": 0.680419833514296, "step": 940}, {"loss": 1.3991, "grad_norm": 0.5197525024414062, "learning_rate": 0.0002, "epoch": 0.6876583423814694, "step": 950}, {"loss": 1.4398, "grad_norm": 0.6490758061408997, "learning_rate": 0.0002, "epoch": 0.6948968512486428, "step": 960}, {"loss": 1.5251, "grad_norm": 0.6450682878494263, "learning_rate": 0.0002, "epoch": 0.7021353601158161, "step": 970}, {"loss": 1.5417, "grad_norm": 0.6203766465187073, "learning_rate": 0.0002, "epoch": 0.7093738689829895, "step": 980}, {"loss": 1.4575, "grad_norm": 0.6023609638214111, "learning_rate": 0.0002, "epoch": 0.7166123778501629, "step": 990}, {"loss": 1.4973, "grad_norm": 0.5765255093574524, "learning_rate": 0.0002, "epoch": 0.7238508867173362, "step": 1000}, {"loss": 1.483, "grad_norm": 0.6650075316429138, "learning_rate": 0.0002, "epoch": 0.7310893955845096, "step": 1010}, {"loss": 1.5959, "grad_norm": 0.5610854029655457, "learning_rate": 0.0002, "epoch": 0.738327904451683, "step": 1020}, {"loss": 1.5248, "grad_norm": 0.7072813510894775, "learning_rate": 0.0002, "epoch": 0.7455664133188563, "step": 1030}, {"loss": 1.5776, "grad_norm": 0.6815407872200012, "learning_rate": 0.0002, "epoch": 0.7528049221860297, "step": 1040}, {"loss": 1.4577, "grad_norm": 0.7932390570640564, "learning_rate": 0.0002, "epoch": 0.760043431053203, "step": 1050}, {"loss": 1.4515, "grad_norm": 0.5798183083534241, "learning_rate": 0.0002, "epoch": 0.7672819399203764, "step": 1060}, {"loss": 1.5053, "grad_norm": 0.7898504137992859, "learning_rate": 0.0002, "epoch": 0.7745204487875498, "step": 1070}, {"loss": 1.4776, "grad_norm": 0.4983280301094055, "learning_rate": 0.0002, "epoch": 0.7817589576547231, "step": 1080}, {"loss": 1.5007, "grad_norm": 0.691403329372406, "learning_rate": 0.0002, "epoch": 0.7889974665218965, "step": 1090}, {"loss": 1.5153, "grad_norm": 0.5394481420516968, "learning_rate": 0.0002, "epoch": 0.7962359753890699, "step": 1100}, {"loss": 1.6892, "grad_norm": 0.5136822462081909, "learning_rate": 0.0002, "epoch": 0.8034744842562432, "step": 1110}, {"loss": 1.4902, "grad_norm": 0.6828126907348633, "learning_rate": 0.0002, "epoch": 0.8107129931234166, "step": 1120}, {"loss": 1.4346, "grad_norm": 0.6799656748771667, "learning_rate": 0.0002, "epoch": 0.81795150199059, "step": 1130}, {"loss": 1.2678, "grad_norm": 0.5428406000137329, "learning_rate": 0.0002, "epoch": 0.8251900108577633, "step": 1140}, {"loss": 1.4072, "grad_norm": 0.4811290502548218, "learning_rate": 0.0002, "epoch": 0.8324285197249367, "step": 1150}, {"loss": 1.4512, "grad_norm": 0.5519434809684753, "learning_rate": 0.0002, "epoch": 0.8396670285921101, "step": 1160}, {"loss": 1.4072, "grad_norm": 0.9748060703277588, "learning_rate": 0.0002, "epoch": 0.8469055374592834, "step": 1170}, {"loss": 1.4309, "grad_norm": 0.712609589099884, "learning_rate": 0.0002, "epoch": 0.8541440463264568, "step": 1180}, {"loss": 1.434, "grad_norm": 0.6866157054901123, "learning_rate": 0.0002, "epoch": 0.8613825551936302, "step": 1190}, {"loss": 1.3704, "grad_norm": 0.5068854093551636, "learning_rate": 0.0002, "epoch": 0.8686210640608035, "step": 1200}, {"loss": 1.5601, "grad_norm": 0.6333245038986206, "learning_rate": 0.0002, "epoch": 0.8758595729279768, "step": 1210}, {"loss": 1.4636, "grad_norm": 0.6424421072006226, "learning_rate": 0.0002, "epoch": 0.8830980817951501, "step": 1220}, {"loss": 1.4186, "grad_norm": 0.4771921932697296, "learning_rate": 0.0002, "epoch": 0.8903365906623235, "step": 1230}, {"loss": 1.6323, "grad_norm": 0.5191764235496521, "learning_rate": 0.0002, "epoch": 0.8975750995294969, "step": 1240}, {"loss": 1.6105, "grad_norm": 0.756222128868103, "learning_rate": 0.0002, "epoch": 0.9048136083966702, "step": 1250}, {"loss": 1.4396, "grad_norm": 0.623823881149292, "learning_rate": 0.0002, "epoch": 0.9120521172638436, "step": 1260}, {"loss": 1.3097, "grad_norm": 0.8166571259498596, "learning_rate": 0.0002, "epoch": 0.919290626131017, "step": 1270}, {"loss": 1.4625, "grad_norm": 0.6059346795082092, "learning_rate": 0.0002, "epoch": 0.9265291349981903, "step": 1280}, {"loss": 1.3555, "grad_norm": 0.5842690467834473, "learning_rate": 0.0002, "epoch": 0.9337676438653637, "step": 1290}, {"loss": 1.5859, "grad_norm": 0.7649800777435303, "learning_rate": 0.0002, "epoch": 0.9410061527325371, "step": 1300}, {"loss": 1.5915, "grad_norm": 0.6420919895172119, "learning_rate": 0.0002, "epoch": 0.9482446615997104, "step": 1310}, {"loss": 1.453, "grad_norm": 0.7011452913284302, "learning_rate": 0.0002, "epoch": 0.9554831704668838, "step": 1320}, {"loss": 1.6766, "grad_norm": 0.5783746242523193, "learning_rate": 0.0002, "epoch": 0.9627216793340572, "step": 1330}, {"loss": 1.6308, "grad_norm": 0.5973192453384399, "learning_rate": 0.0002, "epoch": 0.9699601882012305, "step": 1340}, {"loss": 1.5901, "grad_norm": 0.6181833744049072, "learning_rate": 0.0002, "epoch": 0.9771986970684039, "step": 1350}, {"loss": 1.5258, "grad_norm": 0.5563396215438843, "learning_rate": 0.0002, "epoch": 0.9844372059355773, "step": 1360}, {"loss": 1.4508, "grad_norm": 0.45723360776901245, "learning_rate": 0.0002, "epoch": 0.9916757148027506, "step": 1370}, {"loss": 1.3291, "grad_norm": 0.5947498679161072, "learning_rate": 0.0002, "epoch": 0.998914223669924, "step": 1380}, {"eval_loss": 1.480796456336975, "eval_runtime": 27.3103, "eval_samples_per_second": 15.965, "eval_steps_per_second": 2.014, "epoch": 0.9996380745566413, "step": 1381}, {"loss": 1.3057, "grad_norm": 0.5599952936172485, "learning_rate": 0.0002, "epoch": 1.0061527325370974, "step": 1390}, {"loss": 1.4991, "grad_norm": 0.5932008028030396, "learning_rate": 0.0002, "epoch": 1.0133912414042707, "step": 1400}, {"loss": 1.4506, "grad_norm": 0.6194121837615967, "learning_rate": 0.0002, "epoch": 1.020629750271444, "step": 1410}, {"loss": 1.5966, "grad_norm": 0.6995621919631958, "learning_rate": 0.0002, "epoch": 1.0278682591386175, "step": 1420}, {"loss": 1.4153, "grad_norm": 0.7905810475349426, "learning_rate": 0.0002, "epoch": 1.0351067680057908, "step": 1430}, {"loss": 1.4414, "grad_norm": 0.7221615314483643, "learning_rate": 0.0002, "epoch": 1.0423452768729642, "step": 1440}, {"loss": 1.3859, "grad_norm": 0.6170642375946045, "learning_rate": 0.0002, "epoch": 1.0495837857401376, "step": 1450}, {"loss": 1.3806, "grad_norm": 0.5844094753265381, "learning_rate": 0.0002, "epoch": 1.056822294607311, "step": 1460}, {"loss": 1.4871, "grad_norm": 0.7731822729110718, "learning_rate": 0.0002, "epoch": 1.0640608034744843, "step": 1470}, {"loss": 1.4286, "grad_norm": 0.4554748237133026, "learning_rate": 0.0002, "epoch": 1.0712993123416577, "step": 1480}, {"loss": 1.3977, "grad_norm": 0.6923259496688843, "learning_rate": 0.0002, "epoch": 1.078537821208831, "step": 1490}, {"loss": 1.3936, "grad_norm": 0.6008219122886658, "learning_rate": 0.0002, "epoch": 1.0857763300760044, "step": 1500}, {"loss": 1.4821, "grad_norm": 0.6450045704841614, "learning_rate": 0.0002, "epoch": 1.0930148389431777, "step": 1510}, {"loss": 1.3295, "grad_norm": 0.7833753824234009, "learning_rate": 0.0002, "epoch": 1.1002533478103511, "step": 1520}, {"loss": 1.3424, "grad_norm": 0.5076758861541748, "learning_rate": 0.0002, "epoch": 1.1074918566775245, "step": 1530}, {"loss": 1.4043, "grad_norm": 0.5661332011222839, "learning_rate": 0.0002, "epoch": 1.1147303655446978, "step": 1540}, {"loss": 1.4963, "grad_norm": 0.6526919603347778, "learning_rate": 0.0002, "epoch": 1.1219688744118712, "step": 1550}, {"loss": 1.3671, "grad_norm": 0.5613082647323608, "learning_rate": 0.0002, "epoch": 1.1292073832790446, "step": 1560}, {"loss": 1.4458, "grad_norm": 0.6113885641098022, "learning_rate": 0.0002, "epoch": 1.136445892146218, "step": 1570}, {"loss": 1.3552, "grad_norm": 0.6732510328292847, "learning_rate": 0.0002, "epoch": 1.1436844010133913, "step": 1580}, {"loss": 1.3114, "grad_norm": 0.6146392226219177, "learning_rate": 0.0002, "epoch": 1.1509229098805647, "step": 1590}, {"loss": 1.411, "grad_norm": 0.6766974329948425, "learning_rate": 0.0002, "epoch": 1.158161418747738, "step": 1600}, {"loss": 1.2401, "grad_norm": 0.7621957659721375, "learning_rate": 0.0002, "epoch": 1.1653999276149114, "step": 1610}, {"loss": 1.3758, "grad_norm": 0.6959581971168518, "learning_rate": 0.0002, "epoch": 1.1726384364820848, "step": 1620}, {"loss": 1.382, "grad_norm": 0.6691278219223022, "learning_rate": 0.0002, "epoch": 1.1798769453492581, "step": 1630}, {"loss": 1.4147, "grad_norm": 0.4927774965763092, "learning_rate": 0.0002, "epoch": 1.1871154542164315, "step": 1640}, {"loss": 1.449, "grad_norm": 0.7724234461784363, "learning_rate": 0.0002, "epoch": 1.1943539630836049, "step": 1650}, {"loss": 1.4778, "grad_norm": 0.6817787885665894, "learning_rate": 0.0002, "epoch": 1.2015924719507782, "step": 1660}, {"loss": 1.3776, "grad_norm": 0.6500699520111084, "learning_rate": 0.0002, "epoch": 1.2088309808179516, "step": 1670}, {"loss": 1.3875, "grad_norm": 0.5703568458557129, "learning_rate": 0.0002, "epoch": 1.216069489685125, "step": 1680}, {"loss": 1.4735, "grad_norm": 0.6261579990386963, "learning_rate": 0.0002, "epoch": 1.2233079985522983, "step": 1690}, {"loss": 1.3898, "grad_norm": 0.651713490486145, "learning_rate": 0.0002, "epoch": 1.2305465074194717, "step": 1700}, {"loss": 1.4002, "grad_norm": 0.684399425983429, "learning_rate": 0.0002, "epoch": 1.237785016286645, "step": 1710}, {"loss": 1.5027, "grad_norm": 0.6996857523918152, "learning_rate": 0.0002, "epoch": 1.2450235251538184, "step": 1720}, {"loss": 1.3326, "grad_norm": 0.7102537751197815, "learning_rate": 0.0002, "epoch": 1.2522620340209918, "step": 1730}, {"loss": 1.3675, "grad_norm": 0.45809897780418396, "learning_rate": 0.0002, "epoch": 1.2595005428881652, "step": 1740}, {"loss": 1.4175, "grad_norm": 0.6377046704292297, "learning_rate": 0.0002, "epoch": 1.2667390517553385, "step": 1750}, {"loss": 1.3479, "grad_norm": 0.6965704560279846, "learning_rate": 0.0002, "epoch": 1.2739775606225119, "step": 1760}, {"loss": 1.5647, "grad_norm": 0.5688214302062988, "learning_rate": 0.0002, "epoch": 1.2812160694896852, "step": 1770}, {"loss": 1.3967, "grad_norm": 0.6384190320968628, "learning_rate": 0.0002, "epoch": 1.2884545783568586, "step": 1780}, {"loss": 1.3671, "grad_norm": 0.5629363656044006, "learning_rate": 0.0002, "epoch": 1.295693087224032, "step": 1790}, {"loss": 1.2292, "grad_norm": 0.6148255467414856, "learning_rate": 0.0002, "epoch": 1.3029315960912053, "step": 1800}, {"loss": 1.5806, "grad_norm": 0.655580997467041, "learning_rate": 0.0002, "epoch": 1.3101701049583787, "step": 1810}, {"loss": 1.2398, "grad_norm": 0.5642657279968262, "learning_rate": 0.0002, "epoch": 1.3174086138255519, "step": 1820}, {"loss": 1.3246, "grad_norm": 0.59607994556427, "learning_rate": 0.0002, "epoch": 1.3246471226927252, "step": 1830}, {"loss": 1.3274, "grad_norm": 0.5564199090003967, "learning_rate": 0.0002, "epoch": 1.3318856315598986, "step": 1840}, {"loss": 1.5834, "grad_norm": 0.6949955821037292, "learning_rate": 0.0002, "epoch": 1.339124140427072, "step": 1850}, {"loss": 1.4722, "grad_norm": 0.7036856412887573, "learning_rate": 0.0002, "epoch": 1.3463626492942453, "step": 1860}, {"loss": 1.333, "grad_norm": 0.722062885761261, "learning_rate": 0.0002, "epoch": 1.3536011581614187, "step": 1870}, {"loss": 1.4044, "grad_norm": 0.6098677515983582, "learning_rate": 0.0002, "epoch": 1.360839667028592, "step": 1880}, {"loss": 1.6217, "grad_norm": 0.5376402735710144, "learning_rate": 0.0002, "epoch": 1.3680781758957654, "step": 1890}, {"loss": 1.5071, "grad_norm": 0.6974610090255737, "learning_rate": 0.0002, "epoch": 1.3753166847629388, "step": 1900}, {"loss": 1.5854, "grad_norm": 0.6520763635635376, "learning_rate": 0.0002, "epoch": 1.3825551936301121, "step": 1910}, {"loss": 1.4271, "grad_norm": 0.6604374647140503, "learning_rate": 0.0002, "epoch": 1.3897937024972855, "step": 1920}, {"loss": 1.419, "grad_norm": 0.7364398241043091, "learning_rate": 0.0002, "epoch": 1.3970322113644589, "step": 1930}, {"loss": 1.4585, "grad_norm": 0.6849475502967834, "learning_rate": 0.0002, "epoch": 1.4042707202316322, "step": 1940}, {"loss": 1.5577, "grad_norm": 0.6562670469284058, "learning_rate": 0.0002, "epoch": 1.4115092290988056, "step": 1950}, {"loss": 1.4725, "grad_norm": 0.5695616006851196, "learning_rate": 0.0002, "epoch": 1.418747737965979, "step": 1960}, {"loss": 1.3088, "grad_norm": 0.5244464874267578, "learning_rate": 0.0002, "epoch": 1.4259862468331523, "step": 1970}, {"loss": 1.5069, "grad_norm": 0.6347293257713318, "learning_rate": 0.0002, "epoch": 1.4332247557003257, "step": 1980}, {"loss": 1.3502, "grad_norm": 0.5528361201286316, "learning_rate": 0.0002, "epoch": 1.440463264567499, "step": 1990}, {"loss": 1.3978, "grad_norm": 0.6987585425376892, "learning_rate": 0.0002, "epoch": 1.4477017734346724, "step": 2000}, {"loss": 1.4262, "grad_norm": 0.6568987369537354, "learning_rate": 0.0002, "epoch": 1.4549402823018458, "step": 2010}, {"loss": 1.4175, "grad_norm": 0.7665994763374329, "learning_rate": 0.0002, "epoch": 1.4621787911690192, "step": 2020}, {"loss": 1.244, "grad_norm": 0.5127707123756409, "learning_rate": 0.0002, "epoch": 1.4694173000361925, "step": 2030}, {"loss": 1.3699, "grad_norm": 0.5406824946403503, "learning_rate": 0.0002, "epoch": 1.476655808903366, "step": 2040}, {"loss": 1.3353, "grad_norm": 0.5990166664123535, "learning_rate": 0.0002, "epoch": 1.4838943177705393, "step": 2050}, {"loss": 1.2454, "grad_norm": 0.6186193823814392, "learning_rate": 0.0002, "epoch": 1.4911328266377126, "step": 2060}, {"loss": 1.428, "grad_norm": 0.6154307126998901, "learning_rate": 0.0002, "epoch": 1.498371335504886, "step": 2070}, {"loss": 1.4528, "grad_norm": 0.5606056451797485, "learning_rate": 0.0002, "epoch": 1.5056098443720594, "step": 2080}, {"loss": 1.2405, "grad_norm": 0.5006417036056519, "learning_rate": 0.0002, "epoch": 1.5128483532392327, "step": 2090}, {"loss": 1.4258, "grad_norm": 0.5968486070632935, "learning_rate": 0.0002, "epoch": 1.520086862106406, "step": 2100}, {"loss": 1.2752, "grad_norm": 0.5835496187210083, "learning_rate": 0.0002, "epoch": 1.5273253709735795, "step": 2110}, {"loss": 1.5443, "grad_norm": 0.6753535270690918, "learning_rate": 0.0002, "epoch": 1.5345638798407528, "step": 2120}, {"loss": 1.2139, "grad_norm": 0.7299720644950867, "learning_rate": 0.0002, "epoch": 1.5418023887079262, "step": 2130}, {"loss": 1.2364, "grad_norm": 0.5105988383293152, "learning_rate": 0.0002, "epoch": 1.5490408975750996, "step": 2140}, {"loss": 1.4528, "grad_norm": 0.5675431489944458, "learning_rate": 0.0002, "epoch": 1.556279406442273, "step": 2150}, {"loss": 1.4563, "grad_norm": 0.6246723532676697, "learning_rate": 0.0002, "epoch": 1.5635179153094463, "step": 2160}, {"loss": 1.5255, "grad_norm": 0.7291720509529114, "learning_rate": 0.0002, "epoch": 1.5707564241766196, "step": 2170}, {"loss": 1.5432, "grad_norm": 0.678114116191864, "learning_rate": 0.0002, "epoch": 1.577994933043793, "step": 2180}, {"loss": 1.5212, "grad_norm": 0.5136260986328125, "learning_rate": 0.0002, "epoch": 1.5852334419109664, "step": 2190}, {"loss": 1.3271, "grad_norm": 0.6359935998916626, "learning_rate": 0.0002, "epoch": 1.5924719507781397, "step": 2200}, {"loss": 1.4038, "grad_norm": 0.7650278806686401, "learning_rate": 0.0002, "epoch": 1.599710459645313, "step": 2210}, {"loss": 1.5478, "grad_norm": 0.7256110906600952, "learning_rate": 0.0002, "epoch": 1.6069489685124865, "step": 2220}, {"loss": 1.4387, "grad_norm": 0.688689649105072, "learning_rate": 0.0002, "epoch": 1.6141874773796598, "step": 2230}, {"loss": 1.4096, "grad_norm": 0.6045311093330383, "learning_rate": 0.0002, "epoch": 1.6214259862468332, "step": 2240}, {"loss": 1.4097, "grad_norm": 0.7064604163169861, "learning_rate": 0.0002, "epoch": 1.6286644951140063, "step": 2250}, {"loss": 1.3477, "grad_norm": 0.5309562087059021, "learning_rate": 0.0002, "epoch": 1.6359030039811797, "step": 2260}, {"loss": 1.4022, "grad_norm": 0.5687053203582764, "learning_rate": 0.0002, "epoch": 1.643141512848353, "step": 2270}, {"loss": 1.2977, "grad_norm": 0.535872757434845, "learning_rate": 0.0002, "epoch": 1.6503800217155264, "step": 2280}, {"loss": 1.3844, "grad_norm": 0.5502381920814514, "learning_rate": 0.0002, "epoch": 1.6576185305826998, "step": 2290}, {"loss": 1.3764, "grad_norm": 0.6158602237701416, "learning_rate": 0.0002, "epoch": 1.6648570394498732, "step": 2300}, {"loss": 1.3515, "grad_norm": 0.5804675817489624, "learning_rate": 0.0002, "epoch": 1.6720955483170465, "step": 2310}, {"loss": 1.2532, "grad_norm": 0.600742757320404, "learning_rate": 0.0002, "epoch": 1.67933405718422, "step": 2320}, {"loss": 1.477, "grad_norm": 0.7101941108703613, "learning_rate": 0.0002, "epoch": 1.6865725660513933, "step": 2330}, {"loss": 1.4849, "grad_norm": 0.7507809996604919, "learning_rate": 0.0002, "epoch": 1.6938110749185666, "step": 2340}, {"loss": 1.2703, "grad_norm": 0.768502414226532, "learning_rate": 0.0002, "epoch": 1.70104958378574, "step": 2350}, {"loss": 1.3332, "grad_norm": 0.4801851212978363, "learning_rate": 0.0002, "epoch": 1.7082880926529134, "step": 2360}, {"loss": 1.4158, "grad_norm": 0.5322122573852539, "learning_rate": 0.0002, "epoch": 1.7155266015200867, "step": 2370}, {"loss": 1.4136, "grad_norm": 0.587661862373352, "learning_rate": 0.0002, "epoch": 1.72276511038726, "step": 2380}, {"loss": 1.3771, "grad_norm": 0.6073525547981262, "learning_rate": 0.0002, "epoch": 1.7300036192544335, "step": 2390}, {"loss": 1.2754, "grad_norm": 0.6950460076332092, "learning_rate": 0.0002, "epoch": 1.7372421281216068, "step": 2400}, {"loss": 1.3858, "grad_norm": 0.5981102585792542, "learning_rate": 0.0002, "epoch": 1.7444806369887802, "step": 2410}, {"loss": 1.4075, "grad_norm": 0.544570803642273, "learning_rate": 0.0002, "epoch": 1.7517191458559536, "step": 2420}, {"loss": 1.3861, "grad_norm": 0.5304399728775024, "learning_rate": 0.0002, "epoch": 1.758957654723127, "step": 2430}, {"loss": 1.4244, "grad_norm": 0.7921594977378845, "learning_rate": 0.0002, "epoch": 1.7661961635903003, "step": 2440}, {"loss": 1.3053, "grad_norm": 0.6084808707237244, "learning_rate": 0.0002, "epoch": 1.7734346724574737, "step": 2450}, {"loss": 1.3781, "grad_norm": 0.8844701051712036, "learning_rate": 0.0002, "epoch": 1.780673181324647, "step": 2460}, {"loss": 1.3227, "grad_norm": 0.5729258060455322, "learning_rate": 0.0002, "epoch": 1.7879116901918204, "step": 2470}, {"loss": 1.3422, "grad_norm": 0.6303611993789673, "learning_rate": 0.0002, "epoch": 1.7951501990589938, "step": 2480}, {"loss": 1.3926, "grad_norm": 0.5627942085266113, "learning_rate": 0.0002, "epoch": 1.8023887079261671, "step": 2490}, {"loss": 1.3816, "grad_norm": 0.6724274158477783, "learning_rate": 0.0002, "epoch": 1.8096272167933405, "step": 2500}, {"loss": 1.2951, "grad_norm": 0.5030826330184937, "learning_rate": 0.0002, "epoch": 1.8168657256605139, "step": 2510}, {"loss": 1.2839, "grad_norm": 0.5504099130630493, "learning_rate": 0.0002, "epoch": 1.8241042345276872, "step": 2520}, {"loss": 1.4264, "grad_norm": 0.6338945627212524, "learning_rate": 0.0002, "epoch": 1.8313427433948606, "step": 2530}, {"loss": 1.563, "grad_norm": 0.5902037620544434, "learning_rate": 0.0002, "epoch": 1.838581252262034, "step": 2540}, {"loss": 1.2961, "grad_norm": 0.48814457654953003, "learning_rate": 0.0002, "epoch": 1.8458197611292073, "step": 2550}, {"loss": 1.466, "grad_norm": 0.6216312646865845, "learning_rate": 0.0002, "epoch": 1.8530582699963807, "step": 2560}, {"loss": 1.5123, "grad_norm": 0.635603666305542, "learning_rate": 0.0002, "epoch": 1.860296778863554, "step": 2570}, {"loss": 1.372, "grad_norm": 0.6938216090202332, "learning_rate": 0.0002, "epoch": 1.8675352877307274, "step": 2580}, {"loss": 1.5011, "grad_norm": 0.599557638168335, "learning_rate": 0.0002, "epoch": 1.8747737965979008, "step": 2590}, {"loss": 1.2714, "grad_norm": 0.564424455165863, "learning_rate": 0.0002, "epoch": 1.8820123054650741, "step": 2600}, {"loss": 1.3403, "grad_norm": 0.5430700182914734, "learning_rate": 0.0002, "epoch": 1.8892508143322475, "step": 2610}, {"loss": 1.4347, "grad_norm": 0.6150169372558594, "learning_rate": 0.0002, "epoch": 1.8964893231994209, "step": 2620}, {"loss": 1.2474, "grad_norm": 0.48159119486808777, "learning_rate": 0.0002, "epoch": 1.9037278320665942, "step": 2630}, {"loss": 1.3716, "grad_norm": 0.5608997941017151, "learning_rate": 0.0002, "epoch": 1.9109663409337676, "step": 2640}, {"loss": 1.5787, "grad_norm": 0.6454501748085022, "learning_rate": 0.0002, "epoch": 1.918204849800941, "step": 2650}, {"loss": 1.3238, "grad_norm": 0.5458073616027832, "learning_rate": 0.0002, "epoch": 1.9254433586681143, "step": 2660}, {"loss": 1.3208, "grad_norm": 0.5328490734100342, "learning_rate": 0.0002, "epoch": 1.9326818675352877, "step": 2670}, {"loss": 1.4971, "grad_norm": 0.6444696187973022, "learning_rate": 0.0002, "epoch": 1.939920376402461, "step": 2680}, {"loss": 1.5387, "grad_norm": 0.7126023769378662, "learning_rate": 0.0002, "epoch": 1.9471588852696344, "step": 2690}, {"loss": 1.3637, "grad_norm": 0.5164045095443726, "learning_rate": 0.0002, "epoch": 1.9543973941368078, "step": 2700}, {"loss": 1.5303, "grad_norm": 0.5347061157226562, "learning_rate": 0.0002, "epoch": 1.9616359030039812, "step": 2710}, {"loss": 1.2815, "grad_norm": 0.5297950506210327, "learning_rate": 0.0002, "epoch": 1.9688744118711545, "step": 2720}, {"loss": 1.3566, "grad_norm": 0.6537790298461914, "learning_rate": 0.0002, "epoch": 1.976112920738328, "step": 2730}, {"loss": 1.332, "grad_norm": 0.5536222457885742, "learning_rate": 0.0002, "epoch": 1.9833514296055013, "step": 2740}, {"loss": 1.3333, "grad_norm": 0.4856105446815491, "learning_rate": 0.0002, "epoch": 1.9905899384726746, "step": 2750}, {"loss": 1.3521, "grad_norm": 0.6642730832099915, "learning_rate": 0.0002, "epoch": 1.997828447339848, "step": 2760}]}
+{"epoch": 2.9996380745566413, "step": 4144, "epoch_duration": 1325.2042412757874, "total_accumulated_duration": 4011.802587747574, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 3048.73388671875}, "peak_memory_usage": {"GPU_0": 15079.2998046875}, "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-0.9-num-8828-sd-1/checkpoint-2763", "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": 4.7061, "grad_norm": 1.2523442506790161, "learning_rate": 0.0002, "epoch": 0.007238508867173362, "step": 10}, {"loss": 3.3493, "grad_norm": 1.8887330293655396, "learning_rate": 0.0002, "epoch": 0.014477017734346724, "step": 20}, {"loss": 2.7585, "grad_norm": 0.9668035507202148, "learning_rate": 0.0002, "epoch": 0.021715526601520086, "step": 30}, {"loss": 2.3699, "grad_norm": 2.9167306423187256, "learning_rate": 0.0002, "epoch": 0.028954035468693448, "step": 40}, {"loss": 2.2679, "grad_norm": 2.649867296218872, "learning_rate": 0.0002, "epoch": 0.036192544335866814, "step": 50}, {"loss": 2.2202, "grad_norm": 1.5120655298233032, "learning_rate": 0.0002, "epoch": 0.04343105320304017, "step": 60}, {"loss": 2.2026, "grad_norm": 0.7879868149757385, "learning_rate": 0.0002, "epoch": 0.05066956207021354, "step": 70}, {"loss": 1.9447, "grad_norm": 0.7616953253746033, "learning_rate": 0.0002, "epoch": 0.057908070937386896, "step": 80}, {"loss": 2.0112, "grad_norm": 1.8809149265289307, "learning_rate": 0.0002, "epoch": 0.06514657980456026, "step": 90}, {"loss": 1.8337, "grad_norm": 0.9294016361236572, "learning_rate": 0.0002, "epoch": 0.07238508867173363, "step": 100}, {"loss": 1.8419, "grad_norm": 0.7145281434059143, "learning_rate": 0.0002, "epoch": 0.07962359753890698, "step": 110}, {"loss": 2.0036, "grad_norm": 0.7564446330070496, "learning_rate": 0.0002, "epoch": 0.08686210640608034, "step": 120}, {"loss": 1.9306, "grad_norm": 1.1681925058364868, "learning_rate": 0.0002, "epoch": 0.09410061527325371, "step": 130}, {"loss": 1.7875, "grad_norm": 0.6708641648292542, "learning_rate": 0.0002, "epoch": 0.10133912414042708, "step": 140}, {"loss": 1.786, "grad_norm": 0.7625647783279419, "learning_rate": 0.0002, "epoch": 0.10857763300760044, "step": 150}, {"loss": 1.6687, "grad_norm": 0.8463464975357056, "learning_rate": 0.0002, "epoch": 0.11581614187477379, "step": 160}, {"loss": 1.6214, "grad_norm": 0.7502335906028748, "learning_rate": 0.0002, "epoch": 0.12305465074194716, "step": 170}, {"loss": 1.7433, "grad_norm": 0.6929958462715149, "learning_rate": 0.0002, "epoch": 0.13029315960912052, "step": 180}, {"loss": 1.6009, "grad_norm": 0.6798707842826843, "learning_rate": 0.0002, "epoch": 0.1375316684762939, "step": 190}, {"loss": 1.6208, "grad_norm": 0.7566508650779724, "learning_rate": 0.0002, "epoch": 0.14477017734346725, "step": 200}, {"loss": 1.5823, "grad_norm": 0.7196869850158691, "learning_rate": 0.0002, "epoch": 0.15200868621064062, "step": 210}, {"loss": 1.738, "grad_norm": 0.8401045799255371, "learning_rate": 0.0002, "epoch": 0.15924719507781396, "step": 220}, {"loss": 1.7574, "grad_norm": 0.8503773212432861, "learning_rate": 0.0002, "epoch": 0.16648570394498732, "step": 230}, {"loss": 1.7861, "grad_norm": 0.7183733582496643, "learning_rate": 0.0002, "epoch": 0.1737242128121607, "step": 240}, {"loss": 1.6693, "grad_norm": 0.7082605957984924, "learning_rate": 0.0002, "epoch": 0.18096272167933405, "step": 250}, {"loss": 1.619, "grad_norm": 0.9386326670646667, "learning_rate": 0.0002, "epoch": 0.18820123054650742, "step": 260}, {"loss": 1.6511, "grad_norm": 0.7332451939582825, "learning_rate": 0.0002, "epoch": 0.19543973941368079, "step": 270}, {"loss": 1.6353, "grad_norm": 0.7092869877815247, "learning_rate": 0.0002, "epoch": 0.20267824828085415, "step": 280}, {"loss": 1.5996, "grad_norm": 0.7256413698196411, "learning_rate": 0.0002, "epoch": 0.20991675714802752, "step": 290}, {"loss": 1.6754, "grad_norm": 0.6398681402206421, "learning_rate": 0.0002, "epoch": 0.21715526601520088, "step": 300}, {"loss": 1.397, "grad_norm": 0.6273287534713745, "learning_rate": 0.0002, "epoch": 0.22439377488237422, "step": 310}, {"loss": 1.5115, "grad_norm": 0.511648416519165, "learning_rate": 0.0002, "epoch": 0.23163228374954759, "step": 320}, {"loss": 1.5424, "grad_norm": 0.8677352070808411, "learning_rate": 0.0002, "epoch": 0.23887079261672095, "step": 330}, {"loss": 1.6779, "grad_norm": 0.6270743012428284, "learning_rate": 0.0002, "epoch": 0.24610930148389432, "step": 340}, {"loss": 1.626, "grad_norm": 0.7980281114578247, "learning_rate": 0.0002, "epoch": 0.2533478103510677, "step": 350}, {"loss": 1.5238, "grad_norm": 0.632486879825592, "learning_rate": 0.0002, "epoch": 0.26058631921824105, "step": 360}, {"loss": 1.5175, "grad_norm": 0.6527034640312195, "learning_rate": 0.0002, "epoch": 0.2678248280854144, "step": 370}, {"loss": 1.627, "grad_norm": 0.7672118544578552, "learning_rate": 0.0002, "epoch": 0.2750633369525878, "step": 380}, {"loss": 1.5605, "grad_norm": 0.6035117506980896, "learning_rate": 0.0002, "epoch": 0.28230184581976114, "step": 390}, {"loss": 1.4603, "grad_norm": 0.5955103039741516, "learning_rate": 0.0002, "epoch": 0.2895403546869345, "step": 400}, {"loss": 1.558, "grad_norm": 0.6015191674232483, "learning_rate": 0.0002, "epoch": 0.2967788635541079, "step": 410}, {"loss": 1.6091, "grad_norm": 0.6380982398986816, "learning_rate": 0.0002, "epoch": 0.30401737242128124, "step": 420}, {"loss": 1.5292, "grad_norm": 0.6707863211631775, "learning_rate": 0.0002, "epoch": 0.3112558812884546, "step": 430}, {"loss": 1.4426, "grad_norm": 0.7010176777839661, "learning_rate": 0.0002, "epoch": 0.3184943901556279, "step": 440}, {"loss": 1.5572, "grad_norm": 0.8263739943504333, "learning_rate": 0.0002, "epoch": 0.3257328990228013, "step": 450}, {"loss": 1.5188, "grad_norm": 0.7253276109695435, "learning_rate": 0.0002, "epoch": 0.33297140788997465, "step": 460}, {"loss": 1.584, "grad_norm": 0.5238934755325317, "learning_rate": 0.0002, "epoch": 0.340209916757148, "step": 470}, {"loss": 1.7035, "grad_norm": 0.7869495749473572, "learning_rate": 0.0002, "epoch": 0.3474484256243214, "step": 480}, {"loss": 1.5776, "grad_norm": 0.7485215663909912, "learning_rate": 0.0002, "epoch": 0.35468693449149474, "step": 490}, {"loss": 1.6274, "grad_norm": 0.5413193106651306, "learning_rate": 0.0002, "epoch": 0.3619254433586681, "step": 500}, {"loss": 1.7323, "grad_norm": 0.7615048885345459, "learning_rate": 0.0002, "epoch": 0.3691639522258415, "step": 510}, {"loss": 1.532, "grad_norm": 0.7685340046882629, "learning_rate": 0.0002, "epoch": 0.37640246109301484, "step": 520}, {"loss": 1.6312, "grad_norm": 0.6379081010818481, "learning_rate": 0.0002, "epoch": 0.3836409699601882, "step": 530}, {"loss": 1.5645, "grad_norm": 0.7946939468383789, "learning_rate": 0.0002, "epoch": 0.39087947882736157, "step": 540}, {"loss": 1.4001, "grad_norm": 0.6287278532981873, "learning_rate": 0.0002, "epoch": 0.39811798769453494, "step": 550}, {"loss": 1.5982, "grad_norm": 0.6811642646789551, "learning_rate": 0.0002, "epoch": 0.4053564965617083, "step": 560}, {"loss": 1.4953, "grad_norm": 0.671073317527771, "learning_rate": 0.0002, "epoch": 0.41259500542888167, "step": 570}, {"loss": 1.6753, "grad_norm": 0.6313900351524353, "learning_rate": 0.0002, "epoch": 0.41983351429605503, "step": 580}, {"loss": 1.546, "grad_norm": 0.5291772484779358, "learning_rate": 0.0002, "epoch": 0.4270720231632284, "step": 590}, {"loss": 1.5441, "grad_norm": 0.62503582239151, "learning_rate": 0.0002, "epoch": 0.43431053203040176, "step": 600}, {"loss": 1.6276, "grad_norm": 0.5777305364608765, "learning_rate": 0.0002, "epoch": 0.4415490408975751, "step": 610}, {"loss": 1.4758, "grad_norm": 0.7013497352600098, "learning_rate": 0.0002, "epoch": 0.44878754976474844, "step": 620}, {"loss": 1.4029, "grad_norm": 0.8044822216033936, "learning_rate": 0.0002, "epoch": 0.4560260586319218, "step": 630}, {"loss": 1.7195, "grad_norm": 0.672531247138977, "learning_rate": 0.0002, "epoch": 0.46326456749909517, "step": 640}, {"loss": 1.614, "grad_norm": 0.6233910322189331, "learning_rate": 0.0002, "epoch": 0.47050307636626854, "step": 650}, {"loss": 1.6041, "grad_norm": 0.651524543762207, "learning_rate": 0.0002, "epoch": 0.4777415852334419, "step": 660}, {"loss": 1.5842, "grad_norm": 0.7213939428329468, "learning_rate": 0.0002, "epoch": 0.48498009410061527, "step": 670}, {"loss": 1.5453, "grad_norm": 0.6541454792022705, "learning_rate": 0.0002, "epoch": 0.49221860296778863, "step": 680}, {"loss": 1.662, "grad_norm": 0.6568936109542847, "learning_rate": 0.0002, "epoch": 0.499457111834962, "step": 690}, {"loss": 1.624, "grad_norm": 0.7176415324211121, "learning_rate": 0.0002, "epoch": 0.5066956207021354, "step": 700}, {"loss": 1.6099, "grad_norm": 0.6553855538368225, "learning_rate": 0.0002, "epoch": 0.5139341295693087, "step": 710}, {"loss": 1.5508, "grad_norm": 0.5654335618019104, "learning_rate": 0.0002, "epoch": 0.5211726384364821, "step": 720}, {"loss": 1.392, "grad_norm": 0.5671001672744751, "learning_rate": 0.0002, "epoch": 0.5284111473036555, "step": 730}, {"loss": 1.388, "grad_norm": 0.7914412021636963, "learning_rate": 0.0002, "epoch": 0.5356496561708288, "step": 740}, {"loss": 1.5931, "grad_norm": 0.6172138452529907, "learning_rate": 0.0002, "epoch": 0.5428881650380022, "step": 750}, {"loss": 1.4018, "grad_norm": 0.6132623553276062, "learning_rate": 0.0002, "epoch": 0.5501266739051756, "step": 760}, {"loss": 1.513, "grad_norm": 0.654000461101532, "learning_rate": 0.0002, "epoch": 0.5573651827723489, "step": 770}, {"loss": 1.5035, "grad_norm": 0.5691370964050293, "learning_rate": 0.0002, "epoch": 0.5646036916395223, "step": 780}, {"loss": 1.65, "grad_norm": 0.7922580242156982, "learning_rate": 0.0002, "epoch": 0.5718422005066957, "step": 790}, {"loss": 1.4521, "grad_norm": 0.6831880211830139, "learning_rate": 0.0002, "epoch": 0.579080709373869, "step": 800}, {"loss": 1.4734, "grad_norm": 0.6740124821662903, "learning_rate": 0.0002, "epoch": 0.5863192182410424, "step": 810}, {"loss": 1.6498, "grad_norm": 1.380016803741455, "learning_rate": 0.0002, "epoch": 0.5935577271082157, "step": 820}, {"loss": 1.4642, "grad_norm": 0.6552878022193909, "learning_rate": 0.0002, "epoch": 0.6007962359753891, "step": 830}, {"loss": 1.6271, "grad_norm": 0.6649535298347473, "learning_rate": 0.0002, "epoch": 0.6080347448425625, "step": 840}, {"loss": 1.5886, "grad_norm": 0.561738133430481, "learning_rate": 0.0002, "epoch": 0.6152732537097358, "step": 850}, {"loss": 1.5364, "grad_norm": 0.6133047938346863, "learning_rate": 0.0002, "epoch": 0.6225117625769092, "step": 860}, {"loss": 1.3489, "grad_norm": 0.559843122959137, "learning_rate": 0.0002, "epoch": 0.6297502714440825, "step": 870}, {"loss": 1.4878, "grad_norm": 0.6117811799049377, "learning_rate": 0.0002, "epoch": 0.6369887803112558, "step": 880}, {"loss": 1.56, "grad_norm": 0.6209776401519775, "learning_rate": 0.0002, "epoch": 0.6442272891784292, "step": 890}, {"loss": 1.6747, "grad_norm": 0.6234082579612732, "learning_rate": 0.0002, "epoch": 0.6514657980456026, "step": 900}, {"loss": 1.6963, "grad_norm": 0.7623258233070374, "learning_rate": 0.0002, "epoch": 0.6587043069127759, "step": 910}, {"loss": 1.2424, "grad_norm": 0.6148061752319336, "learning_rate": 0.0002, "epoch": 0.6659428157799493, "step": 920}, {"loss": 1.4319, "grad_norm": 0.6682973504066467, "learning_rate": 0.0002, "epoch": 0.6731813246471227, "step": 930}, {"loss": 1.5377, "grad_norm": 0.5513041615486145, "learning_rate": 0.0002, "epoch": 0.680419833514296, "step": 940}, {"loss": 1.3991, "grad_norm": 0.5197525024414062, "learning_rate": 0.0002, "epoch": 0.6876583423814694, "step": 950}, {"loss": 1.4398, "grad_norm": 0.6490758061408997, "learning_rate": 0.0002, "epoch": 0.6948968512486428, "step": 960}, {"loss": 1.5251, "grad_norm": 0.6450682878494263, "learning_rate": 0.0002, "epoch": 0.7021353601158161, "step": 970}, {"loss": 1.5417, "grad_norm": 0.6203766465187073, "learning_rate": 0.0002, "epoch": 0.7093738689829895, "step": 980}, {"loss": 1.4575, "grad_norm": 0.6023609638214111, "learning_rate": 0.0002, "epoch": 0.7166123778501629, "step": 990}, {"loss": 1.4973, "grad_norm": 0.5765255093574524, "learning_rate": 0.0002, "epoch": 0.7238508867173362, "step": 1000}, {"loss": 1.483, "grad_norm": 0.6650075316429138, "learning_rate": 0.0002, "epoch": 0.7310893955845096, "step": 1010}, {"loss": 1.5959, "grad_norm": 0.5610854029655457, "learning_rate": 0.0002, "epoch": 0.738327904451683, "step": 1020}, {"loss": 1.5248, "grad_norm": 0.7072813510894775, "learning_rate": 0.0002, "epoch": 0.7455664133188563, "step": 1030}, {"loss": 1.5776, "grad_norm": 0.6815407872200012, "learning_rate": 0.0002, "epoch": 0.7528049221860297, "step": 1040}, {"loss": 1.4577, "grad_norm": 0.7932390570640564, "learning_rate": 0.0002, "epoch": 0.760043431053203, "step": 1050}, {"loss": 1.4515, "grad_norm": 0.5798183083534241, "learning_rate": 0.0002, "epoch": 0.7672819399203764, "step": 1060}, {"loss": 1.5053, "grad_norm": 0.7898504137992859, "learning_rate": 0.0002, "epoch": 0.7745204487875498, "step": 1070}, {"loss": 1.4776, "grad_norm": 0.4983280301094055, "learning_rate": 0.0002, "epoch": 0.7817589576547231, "step": 1080}, {"loss": 1.5007, "grad_norm": 0.691403329372406, "learning_rate": 0.0002, "epoch": 0.7889974665218965, "step": 1090}, {"loss": 1.5153, "grad_norm": 0.5394481420516968, "learning_rate": 0.0002, "epoch": 0.7962359753890699, "step": 1100}, {"loss": 1.6892, "grad_norm": 0.5136822462081909, "learning_rate": 0.0002, "epoch": 0.8034744842562432, "step": 1110}, {"loss": 1.4902, "grad_norm": 0.6828126907348633, "learning_rate": 0.0002, "epoch": 0.8107129931234166, "step": 1120}, {"loss": 1.4346, "grad_norm": 0.6799656748771667, "learning_rate": 0.0002, "epoch": 0.81795150199059, "step": 1130}, {"loss": 1.2678, "grad_norm": 0.5428406000137329, "learning_rate": 0.0002, "epoch": 0.8251900108577633, "step": 1140}, {"loss": 1.4072, "grad_norm": 0.4811290502548218, "learning_rate": 0.0002, "epoch": 0.8324285197249367, "step": 1150}, {"loss": 1.4512, "grad_norm": 0.5519434809684753, "learning_rate": 0.0002, "epoch": 0.8396670285921101, "step": 1160}, {"loss": 1.4072, "grad_norm": 0.9748060703277588, "learning_rate": 0.0002, "epoch": 0.8469055374592834, "step": 1170}, {"loss": 1.4309, "grad_norm": 0.712609589099884, "learning_rate": 0.0002, "epoch": 0.8541440463264568, "step": 1180}, {"loss": 1.434, "grad_norm": 0.6866157054901123, "learning_rate": 0.0002, "epoch": 0.8613825551936302, "step": 1190}, {"loss": 1.3704, "grad_norm": 0.5068854093551636, "learning_rate": 0.0002, "epoch": 0.8686210640608035, "step": 1200}, {"loss": 1.5601, "grad_norm": 0.6333245038986206, "learning_rate": 0.0002, "epoch": 0.8758595729279768, "step": 1210}, {"loss": 1.4636, "grad_norm": 0.6424421072006226, "learning_rate": 0.0002, "epoch": 0.8830980817951501, "step": 1220}, {"loss": 1.4186, "grad_norm": 0.4771921932697296, "learning_rate": 0.0002, "epoch": 0.8903365906623235, "step": 1230}, {"loss": 1.6323, "grad_norm": 0.5191764235496521, "learning_rate": 0.0002, "epoch": 0.8975750995294969, "step": 1240}, {"loss": 1.6105, "grad_norm": 0.756222128868103, "learning_rate": 0.0002, "epoch": 0.9048136083966702, "step": 1250}, {"loss": 1.4396, "grad_norm": 0.623823881149292, "learning_rate": 0.0002, "epoch": 0.9120521172638436, "step": 1260}, {"loss": 1.3097, "grad_norm": 0.8166571259498596, "learning_rate": 0.0002, "epoch": 0.919290626131017, "step": 1270}, {"loss": 1.4625, "grad_norm": 0.6059346795082092, "learning_rate": 0.0002, "epoch": 0.9265291349981903, "step": 1280}, {"loss": 1.3555, "grad_norm": 0.5842690467834473, "learning_rate": 0.0002, "epoch": 0.9337676438653637, "step": 1290}, {"loss": 1.5859, "grad_norm": 0.7649800777435303, "learning_rate": 0.0002, "epoch": 0.9410061527325371, "step": 1300}, {"loss": 1.5915, "grad_norm": 0.6420919895172119, "learning_rate": 0.0002, "epoch": 0.9482446615997104, "step": 1310}, {"loss": 1.453, "grad_norm": 0.7011452913284302, "learning_rate": 0.0002, "epoch": 0.9554831704668838, "step": 1320}, {"loss": 1.6766, "grad_norm": 0.5783746242523193, "learning_rate": 0.0002, "epoch": 0.9627216793340572, "step": 1330}, {"loss": 1.6308, "grad_norm": 0.5973192453384399, "learning_rate": 0.0002, "epoch": 0.9699601882012305, "step": 1340}, {"loss": 1.5901, "grad_norm": 0.6181833744049072, "learning_rate": 0.0002, "epoch": 0.9771986970684039, "step": 1350}, {"loss": 1.5258, "grad_norm": 0.5563396215438843, "learning_rate": 0.0002, "epoch": 0.9844372059355773, "step": 1360}, {"loss": 1.4508, "grad_norm": 0.45723360776901245, "learning_rate": 0.0002, "epoch": 0.9916757148027506, "step": 1370}, {"loss": 1.3291, "grad_norm": 0.5947498679161072, "learning_rate": 0.0002, "epoch": 0.998914223669924, "step": 1380}, {"eval_loss": 1.480796456336975, "eval_runtime": 27.3103, "eval_samples_per_second": 15.965, "eval_steps_per_second": 2.014, "epoch": 0.9996380745566413, "step": 1381}, {"loss": 1.3057, "grad_norm": 0.5599952936172485, "learning_rate": 0.0002, "epoch": 1.0061527325370974, "step": 1390}, {"loss": 1.4991, "grad_norm": 0.5932008028030396, "learning_rate": 0.0002, "epoch": 1.0133912414042707, "step": 1400}, {"loss": 1.4506, "grad_norm": 0.6194121837615967, "learning_rate": 0.0002, "epoch": 1.020629750271444, "step": 1410}, {"loss": 1.5966, "grad_norm": 0.6995621919631958, "learning_rate": 0.0002, "epoch": 1.0278682591386175, "step": 1420}, {"loss": 1.4153, "grad_norm": 0.7905810475349426, "learning_rate": 0.0002, "epoch": 1.0351067680057908, "step": 1430}, {"loss": 1.4414, "grad_norm": 0.7221615314483643, "learning_rate": 0.0002, "epoch": 1.0423452768729642, "step": 1440}, {"loss": 1.3859, "grad_norm": 0.6170642375946045, "learning_rate": 0.0002, "epoch": 1.0495837857401376, "step": 1450}, {"loss": 1.3806, "grad_norm": 0.5844094753265381, "learning_rate": 0.0002, "epoch": 1.056822294607311, "step": 1460}, {"loss": 1.4871, "grad_norm": 0.7731822729110718, "learning_rate": 0.0002, "epoch": 1.0640608034744843, "step": 1470}, {"loss": 1.4286, "grad_norm": 0.4554748237133026, "learning_rate": 0.0002, "epoch": 1.0712993123416577, "step": 1480}, {"loss": 1.3977, "grad_norm": 0.6923259496688843, "learning_rate": 0.0002, "epoch": 1.078537821208831, "step": 1490}, {"loss": 1.3936, "grad_norm": 0.6008219122886658, "learning_rate": 0.0002, "epoch": 1.0857763300760044, "step": 1500}, {"loss": 1.4821, "grad_norm": 0.6450045704841614, "learning_rate": 0.0002, "epoch": 1.0930148389431777, "step": 1510}, {"loss": 1.3295, "grad_norm": 0.7833753824234009, "learning_rate": 0.0002, "epoch": 1.1002533478103511, "step": 1520}, {"loss": 1.3424, "grad_norm": 0.5076758861541748, "learning_rate": 0.0002, "epoch": 1.1074918566775245, "step": 1530}, {"loss": 1.4043, "grad_norm": 0.5661332011222839, "learning_rate": 0.0002, "epoch": 1.1147303655446978, "step": 1540}, {"loss": 1.4963, "grad_norm": 0.6526919603347778, "learning_rate": 0.0002, "epoch": 1.1219688744118712, "step": 1550}, {"loss": 1.3671, "grad_norm": 0.5613082647323608, "learning_rate": 0.0002, "epoch": 1.1292073832790446, "step": 1560}, {"loss": 1.4458, "grad_norm": 0.6113885641098022, "learning_rate": 0.0002, "epoch": 1.136445892146218, "step": 1570}, {"loss": 1.3552, "grad_norm": 0.6732510328292847, "learning_rate": 0.0002, "epoch": 1.1436844010133913, "step": 1580}, {"loss": 1.3114, "grad_norm": 0.6146392226219177, "learning_rate": 0.0002, "epoch": 1.1509229098805647, "step": 1590}, {"loss": 1.411, "grad_norm": 0.6766974329948425, "learning_rate": 0.0002, "epoch": 1.158161418747738, "step": 1600}, {"loss": 1.2401, "grad_norm": 0.7621957659721375, "learning_rate": 0.0002, "epoch": 1.1653999276149114, "step": 1610}, {"loss": 1.3758, "grad_norm": 0.6959581971168518, "learning_rate": 0.0002, "epoch": 1.1726384364820848, "step": 1620}, {"loss": 1.382, "grad_norm": 0.6691278219223022, "learning_rate": 0.0002, "epoch": 1.1798769453492581, "step": 1630}, {"loss": 1.4147, "grad_norm": 0.4927774965763092, "learning_rate": 0.0002, "epoch": 1.1871154542164315, "step": 1640}, {"loss": 1.449, "grad_norm": 0.7724234461784363, "learning_rate": 0.0002, "epoch": 1.1943539630836049, "step": 1650}, {"loss": 1.4778, "grad_norm": 0.6817787885665894, "learning_rate": 0.0002, "epoch": 1.2015924719507782, "step": 1660}, {"loss": 1.3776, "grad_norm": 0.6500699520111084, "learning_rate": 0.0002, "epoch": 1.2088309808179516, "step": 1670}, {"loss": 1.3875, "grad_norm": 0.5703568458557129, "learning_rate": 0.0002, "epoch": 1.216069489685125, "step": 1680}, {"loss": 1.4735, "grad_norm": 0.6261579990386963, "learning_rate": 0.0002, "epoch": 1.2233079985522983, "step": 1690}, {"loss": 1.3898, "grad_norm": 0.651713490486145, "learning_rate": 0.0002, "epoch": 1.2305465074194717, "step": 1700}, {"loss": 1.4002, "grad_norm": 0.684399425983429, "learning_rate": 0.0002, "epoch": 1.237785016286645, "step": 1710}, {"loss": 1.5027, "grad_norm": 0.6996857523918152, "learning_rate": 0.0002, "epoch": 1.2450235251538184, "step": 1720}, {"loss": 1.3326, "grad_norm": 0.7102537751197815, "learning_rate": 0.0002, "epoch": 1.2522620340209918, "step": 1730}, {"loss": 1.3675, "grad_norm": 0.45809897780418396, "learning_rate": 0.0002, "epoch": 1.2595005428881652, "step": 1740}, {"loss": 1.4175, "grad_norm": 0.6377046704292297, "learning_rate": 0.0002, "epoch": 1.2667390517553385, "step": 1750}, {"loss": 1.3479, "grad_norm": 0.6965704560279846, "learning_rate": 0.0002, "epoch": 1.2739775606225119, "step": 1760}, {"loss": 1.5647, "grad_norm": 0.5688214302062988, "learning_rate": 0.0002, "epoch": 1.2812160694896852, "step": 1770}, {"loss": 1.3967, "grad_norm": 0.6384190320968628, "learning_rate": 0.0002, "epoch": 1.2884545783568586, "step": 1780}, {"loss": 1.3671, "grad_norm": 0.5629363656044006, "learning_rate": 0.0002, "epoch": 1.295693087224032, "step": 1790}, {"loss": 1.2292, "grad_norm": 0.6148255467414856, "learning_rate": 0.0002, "epoch": 1.3029315960912053, "step": 1800}, {"loss": 1.5806, "grad_norm": 0.655580997467041, "learning_rate": 0.0002, "epoch": 1.3101701049583787, "step": 1810}, {"loss": 1.2398, "grad_norm": 0.5642657279968262, "learning_rate": 0.0002, "epoch": 1.3174086138255519, "step": 1820}, {"loss": 1.3246, "grad_norm": 0.59607994556427, "learning_rate": 0.0002, "epoch": 1.3246471226927252, "step": 1830}, {"loss": 1.3274, "grad_norm": 0.5564199090003967, "learning_rate": 0.0002, "epoch": 1.3318856315598986, "step": 1840}, {"loss": 1.5834, "grad_norm": 0.6949955821037292, "learning_rate": 0.0002, "epoch": 1.339124140427072, "step": 1850}, {"loss": 1.4722, "grad_norm": 0.7036856412887573, "learning_rate": 0.0002, "epoch": 1.3463626492942453, "step": 1860}, {"loss": 1.333, "grad_norm": 0.722062885761261, "learning_rate": 0.0002, "epoch": 1.3536011581614187, "step": 1870}, {"loss": 1.4044, "grad_norm": 0.6098677515983582, "learning_rate": 0.0002, "epoch": 1.360839667028592, "step": 1880}, {"loss": 1.6217, "grad_norm": 0.5376402735710144, "learning_rate": 0.0002, "epoch": 1.3680781758957654, "step": 1890}, {"loss": 1.5071, "grad_norm": 0.6974610090255737, "learning_rate": 0.0002, "epoch": 1.3753166847629388, "step": 1900}, {"loss": 1.5854, "grad_norm": 0.6520763635635376, "learning_rate": 0.0002, "epoch": 1.3825551936301121, "step": 1910}, {"loss": 1.4271, "grad_norm": 0.6604374647140503, "learning_rate": 0.0002, "epoch": 1.3897937024972855, "step": 1920}, {"loss": 1.419, "grad_norm": 0.7364398241043091, "learning_rate": 0.0002, "epoch": 1.3970322113644589, "step": 1930}, {"loss": 1.4585, "grad_norm": 0.6849475502967834, "learning_rate": 0.0002, "epoch": 1.4042707202316322, "step": 1940}, {"loss": 1.5577, "grad_norm": 0.6562670469284058, "learning_rate": 0.0002, "epoch": 1.4115092290988056, "step": 1950}, {"loss": 1.4725, "grad_norm": 0.5695616006851196, "learning_rate": 0.0002, "epoch": 1.418747737965979, "step": 1960}, {"loss": 1.3088, "grad_norm": 0.5244464874267578, "learning_rate": 0.0002, "epoch": 1.4259862468331523, "step": 1970}, {"loss": 1.5069, "grad_norm": 0.6347293257713318, "learning_rate": 0.0002, "epoch": 1.4332247557003257, "step": 1980}, {"loss": 1.3502, "grad_norm": 0.5528361201286316, "learning_rate": 0.0002, "epoch": 1.440463264567499, "step": 1990}, {"loss": 1.3978, "grad_norm": 0.6987585425376892, "learning_rate": 0.0002, "epoch": 1.4477017734346724, "step": 2000}, {"loss": 1.4262, "grad_norm": 0.6568987369537354, "learning_rate": 0.0002, "epoch": 1.4549402823018458, "step": 2010}, {"loss": 1.4175, "grad_norm": 0.7665994763374329, "learning_rate": 0.0002, "epoch": 1.4621787911690192, "step": 2020}, {"loss": 1.244, "grad_norm": 0.5127707123756409, "learning_rate": 0.0002, "epoch": 1.4694173000361925, "step": 2030}, {"loss": 1.3699, "grad_norm": 0.5406824946403503, "learning_rate": 0.0002, "epoch": 1.476655808903366, "step": 2040}, {"loss": 1.3353, "grad_norm": 0.5990166664123535, "learning_rate": 0.0002, "epoch": 1.4838943177705393, "step": 2050}, {"loss": 1.2454, "grad_norm": 0.6186193823814392, "learning_rate": 0.0002, "epoch": 1.4911328266377126, "step": 2060}, {"loss": 1.428, "grad_norm": 0.6154307126998901, "learning_rate": 0.0002, "epoch": 1.498371335504886, "step": 2070}, {"loss": 1.4528, "grad_norm": 0.5606056451797485, "learning_rate": 0.0002, "epoch": 1.5056098443720594, "step": 2080}, {"loss": 1.2405, "grad_norm": 0.5006417036056519, "learning_rate": 0.0002, "epoch": 1.5128483532392327, "step": 2090}, {"loss": 1.4258, "grad_norm": 0.5968486070632935, "learning_rate": 0.0002, "epoch": 1.520086862106406, "step": 2100}, {"loss": 1.2752, "grad_norm": 0.5835496187210083, "learning_rate": 0.0002, "epoch": 1.5273253709735795, "step": 2110}, {"loss": 1.5443, "grad_norm": 0.6753535270690918, "learning_rate": 0.0002, "epoch": 1.5345638798407528, "step": 2120}, {"loss": 1.2139, "grad_norm": 0.7299720644950867, "learning_rate": 0.0002, "epoch": 1.5418023887079262, "step": 2130}, {"loss": 1.2364, "grad_norm": 0.5105988383293152, "learning_rate": 0.0002, "epoch": 1.5490408975750996, "step": 2140}, {"loss": 1.4528, "grad_norm": 0.5675431489944458, "learning_rate": 0.0002, "epoch": 1.556279406442273, "step": 2150}, {"loss": 1.4563, "grad_norm": 0.6246723532676697, "learning_rate": 0.0002, "epoch": 1.5635179153094463, "step": 2160}, {"loss": 1.5255, "grad_norm": 0.7291720509529114, "learning_rate": 0.0002, "epoch": 1.5707564241766196, "step": 2170}, {"loss": 1.5432, "grad_norm": 0.678114116191864, "learning_rate": 0.0002, "epoch": 1.577994933043793, "step": 2180}, {"loss": 1.5212, "grad_norm": 0.5136260986328125, "learning_rate": 0.0002, "epoch": 1.5852334419109664, "step": 2190}, {"loss": 1.3271, "grad_norm": 0.6359935998916626, "learning_rate": 0.0002, "epoch": 1.5924719507781397, "step": 2200}, {"loss": 1.4038, "grad_norm": 0.7650278806686401, "learning_rate": 0.0002, "epoch": 1.599710459645313, "step": 2210}, {"loss": 1.5478, "grad_norm": 0.7256110906600952, "learning_rate": 0.0002, "epoch": 1.6069489685124865, "step": 2220}, {"loss": 1.4387, "grad_norm": 0.688689649105072, "learning_rate": 0.0002, "epoch": 1.6141874773796598, "step": 2230}, {"loss": 1.4096, "grad_norm": 0.6045311093330383, "learning_rate": 0.0002, "epoch": 1.6214259862468332, "step": 2240}, {"loss": 1.4097, "grad_norm": 0.7064604163169861, "learning_rate": 0.0002, "epoch": 1.6286644951140063, "step": 2250}, {"loss": 1.3477, "grad_norm": 0.5309562087059021, "learning_rate": 0.0002, "epoch": 1.6359030039811797, "step": 2260}, {"loss": 1.4022, "grad_norm": 0.5687053203582764, "learning_rate": 0.0002, "epoch": 1.643141512848353, "step": 2270}, {"loss": 1.2977, "grad_norm": 0.535872757434845, "learning_rate": 0.0002, "epoch": 1.6503800217155264, "step": 2280}, {"loss": 1.3844, "grad_norm": 0.5502381920814514, "learning_rate": 0.0002, "epoch": 1.6576185305826998, "step": 2290}, {"loss": 1.3764, "grad_norm": 0.6158602237701416, "learning_rate": 0.0002, "epoch": 1.6648570394498732, "step": 2300}, {"loss": 1.3515, "grad_norm": 0.5804675817489624, "learning_rate": 0.0002, "epoch": 1.6720955483170465, "step": 2310}, {"loss": 1.2532, "grad_norm": 0.600742757320404, "learning_rate": 0.0002, "epoch": 1.67933405718422, "step": 2320}, {"loss": 1.477, "grad_norm": 0.7101941108703613, "learning_rate": 0.0002, "epoch": 1.6865725660513933, "step": 2330}, {"loss": 1.4849, "grad_norm": 0.7507809996604919, "learning_rate": 0.0002, "epoch": 1.6938110749185666, "step": 2340}, {"loss": 1.2703, "grad_norm": 0.768502414226532, "learning_rate": 0.0002, "epoch": 1.70104958378574, "step": 2350}, {"loss": 1.3332, "grad_norm": 0.4801851212978363, "learning_rate": 0.0002, "epoch": 1.7082880926529134, "step": 2360}, {"loss": 1.4158, "grad_norm": 0.5322122573852539, "learning_rate": 0.0002, "epoch": 1.7155266015200867, "step": 2370}, {"loss": 1.4136, "grad_norm": 0.587661862373352, "learning_rate": 0.0002, "epoch": 1.72276511038726, "step": 2380}, {"loss": 1.3771, "grad_norm": 0.6073525547981262, "learning_rate": 0.0002, "epoch": 1.7300036192544335, "step": 2390}, {"loss": 1.2754, "grad_norm": 0.6950460076332092, "learning_rate": 0.0002, "epoch": 1.7372421281216068, "step": 2400}, {"loss": 1.3858, "grad_norm": 0.5981102585792542, "learning_rate": 0.0002, "epoch": 1.7444806369887802, "step": 2410}, {"loss": 1.4075, "grad_norm": 0.544570803642273, "learning_rate": 0.0002, "epoch": 1.7517191458559536, "step": 2420}, {"loss": 1.3861, "grad_norm": 0.5304399728775024, "learning_rate": 0.0002, "epoch": 1.758957654723127, "step": 2430}, {"loss": 1.4244, "grad_norm": 0.7921594977378845, "learning_rate": 0.0002, "epoch": 1.7661961635903003, "step": 2440}, {"loss": 1.3053, "grad_norm": 0.6084808707237244, "learning_rate": 0.0002, "epoch": 1.7734346724574737, "step": 2450}, {"loss": 1.3781, "grad_norm": 0.8844701051712036, "learning_rate": 0.0002, "epoch": 1.780673181324647, "step": 2460}, {"loss": 1.3227, "grad_norm": 0.5729258060455322, "learning_rate": 0.0002, "epoch": 1.7879116901918204, "step": 2470}, {"loss": 1.3422, "grad_norm": 0.6303611993789673, "learning_rate": 0.0002, "epoch": 1.7951501990589938, "step": 2480}, {"loss": 1.3926, "grad_norm": 0.5627942085266113, "learning_rate": 0.0002, "epoch": 1.8023887079261671, "step": 2490}, {"loss": 1.3816, "grad_norm": 0.6724274158477783, "learning_rate": 0.0002, "epoch": 1.8096272167933405, "step": 2500}, {"loss": 1.2951, "grad_norm": 0.5030826330184937, "learning_rate": 0.0002, "epoch": 1.8168657256605139, "step": 2510}, {"loss": 1.2839, "grad_norm": 0.5504099130630493, "learning_rate": 0.0002, "epoch": 1.8241042345276872, "step": 2520}, {"loss": 1.4264, "grad_norm": 0.6338945627212524, "learning_rate": 0.0002, "epoch": 1.8313427433948606, "step": 2530}, {"loss": 1.563, "grad_norm": 0.5902037620544434, "learning_rate": 0.0002, "epoch": 1.838581252262034, "step": 2540}, {"loss": 1.2961, "grad_norm": 0.48814457654953003, "learning_rate": 0.0002, "epoch": 1.8458197611292073, "step": 2550}, {"loss": 1.466, "grad_norm": 0.6216312646865845, "learning_rate": 0.0002, "epoch": 1.8530582699963807, "step": 2560}, {"loss": 1.5123, "grad_norm": 0.635603666305542, "learning_rate": 0.0002, "epoch": 1.860296778863554, "step": 2570}, {"loss": 1.372, "grad_norm": 0.6938216090202332, "learning_rate": 0.0002, "epoch": 1.8675352877307274, "step": 2580}, {"loss": 1.5011, "grad_norm": 0.599557638168335, "learning_rate": 0.0002, "epoch": 1.8747737965979008, "step": 2590}, {"loss": 1.2714, "grad_norm": 0.564424455165863, "learning_rate": 0.0002, "epoch": 1.8820123054650741, "step": 2600}, {"loss": 1.3403, "grad_norm": 0.5430700182914734, "learning_rate": 0.0002, "epoch": 1.8892508143322475, "step": 2610}, {"loss": 1.4347, "grad_norm": 0.6150169372558594, "learning_rate": 0.0002, "epoch": 1.8964893231994209, "step": 2620}, {"loss": 1.2474, "grad_norm": 0.48159119486808777, "learning_rate": 0.0002, "epoch": 1.9037278320665942, "step": 2630}, {"loss": 1.3716, "grad_norm": 0.5608997941017151, "learning_rate": 0.0002, "epoch": 1.9109663409337676, "step": 2640}, {"loss": 1.5787, "grad_norm": 0.6454501748085022, "learning_rate": 0.0002, "epoch": 1.918204849800941, "step": 2650}, {"loss": 1.3238, "grad_norm": 0.5458073616027832, "learning_rate": 0.0002, "epoch": 1.9254433586681143, "step": 2660}, {"loss": 1.3208, "grad_norm": 0.5328490734100342, "learning_rate": 0.0002, "epoch": 1.9326818675352877, "step": 2670}, {"loss": 1.4971, "grad_norm": 0.6444696187973022, "learning_rate": 0.0002, "epoch": 1.939920376402461, "step": 2680}, {"loss": 1.5387, "grad_norm": 0.7126023769378662, "learning_rate": 0.0002, "epoch": 1.9471588852696344, "step": 2690}, {"loss": 1.3637, "grad_norm": 0.5164045095443726, "learning_rate": 0.0002, "epoch": 1.9543973941368078, "step": 2700}, {"loss": 1.5303, "grad_norm": 0.5347061157226562, "learning_rate": 0.0002, "epoch": 1.9616359030039812, "step": 2710}, {"loss": 1.2815, "grad_norm": 0.5297950506210327, "learning_rate": 0.0002, "epoch": 1.9688744118711545, "step": 2720}, {"loss": 1.3566, "grad_norm": 0.6537790298461914, "learning_rate": 0.0002, "epoch": 1.976112920738328, "step": 2730}, {"loss": 1.332, "grad_norm": 0.5536222457885742, "learning_rate": 0.0002, "epoch": 1.9833514296055013, "step": 2740}, {"loss": 1.3333, "grad_norm": 0.4856105446815491, "learning_rate": 0.0002, "epoch": 1.9905899384726746, "step": 2750}, {"loss": 1.3521, "grad_norm": 0.6642730832099915, "learning_rate": 0.0002, "epoch": 1.997828447339848, "step": 2760}, {"eval_loss": 1.4366681575775146, "eval_runtime": 27.3729, "eval_samples_per_second": 15.928, "eval_steps_per_second": 2.009, "epoch": 2.0, "step": 2763}, {"loss": 1.4322, "grad_norm": 0.740253210067749, "learning_rate": 0.0002, "epoch": 2.0050669562070214, "step": 2770}, {"loss": 1.277, "grad_norm": 0.5826276540756226, "learning_rate": 0.0002, "epoch": 2.0123054650741947, "step": 2780}, {"loss": 1.2424, "grad_norm": 0.607356071472168, "learning_rate": 0.0002, "epoch": 2.019543973941368, "step": 2790}, {"loss": 1.2601, "grad_norm": 0.5918063521385193, "learning_rate": 0.0002, "epoch": 2.0267824828085415, "step": 2800}, {"loss": 1.3715, "grad_norm": 0.5610089898109436, "learning_rate": 0.0002, "epoch": 2.034020991675715, "step": 2810}, {"loss": 1.2092, "grad_norm": 0.5869926810264587, "learning_rate": 0.0002, "epoch": 2.041259500542888, "step": 2820}, {"loss": 1.1929, "grad_norm": 0.5753467679023743, "learning_rate": 0.0002, "epoch": 2.0484980094100615, "step": 2830}, {"loss": 1.333, "grad_norm": 0.7096508145332336, "learning_rate": 0.0002, "epoch": 2.055736518277235, "step": 2840}, {"loss": 1.1766, "grad_norm": 0.7653635144233704, "learning_rate": 0.0002, "epoch": 2.0629750271444083, "step": 2850}, {"loss": 1.2331, "grad_norm": 0.6202841997146606, "learning_rate": 0.0002, "epoch": 2.0702135360115816, "step": 2860}, {"loss": 1.3298, "grad_norm": 0.6810227632522583, "learning_rate": 0.0002, "epoch": 2.077452044878755, "step": 2870}, {"loss": 1.2505, "grad_norm": 0.7481493353843689, "learning_rate": 0.0002, "epoch": 2.0846905537459284, "step": 2880}, {"loss": 1.2484, "grad_norm": 0.7089637517929077, "learning_rate": 0.0002, "epoch": 2.0919290626131017, "step": 2890}, {"loss": 1.3095, "grad_norm": 0.7472923398017883, "learning_rate": 0.0002, "epoch": 2.099167571480275, "step": 2900}, {"loss": 1.304, "grad_norm": 0.8135465979576111, "learning_rate": 0.0002, "epoch": 2.1064060803474485, "step": 2910}, {"loss": 1.273, "grad_norm": 0.6097133159637451, "learning_rate": 0.0002, "epoch": 2.113644589214622, "step": 2920}, {"loss": 1.3384, "grad_norm": 0.5970117449760437, "learning_rate": 0.0002, "epoch": 2.120883098081795, "step": 2930}, {"loss": 1.3233, "grad_norm": 0.6169309616088867, "learning_rate": 0.0002, "epoch": 2.1281216069489686, "step": 2940}, {"loss": 1.4246, "grad_norm": 0.9428738355636597, "learning_rate": 0.0002, "epoch": 2.135360115816142, "step": 2950}, {"loss": 1.3527, "grad_norm": 0.5671679973602295, "learning_rate": 0.0002, "epoch": 2.1425986246833153, "step": 2960}, {"loss": 1.1375, "grad_norm": 0.7007262110710144, "learning_rate": 0.0002, "epoch": 2.1498371335504887, "step": 2970}, {"loss": 1.2015, "grad_norm": 0.6294044256210327, "learning_rate": 0.0002, "epoch": 2.157075642417662, "step": 2980}, {"loss": 1.2167, "grad_norm": 0.6105241775512695, "learning_rate": 0.0002, "epoch": 2.1643141512848354, "step": 2990}, {"loss": 1.2065, "grad_norm": 0.557124137878418, "learning_rate": 0.0002, "epoch": 2.1715526601520088, "step": 3000}, {"loss": 1.2515, "grad_norm": 0.6250392198562622, "learning_rate": 0.0002, "epoch": 2.178791169019182, "step": 3010}, {"loss": 1.385, "grad_norm": 0.645218551158905, "learning_rate": 0.0002, "epoch": 2.1860296778863555, "step": 3020}, {"loss": 1.3928, "grad_norm": 0.9033605456352234, "learning_rate": 0.0002, "epoch": 2.193268186753529, "step": 3030}, {"loss": 1.2458, "grad_norm": 0.5325747132301331, "learning_rate": 0.0002, "epoch": 2.2005066956207022, "step": 3040}, {"loss": 1.261, "grad_norm": 0.6334700584411621, "learning_rate": 0.0002, "epoch": 2.2077452044878756, "step": 3050}, {"loss": 1.2385, "grad_norm": 0.5206325054168701, "learning_rate": 0.0002, "epoch": 2.214983713355049, "step": 3060}, {"loss": 1.3103, "grad_norm": 0.5987200140953064, "learning_rate": 0.0002, "epoch": 2.2222222222222223, "step": 3070}, {"loss": 1.1756, "grad_norm": 0.5893264412879944, "learning_rate": 0.0002, "epoch": 2.2294607310893957, "step": 3080}, {"loss": 1.235, "grad_norm": 0.6869237422943115, "learning_rate": 0.0002, "epoch": 2.236699239956569, "step": 3090}, {"loss": 1.3285, "grad_norm": 0.5040048360824585, "learning_rate": 0.0002, "epoch": 2.2439377488237424, "step": 3100}, {"loss": 1.3316, "grad_norm": 0.6660613417625427, "learning_rate": 0.0002, "epoch": 2.251176257690916, "step": 3110}, {"loss": 1.3108, "grad_norm": 0.5890918970108032, "learning_rate": 0.0002, "epoch": 2.258414766558089, "step": 3120}, {"loss": 1.248, "grad_norm": 0.6458896994590759, "learning_rate": 0.0002, "epoch": 2.2656532754252625, "step": 3130}, {"loss": 1.4151, "grad_norm": 0.6832690834999084, "learning_rate": 0.0002, "epoch": 2.272891784292436, "step": 3140}, {"loss": 1.4458, "grad_norm": 0.833908200263977, "learning_rate": 0.0002, "epoch": 2.2801302931596092, "step": 3150}, {"loss": 1.2931, "grad_norm": 0.4596034586429596, "learning_rate": 0.0002, "epoch": 2.2873688020267826, "step": 3160}, {"loss": 1.449, "grad_norm": 0.9130966067314148, "learning_rate": 0.0002, "epoch": 2.294607310893956, "step": 3170}, {"loss": 1.3806, "grad_norm": 0.7143292427062988, "learning_rate": 0.0002, "epoch": 2.3018458197611293, "step": 3180}, {"loss": 1.2692, "grad_norm": 0.5388900637626648, "learning_rate": 0.0002, "epoch": 2.3090843286283027, "step": 3190}, {"loss": 1.2402, "grad_norm": 0.5607513189315796, "learning_rate": 0.0002, "epoch": 2.316322837495476, "step": 3200}, {"loss": 1.3874, "grad_norm": 0.6795142292976379, "learning_rate": 0.0002, "epoch": 2.3235613463626494, "step": 3210}, {"loss": 1.3042, "grad_norm": 0.6561070680618286, "learning_rate": 0.0002, "epoch": 2.330799855229823, "step": 3220}, {"loss": 1.4636, "grad_norm": 0.8858118057250977, "learning_rate": 0.0002, "epoch": 2.338038364096996, "step": 3230}, {"loss": 1.3214, "grad_norm": 0.6604151725769043, "learning_rate": 0.0002, "epoch": 2.3452768729641695, "step": 3240}, {"loss": 1.4004, "grad_norm": 0.6755785346031189, "learning_rate": 0.0002, "epoch": 2.352515381831343, "step": 3250}, {"loss": 1.2503, "grad_norm": 0.6981677412986755, "learning_rate": 0.0002, "epoch": 2.3597538906985163, "step": 3260}, {"loss": 1.3078, "grad_norm": 0.6338568329811096, "learning_rate": 0.0002, "epoch": 2.3669923995656896, "step": 3270}, {"loss": 1.285, "grad_norm": 0.5754265785217285, "learning_rate": 0.0002, "epoch": 2.374230908432863, "step": 3280}, {"loss": 1.2924, "grad_norm": 0.7533153295516968, "learning_rate": 0.0002, "epoch": 2.3814694173000364, "step": 3290}, {"loss": 1.3711, "grad_norm": 0.675065279006958, "learning_rate": 0.0002, "epoch": 2.3887079261672097, "step": 3300}, {"loss": 1.3548, "grad_norm": 0.5686452984809875, "learning_rate": 0.0002, "epoch": 2.395946435034383, "step": 3310}, {"loss": 1.1998, "grad_norm": 0.8129481673240662, "learning_rate": 0.0002, "epoch": 2.4031849439015565, "step": 3320}, {"loss": 1.2584, "grad_norm": 0.6615934371948242, "learning_rate": 0.0002, "epoch": 2.41042345276873, "step": 3330}, {"loss": 1.3691, "grad_norm": 0.6678834557533264, "learning_rate": 0.0002, "epoch": 2.417661961635903, "step": 3340}, {"loss": 1.2381, "grad_norm": 0.5581308007240295, "learning_rate": 0.0002, "epoch": 2.4249004705030766, "step": 3350}, {"loss": 1.3853, "grad_norm": 0.6098920106887817, "learning_rate": 0.0002, "epoch": 2.43213897937025, "step": 3360}, {"loss": 1.3692, "grad_norm": 0.8101736903190613, "learning_rate": 0.0002, "epoch": 2.4393774882374233, "step": 3370}, {"loss": 1.4418, "grad_norm": 0.6621488928794861, "learning_rate": 0.0002, "epoch": 2.4466159971045967, "step": 3380}, {"loss": 1.4579, "grad_norm": 0.8693289160728455, "learning_rate": 0.0002, "epoch": 2.45385450597177, "step": 3390}, {"loss": 1.3644, "grad_norm": 0.6724580526351929, "learning_rate": 0.0002, "epoch": 2.4610930148389434, "step": 3400}, {"loss": 1.2006, "grad_norm": 0.6776891946792603, "learning_rate": 0.0002, "epoch": 2.4683315237061167, "step": 3410}, {"loss": 1.2937, "grad_norm": 0.7214453816413879, "learning_rate": 0.0002, "epoch": 2.47557003257329, "step": 3420}, {"loss": 1.4051, "grad_norm": 0.8390451073646545, "learning_rate": 0.0002, "epoch": 2.4828085414404635, "step": 3430}, {"loss": 1.25, "grad_norm": 0.7130982279777527, "learning_rate": 0.0002, "epoch": 2.490047050307637, "step": 3440}, {"loss": 1.2231, "grad_norm": 0.8873937129974365, "learning_rate": 0.0002, "epoch": 2.49728555917481, "step": 3450}, {"loss": 1.1429, "grad_norm": 0.725185751914978, "learning_rate": 0.0002, "epoch": 2.5045240680419836, "step": 3460}, {"loss": 1.2699, "grad_norm": 0.6120352149009705, "learning_rate": 0.0002, "epoch": 2.511762576909157, "step": 3470}, {"loss": 1.2552, "grad_norm": 0.7713613510131836, "learning_rate": 0.0002, "epoch": 2.5190010857763303, "step": 3480}, {"loss": 1.4648, "grad_norm": 0.895309567451477, "learning_rate": 0.0002, "epoch": 2.5262395946435037, "step": 3490}, {"loss": 1.3043, "grad_norm": 0.9631021022796631, "learning_rate": 0.0002, "epoch": 2.533478103510677, "step": 3500}, {"loss": 1.3492, "grad_norm": 0.7475683093070984, "learning_rate": 0.0002, "epoch": 2.5407166123778504, "step": 3510}, {"loss": 1.3637, "grad_norm": 0.7271341681480408, "learning_rate": 0.0002, "epoch": 2.5479551212450238, "step": 3520}, {"loss": 1.304, "grad_norm": 0.6979510188102722, "learning_rate": 0.0002, "epoch": 2.555193630112197, "step": 3530}, {"loss": 1.2353, "grad_norm": 0.6504196524620056, "learning_rate": 0.0002, "epoch": 2.5624321389793705, "step": 3540}, {"loss": 1.2699, "grad_norm": 0.7226675748825073, "learning_rate": 0.0002, "epoch": 2.569670647846544, "step": 3550}, {"loss": 1.3002, "grad_norm": 0.6143222451210022, "learning_rate": 0.0002, "epoch": 2.5769091567137172, "step": 3560}, {"loss": 1.1585, "grad_norm": 0.7245154976844788, "learning_rate": 0.0002, "epoch": 2.5841476655808906, "step": 3570}, {"loss": 1.3651, "grad_norm": 0.943540632724762, "learning_rate": 0.0002, "epoch": 2.591386174448064, "step": 3580}, {"loss": 1.3034, "grad_norm": 0.7707241773605347, "learning_rate": 0.0002, "epoch": 2.5986246833152373, "step": 3590}, {"loss": 1.3063, "grad_norm": 0.6705001592636108, "learning_rate": 0.0002, "epoch": 2.6058631921824107, "step": 3600}, {"loss": 1.2437, "grad_norm": 0.6360933780670166, "learning_rate": 0.0002, "epoch": 2.613101701049584, "step": 3610}, {"loss": 1.1844, "grad_norm": 0.5846424698829651, "learning_rate": 0.0002, "epoch": 2.6203402099167574, "step": 3620}, {"loss": 1.3674, "grad_norm": 0.5958625674247742, "learning_rate": 0.0002, "epoch": 2.6275787187839303, "step": 3630}, {"loss": 1.3599, "grad_norm": 0.6819243431091309, "learning_rate": 0.0002, "epoch": 2.6348172276511037, "step": 3640}, {"loss": 1.3884, "grad_norm": 0.7033445835113525, "learning_rate": 0.0002, "epoch": 2.642055736518277, "step": 3650}, {"loss": 1.3392, "grad_norm": 0.6134849786758423, "learning_rate": 0.0002, "epoch": 2.6492942453854504, "step": 3660}, {"loss": 1.2661, "grad_norm": 0.658009946346283, "learning_rate": 0.0002, "epoch": 2.656532754252624, "step": 3670}, {"loss": 1.3987, "grad_norm": 0.6280999779701233, "learning_rate": 0.0002, "epoch": 2.663771263119797, "step": 3680}, {"loss": 1.2995, "grad_norm": 0.5536085963249207, "learning_rate": 0.0002, "epoch": 2.6710097719869705, "step": 3690}, {"loss": 1.2044, "grad_norm": 0.8603981733322144, "learning_rate": 0.0002, "epoch": 2.678248280854144, "step": 3700}, {"loss": 1.3879, "grad_norm": 0.5509994626045227, "learning_rate": 0.0002, "epoch": 2.6854867897213173, "step": 3710}, {"loss": 1.3253, "grad_norm": 0.9093621969223022, "learning_rate": 0.0002, "epoch": 2.6927252985884906, "step": 3720}, {"loss": 1.2668, "grad_norm": 0.7525952458381653, "learning_rate": 0.0002, "epoch": 2.699963807455664, "step": 3730}, {"loss": 1.248, "grad_norm": 0.6737023591995239, "learning_rate": 0.0002, "epoch": 2.7072023163228374, "step": 3740}, {"loss": 1.2981, "grad_norm": 0.8656924962997437, "learning_rate": 0.0002, "epoch": 2.7144408251900107, "step": 3750}, {"loss": 1.2342, "grad_norm": 0.7494133114814758, "learning_rate": 0.0002, "epoch": 2.721679334057184, "step": 3760}, {"loss": 1.2417, "grad_norm": 0.5725520849227905, "learning_rate": 0.0002, "epoch": 2.7289178429243575, "step": 3770}, {"loss": 1.28, "grad_norm": 0.836412787437439, "learning_rate": 0.0002, "epoch": 2.736156351791531, "step": 3780}, {"loss": 1.3784, "grad_norm": 0.6893242597579956, "learning_rate": 0.0002, "epoch": 2.743394860658704, "step": 3790}, {"loss": 1.2929, "grad_norm": 0.6696223020553589, "learning_rate": 0.0002, "epoch": 2.7506333695258776, "step": 3800}, {"loss": 1.2449, "grad_norm": 0.6483015418052673, "learning_rate": 0.0002, "epoch": 2.757871878393051, "step": 3810}, {"loss": 1.3282, "grad_norm": 0.8084456920623779, "learning_rate": 0.0002, "epoch": 2.7651103872602243, "step": 3820}, {"loss": 1.3694, "grad_norm": 0.6601949334144592, "learning_rate": 0.0002, "epoch": 2.7723488961273977, "step": 3830}, {"loss": 1.3568, "grad_norm": 0.6905533671379089, "learning_rate": 0.0002, "epoch": 2.779587404994571, "step": 3840}, {"loss": 1.3854, "grad_norm": 0.619318425655365, "learning_rate": 0.0002, "epoch": 2.7868259138617444, "step": 3850}, {"loss": 1.2551, "grad_norm": 0.5994023084640503, "learning_rate": 0.0002, "epoch": 2.7940644227289178, "step": 3860}, {"loss": 1.2022, "grad_norm": 0.5627168416976929, "learning_rate": 0.0002, "epoch": 2.801302931596091, "step": 3870}, {"loss": 1.3921, "grad_norm": 0.6001605987548828, "learning_rate": 0.0002, "epoch": 2.8085414404632645, "step": 3880}, {"loss": 1.3026, "grad_norm": 0.6022412776947021, "learning_rate": 0.0002, "epoch": 2.815779949330438, "step": 3890}, {"loss": 1.2765, "grad_norm": 0.6832426190376282, "learning_rate": 0.0002, "epoch": 2.823018458197611, "step": 3900}, {"loss": 1.1363, "grad_norm": 0.5936811566352844, "learning_rate": 0.0002, "epoch": 2.8302569670647846, "step": 3910}, {"loss": 1.1707, "grad_norm": 0.6960572600364685, "learning_rate": 0.0002, "epoch": 2.837495475931958, "step": 3920}, {"loss": 1.4063, "grad_norm": 0.5913406610488892, "learning_rate": 0.0002, "epoch": 2.8447339847991313, "step": 3930}, {"loss": 1.3245, "grad_norm": 0.678154706954956, "learning_rate": 0.0002, "epoch": 2.8519724936663047, "step": 3940}, {"loss": 1.366, "grad_norm": 0.7898936867713928, "learning_rate": 0.0002, "epoch": 2.859211002533478, "step": 3950}, {"loss": 1.3948, "grad_norm": 0.9234195351600647, "learning_rate": 0.0002, "epoch": 2.8664495114006514, "step": 3960}, {"loss": 1.2773, "grad_norm": 0.5960825085639954, "learning_rate": 0.0002, "epoch": 2.8736880202678248, "step": 3970}, {"loss": 1.3127, "grad_norm": 0.677118182182312, "learning_rate": 0.0002, "epoch": 2.880926529134998, "step": 3980}, {"loss": 1.2652, "grad_norm": 0.6505142450332642, "learning_rate": 0.0002, "epoch": 2.8881650380021715, "step": 3990}, {"loss": 1.2078, "grad_norm": 0.550826907157898, "learning_rate": 0.0002, "epoch": 2.895403546869345, "step": 4000}, {"loss": 1.1811, "grad_norm": 0.6209215521812439, "learning_rate": 0.0002, "epoch": 2.9026420557365182, "step": 4010}, {"loss": 1.4001, "grad_norm": 0.6549018025398254, "learning_rate": 0.0002, "epoch": 2.9098805646036916, "step": 4020}, {"loss": 1.2285, "grad_norm": 0.570682168006897, "learning_rate": 0.0002, "epoch": 2.917119073470865, "step": 4030}, {"loss": 1.0832, "grad_norm": 1.1807632446289062, "learning_rate": 0.0002, "epoch": 2.9243575823380383, "step": 4040}, {"loss": 1.2693, "grad_norm": 0.7058857679367065, "learning_rate": 0.0002, "epoch": 2.9315960912052117, "step": 4050}, {"loss": 1.2905, "grad_norm": 0.5542812943458557, "learning_rate": 0.0002, "epoch": 2.938834600072385, "step": 4060}, {"loss": 1.33, "grad_norm": 0.63167804479599, "learning_rate": 0.0002, "epoch": 2.9460731089395584, "step": 4070}, {"loss": 1.3075, "grad_norm": 0.5702962279319763, "learning_rate": 0.0002, "epoch": 2.953311617806732, "step": 4080}, {"loss": 1.2007, "grad_norm": 0.620944082736969, "learning_rate": 0.0002, "epoch": 2.960550126673905, "step": 4090}, {"loss": 1.2864, "grad_norm": 0.5866289734840393, "learning_rate": 0.0002, "epoch": 2.9677886355410785, "step": 4100}, {"loss": 1.3293, "grad_norm": 0.560170590877533, "learning_rate": 0.0002, "epoch": 2.975027144408252, "step": 4110}, {"loss": 1.2071, "grad_norm": 0.675082802772522, "learning_rate": 0.0002, "epoch": 2.9822656532754253, "step": 4120}, {"loss": 1.2981, "grad_norm": 0.62708580493927, "learning_rate": 0.0002, "epoch": 2.9895041621425986, "step": 4130}, {"loss": 1.2758, "grad_norm": 0.7893929481506348, "learning_rate": 0.0002, "epoch": 2.996742671009772, "step": 4140}]}
+{"epoch": 4.0, "step": 5526, "epoch_duration": 1327.1966397762299, "total_accumulated_duration": 5338.999227523804, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 3020.60888671875}, "peak_memory_usage": {"GPU_0": 15079.2998046875}, "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-0.9-num-8828-sd-1/checkpoint-4144", "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": 4.7061, "grad_norm": 1.2523442506790161, "learning_rate": 0.0002, "epoch": 0.007238508867173362, "step": 10}, {"loss": 3.3493, "grad_norm": 1.8887330293655396, "learning_rate": 0.0002, "epoch": 0.014477017734346724, "step": 20}, {"loss": 2.7585, "grad_norm": 0.9668035507202148, "learning_rate": 0.0002, "epoch": 0.021715526601520086, "step": 30}, {"loss": 2.3699, "grad_norm": 2.9167306423187256, "learning_rate": 0.0002, "epoch": 0.028954035468693448, "step": 40}, {"loss": 2.2679, "grad_norm": 2.649867296218872, "learning_rate": 0.0002, "epoch": 0.036192544335866814, "step": 50}, {"loss": 2.2202, "grad_norm": 1.5120655298233032, "learning_rate": 0.0002, "epoch": 0.04343105320304017, "step": 60}, {"loss": 2.2026, "grad_norm": 0.7879868149757385, "learning_rate": 0.0002, "epoch": 0.05066956207021354, "step": 70}, {"loss": 1.9447, "grad_norm": 0.7616953253746033, "learning_rate": 0.0002, "epoch": 0.057908070937386896, "step": 80}, {"loss": 2.0112, "grad_norm": 1.8809149265289307, "learning_rate": 0.0002, "epoch": 0.06514657980456026, "step": 90}, {"loss": 1.8337, "grad_norm": 0.9294016361236572, "learning_rate": 0.0002, "epoch": 0.07238508867173363, "step": 100}, {"loss": 1.8419, "grad_norm": 0.7145281434059143, "learning_rate": 0.0002, "epoch": 0.07962359753890698, "step": 110}, {"loss": 2.0036, "grad_norm": 0.7564446330070496, "learning_rate": 0.0002, "epoch": 0.08686210640608034, "step": 120}, {"loss": 1.9306, "grad_norm": 1.1681925058364868, "learning_rate": 0.0002, "epoch": 0.09410061527325371, "step": 130}, {"loss": 1.7875, "grad_norm": 0.6708641648292542, "learning_rate": 0.0002, "epoch": 0.10133912414042708, "step": 140}, {"loss": 1.786, "grad_norm": 0.7625647783279419, "learning_rate": 0.0002, "epoch": 0.10857763300760044, "step": 150}, {"loss": 1.6687, "grad_norm": 0.8463464975357056, "learning_rate": 0.0002, "epoch": 0.11581614187477379, "step": 160}, {"loss": 1.6214, "grad_norm": 0.7502335906028748, "learning_rate": 0.0002, "epoch": 0.12305465074194716, "step": 170}, {"loss": 1.7433, "grad_norm": 0.6929958462715149, "learning_rate": 0.0002, "epoch": 0.13029315960912052, "step": 180}, {"loss": 1.6009, "grad_norm": 0.6798707842826843, "learning_rate": 0.0002, "epoch": 0.1375316684762939, "step": 190}, {"loss": 1.6208, "grad_norm": 0.7566508650779724, "learning_rate": 0.0002, "epoch": 0.14477017734346725, "step": 200}, {"loss": 1.5823, "grad_norm": 0.7196869850158691, "learning_rate": 0.0002, "epoch": 0.15200868621064062, "step": 210}, {"loss": 1.738, "grad_norm": 0.8401045799255371, "learning_rate": 0.0002, "epoch": 0.15924719507781396, "step": 220}, {"loss": 1.7574, "grad_norm": 0.8503773212432861, "learning_rate": 0.0002, "epoch": 0.16648570394498732, "step": 230}, {"loss": 1.7861, "grad_norm": 0.7183733582496643, "learning_rate": 0.0002, "epoch": 0.1737242128121607, "step": 240}, {"loss": 1.6693, "grad_norm": 0.7082605957984924, "learning_rate": 0.0002, "epoch": 0.18096272167933405, "step": 250}, {"loss": 1.619, "grad_norm": 0.9386326670646667, "learning_rate": 0.0002, "epoch": 0.18820123054650742, "step": 260}, {"loss": 1.6511, "grad_norm": 0.7332451939582825, "learning_rate": 0.0002, "epoch": 0.19543973941368079, "step": 270}, {"loss": 1.6353, "grad_norm": 0.7092869877815247, "learning_rate": 0.0002, "epoch": 0.20267824828085415, "step": 280}, {"loss": 1.5996, "grad_norm": 0.7256413698196411, "learning_rate": 0.0002, "epoch": 0.20991675714802752, "step": 290}, {"loss": 1.6754, "grad_norm": 0.6398681402206421, "learning_rate": 0.0002, "epoch": 0.21715526601520088, "step": 300}, {"loss": 1.397, "grad_norm": 0.6273287534713745, "learning_rate": 0.0002, "epoch": 0.22439377488237422, "step": 310}, {"loss": 1.5115, "grad_norm": 0.511648416519165, "learning_rate": 0.0002, "epoch": 0.23163228374954759, "step": 320}, {"loss": 1.5424, "grad_norm": 0.8677352070808411, "learning_rate": 0.0002, "epoch": 0.23887079261672095, "step": 330}, {"loss": 1.6779, "grad_norm": 0.6270743012428284, "learning_rate": 0.0002, "epoch": 0.24610930148389432, "step": 340}, {"loss": 1.626, "grad_norm": 0.7980281114578247, "learning_rate": 0.0002, "epoch": 0.2533478103510677, "step": 350}, {"loss": 1.5238, "grad_norm": 0.632486879825592, "learning_rate": 0.0002, "epoch": 0.26058631921824105, "step": 360}, {"loss": 1.5175, "grad_norm": 0.6527034640312195, "learning_rate": 0.0002, "epoch": 0.2678248280854144, "step": 370}, {"loss": 1.627, "grad_norm": 0.7672118544578552, "learning_rate": 0.0002, "epoch": 0.2750633369525878, "step": 380}, {"loss": 1.5605, "grad_norm": 0.6035117506980896, "learning_rate": 0.0002, "epoch": 0.28230184581976114, "step": 390}, {"loss": 1.4603, "grad_norm": 0.5955103039741516, "learning_rate": 0.0002, "epoch": 0.2895403546869345, "step": 400}, {"loss": 1.558, "grad_norm": 0.6015191674232483, "learning_rate": 0.0002, "epoch": 0.2967788635541079, "step": 410}, {"loss": 1.6091, "grad_norm": 0.6380982398986816, "learning_rate": 0.0002, "epoch": 0.30401737242128124, "step": 420}, {"loss": 1.5292, "grad_norm": 0.6707863211631775, "learning_rate": 0.0002, "epoch": 0.3112558812884546, "step": 430}, {"loss": 1.4426, "grad_norm": 0.7010176777839661, "learning_rate": 0.0002, "epoch": 0.3184943901556279, "step": 440}, {"loss": 1.5572, "grad_norm": 0.8263739943504333, "learning_rate": 0.0002, "epoch": 0.3257328990228013, "step": 450}, {"loss": 1.5188, "grad_norm": 0.7253276109695435, "learning_rate": 0.0002, "epoch": 0.33297140788997465, "step": 460}, {"loss": 1.584, "grad_norm": 0.5238934755325317, "learning_rate": 0.0002, "epoch": 0.340209916757148, "step": 470}, {"loss": 1.7035, "grad_norm": 0.7869495749473572, "learning_rate": 0.0002, "epoch": 0.3474484256243214, "step": 480}, {"loss": 1.5776, "grad_norm": 0.7485215663909912, "learning_rate": 0.0002, "epoch": 0.35468693449149474, "step": 490}, {"loss": 1.6274, "grad_norm": 0.5413193106651306, "learning_rate": 0.0002, "epoch": 0.3619254433586681, "step": 500}, {"loss": 1.7323, "grad_norm": 0.7615048885345459, "learning_rate": 0.0002, "epoch": 0.3691639522258415, "step": 510}, {"loss": 1.532, "grad_norm": 0.7685340046882629, "learning_rate": 0.0002, "epoch": 0.37640246109301484, "step": 520}, {"loss": 1.6312, "grad_norm": 0.6379081010818481, "learning_rate": 0.0002, "epoch": 0.3836409699601882, "step": 530}, {"loss": 1.5645, "grad_norm": 0.7946939468383789, "learning_rate": 0.0002, "epoch": 0.39087947882736157, "step": 540}, {"loss": 1.4001, "grad_norm": 0.6287278532981873, "learning_rate": 0.0002, "epoch": 0.39811798769453494, "step": 550}, {"loss": 1.5982, "grad_norm": 0.6811642646789551, "learning_rate": 0.0002, "epoch": 0.4053564965617083, "step": 560}, {"loss": 1.4953, "grad_norm": 0.671073317527771, "learning_rate": 0.0002, "epoch": 0.41259500542888167, "step": 570}, {"loss": 1.6753, "grad_norm": 0.6313900351524353, "learning_rate": 0.0002, "epoch": 0.41983351429605503, "step": 580}, {"loss": 1.546, "grad_norm": 0.5291772484779358, "learning_rate": 0.0002, "epoch": 0.4270720231632284, "step": 590}, {"loss": 1.5441, "grad_norm": 0.62503582239151, "learning_rate": 0.0002, "epoch": 0.43431053203040176, "step": 600}, {"loss": 1.6276, "grad_norm": 0.5777305364608765, "learning_rate": 0.0002, "epoch": 0.4415490408975751, "step": 610}, {"loss": 1.4758, "grad_norm": 0.7013497352600098, "learning_rate": 0.0002, "epoch": 0.44878754976474844, "step": 620}, {"loss": 1.4029, "grad_norm": 0.8044822216033936, "learning_rate": 0.0002, "epoch": 0.4560260586319218, "step": 630}, {"loss": 1.7195, "grad_norm": 0.672531247138977, "learning_rate": 0.0002, "epoch": 0.46326456749909517, "step": 640}, {"loss": 1.614, "grad_norm": 0.6233910322189331, "learning_rate": 0.0002, "epoch": 0.47050307636626854, "step": 650}, {"loss": 1.6041, "grad_norm": 0.651524543762207, "learning_rate": 0.0002, "epoch": 0.4777415852334419, "step": 660}, {"loss": 1.5842, "grad_norm": 0.7213939428329468, "learning_rate": 0.0002, "epoch": 0.48498009410061527, "step": 670}, {"loss": 1.5453, "grad_norm": 0.6541454792022705, "learning_rate": 0.0002, "epoch": 0.49221860296778863, "step": 680}, {"loss": 1.662, "grad_norm": 0.6568936109542847, "learning_rate": 0.0002, "epoch": 0.499457111834962, "step": 690}, {"loss": 1.624, "grad_norm": 0.7176415324211121, "learning_rate": 0.0002, "epoch": 0.5066956207021354, "step": 700}, {"loss": 1.6099, "grad_norm": 0.6553855538368225, "learning_rate": 0.0002, "epoch": 0.5139341295693087, "step": 710}, {"loss": 1.5508, "grad_norm": 0.5654335618019104, "learning_rate": 0.0002, "epoch": 0.5211726384364821, "step": 720}, {"loss": 1.392, "grad_norm": 0.5671001672744751, "learning_rate": 0.0002, "epoch": 0.5284111473036555, "step": 730}, {"loss": 1.388, "grad_norm": 0.7914412021636963, "learning_rate": 0.0002, "epoch": 0.5356496561708288, "step": 740}, {"loss": 1.5931, "grad_norm": 0.6172138452529907, "learning_rate": 0.0002, "epoch": 0.5428881650380022, "step": 750}, {"loss": 1.4018, "grad_norm": 0.6132623553276062, "learning_rate": 0.0002, "epoch": 0.5501266739051756, "step": 760}, {"loss": 1.513, "grad_norm": 0.654000461101532, "learning_rate": 0.0002, "epoch": 0.5573651827723489, "step": 770}, {"loss": 1.5035, "grad_norm": 0.5691370964050293, "learning_rate": 0.0002, "epoch": 0.5646036916395223, "step": 780}, {"loss": 1.65, "grad_norm": 0.7922580242156982, "learning_rate": 0.0002, "epoch": 0.5718422005066957, "step": 790}, {"loss": 1.4521, "grad_norm": 0.6831880211830139, "learning_rate": 0.0002, "epoch": 0.579080709373869, "step": 800}, {"loss": 1.4734, "grad_norm": 0.6740124821662903, "learning_rate": 0.0002, "epoch": 0.5863192182410424, "step": 810}, {"loss": 1.6498, "grad_norm": 1.380016803741455, "learning_rate": 0.0002, "epoch": 0.5935577271082157, "step": 820}, {"loss": 1.4642, "grad_norm": 0.6552878022193909, "learning_rate": 0.0002, "epoch": 0.6007962359753891, "step": 830}, {"loss": 1.6271, "grad_norm": 0.6649535298347473, "learning_rate": 0.0002, "epoch": 0.6080347448425625, "step": 840}, {"loss": 1.5886, "grad_norm": 0.561738133430481, "learning_rate": 0.0002, "epoch": 0.6152732537097358, "step": 850}, {"loss": 1.5364, "grad_norm": 0.6133047938346863, "learning_rate": 0.0002, "epoch": 0.6225117625769092, "step": 860}, {"loss": 1.3489, "grad_norm": 0.559843122959137, "learning_rate": 0.0002, "epoch": 0.6297502714440825, "step": 870}, {"loss": 1.4878, "grad_norm": 0.6117811799049377, "learning_rate": 0.0002, "epoch": 0.6369887803112558, "step": 880}, {"loss": 1.56, "grad_norm": 0.6209776401519775, "learning_rate": 0.0002, "epoch": 0.6442272891784292, "step": 890}, {"loss": 1.6747, "grad_norm": 0.6234082579612732, "learning_rate": 0.0002, "epoch": 0.6514657980456026, "step": 900}, {"loss": 1.6963, "grad_norm": 0.7623258233070374, "learning_rate": 0.0002, "epoch": 0.6587043069127759, "step": 910}, {"loss": 1.2424, "grad_norm": 0.6148061752319336, "learning_rate": 0.0002, "epoch": 0.6659428157799493, "step": 920}, {"loss": 1.4319, "grad_norm": 0.6682973504066467, "learning_rate": 0.0002, "epoch": 0.6731813246471227, "step": 930}, {"loss": 1.5377, "grad_norm": 0.5513041615486145, "learning_rate": 0.0002, "epoch": 0.680419833514296, "step": 940}, {"loss": 1.3991, "grad_norm": 0.5197525024414062, "learning_rate": 0.0002, "epoch": 0.6876583423814694, "step": 950}, {"loss": 1.4398, "grad_norm": 0.6490758061408997, "learning_rate": 0.0002, "epoch": 0.6948968512486428, "step": 960}, {"loss": 1.5251, "grad_norm": 0.6450682878494263, "learning_rate": 0.0002, "epoch": 0.7021353601158161, "step": 970}, {"loss": 1.5417, "grad_norm": 0.6203766465187073, "learning_rate": 0.0002, "epoch": 0.7093738689829895, "step": 980}, {"loss": 1.4575, "grad_norm": 0.6023609638214111, "learning_rate": 0.0002, "epoch": 0.7166123778501629, "step": 990}, {"loss": 1.4973, "grad_norm": 0.5765255093574524, "learning_rate": 0.0002, "epoch": 0.7238508867173362, "step": 1000}, {"loss": 1.483, "grad_norm": 0.6650075316429138, "learning_rate": 0.0002, "epoch": 0.7310893955845096, "step": 1010}, {"loss": 1.5959, "grad_norm": 0.5610854029655457, "learning_rate": 0.0002, "epoch": 0.738327904451683, "step": 1020}, {"loss": 1.5248, "grad_norm": 0.7072813510894775, "learning_rate": 0.0002, "epoch": 0.7455664133188563, "step": 1030}, {"loss": 1.5776, "grad_norm": 0.6815407872200012, "learning_rate": 0.0002, "epoch": 0.7528049221860297, "step": 1040}, {"loss": 1.4577, "grad_norm": 0.7932390570640564, "learning_rate": 0.0002, "epoch": 0.760043431053203, "step": 1050}, {"loss": 1.4515, "grad_norm": 0.5798183083534241, "learning_rate": 0.0002, "epoch": 0.7672819399203764, "step": 1060}, {"loss": 1.5053, "grad_norm": 0.7898504137992859, "learning_rate": 0.0002, "epoch": 0.7745204487875498, "step": 1070}, {"loss": 1.4776, "grad_norm": 0.4983280301094055, "learning_rate": 0.0002, "epoch": 0.7817589576547231, "step": 1080}, {"loss": 1.5007, "grad_norm": 0.691403329372406, "learning_rate": 0.0002, "epoch": 0.7889974665218965, "step": 1090}, {"loss": 1.5153, "grad_norm": 0.5394481420516968, "learning_rate": 0.0002, "epoch": 0.7962359753890699, "step": 1100}, {"loss": 1.6892, "grad_norm": 0.5136822462081909, "learning_rate": 0.0002, "epoch": 0.8034744842562432, "step": 1110}, {"loss": 1.4902, "grad_norm": 0.6828126907348633, "learning_rate": 0.0002, "epoch": 0.8107129931234166, "step": 1120}, {"loss": 1.4346, "grad_norm": 0.6799656748771667, "learning_rate": 0.0002, "epoch": 0.81795150199059, "step": 1130}, {"loss": 1.2678, "grad_norm": 0.5428406000137329, "learning_rate": 0.0002, "epoch": 0.8251900108577633, "step": 1140}, {"loss": 1.4072, "grad_norm": 0.4811290502548218, "learning_rate": 0.0002, "epoch": 0.8324285197249367, "step": 1150}, {"loss": 1.4512, "grad_norm": 0.5519434809684753, "learning_rate": 0.0002, "epoch": 0.8396670285921101, "step": 1160}, {"loss": 1.4072, "grad_norm": 0.9748060703277588, "learning_rate": 0.0002, "epoch": 0.8469055374592834, "step": 1170}, {"loss": 1.4309, "grad_norm": 0.712609589099884, "learning_rate": 0.0002, "epoch": 0.8541440463264568, "step": 1180}, {"loss": 1.434, "grad_norm": 0.6866157054901123, "learning_rate": 0.0002, "epoch": 0.8613825551936302, "step": 1190}, {"loss": 1.3704, "grad_norm": 0.5068854093551636, "learning_rate": 0.0002, "epoch": 0.8686210640608035, "step": 1200}, {"loss": 1.5601, "grad_norm": 0.6333245038986206, "learning_rate": 0.0002, "epoch": 0.8758595729279768, "step": 1210}, {"loss": 1.4636, "grad_norm": 0.6424421072006226, "learning_rate": 0.0002, "epoch": 0.8830980817951501, "step": 1220}, {"loss": 1.4186, "grad_norm": 0.4771921932697296, "learning_rate": 0.0002, "epoch": 0.8903365906623235, "step": 1230}, {"loss": 1.6323, "grad_norm": 0.5191764235496521, "learning_rate": 0.0002, "epoch": 0.8975750995294969, "step": 1240}, {"loss": 1.6105, "grad_norm": 0.756222128868103, "learning_rate": 0.0002, "epoch": 0.9048136083966702, "step": 1250}, {"loss": 1.4396, "grad_norm": 0.623823881149292, "learning_rate": 0.0002, "epoch": 0.9120521172638436, "step": 1260}, {"loss": 1.3097, "grad_norm": 0.8166571259498596, "learning_rate": 0.0002, "epoch": 0.919290626131017, "step": 1270}, {"loss": 1.4625, "grad_norm": 0.6059346795082092, "learning_rate": 0.0002, "epoch": 0.9265291349981903, "step": 1280}, {"loss": 1.3555, "grad_norm": 0.5842690467834473, "learning_rate": 0.0002, "epoch": 0.9337676438653637, "step": 1290}, {"loss": 1.5859, "grad_norm": 0.7649800777435303, "learning_rate": 0.0002, "epoch": 0.9410061527325371, "step": 1300}, {"loss": 1.5915, "grad_norm": 0.6420919895172119, "learning_rate": 0.0002, "epoch": 0.9482446615997104, "step": 1310}, {"loss": 1.453, "grad_norm": 0.7011452913284302, "learning_rate": 0.0002, "epoch": 0.9554831704668838, "step": 1320}, {"loss": 1.6766, "grad_norm": 0.5783746242523193, "learning_rate": 0.0002, "epoch": 0.9627216793340572, "step": 1330}, {"loss": 1.6308, "grad_norm": 0.5973192453384399, "learning_rate": 0.0002, "epoch": 0.9699601882012305, "step": 1340}, {"loss": 1.5901, "grad_norm": 0.6181833744049072, "learning_rate": 0.0002, "epoch": 0.9771986970684039, "step": 1350}, {"loss": 1.5258, "grad_norm": 0.5563396215438843, "learning_rate": 0.0002, "epoch": 0.9844372059355773, "step": 1360}, {"loss": 1.4508, "grad_norm": 0.45723360776901245, "learning_rate": 0.0002, "epoch": 0.9916757148027506, "step": 1370}, {"loss": 1.3291, "grad_norm": 0.5947498679161072, "learning_rate": 0.0002, "epoch": 0.998914223669924, "step": 1380}, {"eval_loss": 1.480796456336975, "eval_runtime": 27.3103, "eval_samples_per_second": 15.965, "eval_steps_per_second": 2.014, "epoch": 0.9996380745566413, "step": 1381}, {"loss": 1.3057, "grad_norm": 0.5599952936172485, "learning_rate": 0.0002, "epoch": 1.0061527325370974, "step": 1390}, {"loss": 1.4991, "grad_norm": 0.5932008028030396, "learning_rate": 0.0002, "epoch": 1.0133912414042707, "step": 1400}, {"loss": 1.4506, "grad_norm": 0.6194121837615967, "learning_rate": 0.0002, "epoch": 1.020629750271444, "step": 1410}, {"loss": 1.5966, "grad_norm": 0.6995621919631958, "learning_rate": 0.0002, "epoch": 1.0278682591386175, "step": 1420}, {"loss": 1.4153, "grad_norm": 0.7905810475349426, "learning_rate": 0.0002, "epoch": 1.0351067680057908, "step": 1430}, {"loss": 1.4414, "grad_norm": 0.7221615314483643, "learning_rate": 0.0002, "epoch": 1.0423452768729642, "step": 1440}, {"loss": 1.3859, "grad_norm": 0.6170642375946045, "learning_rate": 0.0002, "epoch": 1.0495837857401376, "step": 1450}, {"loss": 1.3806, "grad_norm": 0.5844094753265381, "learning_rate": 0.0002, "epoch": 1.056822294607311, "step": 1460}, {"loss": 1.4871, "grad_norm": 0.7731822729110718, "learning_rate": 0.0002, "epoch": 1.0640608034744843, "step": 1470}, {"loss": 1.4286, "grad_norm": 0.4554748237133026, "learning_rate": 0.0002, "epoch": 1.0712993123416577, "step": 1480}, {"loss": 1.3977, "grad_norm": 0.6923259496688843, "learning_rate": 0.0002, "epoch": 1.078537821208831, "step": 1490}, {"loss": 1.3936, "grad_norm": 0.6008219122886658, "learning_rate": 0.0002, "epoch": 1.0857763300760044, "step": 1500}, {"loss": 1.4821, "grad_norm": 0.6450045704841614, "learning_rate": 0.0002, "epoch": 1.0930148389431777, "step": 1510}, {"loss": 1.3295, "grad_norm": 0.7833753824234009, "learning_rate": 0.0002, "epoch": 1.1002533478103511, "step": 1520}, {"loss": 1.3424, "grad_norm": 0.5076758861541748, "learning_rate": 0.0002, "epoch": 1.1074918566775245, "step": 1530}, {"loss": 1.4043, "grad_norm": 0.5661332011222839, "learning_rate": 0.0002, "epoch": 1.1147303655446978, "step": 1540}, {"loss": 1.4963, "grad_norm": 0.6526919603347778, "learning_rate": 0.0002, "epoch": 1.1219688744118712, "step": 1550}, {"loss": 1.3671, "grad_norm": 0.5613082647323608, "learning_rate": 0.0002, "epoch": 1.1292073832790446, "step": 1560}, {"loss": 1.4458, "grad_norm": 0.6113885641098022, "learning_rate": 0.0002, "epoch": 1.136445892146218, "step": 1570}, {"loss": 1.3552, "grad_norm": 0.6732510328292847, "learning_rate": 0.0002, "epoch": 1.1436844010133913, "step": 1580}, {"loss": 1.3114, "grad_norm": 0.6146392226219177, "learning_rate": 0.0002, "epoch": 1.1509229098805647, "step": 1590}, {"loss": 1.411, "grad_norm": 0.6766974329948425, "learning_rate": 0.0002, "epoch": 1.158161418747738, "step": 1600}, {"loss": 1.2401, "grad_norm": 0.7621957659721375, "learning_rate": 0.0002, "epoch": 1.1653999276149114, "step": 1610}, {"loss": 1.3758, "grad_norm": 0.6959581971168518, "learning_rate": 0.0002, "epoch": 1.1726384364820848, "step": 1620}, {"loss": 1.382, "grad_norm": 0.6691278219223022, "learning_rate": 0.0002, "epoch": 1.1798769453492581, "step": 1630}, {"loss": 1.4147, "grad_norm": 0.4927774965763092, "learning_rate": 0.0002, "epoch": 1.1871154542164315, "step": 1640}, {"loss": 1.449, "grad_norm": 0.7724234461784363, "learning_rate": 0.0002, "epoch": 1.1943539630836049, "step": 1650}, {"loss": 1.4778, "grad_norm": 0.6817787885665894, "learning_rate": 0.0002, "epoch": 1.2015924719507782, "step": 1660}, {"loss": 1.3776, "grad_norm": 0.6500699520111084, "learning_rate": 0.0002, "epoch": 1.2088309808179516, "step": 1670}, {"loss": 1.3875, "grad_norm": 0.5703568458557129, "learning_rate": 0.0002, "epoch": 1.216069489685125, "step": 1680}, {"loss": 1.4735, "grad_norm": 0.6261579990386963, "learning_rate": 0.0002, "epoch": 1.2233079985522983, "step": 1690}, {"loss": 1.3898, "grad_norm": 0.651713490486145, "learning_rate": 0.0002, "epoch": 1.2305465074194717, "step": 1700}, {"loss": 1.4002, "grad_norm": 0.684399425983429, "learning_rate": 0.0002, "epoch": 1.237785016286645, "step": 1710}, {"loss": 1.5027, "grad_norm": 0.6996857523918152, "learning_rate": 0.0002, "epoch": 1.2450235251538184, "step": 1720}, {"loss": 1.3326, "grad_norm": 0.7102537751197815, "learning_rate": 0.0002, "epoch": 1.2522620340209918, "step": 1730}, {"loss": 1.3675, "grad_norm": 0.45809897780418396, "learning_rate": 0.0002, "epoch": 1.2595005428881652, "step": 1740}, {"loss": 1.4175, "grad_norm": 0.6377046704292297, "learning_rate": 0.0002, "epoch": 1.2667390517553385, "step": 1750}, {"loss": 1.3479, "grad_norm": 0.6965704560279846, "learning_rate": 0.0002, "epoch": 1.2739775606225119, "step": 1760}, {"loss": 1.5647, "grad_norm": 0.5688214302062988, "learning_rate": 0.0002, "epoch": 1.2812160694896852, "step": 1770}, {"loss": 1.3967, "grad_norm": 0.6384190320968628, "learning_rate": 0.0002, "epoch": 1.2884545783568586, "step": 1780}, {"loss": 1.3671, "grad_norm": 0.5629363656044006, "learning_rate": 0.0002, "epoch": 1.295693087224032, "step": 1790}, {"loss": 1.2292, "grad_norm": 0.6148255467414856, "learning_rate": 0.0002, "epoch": 1.3029315960912053, "step": 1800}, {"loss": 1.5806, "grad_norm": 0.655580997467041, "learning_rate": 0.0002, "epoch": 1.3101701049583787, "step": 1810}, {"loss": 1.2398, "grad_norm": 0.5642657279968262, "learning_rate": 0.0002, "epoch": 1.3174086138255519, "step": 1820}, {"loss": 1.3246, "grad_norm": 0.59607994556427, "learning_rate": 0.0002, "epoch": 1.3246471226927252, "step": 1830}, {"loss": 1.3274, "grad_norm": 0.5564199090003967, "learning_rate": 0.0002, "epoch": 1.3318856315598986, "step": 1840}, {"loss": 1.5834, "grad_norm": 0.6949955821037292, "learning_rate": 0.0002, "epoch": 1.339124140427072, "step": 1850}, {"loss": 1.4722, "grad_norm": 0.7036856412887573, "learning_rate": 0.0002, "epoch": 1.3463626492942453, "step": 1860}, {"loss": 1.333, "grad_norm": 0.722062885761261, "learning_rate": 0.0002, "epoch": 1.3536011581614187, "step": 1870}, {"loss": 1.4044, "grad_norm": 0.6098677515983582, "learning_rate": 0.0002, "epoch": 1.360839667028592, "step": 1880}, {"loss": 1.6217, "grad_norm": 0.5376402735710144, "learning_rate": 0.0002, "epoch": 1.3680781758957654, "step": 1890}, {"loss": 1.5071, "grad_norm": 0.6974610090255737, "learning_rate": 0.0002, "epoch": 1.3753166847629388, "step": 1900}, {"loss": 1.5854, "grad_norm": 0.6520763635635376, "learning_rate": 0.0002, "epoch": 1.3825551936301121, "step": 1910}, {"loss": 1.4271, "grad_norm": 0.6604374647140503, "learning_rate": 0.0002, "epoch": 1.3897937024972855, "step": 1920}, {"loss": 1.419, "grad_norm": 0.7364398241043091, "learning_rate": 0.0002, "epoch": 1.3970322113644589, "step": 1930}, {"loss": 1.4585, "grad_norm": 0.6849475502967834, "learning_rate": 0.0002, "epoch": 1.4042707202316322, "step": 1940}, {"loss": 1.5577, "grad_norm": 0.6562670469284058, "learning_rate": 0.0002, "epoch": 1.4115092290988056, "step": 1950}, {"loss": 1.4725, "grad_norm": 0.5695616006851196, "learning_rate": 0.0002, "epoch": 1.418747737965979, "step": 1960}, {"loss": 1.3088, "grad_norm": 0.5244464874267578, "learning_rate": 0.0002, "epoch": 1.4259862468331523, "step": 1970}, {"loss": 1.5069, "grad_norm": 0.6347293257713318, "learning_rate": 0.0002, "epoch": 1.4332247557003257, "step": 1980}, {"loss": 1.3502, "grad_norm": 0.5528361201286316, "learning_rate": 0.0002, "epoch": 1.440463264567499, "step": 1990}, {"loss": 1.3978, "grad_norm": 0.6987585425376892, "learning_rate": 0.0002, "epoch": 1.4477017734346724, "step": 2000}, {"loss": 1.4262, "grad_norm": 0.6568987369537354, "learning_rate": 0.0002, "epoch": 1.4549402823018458, "step": 2010}, {"loss": 1.4175, "grad_norm": 0.7665994763374329, "learning_rate": 0.0002, "epoch": 1.4621787911690192, "step": 2020}, {"loss": 1.244, "grad_norm": 0.5127707123756409, "learning_rate": 0.0002, "epoch": 1.4694173000361925, "step": 2030}, {"loss": 1.3699, "grad_norm": 0.5406824946403503, "learning_rate": 0.0002, "epoch": 1.476655808903366, "step": 2040}, {"loss": 1.3353, "grad_norm": 0.5990166664123535, "learning_rate": 0.0002, "epoch": 1.4838943177705393, "step": 2050}, {"loss": 1.2454, "grad_norm": 0.6186193823814392, "learning_rate": 0.0002, "epoch": 1.4911328266377126, "step": 2060}, {"loss": 1.428, "grad_norm": 0.6154307126998901, "learning_rate": 0.0002, "epoch": 1.498371335504886, "step": 2070}, {"loss": 1.4528, "grad_norm": 0.5606056451797485, "learning_rate": 0.0002, "epoch": 1.5056098443720594, "step": 2080}, {"loss": 1.2405, "grad_norm": 0.5006417036056519, "learning_rate": 0.0002, "epoch": 1.5128483532392327, "step": 2090}, {"loss": 1.4258, "grad_norm": 0.5968486070632935, "learning_rate": 0.0002, "epoch": 1.520086862106406, "step": 2100}, {"loss": 1.2752, "grad_norm": 0.5835496187210083, "learning_rate": 0.0002, "epoch": 1.5273253709735795, "step": 2110}, {"loss": 1.5443, "grad_norm": 0.6753535270690918, "learning_rate": 0.0002, "epoch": 1.5345638798407528, "step": 2120}, {"loss": 1.2139, "grad_norm": 0.7299720644950867, "learning_rate": 0.0002, "epoch": 1.5418023887079262, "step": 2130}, {"loss": 1.2364, "grad_norm": 0.5105988383293152, "learning_rate": 0.0002, "epoch": 1.5490408975750996, "step": 2140}, {"loss": 1.4528, "grad_norm": 0.5675431489944458, "learning_rate": 0.0002, "epoch": 1.556279406442273, "step": 2150}, {"loss": 1.4563, "grad_norm": 0.6246723532676697, "learning_rate": 0.0002, "epoch": 1.5635179153094463, "step": 2160}, {"loss": 1.5255, "grad_norm": 0.7291720509529114, "learning_rate": 0.0002, "epoch": 1.5707564241766196, "step": 2170}, {"loss": 1.5432, "grad_norm": 0.678114116191864, "learning_rate": 0.0002, "epoch": 1.577994933043793, "step": 2180}, {"loss": 1.5212, "grad_norm": 0.5136260986328125, "learning_rate": 0.0002, "epoch": 1.5852334419109664, "step": 2190}, {"loss": 1.3271, "grad_norm": 0.6359935998916626, "learning_rate": 0.0002, "epoch": 1.5924719507781397, "step": 2200}, {"loss": 1.4038, "grad_norm": 0.7650278806686401, "learning_rate": 0.0002, "epoch": 1.599710459645313, "step": 2210}, {"loss": 1.5478, "grad_norm": 0.7256110906600952, "learning_rate": 0.0002, "epoch": 1.6069489685124865, "step": 2220}, {"loss": 1.4387, "grad_norm": 0.688689649105072, "learning_rate": 0.0002, "epoch": 1.6141874773796598, "step": 2230}, {"loss": 1.4096, "grad_norm": 0.6045311093330383, "learning_rate": 0.0002, "epoch": 1.6214259862468332, "step": 2240}, {"loss": 1.4097, "grad_norm": 0.7064604163169861, "learning_rate": 0.0002, "epoch": 1.6286644951140063, "step": 2250}, {"loss": 1.3477, "grad_norm": 0.5309562087059021, "learning_rate": 0.0002, "epoch": 1.6359030039811797, "step": 2260}, {"loss": 1.4022, "grad_norm": 0.5687053203582764, "learning_rate": 0.0002, "epoch": 1.643141512848353, "step": 2270}, {"loss": 1.2977, "grad_norm": 0.535872757434845, "learning_rate": 0.0002, "epoch": 1.6503800217155264, "step": 2280}, {"loss": 1.3844, "grad_norm": 0.5502381920814514, "learning_rate": 0.0002, "epoch": 1.6576185305826998, "step": 2290}, {"loss": 1.3764, "grad_norm": 0.6158602237701416, "learning_rate": 0.0002, "epoch": 1.6648570394498732, "step": 2300}, {"loss": 1.3515, "grad_norm": 0.5804675817489624, "learning_rate": 0.0002, "epoch": 1.6720955483170465, "step": 2310}, {"loss": 1.2532, "grad_norm": 0.600742757320404, "learning_rate": 0.0002, "epoch": 1.67933405718422, "step": 2320}, {"loss": 1.477, "grad_norm": 0.7101941108703613, "learning_rate": 0.0002, "epoch": 1.6865725660513933, "step": 2330}, {"loss": 1.4849, "grad_norm": 0.7507809996604919, "learning_rate": 0.0002, "epoch": 1.6938110749185666, "step": 2340}, {"loss": 1.2703, "grad_norm": 0.768502414226532, "learning_rate": 0.0002, "epoch": 1.70104958378574, "step": 2350}, {"loss": 1.3332, "grad_norm": 0.4801851212978363, "learning_rate": 0.0002, "epoch": 1.7082880926529134, "step": 2360}, {"loss": 1.4158, "grad_norm": 0.5322122573852539, "learning_rate": 0.0002, "epoch": 1.7155266015200867, "step": 2370}, {"loss": 1.4136, "grad_norm": 0.587661862373352, "learning_rate": 0.0002, "epoch": 1.72276511038726, "step": 2380}, {"loss": 1.3771, "grad_norm": 0.6073525547981262, "learning_rate": 0.0002, "epoch": 1.7300036192544335, "step": 2390}, {"loss": 1.2754, "grad_norm": 0.6950460076332092, "learning_rate": 0.0002, "epoch": 1.7372421281216068, "step": 2400}, {"loss": 1.3858, "grad_norm": 0.5981102585792542, "learning_rate": 0.0002, "epoch": 1.7444806369887802, "step": 2410}, {"loss": 1.4075, "grad_norm": 0.544570803642273, "learning_rate": 0.0002, "epoch": 1.7517191458559536, "step": 2420}, {"loss": 1.3861, "grad_norm": 0.5304399728775024, "learning_rate": 0.0002, "epoch": 1.758957654723127, "step": 2430}, {"loss": 1.4244, "grad_norm": 0.7921594977378845, "learning_rate": 0.0002, "epoch": 1.7661961635903003, "step": 2440}, {"loss": 1.3053, "grad_norm": 0.6084808707237244, "learning_rate": 0.0002, "epoch": 1.7734346724574737, "step": 2450}, {"loss": 1.3781, "grad_norm": 0.8844701051712036, "learning_rate": 0.0002, "epoch": 1.780673181324647, "step": 2460}, {"loss": 1.3227, "grad_norm": 0.5729258060455322, "learning_rate": 0.0002, "epoch": 1.7879116901918204, "step": 2470}, {"loss": 1.3422, "grad_norm": 0.6303611993789673, "learning_rate": 0.0002, "epoch": 1.7951501990589938, "step": 2480}, {"loss": 1.3926, "grad_norm": 0.5627942085266113, "learning_rate": 0.0002, "epoch": 1.8023887079261671, "step": 2490}, {"loss": 1.3816, "grad_norm": 0.6724274158477783, "learning_rate": 0.0002, "epoch": 1.8096272167933405, "step": 2500}, {"loss": 1.2951, "grad_norm": 0.5030826330184937, "learning_rate": 0.0002, "epoch": 1.8168657256605139, "step": 2510}, {"loss": 1.2839, "grad_norm": 0.5504099130630493, "learning_rate": 0.0002, "epoch": 1.8241042345276872, "step": 2520}, {"loss": 1.4264, "grad_norm": 0.6338945627212524, "learning_rate": 0.0002, "epoch": 1.8313427433948606, "step": 2530}, {"loss": 1.563, "grad_norm": 0.5902037620544434, "learning_rate": 0.0002, "epoch": 1.838581252262034, "step": 2540}, {"loss": 1.2961, "grad_norm": 0.48814457654953003, "learning_rate": 0.0002, "epoch": 1.8458197611292073, "step": 2550}, {"loss": 1.466, "grad_norm": 0.6216312646865845, "learning_rate": 0.0002, "epoch": 1.8530582699963807, "step": 2560}, {"loss": 1.5123, "grad_norm": 0.635603666305542, "learning_rate": 0.0002, "epoch": 1.860296778863554, "step": 2570}, {"loss": 1.372, "grad_norm": 0.6938216090202332, "learning_rate": 0.0002, "epoch": 1.8675352877307274, "step": 2580}, {"loss": 1.5011, "grad_norm": 0.599557638168335, "learning_rate": 0.0002, "epoch": 1.8747737965979008, "step": 2590}, {"loss": 1.2714, "grad_norm": 0.564424455165863, "learning_rate": 0.0002, "epoch": 1.8820123054650741, "step": 2600}, {"loss": 1.3403, "grad_norm": 0.5430700182914734, "learning_rate": 0.0002, "epoch": 1.8892508143322475, "step": 2610}, {"loss": 1.4347, "grad_norm": 0.6150169372558594, "learning_rate": 0.0002, "epoch": 1.8964893231994209, "step": 2620}, {"loss": 1.2474, "grad_norm": 0.48159119486808777, "learning_rate": 0.0002, "epoch": 1.9037278320665942, "step": 2630}, {"loss": 1.3716, "grad_norm": 0.5608997941017151, "learning_rate": 0.0002, "epoch": 1.9109663409337676, "step": 2640}, {"loss": 1.5787, "grad_norm": 0.6454501748085022, "learning_rate": 0.0002, "epoch": 1.918204849800941, "step": 2650}, {"loss": 1.3238, "grad_norm": 0.5458073616027832, "learning_rate": 0.0002, "epoch": 1.9254433586681143, "step": 2660}, {"loss": 1.3208, "grad_norm": 0.5328490734100342, "learning_rate": 0.0002, "epoch": 1.9326818675352877, "step": 2670}, {"loss": 1.4971, "grad_norm": 0.6444696187973022, "learning_rate": 0.0002, "epoch": 1.939920376402461, "step": 2680}, {"loss": 1.5387, "grad_norm": 0.7126023769378662, "learning_rate": 0.0002, "epoch": 1.9471588852696344, "step": 2690}, {"loss": 1.3637, "grad_norm": 0.5164045095443726, "learning_rate": 0.0002, "epoch": 1.9543973941368078, "step": 2700}, {"loss": 1.5303, "grad_norm": 0.5347061157226562, "learning_rate": 0.0002, "epoch": 1.9616359030039812, "step": 2710}, {"loss": 1.2815, "grad_norm": 0.5297950506210327, "learning_rate": 0.0002, "epoch": 1.9688744118711545, "step": 2720}, {"loss": 1.3566, "grad_norm": 0.6537790298461914, "learning_rate": 0.0002, "epoch": 1.976112920738328, "step": 2730}, {"loss": 1.332, "grad_norm": 0.5536222457885742, "learning_rate": 0.0002, "epoch": 1.9833514296055013, "step": 2740}, {"loss": 1.3333, "grad_norm": 0.4856105446815491, "learning_rate": 0.0002, "epoch": 1.9905899384726746, "step": 2750}, {"loss": 1.3521, "grad_norm": 0.6642730832099915, "learning_rate": 0.0002, "epoch": 1.997828447339848, "step": 2760}, {"eval_loss": 1.4366681575775146, "eval_runtime": 27.3729, "eval_samples_per_second": 15.928, "eval_steps_per_second": 2.009, "epoch": 2.0, "step": 2763}, {"loss": 1.4322, "grad_norm": 0.740253210067749, "learning_rate": 0.0002, "epoch": 2.0050669562070214, "step": 2770}, {"loss": 1.277, "grad_norm": 0.5826276540756226, "learning_rate": 0.0002, "epoch": 2.0123054650741947, "step": 2780}, {"loss": 1.2424, "grad_norm": 0.607356071472168, "learning_rate": 0.0002, "epoch": 2.019543973941368, "step": 2790}, {"loss": 1.2601, "grad_norm": 0.5918063521385193, "learning_rate": 0.0002, "epoch": 2.0267824828085415, "step": 2800}, {"loss": 1.3715, "grad_norm": 0.5610089898109436, "learning_rate": 0.0002, "epoch": 2.034020991675715, "step": 2810}, {"loss": 1.2092, "grad_norm": 0.5869926810264587, "learning_rate": 0.0002, "epoch": 2.041259500542888, "step": 2820}, {"loss": 1.1929, "grad_norm": 0.5753467679023743, "learning_rate": 0.0002, "epoch": 2.0484980094100615, "step": 2830}, {"loss": 1.333, "grad_norm": 0.7096508145332336, "learning_rate": 0.0002, "epoch": 2.055736518277235, "step": 2840}, {"loss": 1.1766, "grad_norm": 0.7653635144233704, "learning_rate": 0.0002, "epoch": 2.0629750271444083, "step": 2850}, {"loss": 1.2331, "grad_norm": 0.6202841997146606, "learning_rate": 0.0002, "epoch": 2.0702135360115816, "step": 2860}, {"loss": 1.3298, "grad_norm": 0.6810227632522583, "learning_rate": 0.0002, "epoch": 2.077452044878755, "step": 2870}, {"loss": 1.2505, "grad_norm": 0.7481493353843689, "learning_rate": 0.0002, "epoch": 2.0846905537459284, "step": 2880}, {"loss": 1.2484, "grad_norm": 0.7089637517929077, "learning_rate": 0.0002, "epoch": 2.0919290626131017, "step": 2890}, {"loss": 1.3095, "grad_norm": 0.7472923398017883, "learning_rate": 0.0002, "epoch": 2.099167571480275, "step": 2900}, {"loss": 1.304, "grad_norm": 0.8135465979576111, "learning_rate": 0.0002, "epoch": 2.1064060803474485, "step": 2910}, {"loss": 1.273, "grad_norm": 0.6097133159637451, "learning_rate": 0.0002, "epoch": 2.113644589214622, "step": 2920}, {"loss": 1.3384, "grad_norm": 0.5970117449760437, "learning_rate": 0.0002, "epoch": 2.120883098081795, "step": 2930}, {"loss": 1.3233, "grad_norm": 0.6169309616088867, "learning_rate": 0.0002, "epoch": 2.1281216069489686, "step": 2940}, {"loss": 1.4246, "grad_norm": 0.9428738355636597, "learning_rate": 0.0002, "epoch": 2.135360115816142, "step": 2950}, {"loss": 1.3527, "grad_norm": 0.5671679973602295, "learning_rate": 0.0002, "epoch": 2.1425986246833153, "step": 2960}, {"loss": 1.1375, "grad_norm": 0.7007262110710144, "learning_rate": 0.0002, "epoch": 2.1498371335504887, "step": 2970}, {"loss": 1.2015, "grad_norm": 0.6294044256210327, "learning_rate": 0.0002, "epoch": 2.157075642417662, "step": 2980}, {"loss": 1.2167, "grad_norm": 0.6105241775512695, "learning_rate": 0.0002, "epoch": 2.1643141512848354, "step": 2990}, {"loss": 1.2065, "grad_norm": 0.557124137878418, "learning_rate": 0.0002, "epoch": 2.1715526601520088, "step": 3000}, {"loss": 1.2515, "grad_norm": 0.6250392198562622, "learning_rate": 0.0002, "epoch": 2.178791169019182, "step": 3010}, {"loss": 1.385, "grad_norm": 0.645218551158905, "learning_rate": 0.0002, "epoch": 2.1860296778863555, "step": 3020}, {"loss": 1.3928, "grad_norm": 0.9033605456352234, "learning_rate": 0.0002, "epoch": 2.193268186753529, "step": 3030}, {"loss": 1.2458, "grad_norm": 0.5325747132301331, "learning_rate": 0.0002, "epoch": 2.2005066956207022, "step": 3040}, {"loss": 1.261, "grad_norm": 0.6334700584411621, "learning_rate": 0.0002, "epoch": 2.2077452044878756, "step": 3050}, {"loss": 1.2385, "grad_norm": 0.5206325054168701, "learning_rate": 0.0002, "epoch": 2.214983713355049, "step": 3060}, {"loss": 1.3103, "grad_norm": 0.5987200140953064, "learning_rate": 0.0002, "epoch": 2.2222222222222223, "step": 3070}, {"loss": 1.1756, "grad_norm": 0.5893264412879944, "learning_rate": 0.0002, "epoch": 2.2294607310893957, "step": 3080}, {"loss": 1.235, "grad_norm": 0.6869237422943115, "learning_rate": 0.0002, "epoch": 2.236699239956569, "step": 3090}, {"loss": 1.3285, "grad_norm": 0.5040048360824585, "learning_rate": 0.0002, "epoch": 2.2439377488237424, "step": 3100}, {"loss": 1.3316, "grad_norm": 0.6660613417625427, "learning_rate": 0.0002, "epoch": 2.251176257690916, "step": 3110}, {"loss": 1.3108, "grad_norm": 0.5890918970108032, "learning_rate": 0.0002, "epoch": 2.258414766558089, "step": 3120}, {"loss": 1.248, "grad_norm": 0.6458896994590759, "learning_rate": 0.0002, "epoch": 2.2656532754252625, "step": 3130}, {"loss": 1.4151, "grad_norm": 0.6832690834999084, "learning_rate": 0.0002, "epoch": 2.272891784292436, "step": 3140}, {"loss": 1.4458, "grad_norm": 0.833908200263977, "learning_rate": 0.0002, "epoch": 2.2801302931596092, "step": 3150}, {"loss": 1.2931, "grad_norm": 0.4596034586429596, "learning_rate": 0.0002, "epoch": 2.2873688020267826, "step": 3160}, {"loss": 1.449, "grad_norm": 0.9130966067314148, "learning_rate": 0.0002, "epoch": 2.294607310893956, "step": 3170}, {"loss": 1.3806, "grad_norm": 0.7143292427062988, "learning_rate": 0.0002, "epoch": 2.3018458197611293, "step": 3180}, {"loss": 1.2692, "grad_norm": 0.5388900637626648, "learning_rate": 0.0002, "epoch": 2.3090843286283027, "step": 3190}, {"loss": 1.2402, "grad_norm": 0.5607513189315796, "learning_rate": 0.0002, "epoch": 2.316322837495476, "step": 3200}, {"loss": 1.3874, "grad_norm": 0.6795142292976379, "learning_rate": 0.0002, "epoch": 2.3235613463626494, "step": 3210}, {"loss": 1.3042, "grad_norm": 0.6561070680618286, "learning_rate": 0.0002, "epoch": 2.330799855229823, "step": 3220}, {"loss": 1.4636, "grad_norm": 0.8858118057250977, "learning_rate": 0.0002, "epoch": 2.338038364096996, "step": 3230}, {"loss": 1.3214, "grad_norm": 0.6604151725769043, "learning_rate": 0.0002, "epoch": 2.3452768729641695, "step": 3240}, {"loss": 1.4004, "grad_norm": 0.6755785346031189, "learning_rate": 0.0002, "epoch": 2.352515381831343, "step": 3250}, {"loss": 1.2503, "grad_norm": 0.6981677412986755, "learning_rate": 0.0002, "epoch": 2.3597538906985163, "step": 3260}, {"loss": 1.3078, "grad_norm": 0.6338568329811096, "learning_rate": 0.0002, "epoch": 2.3669923995656896, "step": 3270}, {"loss": 1.285, "grad_norm": 0.5754265785217285, "learning_rate": 0.0002, "epoch": 2.374230908432863, "step": 3280}, {"loss": 1.2924, "grad_norm": 0.7533153295516968, "learning_rate": 0.0002, "epoch": 2.3814694173000364, "step": 3290}, {"loss": 1.3711, "grad_norm": 0.675065279006958, "learning_rate": 0.0002, "epoch": 2.3887079261672097, "step": 3300}, {"loss": 1.3548, "grad_norm": 0.5686452984809875, "learning_rate": 0.0002, "epoch": 2.395946435034383, "step": 3310}, {"loss": 1.1998, "grad_norm": 0.8129481673240662, "learning_rate": 0.0002, "epoch": 2.4031849439015565, "step": 3320}, {"loss": 1.2584, "grad_norm": 0.6615934371948242, "learning_rate": 0.0002, "epoch": 2.41042345276873, "step": 3330}, {"loss": 1.3691, "grad_norm": 0.6678834557533264, "learning_rate": 0.0002, "epoch": 2.417661961635903, "step": 3340}, {"loss": 1.2381, "grad_norm": 0.5581308007240295, "learning_rate": 0.0002, "epoch": 2.4249004705030766, "step": 3350}, {"loss": 1.3853, "grad_norm": 0.6098920106887817, "learning_rate": 0.0002, "epoch": 2.43213897937025, "step": 3360}, {"loss": 1.3692, "grad_norm": 0.8101736903190613, "learning_rate": 0.0002, "epoch": 2.4393774882374233, "step": 3370}, {"loss": 1.4418, "grad_norm": 0.6621488928794861, "learning_rate": 0.0002, "epoch": 2.4466159971045967, "step": 3380}, {"loss": 1.4579, "grad_norm": 0.8693289160728455, "learning_rate": 0.0002, "epoch": 2.45385450597177, "step": 3390}, {"loss": 1.3644, "grad_norm": 0.6724580526351929, "learning_rate": 0.0002, "epoch": 2.4610930148389434, "step": 3400}, {"loss": 1.2006, "grad_norm": 0.6776891946792603, "learning_rate": 0.0002, "epoch": 2.4683315237061167, "step": 3410}, {"loss": 1.2937, "grad_norm": 0.7214453816413879, "learning_rate": 0.0002, "epoch": 2.47557003257329, "step": 3420}, {"loss": 1.4051, "grad_norm": 0.8390451073646545, "learning_rate": 0.0002, "epoch": 2.4828085414404635, "step": 3430}, {"loss": 1.25, "grad_norm": 0.7130982279777527, "learning_rate": 0.0002, "epoch": 2.490047050307637, "step": 3440}, {"loss": 1.2231, "grad_norm": 0.8873937129974365, "learning_rate": 0.0002, "epoch": 2.49728555917481, "step": 3450}, {"loss": 1.1429, "grad_norm": 0.725185751914978, "learning_rate": 0.0002, "epoch": 2.5045240680419836, "step": 3460}, {"loss": 1.2699, "grad_norm": 0.6120352149009705, "learning_rate": 0.0002, "epoch": 2.511762576909157, "step": 3470}, {"loss": 1.2552, "grad_norm": 0.7713613510131836, "learning_rate": 0.0002, "epoch": 2.5190010857763303, "step": 3480}, {"loss": 1.4648, "grad_norm": 0.895309567451477, "learning_rate": 0.0002, "epoch": 2.5262395946435037, "step": 3490}, {"loss": 1.3043, "grad_norm": 0.9631021022796631, "learning_rate": 0.0002, "epoch": 2.533478103510677, "step": 3500}, {"loss": 1.3492, "grad_norm": 0.7475683093070984, "learning_rate": 0.0002, "epoch": 2.5407166123778504, "step": 3510}, {"loss": 1.3637, "grad_norm": 0.7271341681480408, "learning_rate": 0.0002, "epoch": 2.5479551212450238, "step": 3520}, {"loss": 1.304, "grad_norm": 0.6979510188102722, "learning_rate": 0.0002, "epoch": 2.555193630112197, "step": 3530}, {"loss": 1.2353, "grad_norm": 0.6504196524620056, "learning_rate": 0.0002, "epoch": 2.5624321389793705, "step": 3540}, {"loss": 1.2699, "grad_norm": 0.7226675748825073, "learning_rate": 0.0002, "epoch": 2.569670647846544, "step": 3550}, {"loss": 1.3002, "grad_norm": 0.6143222451210022, "learning_rate": 0.0002, "epoch": 2.5769091567137172, "step": 3560}, {"loss": 1.1585, "grad_norm": 0.7245154976844788, "learning_rate": 0.0002, "epoch": 2.5841476655808906, "step": 3570}, {"loss": 1.3651, "grad_norm": 0.943540632724762, "learning_rate": 0.0002, "epoch": 2.591386174448064, "step": 3580}, {"loss": 1.3034, "grad_norm": 0.7707241773605347, "learning_rate": 0.0002, "epoch": 2.5986246833152373, "step": 3590}, {"loss": 1.3063, "grad_norm": 0.6705001592636108, "learning_rate": 0.0002, "epoch": 2.6058631921824107, "step": 3600}, {"loss": 1.2437, "grad_norm": 0.6360933780670166, "learning_rate": 0.0002, "epoch": 2.613101701049584, "step": 3610}, {"loss": 1.1844, "grad_norm": 0.5846424698829651, "learning_rate": 0.0002, "epoch": 2.6203402099167574, "step": 3620}, {"loss": 1.3674, "grad_norm": 0.5958625674247742, "learning_rate": 0.0002, "epoch": 2.6275787187839303, "step": 3630}, {"loss": 1.3599, "grad_norm": 0.6819243431091309, "learning_rate": 0.0002, "epoch": 2.6348172276511037, "step": 3640}, {"loss": 1.3884, "grad_norm": 0.7033445835113525, "learning_rate": 0.0002, "epoch": 2.642055736518277, "step": 3650}, {"loss": 1.3392, "grad_norm": 0.6134849786758423, "learning_rate": 0.0002, "epoch": 2.6492942453854504, "step": 3660}, {"loss": 1.2661, "grad_norm": 0.658009946346283, "learning_rate": 0.0002, "epoch": 2.656532754252624, "step": 3670}, {"loss": 1.3987, "grad_norm": 0.6280999779701233, "learning_rate": 0.0002, "epoch": 2.663771263119797, "step": 3680}, {"loss": 1.2995, "grad_norm": 0.5536085963249207, "learning_rate": 0.0002, "epoch": 2.6710097719869705, "step": 3690}, {"loss": 1.2044, "grad_norm": 0.8603981733322144, "learning_rate": 0.0002, "epoch": 2.678248280854144, "step": 3700}, {"loss": 1.3879, "grad_norm": 0.5509994626045227, "learning_rate": 0.0002, "epoch": 2.6854867897213173, "step": 3710}, {"loss": 1.3253, "grad_norm": 0.9093621969223022, "learning_rate": 0.0002, "epoch": 2.6927252985884906, "step": 3720}, {"loss": 1.2668, "grad_norm": 0.7525952458381653, "learning_rate": 0.0002, "epoch": 2.699963807455664, "step": 3730}, {"loss": 1.248, "grad_norm": 0.6737023591995239, "learning_rate": 0.0002, "epoch": 2.7072023163228374, "step": 3740}, {"loss": 1.2981, "grad_norm": 0.8656924962997437, "learning_rate": 0.0002, "epoch": 2.7144408251900107, "step": 3750}, {"loss": 1.2342, "grad_norm": 0.7494133114814758, "learning_rate": 0.0002, "epoch": 2.721679334057184, "step": 3760}, {"loss": 1.2417, "grad_norm": 0.5725520849227905, "learning_rate": 0.0002, "epoch": 2.7289178429243575, "step": 3770}, {"loss": 1.28, "grad_norm": 0.836412787437439, "learning_rate": 0.0002, "epoch": 2.736156351791531, "step": 3780}, {"loss": 1.3784, "grad_norm": 0.6893242597579956, "learning_rate": 0.0002, "epoch": 2.743394860658704, "step": 3790}, {"loss": 1.2929, "grad_norm": 0.6696223020553589, "learning_rate": 0.0002, "epoch": 2.7506333695258776, "step": 3800}, {"loss": 1.2449, "grad_norm": 0.6483015418052673, "learning_rate": 0.0002, "epoch": 2.757871878393051, "step": 3810}, {"loss": 1.3282, "grad_norm": 0.8084456920623779, "learning_rate": 0.0002, "epoch": 2.7651103872602243, "step": 3820}, {"loss": 1.3694, "grad_norm": 0.6601949334144592, "learning_rate": 0.0002, "epoch": 2.7723488961273977, "step": 3830}, {"loss": 1.3568, "grad_norm": 0.6905533671379089, "learning_rate": 0.0002, "epoch": 2.779587404994571, "step": 3840}, {"loss": 1.3854, "grad_norm": 0.619318425655365, "learning_rate": 0.0002, "epoch": 2.7868259138617444, "step": 3850}, {"loss": 1.2551, "grad_norm": 0.5994023084640503, "learning_rate": 0.0002, "epoch": 2.7940644227289178, "step": 3860}, {"loss": 1.2022, "grad_norm": 0.5627168416976929, "learning_rate": 0.0002, "epoch": 2.801302931596091, "step": 3870}, {"loss": 1.3921, "grad_norm": 0.6001605987548828, "learning_rate": 0.0002, "epoch": 2.8085414404632645, "step": 3880}, {"loss": 1.3026, "grad_norm": 0.6022412776947021, "learning_rate": 0.0002, "epoch": 2.815779949330438, "step": 3890}, {"loss": 1.2765, "grad_norm": 0.6832426190376282, "learning_rate": 0.0002, "epoch": 2.823018458197611, "step": 3900}, {"loss": 1.1363, "grad_norm": 0.5936811566352844, "learning_rate": 0.0002, "epoch": 2.8302569670647846, "step": 3910}, {"loss": 1.1707, "grad_norm": 0.6960572600364685, "learning_rate": 0.0002, "epoch": 2.837495475931958, "step": 3920}, {"loss": 1.4063, "grad_norm": 0.5913406610488892, "learning_rate": 0.0002, "epoch": 2.8447339847991313, "step": 3930}, {"loss": 1.3245, "grad_norm": 0.678154706954956, "learning_rate": 0.0002, "epoch": 2.8519724936663047, "step": 3940}, {"loss": 1.366, "grad_norm": 0.7898936867713928, "learning_rate": 0.0002, "epoch": 2.859211002533478, "step": 3950}, {"loss": 1.3948, "grad_norm": 0.9234195351600647, "learning_rate": 0.0002, "epoch": 2.8664495114006514, "step": 3960}, {"loss": 1.2773, "grad_norm": 0.5960825085639954, "learning_rate": 0.0002, "epoch": 2.8736880202678248, "step": 3970}, {"loss": 1.3127, "grad_norm": 0.677118182182312, "learning_rate": 0.0002, "epoch": 2.880926529134998, "step": 3980}, {"loss": 1.2652, "grad_norm": 0.6505142450332642, "learning_rate": 0.0002, "epoch": 2.8881650380021715, "step": 3990}, {"loss": 1.2078, "grad_norm": 0.550826907157898, "learning_rate": 0.0002, "epoch": 2.895403546869345, "step": 4000}, {"loss": 1.1811, "grad_norm": 0.6209215521812439, "learning_rate": 0.0002, "epoch": 2.9026420557365182, "step": 4010}, {"loss": 1.4001, "grad_norm": 0.6549018025398254, "learning_rate": 0.0002, "epoch": 2.9098805646036916, "step": 4020}, {"loss": 1.2285, "grad_norm": 0.570682168006897, "learning_rate": 0.0002, "epoch": 2.917119073470865, "step": 4030}, {"loss": 1.0832, "grad_norm": 1.1807632446289062, "learning_rate": 0.0002, "epoch": 2.9243575823380383, "step": 4040}, {"loss": 1.2693, "grad_norm": 0.7058857679367065, "learning_rate": 0.0002, "epoch": 2.9315960912052117, "step": 4050}, {"loss": 1.2905, "grad_norm": 0.5542812943458557, "learning_rate": 0.0002, "epoch": 2.938834600072385, "step": 4060}, {"loss": 1.33, "grad_norm": 0.63167804479599, "learning_rate": 0.0002, "epoch": 2.9460731089395584, "step": 4070}, {"loss": 1.3075, "grad_norm": 0.5702962279319763, "learning_rate": 0.0002, "epoch": 2.953311617806732, "step": 4080}, {"loss": 1.2007, "grad_norm": 0.620944082736969, "learning_rate": 0.0002, "epoch": 2.960550126673905, "step": 4090}, {"loss": 1.2864, "grad_norm": 0.5866289734840393, "learning_rate": 0.0002, "epoch": 2.9677886355410785, "step": 4100}, {"loss": 1.3293, "grad_norm": 0.560170590877533, "learning_rate": 0.0002, "epoch": 2.975027144408252, "step": 4110}, {"loss": 1.2071, "grad_norm": 0.675082802772522, "learning_rate": 0.0002, "epoch": 2.9822656532754253, "step": 4120}, {"loss": 1.2981, "grad_norm": 0.62708580493927, "learning_rate": 0.0002, "epoch": 2.9895041621425986, "step": 4130}, {"loss": 1.2758, "grad_norm": 0.7893929481506348, "learning_rate": 0.0002, "epoch": 2.996742671009772, "step": 4140}, {"eval_loss": 1.4217946529388428, "eval_runtime": 27.1596, "eval_samples_per_second": 16.053, "eval_steps_per_second": 2.025, "epoch": 2.9996380745566413, "step": 4144}, {"loss": 1.2152, "grad_norm": 0.7043836116790771, "learning_rate": 0.0002, "epoch": 3.0039811798769454, "step": 4150}, {"loss": 1.1664, "grad_norm": 0.6806283593177795, "learning_rate": 0.0002, "epoch": 3.0112196887441187, "step": 4160}, {"loss": 1.292, "grad_norm": 0.7684550285339355, "learning_rate": 0.0002, "epoch": 3.018458197611292, "step": 4170}, {"loss": 1.3467, "grad_norm": 0.7895237803459167, "learning_rate": 0.0002, "epoch": 3.0256967064784654, "step": 4180}, {"loss": 1.1324, "grad_norm": 0.7464531064033508, "learning_rate": 0.0002, "epoch": 3.032935215345639, "step": 4190}, {"loss": 1.1614, "grad_norm": 0.9358500838279724, "learning_rate": 0.0002, "epoch": 3.040173724212812, "step": 4200}, {"loss": 1.1834, "grad_norm": 1.1066628694534302, "learning_rate": 0.0002, "epoch": 3.0474122330799855, "step": 4210}, {"loss": 1.1557, "grad_norm": 0.6663267612457275, "learning_rate": 0.0002, "epoch": 3.054650741947159, "step": 4220}, {"loss": 1.1707, "grad_norm": 0.6669464707374573, "learning_rate": 0.0002, "epoch": 3.0618892508143323, "step": 4230}, {"loss": 1.1841, "grad_norm": 0.7052164077758789, "learning_rate": 0.0002, "epoch": 3.0691277596815056, "step": 4240}, {"loss": 1.2913, "grad_norm": 0.6118432879447937, "learning_rate": 0.0002, "epoch": 3.076366268548679, "step": 4250}, {"loss": 1.1526, "grad_norm": 0.6915903687477112, "learning_rate": 0.0002, "epoch": 3.0836047774158524, "step": 4260}, {"loss": 1.1348, "grad_norm": 0.7441644668579102, "learning_rate": 0.0002, "epoch": 3.0908432862830257, "step": 4270}, {"loss": 1.1672, "grad_norm": 0.823850691318512, "learning_rate": 0.0002, "epoch": 3.098081795150199, "step": 4280}, {"loss": 1.2655, "grad_norm": 0.9677883386611938, "learning_rate": 0.0002, "epoch": 3.1053203040173725, "step": 4290}, {"loss": 1.1794, "grad_norm": 0.7002579569816589, "learning_rate": 0.0002, "epoch": 3.112558812884546, "step": 4300}, {"loss": 1.135, "grad_norm": 0.778789758682251, "learning_rate": 0.0002, "epoch": 3.119797321751719, "step": 4310}, {"loss": 1.0818, "grad_norm": 0.7236007452011108, "learning_rate": 0.0002, "epoch": 3.1270358306188926, "step": 4320}, {"loss": 1.1803, "grad_norm": 0.8809133768081665, "learning_rate": 0.0002, "epoch": 3.134274339486066, "step": 4330}, {"loss": 1.2571, "grad_norm": 0.7924913167953491, "learning_rate": 0.0002, "epoch": 3.1415128483532393, "step": 4340}, {"loss": 1.1413, "grad_norm": 0.7437422275543213, "learning_rate": 0.0002, "epoch": 3.1487513572204127, "step": 4350}, {"loss": 1.2088, "grad_norm": 0.6428450345993042, "learning_rate": 0.0002, "epoch": 3.155989866087586, "step": 4360}, {"loss": 1.3032, "grad_norm": 0.7922873497009277, "learning_rate": 0.0002, "epoch": 3.1632283749547594, "step": 4370}, {"loss": 1.216, "grad_norm": 0.5252506732940674, "learning_rate": 0.0002, "epoch": 3.1704668838219328, "step": 4380}, {"loss": 1.1297, "grad_norm": 0.8570457696914673, "learning_rate": 0.0002, "epoch": 3.177705392689106, "step": 4390}, {"loss": 1.0994, "grad_norm": 0.7218987345695496, "learning_rate": 0.0002, "epoch": 3.1849439015562795, "step": 4400}, {"loss": 1.2891, "grad_norm": 0.6921393275260925, "learning_rate": 0.0002, "epoch": 3.192182410423453, "step": 4410}, {"loss": 1.2668, "grad_norm": 0.7386137843132019, "learning_rate": 0.0002, "epoch": 3.199420919290626, "step": 4420}, {"loss": 1.1654, "grad_norm": 0.6227759122848511, "learning_rate": 0.0002, "epoch": 3.2066594281577996, "step": 4430}, {"loss": 1.1752, "grad_norm": 0.7180278897285461, "learning_rate": 0.0002, "epoch": 3.213897937024973, "step": 4440}, {"loss": 1.1757, "grad_norm": 0.745830774307251, "learning_rate": 0.0002, "epoch": 3.2211364458921463, "step": 4450}, {"loss": 1.234, "grad_norm": 0.6766072511672974, "learning_rate": 0.0002, "epoch": 3.2283749547593197, "step": 4460}, {"loss": 1.1999, "grad_norm": 0.8325067162513733, "learning_rate": 0.0002, "epoch": 3.235613463626493, "step": 4470}, {"loss": 1.1606, "grad_norm": 0.7148305177688599, "learning_rate": 0.0002, "epoch": 3.2428519724936664, "step": 4480}, {"loss": 1.1383, "grad_norm": 0.7752676010131836, "learning_rate": 0.0002, "epoch": 3.25009048136084, "step": 4490}, {"loss": 1.3006, "grad_norm": 0.6776860952377319, "learning_rate": 0.0002, "epoch": 3.257328990228013, "step": 4500}, {"loss": 1.0796, "grad_norm": 0.704359769821167, "learning_rate": 0.0002, "epoch": 3.2645674990951865, "step": 4510}, {"loss": 1.2496, "grad_norm": 0.6880282163619995, "learning_rate": 0.0002, "epoch": 3.27180600796236, "step": 4520}, {"loss": 1.0947, "grad_norm": 0.8179270029067993, "learning_rate": 0.0002, "epoch": 3.2790445168295332, "step": 4530}, {"loss": 1.1909, "grad_norm": 0.6718448996543884, "learning_rate": 0.0002, "epoch": 3.2862830256967066, "step": 4540}, {"loss": 1.2708, "grad_norm": 0.8300657868385315, "learning_rate": 0.0002, "epoch": 3.29352153456388, "step": 4550}, {"loss": 1.2594, "grad_norm": 0.6433690786361694, "learning_rate": 0.0002, "epoch": 3.3007600434310533, "step": 4560}, {"loss": 1.2479, "grad_norm": 0.690262496471405, "learning_rate": 0.0002, "epoch": 3.3079985522982267, "step": 4570}, {"loss": 1.1342, "grad_norm": 0.7022852301597595, "learning_rate": 0.0002, "epoch": 3.3152370611654, "step": 4580}, {"loss": 1.0844, "grad_norm": 0.6438387632369995, "learning_rate": 0.0002, "epoch": 3.3224755700325734, "step": 4590}, {"loss": 1.17, "grad_norm": 0.6866899132728577, "learning_rate": 0.0002, "epoch": 3.329714078899747, "step": 4600}, {"loss": 1.1289, "grad_norm": 0.8233968019485474, "learning_rate": 0.0002, "epoch": 3.33695258776692, "step": 4610}, {"loss": 1.1855, "grad_norm": 0.7251574993133545, "learning_rate": 0.0002, "epoch": 3.3441910966340935, "step": 4620}, {"loss": 1.3403, "grad_norm": 0.7855110168457031, "learning_rate": 0.0002, "epoch": 3.351429605501267, "step": 4630}, {"loss": 1.2922, "grad_norm": 0.8487356305122375, "learning_rate": 0.0002, "epoch": 3.3586681143684403, "step": 4640}, {"loss": 1.2462, "grad_norm": 0.6429011225700378, "learning_rate": 0.0002, "epoch": 3.3659066232356136, "step": 4650}, {"loss": 1.129, "grad_norm": 0.7095270156860352, "learning_rate": 0.0002, "epoch": 3.373145132102787, "step": 4660}, {"loss": 1.262, "grad_norm": 0.6792303323745728, "learning_rate": 0.0002, "epoch": 3.3803836409699604, "step": 4670}, {"loss": 1.256, "grad_norm": 0.6784825921058655, "learning_rate": 0.0002, "epoch": 3.3876221498371337, "step": 4680}, {"loss": 1.0838, "grad_norm": 0.6362888216972351, "learning_rate": 0.0002, "epoch": 3.394860658704307, "step": 4690}, {"loss": 1.2165, "grad_norm": 0.7794778943061829, "learning_rate": 0.0002, "epoch": 3.4020991675714805, "step": 4700}, {"loss": 1.0644, "grad_norm": 0.7287485003471375, "learning_rate": 0.0002, "epoch": 3.409337676438654, "step": 4710}, {"loss": 1.2925, "grad_norm": 0.6481451392173767, "learning_rate": 0.0002, "epoch": 3.416576185305827, "step": 4720}, {"loss": 1.2121, "grad_norm": 0.9200371503829956, "learning_rate": 0.0002, "epoch": 3.4238146941730006, "step": 4730}, {"loss": 1.072, "grad_norm": 1.074180245399475, "learning_rate": 0.0002, "epoch": 3.431053203040174, "step": 4740}, {"loss": 1.0421, "grad_norm": 0.6722986698150635, "learning_rate": 0.0002, "epoch": 3.438291711907347, "step": 4750}, {"loss": 1.2258, "grad_norm": 0.7945933938026428, "learning_rate": 0.0002, "epoch": 3.44553022077452, "step": 4760}, {"loss": 1.0927, "grad_norm": 0.7624640464782715, "learning_rate": 0.0002, "epoch": 3.4527687296416936, "step": 4770}, {"loss": 1.2428, "grad_norm": 0.7763656377792358, "learning_rate": 0.0002, "epoch": 3.460007238508867, "step": 4780}, {"loss": 1.2584, "grad_norm": 0.7736947536468506, "learning_rate": 0.0002, "epoch": 3.4672457473760403, "step": 4790}, {"loss": 1.1953, "grad_norm": 0.8450354933738708, "learning_rate": 0.0002, "epoch": 3.4744842562432137, "step": 4800}, {"loss": 1.1362, "grad_norm": 0.6480133533477783, "learning_rate": 0.0002, "epoch": 3.481722765110387, "step": 4810}, {"loss": 1.1882, "grad_norm": 0.8437445759773254, "learning_rate": 0.0002, "epoch": 3.4889612739775604, "step": 4820}, {"loss": 1.1519, "grad_norm": 0.7781730890274048, "learning_rate": 0.0002, "epoch": 3.4961997828447338, "step": 4830}, {"loss": 1.1836, "grad_norm": 0.8523228168487549, "learning_rate": 0.0002, "epoch": 3.503438291711907, "step": 4840}, {"loss": 1.1672, "grad_norm": 0.6236732006072998, "learning_rate": 0.0002, "epoch": 3.5106768005790805, "step": 4850}, {"loss": 1.1926, "grad_norm": 0.7500787377357483, "learning_rate": 0.0002, "epoch": 3.517915309446254, "step": 4860}, {"loss": 1.1998, "grad_norm": 0.7665374875068665, "learning_rate": 0.0002, "epoch": 3.5251538183134272, "step": 4870}, {"loss": 1.1551, "grad_norm": 0.787857711315155, "learning_rate": 0.0002, "epoch": 3.5323923271806006, "step": 4880}, {"loss": 1.2758, "grad_norm": 0.970595121383667, "learning_rate": 0.0002, "epoch": 3.539630836047774, "step": 4890}, {"loss": 1.1274, "grad_norm": 0.6409347057342529, "learning_rate": 0.0002, "epoch": 3.5468693449149473, "step": 4900}, {"loss": 1.1596, "grad_norm": 0.888551652431488, "learning_rate": 0.0002, "epoch": 3.5541078537821207, "step": 4910}, {"loss": 1.1644, "grad_norm": 1.0808377265930176, "learning_rate": 0.0002, "epoch": 3.561346362649294, "step": 4920}, {"loss": 1.2564, "grad_norm": 0.7501053214073181, "learning_rate": 0.0002, "epoch": 3.5685848715164674, "step": 4930}, {"loss": 1.2351, "grad_norm": 0.7375240325927734, "learning_rate": 0.0002, "epoch": 3.575823380383641, "step": 4940}, {"loss": 1.3568, "grad_norm": 0.7075039744377136, "learning_rate": 0.0002, "epoch": 3.583061889250814, "step": 4950}, {"loss": 1.3355, "grad_norm": 0.939337432384491, "learning_rate": 0.0002, "epoch": 3.5903003981179875, "step": 4960}, {"loss": 1.1722, "grad_norm": 0.6717396974563599, "learning_rate": 0.0002, "epoch": 3.597538906985161, "step": 4970}, {"loss": 1.1186, "grad_norm": 0.7141643762588501, "learning_rate": 0.0002, "epoch": 3.6047774158523342, "step": 4980}, {"loss": 1.1011, "grad_norm": 0.7109216451644897, "learning_rate": 0.0002, "epoch": 3.6120159247195076, "step": 4990}, {"loss": 1.2178, "grad_norm": 0.7020776867866516, "learning_rate": 0.0002, "epoch": 3.619254433586681, "step": 5000}, {"loss": 1.1939, "grad_norm": 0.7158873677253723, "learning_rate": 0.0002, "epoch": 3.6264929424538543, "step": 5010}, {"loss": 1.2624, "grad_norm": 0.7062035202980042, "learning_rate": 0.0002, "epoch": 3.6337314513210277, "step": 5020}, {"loss": 1.0224, "grad_norm": 0.7081155776977539, "learning_rate": 0.0002, "epoch": 3.640969960188201, "step": 5030}, {"loss": 1.2195, "grad_norm": 1.2210607528686523, "learning_rate": 0.0002, "epoch": 3.6482084690553744, "step": 5040}, {"loss": 1.2596, "grad_norm": 0.6650236248970032, "learning_rate": 0.0002, "epoch": 3.655446977922548, "step": 5050}, {"loss": 1.1072, "grad_norm": 0.6884829998016357, "learning_rate": 0.0002, "epoch": 3.662685486789721, "step": 5060}, {"loss": 1.2292, "grad_norm": 0.7317819595336914, "learning_rate": 0.0002, "epoch": 3.6699239956568945, "step": 5070}, {"loss": 1.1917, "grad_norm": 0.7406691908836365, "learning_rate": 0.0002, "epoch": 3.677162504524068, "step": 5080}, {"loss": 1.2949, "grad_norm": 0.9009454250335693, "learning_rate": 0.0002, "epoch": 3.6844010133912413, "step": 5090}, {"loss": 1.1528, "grad_norm": 0.8189385533332825, "learning_rate": 0.0002, "epoch": 3.6916395222584146, "step": 5100}, {"loss": 1.3408, "grad_norm": 1.0793628692626953, "learning_rate": 0.0002, "epoch": 3.698878031125588, "step": 5110}, {"loss": 1.2417, "grad_norm": 0.8593027591705322, "learning_rate": 0.0002, "epoch": 3.7061165399927614, "step": 5120}, {"loss": 1.2141, "grad_norm": 0.8481812477111816, "learning_rate": 0.0002, "epoch": 3.7133550488599347, "step": 5130}, {"loss": 1.125, "grad_norm": 0.6527451276779175, "learning_rate": 0.0002, "epoch": 3.720593557727108, "step": 5140}, {"loss": 1.1584, "grad_norm": 0.9220114350318909, "learning_rate": 0.0002, "epoch": 3.7278320665942815, "step": 5150}, {"loss": 1.2267, "grad_norm": 1.0842019319534302, "learning_rate": 0.0002, "epoch": 3.735070575461455, "step": 5160}, {"loss": 1.3083, "grad_norm": 0.965453565120697, "learning_rate": 0.0002, "epoch": 3.742309084328628, "step": 5170}, {"loss": 1.1772, "grad_norm": 0.9903319478034973, "learning_rate": 0.0002, "epoch": 3.7495475931958016, "step": 5180}, {"loss": 1.2515, "grad_norm": 0.7434818148612976, "learning_rate": 0.0002, "epoch": 3.756786102062975, "step": 5190}, {"loss": 1.2631, "grad_norm": 0.6717280745506287, "learning_rate": 0.0002, "epoch": 3.7640246109301483, "step": 5200}, {"loss": 1.2012, "grad_norm": 0.7754665613174438, "learning_rate": 0.0002, "epoch": 3.7712631197973217, "step": 5210}, {"loss": 1.305, "grad_norm": 1.028374433517456, "learning_rate": 0.0002, "epoch": 3.778501628664495, "step": 5220}, {"loss": 1.1866, "grad_norm": 0.6026996374130249, "learning_rate": 0.0002, "epoch": 3.7857401375316684, "step": 5230}, {"loss": 1.1901, "grad_norm": 0.6978490948677063, "learning_rate": 0.0002, "epoch": 3.7929786463988417, "step": 5240}, {"loss": 1.2576, "grad_norm": 0.7303446531295776, "learning_rate": 0.0002, "epoch": 3.800217155266015, "step": 5250}, {"loss": 1.3173, "grad_norm": 1.0734210014343262, "learning_rate": 0.0002, "epoch": 3.8074556641331885, "step": 5260}, {"loss": 1.1137, "grad_norm": 0.6383201479911804, "learning_rate": 0.0002, "epoch": 3.814694173000362, "step": 5270}, {"loss": 1.0904, "grad_norm": 0.7742630243301392, "learning_rate": 0.0002, "epoch": 3.821932681867535, "step": 5280}, {"loss": 1.2232, "grad_norm": 0.8477074503898621, "learning_rate": 0.0002, "epoch": 3.8291711907347086, "step": 5290}, {"loss": 1.2047, "grad_norm": 0.6675317883491516, "learning_rate": 0.0002, "epoch": 3.836409699601882, "step": 5300}, {"loss": 1.2275, "grad_norm": 0.7515445351600647, "learning_rate": 0.0002, "epoch": 3.8436482084690553, "step": 5310}, {"loss": 1.2569, "grad_norm": 1.1441220045089722, "learning_rate": 0.0002, "epoch": 3.8508867173362287, "step": 5320}, {"loss": 1.1512, "grad_norm": 0.7968795895576477, "learning_rate": 0.0002, "epoch": 3.858125226203402, "step": 5330}, {"loss": 1.232, "grad_norm": 0.7842824459075928, "learning_rate": 0.0002, "epoch": 3.8653637350705754, "step": 5340}, {"loss": 1.1847, "grad_norm": 0.8272225260734558, "learning_rate": 0.0002, "epoch": 3.8726022439377488, "step": 5350}, {"loss": 1.1381, "grad_norm": 0.8413397669792175, "learning_rate": 0.0002, "epoch": 3.879840752804922, "step": 5360}, {"loss": 1.2349, "grad_norm": 1.141764760017395, "learning_rate": 0.0002, "epoch": 3.8870792616720955, "step": 5370}, {"loss": 1.212, "grad_norm": 0.9826975464820862, "learning_rate": 0.0002, "epoch": 3.894317770539269, "step": 5380}, {"loss": 1.1833, "grad_norm": 0.8598255515098572, "learning_rate": 0.0002, "epoch": 3.9015562794064422, "step": 5390}, {"loss": 1.1247, "grad_norm": 0.6271058320999146, "learning_rate": 0.0002, "epoch": 3.9087947882736156, "step": 5400}, {"loss": 1.2212, "grad_norm": 0.6379870772361755, "learning_rate": 0.0002, "epoch": 3.916033297140789, "step": 5410}, {"loss": 1.2481, "grad_norm": 1.0313376188278198, "learning_rate": 0.0002, "epoch": 3.9232718060079623, "step": 5420}, {"loss": 1.1872, "grad_norm": 0.8220619559288025, "learning_rate": 0.0002, "epoch": 3.9305103148751357, "step": 5430}, {"loss": 1.2006, "grad_norm": 0.7576116919517517, "learning_rate": 0.0002, "epoch": 3.937748823742309, "step": 5440}, {"loss": 1.1969, "grad_norm": 1.226235032081604, "learning_rate": 0.0002, "epoch": 3.9449873326094824, "step": 5450}, {"loss": 1.2945, "grad_norm": 0.7979229688644409, "learning_rate": 0.0002, "epoch": 3.952225841476656, "step": 5460}, {"loss": 1.1922, "grad_norm": 0.9911929965019226, "learning_rate": 0.0002, "epoch": 3.959464350343829, "step": 5470}, {"loss": 1.0924, "grad_norm": 0.643738865852356, "learning_rate": 0.0002, "epoch": 3.9667028592110025, "step": 5480}, {"loss": 1.0607, "grad_norm": 0.682305634021759, "learning_rate": 0.0002, "epoch": 3.973941368078176, "step": 5490}, {"loss": 1.2908, "grad_norm": 1.18373441696167, "learning_rate": 0.0002, "epoch": 3.9811798769453492, "step": 5500}, {"loss": 1.0889, "grad_norm": 0.7190203070640564, "learning_rate": 0.0002, "epoch": 3.9884183858125226, "step": 5510}, {"loss": 1.2745, "grad_norm": 0.7516948580741882, "learning_rate": 0.0002, "epoch": 3.995656894679696, "step": 5520}]}
+{"epoch": 4.999638074556641, "step": 6907, "epoch_duration": 1329.0368676185608, "total_accumulated_duration": 6668.0360951423645, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 3048.73388671875}, "peak_memory_usage": {"GPU_0": 15079.2998046875}, "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-0.9-num-8828-sd-1/checkpoint-4144", "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": 4.7061, "grad_norm": 1.2523442506790161, "learning_rate": 0.0002, "epoch": 0.007238508867173362, "step": 10}, {"loss": 3.3493, "grad_norm": 1.8887330293655396, "learning_rate": 0.0002, "epoch": 0.014477017734346724, "step": 20}, {"loss": 2.7585, "grad_norm": 0.9668035507202148, "learning_rate": 0.0002, "epoch": 0.021715526601520086, "step": 30}, {"loss": 2.3699, "grad_norm": 2.9167306423187256, "learning_rate": 0.0002, "epoch": 0.028954035468693448, "step": 40}, {"loss": 2.2679, "grad_norm": 2.649867296218872, "learning_rate": 0.0002, "epoch": 0.036192544335866814, "step": 50}, {"loss": 2.2202, "grad_norm": 1.5120655298233032, "learning_rate": 0.0002, "epoch": 0.04343105320304017, "step": 60}, {"loss": 2.2026, "grad_norm": 0.7879868149757385, "learning_rate": 0.0002, "epoch": 0.05066956207021354, "step": 70}, {"loss": 1.9447, "grad_norm": 0.7616953253746033, "learning_rate": 0.0002, "epoch": 0.057908070937386896, "step": 80}, {"loss": 2.0112, "grad_norm": 1.8809149265289307, "learning_rate": 0.0002, "epoch": 0.06514657980456026, "step": 90}, {"loss": 1.8337, "grad_norm": 0.9294016361236572, "learning_rate": 0.0002, "epoch": 0.07238508867173363, "step": 100}, {"loss": 1.8419, "grad_norm": 0.7145281434059143, "learning_rate": 0.0002, "epoch": 0.07962359753890698, "step": 110}, {"loss": 2.0036, "grad_norm": 0.7564446330070496, "learning_rate": 0.0002, "epoch": 0.08686210640608034, "step": 120}, {"loss": 1.9306, "grad_norm": 1.1681925058364868, "learning_rate": 0.0002, "epoch": 0.09410061527325371, "step": 130}, {"loss": 1.7875, "grad_norm": 0.6708641648292542, "learning_rate": 0.0002, "epoch": 0.10133912414042708, "step": 140}, {"loss": 1.786, "grad_norm": 0.7625647783279419, "learning_rate": 0.0002, "epoch": 0.10857763300760044, "step": 150}, {"loss": 1.6687, "grad_norm": 0.8463464975357056, "learning_rate": 0.0002, "epoch": 0.11581614187477379, "step": 160}, {"loss": 1.6214, "grad_norm": 0.7502335906028748, "learning_rate": 0.0002, "epoch": 0.12305465074194716, "step": 170}, {"loss": 1.7433, "grad_norm": 0.6929958462715149, "learning_rate": 0.0002, "epoch": 0.13029315960912052, "step": 180}, {"loss": 1.6009, "grad_norm": 0.6798707842826843, "learning_rate": 0.0002, "epoch": 0.1375316684762939, "step": 190}, {"loss": 1.6208, "grad_norm": 0.7566508650779724, "learning_rate": 0.0002, "epoch": 0.14477017734346725, "step": 200}, {"loss": 1.5823, "grad_norm": 0.7196869850158691, "learning_rate": 0.0002, "epoch": 0.15200868621064062, "step": 210}, {"loss": 1.738, "grad_norm": 0.8401045799255371, "learning_rate": 0.0002, "epoch": 0.15924719507781396, "step": 220}, {"loss": 1.7574, "grad_norm": 0.8503773212432861, "learning_rate": 0.0002, "epoch": 0.16648570394498732, "step": 230}, {"loss": 1.7861, "grad_norm": 0.7183733582496643, "learning_rate": 0.0002, "epoch": 0.1737242128121607, "step": 240}, {"loss": 1.6693, "grad_norm": 0.7082605957984924, "learning_rate": 0.0002, "epoch": 0.18096272167933405, "step": 250}, {"loss": 1.619, "grad_norm": 0.9386326670646667, "learning_rate": 0.0002, "epoch": 0.18820123054650742, "step": 260}, {"loss": 1.6511, "grad_norm": 0.7332451939582825, "learning_rate": 0.0002, "epoch": 0.19543973941368079, "step": 270}, {"loss": 1.6353, "grad_norm": 0.7092869877815247, "learning_rate": 0.0002, "epoch": 0.20267824828085415, "step": 280}, {"loss": 1.5996, "grad_norm": 0.7256413698196411, "learning_rate": 0.0002, "epoch": 0.20991675714802752, "step": 290}, {"loss": 1.6754, "grad_norm": 0.6398681402206421, "learning_rate": 0.0002, "epoch": 0.21715526601520088, "step": 300}, {"loss": 1.397, "grad_norm": 0.6273287534713745, "learning_rate": 0.0002, "epoch": 0.22439377488237422, "step": 310}, {"loss": 1.5115, "grad_norm": 0.511648416519165, "learning_rate": 0.0002, "epoch": 0.23163228374954759, "step": 320}, {"loss": 1.5424, "grad_norm": 0.8677352070808411, "learning_rate": 0.0002, "epoch": 0.23887079261672095, "step": 330}, {"loss": 1.6779, "grad_norm": 0.6270743012428284, "learning_rate": 0.0002, "epoch": 0.24610930148389432, "step": 340}, {"loss": 1.626, "grad_norm": 0.7980281114578247, "learning_rate": 0.0002, "epoch": 0.2533478103510677, "step": 350}, {"loss": 1.5238, "grad_norm": 0.632486879825592, "learning_rate": 0.0002, "epoch": 0.26058631921824105, "step": 360}, {"loss": 1.5175, "grad_norm": 0.6527034640312195, "learning_rate": 0.0002, "epoch": 0.2678248280854144, "step": 370}, {"loss": 1.627, "grad_norm": 0.7672118544578552, "learning_rate": 0.0002, "epoch": 0.2750633369525878, "step": 380}, {"loss": 1.5605, "grad_norm": 0.6035117506980896, "learning_rate": 0.0002, "epoch": 0.28230184581976114, "step": 390}, {"loss": 1.4603, "grad_norm": 0.5955103039741516, "learning_rate": 0.0002, "epoch": 0.2895403546869345, "step": 400}, {"loss": 1.558, "grad_norm": 0.6015191674232483, "learning_rate": 0.0002, "epoch": 0.2967788635541079, "step": 410}, {"loss": 1.6091, "grad_norm": 0.6380982398986816, "learning_rate": 0.0002, "epoch": 0.30401737242128124, "step": 420}, {"loss": 1.5292, "grad_norm": 0.6707863211631775, "learning_rate": 0.0002, "epoch": 0.3112558812884546, "step": 430}, {"loss": 1.4426, "grad_norm": 0.7010176777839661, "learning_rate": 0.0002, "epoch": 0.3184943901556279, "step": 440}, {"loss": 1.5572, "grad_norm": 0.8263739943504333, "learning_rate": 0.0002, "epoch": 0.3257328990228013, "step": 450}, {"loss": 1.5188, "grad_norm": 0.7253276109695435, "learning_rate": 0.0002, "epoch": 0.33297140788997465, "step": 460}, {"loss": 1.584, "grad_norm": 0.5238934755325317, "learning_rate": 0.0002, "epoch": 0.340209916757148, "step": 470}, {"loss": 1.7035, "grad_norm": 0.7869495749473572, "learning_rate": 0.0002, "epoch": 0.3474484256243214, "step": 480}, {"loss": 1.5776, "grad_norm": 0.7485215663909912, "learning_rate": 0.0002, "epoch": 0.35468693449149474, "step": 490}, {"loss": 1.6274, "grad_norm": 0.5413193106651306, "learning_rate": 0.0002, "epoch": 0.3619254433586681, "step": 500}, {"loss": 1.7323, "grad_norm": 0.7615048885345459, "learning_rate": 0.0002, "epoch": 0.3691639522258415, "step": 510}, {"loss": 1.532, "grad_norm": 0.7685340046882629, "learning_rate": 0.0002, "epoch": 0.37640246109301484, "step": 520}, {"loss": 1.6312, "grad_norm": 0.6379081010818481, "learning_rate": 0.0002, "epoch": 0.3836409699601882, "step": 530}, {"loss": 1.5645, "grad_norm": 0.7946939468383789, "learning_rate": 0.0002, "epoch": 0.39087947882736157, "step": 540}, {"loss": 1.4001, "grad_norm": 0.6287278532981873, "learning_rate": 0.0002, "epoch": 0.39811798769453494, "step": 550}, {"loss": 1.5982, "grad_norm": 0.6811642646789551, "learning_rate": 0.0002, "epoch": 0.4053564965617083, "step": 560}, {"loss": 1.4953, "grad_norm": 0.671073317527771, "learning_rate": 0.0002, "epoch": 0.41259500542888167, "step": 570}, {"loss": 1.6753, "grad_norm": 0.6313900351524353, "learning_rate": 0.0002, "epoch": 0.41983351429605503, "step": 580}, {"loss": 1.546, "grad_norm": 0.5291772484779358, "learning_rate": 0.0002, "epoch": 0.4270720231632284, "step": 590}, {"loss": 1.5441, "grad_norm": 0.62503582239151, "learning_rate": 0.0002, "epoch": 0.43431053203040176, "step": 600}, {"loss": 1.6276, "grad_norm": 0.5777305364608765, "learning_rate": 0.0002, "epoch": 0.4415490408975751, "step": 610}, {"loss": 1.4758, "grad_norm": 0.7013497352600098, "learning_rate": 0.0002, "epoch": 0.44878754976474844, "step": 620}, {"loss": 1.4029, "grad_norm": 0.8044822216033936, "learning_rate": 0.0002, "epoch": 0.4560260586319218, "step": 630}, {"loss": 1.7195, "grad_norm": 0.672531247138977, "learning_rate": 0.0002, "epoch": 0.46326456749909517, "step": 640}, {"loss": 1.614, "grad_norm": 0.6233910322189331, "learning_rate": 0.0002, "epoch": 0.47050307636626854, "step": 650}, {"loss": 1.6041, "grad_norm": 0.651524543762207, "learning_rate": 0.0002, "epoch": 0.4777415852334419, "step": 660}, {"loss": 1.5842, "grad_norm": 0.7213939428329468, "learning_rate": 0.0002, "epoch": 0.48498009410061527, "step": 670}, {"loss": 1.5453, "grad_norm": 0.6541454792022705, "learning_rate": 0.0002, "epoch": 0.49221860296778863, "step": 680}, {"loss": 1.662, "grad_norm": 0.6568936109542847, "learning_rate": 0.0002, "epoch": 0.499457111834962, "step": 690}, {"loss": 1.624, "grad_norm": 0.7176415324211121, "learning_rate": 0.0002, "epoch": 0.5066956207021354, "step": 700}, {"loss": 1.6099, "grad_norm": 0.6553855538368225, "learning_rate": 0.0002, "epoch": 0.5139341295693087, "step": 710}, {"loss": 1.5508, "grad_norm": 0.5654335618019104, "learning_rate": 0.0002, "epoch": 0.5211726384364821, "step": 720}, {"loss": 1.392, "grad_norm": 0.5671001672744751, "learning_rate": 0.0002, "epoch": 0.5284111473036555, "step": 730}, {"loss": 1.388, "grad_norm": 0.7914412021636963, "learning_rate": 0.0002, "epoch": 0.5356496561708288, "step": 740}, {"loss": 1.5931, "grad_norm": 0.6172138452529907, "learning_rate": 0.0002, "epoch": 0.5428881650380022, "step": 750}, {"loss": 1.4018, "grad_norm": 0.6132623553276062, "learning_rate": 0.0002, "epoch": 0.5501266739051756, "step": 760}, {"loss": 1.513, "grad_norm": 0.654000461101532, "learning_rate": 0.0002, "epoch": 0.5573651827723489, "step": 770}, {"loss": 1.5035, "grad_norm": 0.5691370964050293, "learning_rate": 0.0002, "epoch": 0.5646036916395223, "step": 780}, {"loss": 1.65, "grad_norm": 0.7922580242156982, "learning_rate": 0.0002, "epoch": 0.5718422005066957, "step": 790}, {"loss": 1.4521, "grad_norm": 0.6831880211830139, "learning_rate": 0.0002, "epoch": 0.579080709373869, "step": 800}, {"loss": 1.4734, "grad_norm": 0.6740124821662903, "learning_rate": 0.0002, "epoch": 0.5863192182410424, "step": 810}, {"loss": 1.6498, "grad_norm": 1.380016803741455, "learning_rate": 0.0002, "epoch": 0.5935577271082157, "step": 820}, {"loss": 1.4642, "grad_norm": 0.6552878022193909, "learning_rate": 0.0002, "epoch": 0.6007962359753891, "step": 830}, {"loss": 1.6271, "grad_norm": 0.6649535298347473, "learning_rate": 0.0002, "epoch": 0.6080347448425625, "step": 840}, {"loss": 1.5886, "grad_norm": 0.561738133430481, "learning_rate": 0.0002, "epoch": 0.6152732537097358, "step": 850}, {"loss": 1.5364, "grad_norm": 0.6133047938346863, "learning_rate": 0.0002, "epoch": 0.6225117625769092, "step": 860}, {"loss": 1.3489, "grad_norm": 0.559843122959137, "learning_rate": 0.0002, "epoch": 0.6297502714440825, "step": 870}, {"loss": 1.4878, "grad_norm": 0.6117811799049377, "learning_rate": 0.0002, "epoch": 0.6369887803112558, "step": 880}, {"loss": 1.56, "grad_norm": 0.6209776401519775, "learning_rate": 0.0002, "epoch": 0.6442272891784292, "step": 890}, {"loss": 1.6747, "grad_norm": 0.6234082579612732, "learning_rate": 0.0002, "epoch": 0.6514657980456026, "step": 900}, {"loss": 1.6963, "grad_norm": 0.7623258233070374, "learning_rate": 0.0002, "epoch": 0.6587043069127759, "step": 910}, {"loss": 1.2424, "grad_norm": 0.6148061752319336, "learning_rate": 0.0002, "epoch": 0.6659428157799493, "step": 920}, {"loss": 1.4319, "grad_norm": 0.6682973504066467, "learning_rate": 0.0002, "epoch": 0.6731813246471227, "step": 930}, {"loss": 1.5377, "grad_norm": 0.5513041615486145, "learning_rate": 0.0002, "epoch": 0.680419833514296, "step": 940}, {"loss": 1.3991, "grad_norm": 0.5197525024414062, "learning_rate": 0.0002, "epoch": 0.6876583423814694, "step": 950}, {"loss": 1.4398, "grad_norm": 0.6490758061408997, "learning_rate": 0.0002, "epoch": 0.6948968512486428, "step": 960}, {"loss": 1.5251, "grad_norm": 0.6450682878494263, "learning_rate": 0.0002, "epoch": 0.7021353601158161, "step": 970}, {"loss": 1.5417, "grad_norm": 0.6203766465187073, "learning_rate": 0.0002, "epoch": 0.7093738689829895, "step": 980}, {"loss": 1.4575, "grad_norm": 0.6023609638214111, "learning_rate": 0.0002, "epoch": 0.7166123778501629, "step": 990}, {"loss": 1.4973, "grad_norm": 0.5765255093574524, "learning_rate": 0.0002, "epoch": 0.7238508867173362, "step": 1000}, {"loss": 1.483, "grad_norm": 0.6650075316429138, "learning_rate": 0.0002, "epoch": 0.7310893955845096, "step": 1010}, {"loss": 1.5959, "grad_norm": 0.5610854029655457, "learning_rate": 0.0002, "epoch": 0.738327904451683, "step": 1020}, {"loss": 1.5248, "grad_norm": 0.7072813510894775, "learning_rate": 0.0002, "epoch": 0.7455664133188563, "step": 1030}, {"loss": 1.5776, "grad_norm": 0.6815407872200012, "learning_rate": 0.0002, "epoch": 0.7528049221860297, "step": 1040}, {"loss": 1.4577, "grad_norm": 0.7932390570640564, "learning_rate": 0.0002, "epoch": 0.760043431053203, "step": 1050}, {"loss": 1.4515, "grad_norm": 0.5798183083534241, "learning_rate": 0.0002, "epoch": 0.7672819399203764, "step": 1060}, {"loss": 1.5053, "grad_norm": 0.7898504137992859, "learning_rate": 0.0002, "epoch": 0.7745204487875498, "step": 1070}, {"loss": 1.4776, "grad_norm": 0.4983280301094055, "learning_rate": 0.0002, "epoch": 0.7817589576547231, "step": 1080}, {"loss": 1.5007, "grad_norm": 0.691403329372406, "learning_rate": 0.0002, "epoch": 0.7889974665218965, "step": 1090}, {"loss": 1.5153, "grad_norm": 0.5394481420516968, "learning_rate": 0.0002, "epoch": 0.7962359753890699, "step": 1100}, {"loss": 1.6892, "grad_norm": 0.5136822462081909, "learning_rate": 0.0002, "epoch": 0.8034744842562432, "step": 1110}, {"loss": 1.4902, "grad_norm": 0.6828126907348633, "learning_rate": 0.0002, "epoch": 0.8107129931234166, "step": 1120}, {"loss": 1.4346, "grad_norm": 0.6799656748771667, "learning_rate": 0.0002, "epoch": 0.81795150199059, "step": 1130}, {"loss": 1.2678, "grad_norm": 0.5428406000137329, "learning_rate": 0.0002, "epoch": 0.8251900108577633, "step": 1140}, {"loss": 1.4072, "grad_norm": 0.4811290502548218, "learning_rate": 0.0002, "epoch": 0.8324285197249367, "step": 1150}, {"loss": 1.4512, "grad_norm": 0.5519434809684753, "learning_rate": 0.0002, "epoch": 0.8396670285921101, "step": 1160}, {"loss": 1.4072, "grad_norm": 0.9748060703277588, "learning_rate": 0.0002, "epoch": 0.8469055374592834, "step": 1170}, {"loss": 1.4309, "grad_norm": 0.712609589099884, "learning_rate": 0.0002, "epoch": 0.8541440463264568, "step": 1180}, {"loss": 1.434, "grad_norm": 0.6866157054901123, "learning_rate": 0.0002, "epoch": 0.8613825551936302, "step": 1190}, {"loss": 1.3704, "grad_norm": 0.5068854093551636, "learning_rate": 0.0002, "epoch": 0.8686210640608035, "step": 1200}, {"loss": 1.5601, "grad_norm": 0.6333245038986206, "learning_rate": 0.0002, "epoch": 0.8758595729279768, "step": 1210}, {"loss": 1.4636, "grad_norm": 0.6424421072006226, "learning_rate": 0.0002, "epoch": 0.8830980817951501, "step": 1220}, {"loss": 1.4186, "grad_norm": 0.4771921932697296, "learning_rate": 0.0002, "epoch": 0.8903365906623235, "step": 1230}, {"loss": 1.6323, "grad_norm": 0.5191764235496521, "learning_rate": 0.0002, "epoch": 0.8975750995294969, "step": 1240}, {"loss": 1.6105, "grad_norm": 0.756222128868103, "learning_rate": 0.0002, "epoch": 0.9048136083966702, "step": 1250}, {"loss": 1.4396, "grad_norm": 0.623823881149292, "learning_rate": 0.0002, "epoch": 0.9120521172638436, "step": 1260}, {"loss": 1.3097, "grad_norm": 0.8166571259498596, "learning_rate": 0.0002, "epoch": 0.919290626131017, "step": 1270}, {"loss": 1.4625, "grad_norm": 0.6059346795082092, "learning_rate": 0.0002, "epoch": 0.9265291349981903, "step": 1280}, {"loss": 1.3555, "grad_norm": 0.5842690467834473, "learning_rate": 0.0002, "epoch": 0.9337676438653637, "step": 1290}, {"loss": 1.5859, "grad_norm": 0.7649800777435303, "learning_rate": 0.0002, "epoch": 0.9410061527325371, "step": 1300}, {"loss": 1.5915, "grad_norm": 0.6420919895172119, "learning_rate": 0.0002, "epoch": 0.9482446615997104, "step": 1310}, {"loss": 1.453, "grad_norm": 0.7011452913284302, "learning_rate": 0.0002, "epoch": 0.9554831704668838, "step": 1320}, {"loss": 1.6766, "grad_norm": 0.5783746242523193, "learning_rate": 0.0002, "epoch": 0.9627216793340572, "step": 1330}, {"loss": 1.6308, "grad_norm": 0.5973192453384399, "learning_rate": 0.0002, "epoch": 0.9699601882012305, "step": 1340}, {"loss": 1.5901, "grad_norm": 0.6181833744049072, "learning_rate": 0.0002, "epoch": 0.9771986970684039, "step": 1350}, {"loss": 1.5258, "grad_norm": 0.5563396215438843, "learning_rate": 0.0002, "epoch": 0.9844372059355773, "step": 1360}, {"loss": 1.4508, "grad_norm": 0.45723360776901245, "learning_rate": 0.0002, "epoch": 0.9916757148027506, "step": 1370}, {"loss": 1.3291, "grad_norm": 0.5947498679161072, "learning_rate": 0.0002, "epoch": 0.998914223669924, "step": 1380}, {"eval_loss": 1.480796456336975, "eval_runtime": 27.3103, "eval_samples_per_second": 15.965, "eval_steps_per_second": 2.014, "epoch": 0.9996380745566413, "step": 1381}, {"loss": 1.3057, "grad_norm": 0.5599952936172485, "learning_rate": 0.0002, "epoch": 1.0061527325370974, "step": 1390}, {"loss": 1.4991, "grad_norm": 0.5932008028030396, "learning_rate": 0.0002, "epoch": 1.0133912414042707, "step": 1400}, {"loss": 1.4506, "grad_norm": 0.6194121837615967, "learning_rate": 0.0002, "epoch": 1.020629750271444, "step": 1410}, {"loss": 1.5966, "grad_norm": 0.6995621919631958, "learning_rate": 0.0002, "epoch": 1.0278682591386175, "step": 1420}, {"loss": 1.4153, "grad_norm": 0.7905810475349426, "learning_rate": 0.0002, "epoch": 1.0351067680057908, "step": 1430}, {"loss": 1.4414, "grad_norm": 0.7221615314483643, "learning_rate": 0.0002, "epoch": 1.0423452768729642, "step": 1440}, {"loss": 1.3859, "grad_norm": 0.6170642375946045, "learning_rate": 0.0002, "epoch": 1.0495837857401376, "step": 1450}, {"loss": 1.3806, "grad_norm": 0.5844094753265381, "learning_rate": 0.0002, "epoch": 1.056822294607311, "step": 1460}, {"loss": 1.4871, "grad_norm": 0.7731822729110718, "learning_rate": 0.0002, "epoch": 1.0640608034744843, "step": 1470}, {"loss": 1.4286, "grad_norm": 0.4554748237133026, "learning_rate": 0.0002, "epoch": 1.0712993123416577, "step": 1480}, {"loss": 1.3977, "grad_norm": 0.6923259496688843, "learning_rate": 0.0002, "epoch": 1.078537821208831, "step": 1490}, {"loss": 1.3936, "grad_norm": 0.6008219122886658, "learning_rate": 0.0002, "epoch": 1.0857763300760044, "step": 1500}, {"loss": 1.4821, "grad_norm": 0.6450045704841614, "learning_rate": 0.0002, "epoch": 1.0930148389431777, "step": 1510}, {"loss": 1.3295, "grad_norm": 0.7833753824234009, "learning_rate": 0.0002, "epoch": 1.1002533478103511, "step": 1520}, {"loss": 1.3424, "grad_norm": 0.5076758861541748, "learning_rate": 0.0002, "epoch": 1.1074918566775245, "step": 1530}, {"loss": 1.4043, "grad_norm": 0.5661332011222839, "learning_rate": 0.0002, "epoch": 1.1147303655446978, "step": 1540}, {"loss": 1.4963, "grad_norm": 0.6526919603347778, "learning_rate": 0.0002, "epoch": 1.1219688744118712, "step": 1550}, {"loss": 1.3671, "grad_norm": 0.5613082647323608, "learning_rate": 0.0002, "epoch": 1.1292073832790446, "step": 1560}, {"loss": 1.4458, "grad_norm": 0.6113885641098022, "learning_rate": 0.0002, "epoch": 1.136445892146218, "step": 1570}, {"loss": 1.3552, "grad_norm": 0.6732510328292847, "learning_rate": 0.0002, "epoch": 1.1436844010133913, "step": 1580}, {"loss": 1.3114, "grad_norm": 0.6146392226219177, "learning_rate": 0.0002, "epoch": 1.1509229098805647, "step": 1590}, {"loss": 1.411, "grad_norm": 0.6766974329948425, "learning_rate": 0.0002, "epoch": 1.158161418747738, "step": 1600}, {"loss": 1.2401, "grad_norm": 0.7621957659721375, "learning_rate": 0.0002, "epoch": 1.1653999276149114, "step": 1610}, {"loss": 1.3758, "grad_norm": 0.6959581971168518, "learning_rate": 0.0002, "epoch": 1.1726384364820848, "step": 1620}, {"loss": 1.382, "grad_norm": 0.6691278219223022, "learning_rate": 0.0002, "epoch": 1.1798769453492581, "step": 1630}, {"loss": 1.4147, "grad_norm": 0.4927774965763092, "learning_rate": 0.0002, "epoch": 1.1871154542164315, "step": 1640}, {"loss": 1.449, "grad_norm": 0.7724234461784363, "learning_rate": 0.0002, "epoch": 1.1943539630836049, "step": 1650}, {"loss": 1.4778, "grad_norm": 0.6817787885665894, "learning_rate": 0.0002, "epoch": 1.2015924719507782, "step": 1660}, {"loss": 1.3776, "grad_norm": 0.6500699520111084, "learning_rate": 0.0002, "epoch": 1.2088309808179516, "step": 1670}, {"loss": 1.3875, "grad_norm": 0.5703568458557129, "learning_rate": 0.0002, "epoch": 1.216069489685125, "step": 1680}, {"loss": 1.4735, "grad_norm": 0.6261579990386963, "learning_rate": 0.0002, "epoch": 1.2233079985522983, "step": 1690}, {"loss": 1.3898, "grad_norm": 0.651713490486145, "learning_rate": 0.0002, "epoch": 1.2305465074194717, "step": 1700}, {"loss": 1.4002, "grad_norm": 0.684399425983429, "learning_rate": 0.0002, "epoch": 1.237785016286645, "step": 1710}, {"loss": 1.5027, "grad_norm": 0.6996857523918152, "learning_rate": 0.0002, "epoch": 1.2450235251538184, "step": 1720}, {"loss": 1.3326, "grad_norm": 0.7102537751197815, "learning_rate": 0.0002, "epoch": 1.2522620340209918, "step": 1730}, {"loss": 1.3675, "grad_norm": 0.45809897780418396, "learning_rate": 0.0002, "epoch": 1.2595005428881652, "step": 1740}, {"loss": 1.4175, "grad_norm": 0.6377046704292297, "learning_rate": 0.0002, "epoch": 1.2667390517553385, "step": 1750}, {"loss": 1.3479, "grad_norm": 0.6965704560279846, "learning_rate": 0.0002, "epoch": 1.2739775606225119, "step": 1760}, {"loss": 1.5647, "grad_norm": 0.5688214302062988, "learning_rate": 0.0002, "epoch": 1.2812160694896852, "step": 1770}, {"loss": 1.3967, "grad_norm": 0.6384190320968628, "learning_rate": 0.0002, "epoch": 1.2884545783568586, "step": 1780}, {"loss": 1.3671, "grad_norm": 0.5629363656044006, "learning_rate": 0.0002, "epoch": 1.295693087224032, "step": 1790}, {"loss": 1.2292, "grad_norm": 0.6148255467414856, "learning_rate": 0.0002, "epoch": 1.3029315960912053, "step": 1800}, {"loss": 1.5806, "grad_norm": 0.655580997467041, "learning_rate": 0.0002, "epoch": 1.3101701049583787, "step": 1810}, {"loss": 1.2398, "grad_norm": 0.5642657279968262, "learning_rate": 0.0002, "epoch": 1.3174086138255519, "step": 1820}, {"loss": 1.3246, "grad_norm": 0.59607994556427, "learning_rate": 0.0002, "epoch": 1.3246471226927252, "step": 1830}, {"loss": 1.3274, "grad_norm": 0.5564199090003967, "learning_rate": 0.0002, "epoch": 1.3318856315598986, "step": 1840}, {"loss": 1.5834, "grad_norm": 0.6949955821037292, "learning_rate": 0.0002, "epoch": 1.339124140427072, "step": 1850}, {"loss": 1.4722, "grad_norm": 0.7036856412887573, "learning_rate": 0.0002, "epoch": 1.3463626492942453, "step": 1860}, {"loss": 1.333, "grad_norm": 0.722062885761261, "learning_rate": 0.0002, "epoch": 1.3536011581614187, "step": 1870}, {"loss": 1.4044, "grad_norm": 0.6098677515983582, "learning_rate": 0.0002, "epoch": 1.360839667028592, "step": 1880}, {"loss": 1.6217, "grad_norm": 0.5376402735710144, "learning_rate": 0.0002, "epoch": 1.3680781758957654, "step": 1890}, {"loss": 1.5071, "grad_norm": 0.6974610090255737, "learning_rate": 0.0002, "epoch": 1.3753166847629388, "step": 1900}, {"loss": 1.5854, "grad_norm": 0.6520763635635376, "learning_rate": 0.0002, "epoch": 1.3825551936301121, "step": 1910}, {"loss": 1.4271, "grad_norm": 0.6604374647140503, "learning_rate": 0.0002, "epoch": 1.3897937024972855, "step": 1920}, {"loss": 1.419, "grad_norm": 0.7364398241043091, "learning_rate": 0.0002, "epoch": 1.3970322113644589, "step": 1930}, {"loss": 1.4585, "grad_norm": 0.6849475502967834, "learning_rate": 0.0002, "epoch": 1.4042707202316322, "step": 1940}, {"loss": 1.5577, "grad_norm": 0.6562670469284058, "learning_rate": 0.0002, "epoch": 1.4115092290988056, "step": 1950}, {"loss": 1.4725, "grad_norm": 0.5695616006851196, "learning_rate": 0.0002, "epoch": 1.418747737965979, "step": 1960}, {"loss": 1.3088, "grad_norm": 0.5244464874267578, "learning_rate": 0.0002, "epoch": 1.4259862468331523, "step": 1970}, {"loss": 1.5069, "grad_norm": 0.6347293257713318, "learning_rate": 0.0002, "epoch": 1.4332247557003257, "step": 1980}, {"loss": 1.3502, "grad_norm": 0.5528361201286316, "learning_rate": 0.0002, "epoch": 1.440463264567499, "step": 1990}, {"loss": 1.3978, "grad_norm": 0.6987585425376892, "learning_rate": 0.0002, "epoch": 1.4477017734346724, "step": 2000}, {"loss": 1.4262, "grad_norm": 0.6568987369537354, "learning_rate": 0.0002, "epoch": 1.4549402823018458, "step": 2010}, {"loss": 1.4175, "grad_norm": 0.7665994763374329, "learning_rate": 0.0002, "epoch": 1.4621787911690192, "step": 2020}, {"loss": 1.244, "grad_norm": 0.5127707123756409, "learning_rate": 0.0002, "epoch": 1.4694173000361925, "step": 2030}, {"loss": 1.3699, "grad_norm": 0.5406824946403503, "learning_rate": 0.0002, "epoch": 1.476655808903366, "step": 2040}, {"loss": 1.3353, "grad_norm": 0.5990166664123535, "learning_rate": 0.0002, "epoch": 1.4838943177705393, "step": 2050}, {"loss": 1.2454, "grad_norm": 0.6186193823814392, "learning_rate": 0.0002, "epoch": 1.4911328266377126, "step": 2060}, {"loss": 1.428, "grad_norm": 0.6154307126998901, "learning_rate": 0.0002, "epoch": 1.498371335504886, "step": 2070}, {"loss": 1.4528, "grad_norm": 0.5606056451797485, "learning_rate": 0.0002, "epoch": 1.5056098443720594, "step": 2080}, {"loss": 1.2405, "grad_norm": 0.5006417036056519, "learning_rate": 0.0002, "epoch": 1.5128483532392327, "step": 2090}, {"loss": 1.4258, "grad_norm": 0.5968486070632935, "learning_rate": 0.0002, "epoch": 1.520086862106406, "step": 2100}, {"loss": 1.2752, "grad_norm": 0.5835496187210083, "learning_rate": 0.0002, "epoch": 1.5273253709735795, "step": 2110}, {"loss": 1.5443, "grad_norm": 0.6753535270690918, "learning_rate": 0.0002, "epoch": 1.5345638798407528, "step": 2120}, {"loss": 1.2139, "grad_norm": 0.7299720644950867, "learning_rate": 0.0002, "epoch": 1.5418023887079262, "step": 2130}, {"loss": 1.2364, "grad_norm": 0.5105988383293152, "learning_rate": 0.0002, "epoch": 1.5490408975750996, "step": 2140}, {"loss": 1.4528, "grad_norm": 0.5675431489944458, "learning_rate": 0.0002, "epoch": 1.556279406442273, "step": 2150}, {"loss": 1.4563, "grad_norm": 0.6246723532676697, "learning_rate": 0.0002, "epoch": 1.5635179153094463, "step": 2160}, {"loss": 1.5255, "grad_norm": 0.7291720509529114, "learning_rate": 0.0002, "epoch": 1.5707564241766196, "step": 2170}, {"loss": 1.5432, "grad_norm": 0.678114116191864, "learning_rate": 0.0002, "epoch": 1.577994933043793, "step": 2180}, {"loss": 1.5212, "grad_norm": 0.5136260986328125, "learning_rate": 0.0002, "epoch": 1.5852334419109664, "step": 2190}, {"loss": 1.3271, "grad_norm": 0.6359935998916626, "learning_rate": 0.0002, "epoch": 1.5924719507781397, "step": 2200}, {"loss": 1.4038, "grad_norm": 0.7650278806686401, "learning_rate": 0.0002, "epoch": 1.599710459645313, "step": 2210}, {"loss": 1.5478, "grad_norm": 0.7256110906600952, "learning_rate": 0.0002, "epoch": 1.6069489685124865, "step": 2220}, {"loss": 1.4387, "grad_norm": 0.688689649105072, "learning_rate": 0.0002, "epoch": 1.6141874773796598, "step": 2230}, {"loss": 1.4096, "grad_norm": 0.6045311093330383, "learning_rate": 0.0002, "epoch": 1.6214259862468332, "step": 2240}, {"loss": 1.4097, "grad_norm": 0.7064604163169861, "learning_rate": 0.0002, "epoch": 1.6286644951140063, "step": 2250}, {"loss": 1.3477, "grad_norm": 0.5309562087059021, "learning_rate": 0.0002, "epoch": 1.6359030039811797, "step": 2260}, {"loss": 1.4022, "grad_norm": 0.5687053203582764, "learning_rate": 0.0002, "epoch": 1.643141512848353, "step": 2270}, {"loss": 1.2977, "grad_norm": 0.535872757434845, "learning_rate": 0.0002, "epoch": 1.6503800217155264, "step": 2280}, {"loss": 1.3844, "grad_norm": 0.5502381920814514, "learning_rate": 0.0002, "epoch": 1.6576185305826998, "step": 2290}, {"loss": 1.3764, "grad_norm": 0.6158602237701416, "learning_rate": 0.0002, "epoch": 1.6648570394498732, "step": 2300}, {"loss": 1.3515, "grad_norm": 0.5804675817489624, "learning_rate": 0.0002, "epoch": 1.6720955483170465, "step": 2310}, {"loss": 1.2532, "grad_norm": 0.600742757320404, "learning_rate": 0.0002, "epoch": 1.67933405718422, "step": 2320}, {"loss": 1.477, "grad_norm": 0.7101941108703613, "learning_rate": 0.0002, "epoch": 1.6865725660513933, "step": 2330}, {"loss": 1.4849, "grad_norm": 0.7507809996604919, "learning_rate": 0.0002, "epoch": 1.6938110749185666, "step": 2340}, {"loss": 1.2703, "grad_norm": 0.768502414226532, "learning_rate": 0.0002, "epoch": 1.70104958378574, "step": 2350}, {"loss": 1.3332, "grad_norm": 0.4801851212978363, "learning_rate": 0.0002, "epoch": 1.7082880926529134, "step": 2360}, {"loss": 1.4158, "grad_norm": 0.5322122573852539, "learning_rate": 0.0002, "epoch": 1.7155266015200867, "step": 2370}, {"loss": 1.4136, "grad_norm": 0.587661862373352, "learning_rate": 0.0002, "epoch": 1.72276511038726, "step": 2380}, {"loss": 1.3771, "grad_norm": 0.6073525547981262, "learning_rate": 0.0002, "epoch": 1.7300036192544335, "step": 2390}, {"loss": 1.2754, "grad_norm": 0.6950460076332092, "learning_rate": 0.0002, "epoch": 1.7372421281216068, "step": 2400}, {"loss": 1.3858, "grad_norm": 0.5981102585792542, "learning_rate": 0.0002, "epoch": 1.7444806369887802, "step": 2410}, {"loss": 1.4075, "grad_norm": 0.544570803642273, "learning_rate": 0.0002, "epoch": 1.7517191458559536, "step": 2420}, {"loss": 1.3861, "grad_norm": 0.5304399728775024, "learning_rate": 0.0002, "epoch": 1.758957654723127, "step": 2430}, {"loss": 1.4244, "grad_norm": 0.7921594977378845, "learning_rate": 0.0002, "epoch": 1.7661961635903003, "step": 2440}, {"loss": 1.3053, "grad_norm": 0.6084808707237244, "learning_rate": 0.0002, "epoch": 1.7734346724574737, "step": 2450}, {"loss": 1.3781, "grad_norm": 0.8844701051712036, "learning_rate": 0.0002, "epoch": 1.780673181324647, "step": 2460}, {"loss": 1.3227, "grad_norm": 0.5729258060455322, "learning_rate": 0.0002, "epoch": 1.7879116901918204, "step": 2470}, {"loss": 1.3422, "grad_norm": 0.6303611993789673, "learning_rate": 0.0002, "epoch": 1.7951501990589938, "step": 2480}, {"loss": 1.3926, "grad_norm": 0.5627942085266113, "learning_rate": 0.0002, "epoch": 1.8023887079261671, "step": 2490}, {"loss": 1.3816, "grad_norm": 0.6724274158477783, "learning_rate": 0.0002, "epoch": 1.8096272167933405, "step": 2500}, {"loss": 1.2951, "grad_norm": 0.5030826330184937, "learning_rate": 0.0002, "epoch": 1.8168657256605139, "step": 2510}, {"loss": 1.2839, "grad_norm": 0.5504099130630493, "learning_rate": 0.0002, "epoch": 1.8241042345276872, "step": 2520}, {"loss": 1.4264, "grad_norm": 0.6338945627212524, "learning_rate": 0.0002, "epoch": 1.8313427433948606, "step": 2530}, {"loss": 1.563, "grad_norm": 0.5902037620544434, "learning_rate": 0.0002, "epoch": 1.838581252262034, "step": 2540}, {"loss": 1.2961, "grad_norm": 0.48814457654953003, "learning_rate": 0.0002, "epoch": 1.8458197611292073, "step": 2550}, {"loss": 1.466, "grad_norm": 0.6216312646865845, "learning_rate": 0.0002, "epoch": 1.8530582699963807, "step": 2560}, {"loss": 1.5123, "grad_norm": 0.635603666305542, "learning_rate": 0.0002, "epoch": 1.860296778863554, "step": 2570}, {"loss": 1.372, "grad_norm": 0.6938216090202332, "learning_rate": 0.0002, "epoch": 1.8675352877307274, "step": 2580}, {"loss": 1.5011, "grad_norm": 0.599557638168335, "learning_rate": 0.0002, "epoch": 1.8747737965979008, "step": 2590}, {"loss": 1.2714, "grad_norm": 0.564424455165863, "learning_rate": 0.0002, "epoch": 1.8820123054650741, "step": 2600}, {"loss": 1.3403, "grad_norm": 0.5430700182914734, "learning_rate": 0.0002, "epoch": 1.8892508143322475, "step": 2610}, {"loss": 1.4347, "grad_norm": 0.6150169372558594, "learning_rate": 0.0002, "epoch": 1.8964893231994209, "step": 2620}, {"loss": 1.2474, "grad_norm": 0.48159119486808777, "learning_rate": 0.0002, "epoch": 1.9037278320665942, "step": 2630}, {"loss": 1.3716, "grad_norm": 0.5608997941017151, "learning_rate": 0.0002, "epoch": 1.9109663409337676, "step": 2640}, {"loss": 1.5787, "grad_norm": 0.6454501748085022, "learning_rate": 0.0002, "epoch": 1.918204849800941, "step": 2650}, {"loss": 1.3238, "grad_norm": 0.5458073616027832, "learning_rate": 0.0002, "epoch": 1.9254433586681143, "step": 2660}, {"loss": 1.3208, "grad_norm": 0.5328490734100342, "learning_rate": 0.0002, "epoch": 1.9326818675352877, "step": 2670}, {"loss": 1.4971, "grad_norm": 0.6444696187973022, "learning_rate": 0.0002, "epoch": 1.939920376402461, "step": 2680}, {"loss": 1.5387, "grad_norm": 0.7126023769378662, "learning_rate": 0.0002, "epoch": 1.9471588852696344, "step": 2690}, {"loss": 1.3637, "grad_norm": 0.5164045095443726, "learning_rate": 0.0002, "epoch": 1.9543973941368078, "step": 2700}, {"loss": 1.5303, "grad_norm": 0.5347061157226562, "learning_rate": 0.0002, "epoch": 1.9616359030039812, "step": 2710}, {"loss": 1.2815, "grad_norm": 0.5297950506210327, "learning_rate": 0.0002, "epoch": 1.9688744118711545, "step": 2720}, {"loss": 1.3566, "grad_norm": 0.6537790298461914, "learning_rate": 0.0002, "epoch": 1.976112920738328, "step": 2730}, {"loss": 1.332, "grad_norm": 0.5536222457885742, "learning_rate": 0.0002, "epoch": 1.9833514296055013, "step": 2740}, {"loss": 1.3333, "grad_norm": 0.4856105446815491, "learning_rate": 0.0002, "epoch": 1.9905899384726746, "step": 2750}, {"loss": 1.3521, "grad_norm": 0.6642730832099915, "learning_rate": 0.0002, "epoch": 1.997828447339848, "step": 2760}, {"eval_loss": 1.4366681575775146, "eval_runtime": 27.3729, "eval_samples_per_second": 15.928, "eval_steps_per_second": 2.009, "epoch": 2.0, "step": 2763}, {"loss": 1.4322, "grad_norm": 0.740253210067749, "learning_rate": 0.0002, "epoch": 2.0050669562070214, "step": 2770}, {"loss": 1.277, "grad_norm": 0.5826276540756226, "learning_rate": 0.0002, "epoch": 2.0123054650741947, "step": 2780}, {"loss": 1.2424, "grad_norm": 0.607356071472168, "learning_rate": 0.0002, "epoch": 2.019543973941368, "step": 2790}, {"loss": 1.2601, "grad_norm": 0.5918063521385193, "learning_rate": 0.0002, "epoch": 2.0267824828085415, "step": 2800}, {"loss": 1.3715, "grad_norm": 0.5610089898109436, "learning_rate": 0.0002, "epoch": 2.034020991675715, "step": 2810}, {"loss": 1.2092, "grad_norm": 0.5869926810264587, "learning_rate": 0.0002, "epoch": 2.041259500542888, "step": 2820}, {"loss": 1.1929, "grad_norm": 0.5753467679023743, "learning_rate": 0.0002, "epoch": 2.0484980094100615, "step": 2830}, {"loss": 1.333, "grad_norm": 0.7096508145332336, "learning_rate": 0.0002, "epoch": 2.055736518277235, "step": 2840}, {"loss": 1.1766, "grad_norm": 0.7653635144233704, "learning_rate": 0.0002, "epoch": 2.0629750271444083, "step": 2850}, {"loss": 1.2331, "grad_norm": 0.6202841997146606, "learning_rate": 0.0002, "epoch": 2.0702135360115816, "step": 2860}, {"loss": 1.3298, "grad_norm": 0.6810227632522583, "learning_rate": 0.0002, "epoch": 2.077452044878755, "step": 2870}, {"loss": 1.2505, "grad_norm": 0.7481493353843689, "learning_rate": 0.0002, "epoch": 2.0846905537459284, "step": 2880}, {"loss": 1.2484, "grad_norm": 0.7089637517929077, "learning_rate": 0.0002, "epoch": 2.0919290626131017, "step": 2890}, {"loss": 1.3095, "grad_norm": 0.7472923398017883, "learning_rate": 0.0002, "epoch": 2.099167571480275, "step": 2900}, {"loss": 1.304, "grad_norm": 0.8135465979576111, "learning_rate": 0.0002, "epoch": 2.1064060803474485, "step": 2910}, {"loss": 1.273, "grad_norm": 0.6097133159637451, "learning_rate": 0.0002, "epoch": 2.113644589214622, "step": 2920}, {"loss": 1.3384, "grad_norm": 0.5970117449760437, "learning_rate": 0.0002, "epoch": 2.120883098081795, "step": 2930}, {"loss": 1.3233, "grad_norm": 0.6169309616088867, "learning_rate": 0.0002, "epoch": 2.1281216069489686, "step": 2940}, {"loss": 1.4246, "grad_norm": 0.9428738355636597, "learning_rate": 0.0002, "epoch": 2.135360115816142, "step": 2950}, {"loss": 1.3527, "grad_norm": 0.5671679973602295, "learning_rate": 0.0002, "epoch": 2.1425986246833153, "step": 2960}, {"loss": 1.1375, "grad_norm": 0.7007262110710144, "learning_rate": 0.0002, "epoch": 2.1498371335504887, "step": 2970}, {"loss": 1.2015, "grad_norm": 0.6294044256210327, "learning_rate": 0.0002, "epoch": 2.157075642417662, "step": 2980}, {"loss": 1.2167, "grad_norm": 0.6105241775512695, "learning_rate": 0.0002, "epoch": 2.1643141512848354, "step": 2990}, {"loss": 1.2065, "grad_norm": 0.557124137878418, "learning_rate": 0.0002, "epoch": 2.1715526601520088, "step": 3000}, {"loss": 1.2515, "grad_norm": 0.6250392198562622, "learning_rate": 0.0002, "epoch": 2.178791169019182, "step": 3010}, {"loss": 1.385, "grad_norm": 0.645218551158905, "learning_rate": 0.0002, "epoch": 2.1860296778863555, "step": 3020}, {"loss": 1.3928, "grad_norm": 0.9033605456352234, "learning_rate": 0.0002, "epoch": 2.193268186753529, "step": 3030}, {"loss": 1.2458, "grad_norm": 0.5325747132301331, "learning_rate": 0.0002, "epoch": 2.2005066956207022, "step": 3040}, {"loss": 1.261, "grad_norm": 0.6334700584411621, "learning_rate": 0.0002, "epoch": 2.2077452044878756, "step": 3050}, {"loss": 1.2385, "grad_norm": 0.5206325054168701, "learning_rate": 0.0002, "epoch": 2.214983713355049, "step": 3060}, {"loss": 1.3103, "grad_norm": 0.5987200140953064, "learning_rate": 0.0002, "epoch": 2.2222222222222223, "step": 3070}, {"loss": 1.1756, "grad_norm": 0.5893264412879944, "learning_rate": 0.0002, "epoch": 2.2294607310893957, "step": 3080}, {"loss": 1.235, "grad_norm": 0.6869237422943115, "learning_rate": 0.0002, "epoch": 2.236699239956569, "step": 3090}, {"loss": 1.3285, "grad_norm": 0.5040048360824585, "learning_rate": 0.0002, "epoch": 2.2439377488237424, "step": 3100}, {"loss": 1.3316, "grad_norm": 0.6660613417625427, "learning_rate": 0.0002, "epoch": 2.251176257690916, "step": 3110}, {"loss": 1.3108, "grad_norm": 0.5890918970108032, "learning_rate": 0.0002, "epoch": 2.258414766558089, "step": 3120}, {"loss": 1.248, "grad_norm": 0.6458896994590759, "learning_rate": 0.0002, "epoch": 2.2656532754252625, "step": 3130}, {"loss": 1.4151, "grad_norm": 0.6832690834999084, "learning_rate": 0.0002, "epoch": 2.272891784292436, "step": 3140}, {"loss": 1.4458, "grad_norm": 0.833908200263977, "learning_rate": 0.0002, "epoch": 2.2801302931596092, "step": 3150}, {"loss": 1.2931, "grad_norm": 0.4596034586429596, "learning_rate": 0.0002, "epoch": 2.2873688020267826, "step": 3160}, {"loss": 1.449, "grad_norm": 0.9130966067314148, "learning_rate": 0.0002, "epoch": 2.294607310893956, "step": 3170}, {"loss": 1.3806, "grad_norm": 0.7143292427062988, "learning_rate": 0.0002, "epoch": 2.3018458197611293, "step": 3180}, {"loss": 1.2692, "grad_norm": 0.5388900637626648, "learning_rate": 0.0002, "epoch": 2.3090843286283027, "step": 3190}, {"loss": 1.2402, "grad_norm": 0.5607513189315796, "learning_rate": 0.0002, "epoch": 2.316322837495476, "step": 3200}, {"loss": 1.3874, "grad_norm": 0.6795142292976379, "learning_rate": 0.0002, "epoch": 2.3235613463626494, "step": 3210}, {"loss": 1.3042, "grad_norm": 0.6561070680618286, "learning_rate": 0.0002, "epoch": 2.330799855229823, "step": 3220}, {"loss": 1.4636, "grad_norm": 0.8858118057250977, "learning_rate": 0.0002, "epoch": 2.338038364096996, "step": 3230}, {"loss": 1.3214, "grad_norm": 0.6604151725769043, "learning_rate": 0.0002, "epoch": 2.3452768729641695, "step": 3240}, {"loss": 1.4004, "grad_norm": 0.6755785346031189, "learning_rate": 0.0002, "epoch": 2.352515381831343, "step": 3250}, {"loss": 1.2503, "grad_norm": 0.6981677412986755, "learning_rate": 0.0002, "epoch": 2.3597538906985163, "step": 3260}, {"loss": 1.3078, "grad_norm": 0.6338568329811096, "learning_rate": 0.0002, "epoch": 2.3669923995656896, "step": 3270}, {"loss": 1.285, "grad_norm": 0.5754265785217285, "learning_rate": 0.0002, "epoch": 2.374230908432863, "step": 3280}, {"loss": 1.2924, "grad_norm": 0.7533153295516968, "learning_rate": 0.0002, "epoch": 2.3814694173000364, "step": 3290}, {"loss": 1.3711, "grad_norm": 0.675065279006958, "learning_rate": 0.0002, "epoch": 2.3887079261672097, "step": 3300}, {"loss": 1.3548, "grad_norm": 0.5686452984809875, "learning_rate": 0.0002, "epoch": 2.395946435034383, "step": 3310}, {"loss": 1.1998, "grad_norm": 0.8129481673240662, "learning_rate": 0.0002, "epoch": 2.4031849439015565, "step": 3320}, {"loss": 1.2584, "grad_norm": 0.6615934371948242, "learning_rate": 0.0002, "epoch": 2.41042345276873, "step": 3330}, {"loss": 1.3691, "grad_norm": 0.6678834557533264, "learning_rate": 0.0002, "epoch": 2.417661961635903, "step": 3340}, {"loss": 1.2381, "grad_norm": 0.5581308007240295, "learning_rate": 0.0002, "epoch": 2.4249004705030766, "step": 3350}, {"loss": 1.3853, "grad_norm": 0.6098920106887817, "learning_rate": 0.0002, "epoch": 2.43213897937025, "step": 3360}, {"loss": 1.3692, "grad_norm": 0.8101736903190613, "learning_rate": 0.0002, "epoch": 2.4393774882374233, "step": 3370}, {"loss": 1.4418, "grad_norm": 0.6621488928794861, "learning_rate": 0.0002, "epoch": 2.4466159971045967, "step": 3380}, {"loss": 1.4579, "grad_norm": 0.8693289160728455, "learning_rate": 0.0002, "epoch": 2.45385450597177, "step": 3390}, {"loss": 1.3644, "grad_norm": 0.6724580526351929, "learning_rate": 0.0002, "epoch": 2.4610930148389434, "step": 3400}, {"loss": 1.2006, "grad_norm": 0.6776891946792603, "learning_rate": 0.0002, "epoch": 2.4683315237061167, "step": 3410}, {"loss": 1.2937, "grad_norm": 0.7214453816413879, "learning_rate": 0.0002, "epoch": 2.47557003257329, "step": 3420}, {"loss": 1.4051, "grad_norm": 0.8390451073646545, "learning_rate": 0.0002, "epoch": 2.4828085414404635, "step": 3430}, {"loss": 1.25, "grad_norm": 0.7130982279777527, "learning_rate": 0.0002, "epoch": 2.490047050307637, "step": 3440}, {"loss": 1.2231, "grad_norm": 0.8873937129974365, "learning_rate": 0.0002, "epoch": 2.49728555917481, "step": 3450}, {"loss": 1.1429, "grad_norm": 0.725185751914978, "learning_rate": 0.0002, "epoch": 2.5045240680419836, "step": 3460}, {"loss": 1.2699, "grad_norm": 0.6120352149009705, "learning_rate": 0.0002, "epoch": 2.511762576909157, "step": 3470}, {"loss": 1.2552, "grad_norm": 0.7713613510131836, "learning_rate": 0.0002, "epoch": 2.5190010857763303, "step": 3480}, {"loss": 1.4648, "grad_norm": 0.895309567451477, "learning_rate": 0.0002, "epoch": 2.5262395946435037, "step": 3490}, {"loss": 1.3043, "grad_norm": 0.9631021022796631, "learning_rate": 0.0002, "epoch": 2.533478103510677, "step": 3500}, {"loss": 1.3492, "grad_norm": 0.7475683093070984, "learning_rate": 0.0002, "epoch": 2.5407166123778504, "step": 3510}, {"loss": 1.3637, "grad_norm": 0.7271341681480408, "learning_rate": 0.0002, "epoch": 2.5479551212450238, "step": 3520}, {"loss": 1.304, "grad_norm": 0.6979510188102722, "learning_rate": 0.0002, "epoch": 2.555193630112197, "step": 3530}, {"loss": 1.2353, "grad_norm": 0.6504196524620056, "learning_rate": 0.0002, "epoch": 2.5624321389793705, "step": 3540}, {"loss": 1.2699, "grad_norm": 0.7226675748825073, "learning_rate": 0.0002, "epoch": 2.569670647846544, "step": 3550}, {"loss": 1.3002, "grad_norm": 0.6143222451210022, "learning_rate": 0.0002, "epoch": 2.5769091567137172, "step": 3560}, {"loss": 1.1585, "grad_norm": 0.7245154976844788, "learning_rate": 0.0002, "epoch": 2.5841476655808906, "step": 3570}, {"loss": 1.3651, "grad_norm": 0.943540632724762, "learning_rate": 0.0002, "epoch": 2.591386174448064, "step": 3580}, {"loss": 1.3034, "grad_norm": 0.7707241773605347, "learning_rate": 0.0002, "epoch": 2.5986246833152373, "step": 3590}, {"loss": 1.3063, "grad_norm": 0.6705001592636108, "learning_rate": 0.0002, "epoch": 2.6058631921824107, "step": 3600}, {"loss": 1.2437, "grad_norm": 0.6360933780670166, "learning_rate": 0.0002, "epoch": 2.613101701049584, "step": 3610}, {"loss": 1.1844, "grad_norm": 0.5846424698829651, "learning_rate": 0.0002, "epoch": 2.6203402099167574, "step": 3620}, {"loss": 1.3674, "grad_norm": 0.5958625674247742, "learning_rate": 0.0002, "epoch": 2.6275787187839303, "step": 3630}, {"loss": 1.3599, "grad_norm": 0.6819243431091309, "learning_rate": 0.0002, "epoch": 2.6348172276511037, "step": 3640}, {"loss": 1.3884, "grad_norm": 0.7033445835113525, "learning_rate": 0.0002, "epoch": 2.642055736518277, "step": 3650}, {"loss": 1.3392, "grad_norm": 0.6134849786758423, "learning_rate": 0.0002, "epoch": 2.6492942453854504, "step": 3660}, {"loss": 1.2661, "grad_norm": 0.658009946346283, "learning_rate": 0.0002, "epoch": 2.656532754252624, "step": 3670}, {"loss": 1.3987, "grad_norm": 0.6280999779701233, "learning_rate": 0.0002, "epoch": 2.663771263119797, "step": 3680}, {"loss": 1.2995, "grad_norm": 0.5536085963249207, "learning_rate": 0.0002, "epoch": 2.6710097719869705, "step": 3690}, {"loss": 1.2044, "grad_norm": 0.8603981733322144, "learning_rate": 0.0002, "epoch": 2.678248280854144, "step": 3700}, {"loss": 1.3879, "grad_norm": 0.5509994626045227, "learning_rate": 0.0002, "epoch": 2.6854867897213173, "step": 3710}, {"loss": 1.3253, "grad_norm": 0.9093621969223022, "learning_rate": 0.0002, "epoch": 2.6927252985884906, "step": 3720}, {"loss": 1.2668, "grad_norm": 0.7525952458381653, "learning_rate": 0.0002, "epoch": 2.699963807455664, "step": 3730}, {"loss": 1.248, "grad_norm": 0.6737023591995239, "learning_rate": 0.0002, "epoch": 2.7072023163228374, "step": 3740}, {"loss": 1.2981, "grad_norm": 0.8656924962997437, "learning_rate": 0.0002, "epoch": 2.7144408251900107, "step": 3750}, {"loss": 1.2342, "grad_norm": 0.7494133114814758, "learning_rate": 0.0002, "epoch": 2.721679334057184, "step": 3760}, {"loss": 1.2417, "grad_norm": 0.5725520849227905, "learning_rate": 0.0002, "epoch": 2.7289178429243575, "step": 3770}, {"loss": 1.28, "grad_norm": 0.836412787437439, "learning_rate": 0.0002, "epoch": 2.736156351791531, "step": 3780}, {"loss": 1.3784, "grad_norm": 0.6893242597579956, "learning_rate": 0.0002, "epoch": 2.743394860658704, "step": 3790}, {"loss": 1.2929, "grad_norm": 0.6696223020553589, "learning_rate": 0.0002, "epoch": 2.7506333695258776, "step": 3800}, {"loss": 1.2449, "grad_norm": 0.6483015418052673, "learning_rate": 0.0002, "epoch": 2.757871878393051, "step": 3810}, {"loss": 1.3282, "grad_norm": 0.8084456920623779, "learning_rate": 0.0002, "epoch": 2.7651103872602243, "step": 3820}, {"loss": 1.3694, "grad_norm": 0.6601949334144592, "learning_rate": 0.0002, "epoch": 2.7723488961273977, "step": 3830}, {"loss": 1.3568, "grad_norm": 0.6905533671379089, "learning_rate": 0.0002, "epoch": 2.779587404994571, "step": 3840}, {"loss": 1.3854, "grad_norm": 0.619318425655365, "learning_rate": 0.0002, "epoch": 2.7868259138617444, "step": 3850}, {"loss": 1.2551, "grad_norm": 0.5994023084640503, "learning_rate": 0.0002, "epoch": 2.7940644227289178, "step": 3860}, {"loss": 1.2022, "grad_norm": 0.5627168416976929, "learning_rate": 0.0002, "epoch": 2.801302931596091, "step": 3870}, {"loss": 1.3921, "grad_norm": 0.6001605987548828, "learning_rate": 0.0002, "epoch": 2.8085414404632645, "step": 3880}, {"loss": 1.3026, "grad_norm": 0.6022412776947021, "learning_rate": 0.0002, "epoch": 2.815779949330438, "step": 3890}, {"loss": 1.2765, "grad_norm": 0.6832426190376282, "learning_rate": 0.0002, "epoch": 2.823018458197611, "step": 3900}, {"loss": 1.1363, "grad_norm": 0.5936811566352844, "learning_rate": 0.0002, "epoch": 2.8302569670647846, "step": 3910}, {"loss": 1.1707, "grad_norm": 0.6960572600364685, "learning_rate": 0.0002, "epoch": 2.837495475931958, "step": 3920}, {"loss": 1.4063, "grad_norm": 0.5913406610488892, "learning_rate": 0.0002, "epoch": 2.8447339847991313, "step": 3930}, {"loss": 1.3245, "grad_norm": 0.678154706954956, "learning_rate": 0.0002, "epoch": 2.8519724936663047, "step": 3940}, {"loss": 1.366, "grad_norm": 0.7898936867713928, "learning_rate": 0.0002, "epoch": 2.859211002533478, "step": 3950}, {"loss": 1.3948, "grad_norm": 0.9234195351600647, "learning_rate": 0.0002, "epoch": 2.8664495114006514, "step": 3960}, {"loss": 1.2773, "grad_norm": 0.5960825085639954, "learning_rate": 0.0002, "epoch": 2.8736880202678248, "step": 3970}, {"loss": 1.3127, "grad_norm": 0.677118182182312, "learning_rate": 0.0002, "epoch": 2.880926529134998, "step": 3980}, {"loss": 1.2652, "grad_norm": 0.6505142450332642, "learning_rate": 0.0002, "epoch": 2.8881650380021715, "step": 3990}, {"loss": 1.2078, "grad_norm": 0.550826907157898, "learning_rate": 0.0002, "epoch": 2.895403546869345, "step": 4000}, {"loss": 1.1811, "grad_norm": 0.6209215521812439, "learning_rate": 0.0002, "epoch": 2.9026420557365182, "step": 4010}, {"loss": 1.4001, "grad_norm": 0.6549018025398254, "learning_rate": 0.0002, "epoch": 2.9098805646036916, "step": 4020}, {"loss": 1.2285, "grad_norm": 0.570682168006897, "learning_rate": 0.0002, "epoch": 2.917119073470865, "step": 4030}, {"loss": 1.0832, "grad_norm": 1.1807632446289062, "learning_rate": 0.0002, "epoch": 2.9243575823380383, "step": 4040}, {"loss": 1.2693, "grad_norm": 0.7058857679367065, "learning_rate": 0.0002, "epoch": 2.9315960912052117, "step": 4050}, {"loss": 1.2905, "grad_norm": 0.5542812943458557, "learning_rate": 0.0002, "epoch": 2.938834600072385, "step": 4060}, {"loss": 1.33, "grad_norm": 0.63167804479599, "learning_rate": 0.0002, "epoch": 2.9460731089395584, "step": 4070}, {"loss": 1.3075, "grad_norm": 0.5702962279319763, "learning_rate": 0.0002, "epoch": 2.953311617806732, "step": 4080}, {"loss": 1.2007, "grad_norm": 0.620944082736969, "learning_rate": 0.0002, "epoch": 2.960550126673905, "step": 4090}, {"loss": 1.2864, "grad_norm": 0.5866289734840393, "learning_rate": 0.0002, "epoch": 2.9677886355410785, "step": 4100}, {"loss": 1.3293, "grad_norm": 0.560170590877533, "learning_rate": 0.0002, "epoch": 2.975027144408252, "step": 4110}, {"loss": 1.2071, "grad_norm": 0.675082802772522, "learning_rate": 0.0002, "epoch": 2.9822656532754253, "step": 4120}, {"loss": 1.2981, "grad_norm": 0.62708580493927, "learning_rate": 0.0002, "epoch": 2.9895041621425986, "step": 4130}, {"loss": 1.2758, "grad_norm": 0.7893929481506348, "learning_rate": 0.0002, "epoch": 2.996742671009772, "step": 4140}, {"eval_loss": 1.4217946529388428, "eval_runtime": 27.1596, "eval_samples_per_second": 16.053, "eval_steps_per_second": 2.025, "epoch": 2.9996380745566413, "step": 4144}, {"loss": 1.2152, "grad_norm": 0.7043836116790771, "learning_rate": 0.0002, "epoch": 3.0039811798769454, "step": 4150}, {"loss": 1.1664, "grad_norm": 0.6806283593177795, "learning_rate": 0.0002, "epoch": 3.0112196887441187, "step": 4160}, {"loss": 1.292, "grad_norm": 0.7684550285339355, "learning_rate": 0.0002, "epoch": 3.018458197611292, "step": 4170}, {"loss": 1.3467, "grad_norm": 0.7895237803459167, "learning_rate": 0.0002, "epoch": 3.0256967064784654, "step": 4180}, {"loss": 1.1324, "grad_norm": 0.7464531064033508, "learning_rate": 0.0002, "epoch": 3.032935215345639, "step": 4190}, {"loss": 1.1614, "grad_norm": 0.9358500838279724, "learning_rate": 0.0002, "epoch": 3.040173724212812, "step": 4200}, {"loss": 1.1834, "grad_norm": 1.1066628694534302, "learning_rate": 0.0002, "epoch": 3.0474122330799855, "step": 4210}, {"loss": 1.1557, "grad_norm": 0.6663267612457275, "learning_rate": 0.0002, "epoch": 3.054650741947159, "step": 4220}, {"loss": 1.1707, "grad_norm": 0.6669464707374573, "learning_rate": 0.0002, "epoch": 3.0618892508143323, "step": 4230}, {"loss": 1.1841, "grad_norm": 0.7052164077758789, "learning_rate": 0.0002, "epoch": 3.0691277596815056, "step": 4240}, {"loss": 1.2913, "grad_norm": 0.6118432879447937, "learning_rate": 0.0002, "epoch": 3.076366268548679, "step": 4250}, {"loss": 1.1526, "grad_norm": 0.6915903687477112, "learning_rate": 0.0002, "epoch": 3.0836047774158524, "step": 4260}, {"loss": 1.1348, "grad_norm": 0.7441644668579102, "learning_rate": 0.0002, "epoch": 3.0908432862830257, "step": 4270}, {"loss": 1.1672, "grad_norm": 0.823850691318512, "learning_rate": 0.0002, "epoch": 3.098081795150199, "step": 4280}, {"loss": 1.2655, "grad_norm": 0.9677883386611938, "learning_rate": 0.0002, "epoch": 3.1053203040173725, "step": 4290}, {"loss": 1.1794, "grad_norm": 0.7002579569816589, "learning_rate": 0.0002, "epoch": 3.112558812884546, "step": 4300}, {"loss": 1.135, "grad_norm": 0.778789758682251, "learning_rate": 0.0002, "epoch": 3.119797321751719, "step": 4310}, {"loss": 1.0818, "grad_norm": 0.7236007452011108, "learning_rate": 0.0002, "epoch": 3.1270358306188926, "step": 4320}, {"loss": 1.1803, "grad_norm": 0.8809133768081665, "learning_rate": 0.0002, "epoch": 3.134274339486066, "step": 4330}, {"loss": 1.2571, "grad_norm": 0.7924913167953491, "learning_rate": 0.0002, "epoch": 3.1415128483532393, "step": 4340}, {"loss": 1.1413, "grad_norm": 0.7437422275543213, "learning_rate": 0.0002, "epoch": 3.1487513572204127, "step": 4350}, {"loss": 1.2088, "grad_norm": 0.6428450345993042, "learning_rate": 0.0002, "epoch": 3.155989866087586, "step": 4360}, {"loss": 1.3032, "grad_norm": 0.7922873497009277, "learning_rate": 0.0002, "epoch": 3.1632283749547594, "step": 4370}, {"loss": 1.216, "grad_norm": 0.5252506732940674, "learning_rate": 0.0002, "epoch": 3.1704668838219328, "step": 4380}, {"loss": 1.1297, "grad_norm": 0.8570457696914673, "learning_rate": 0.0002, "epoch": 3.177705392689106, "step": 4390}, {"loss": 1.0994, "grad_norm": 0.7218987345695496, "learning_rate": 0.0002, "epoch": 3.1849439015562795, "step": 4400}, {"loss": 1.2891, "grad_norm": 0.6921393275260925, "learning_rate": 0.0002, "epoch": 3.192182410423453, "step": 4410}, {"loss": 1.2668, "grad_norm": 0.7386137843132019, "learning_rate": 0.0002, "epoch": 3.199420919290626, "step": 4420}, {"loss": 1.1654, "grad_norm": 0.6227759122848511, "learning_rate": 0.0002, "epoch": 3.2066594281577996, "step": 4430}, {"loss": 1.1752, "grad_norm": 0.7180278897285461, "learning_rate": 0.0002, "epoch": 3.213897937024973, "step": 4440}, {"loss": 1.1757, "grad_norm": 0.745830774307251, "learning_rate": 0.0002, "epoch": 3.2211364458921463, "step": 4450}, {"loss": 1.234, "grad_norm": 0.6766072511672974, "learning_rate": 0.0002, "epoch": 3.2283749547593197, "step": 4460}, {"loss": 1.1999, "grad_norm": 0.8325067162513733, "learning_rate": 0.0002, "epoch": 3.235613463626493, "step": 4470}, {"loss": 1.1606, "grad_norm": 0.7148305177688599, "learning_rate": 0.0002, "epoch": 3.2428519724936664, "step": 4480}, {"loss": 1.1383, "grad_norm": 0.7752676010131836, "learning_rate": 0.0002, "epoch": 3.25009048136084, "step": 4490}, {"loss": 1.3006, "grad_norm": 0.6776860952377319, "learning_rate": 0.0002, "epoch": 3.257328990228013, "step": 4500}, {"loss": 1.0796, "grad_norm": 0.704359769821167, "learning_rate": 0.0002, "epoch": 3.2645674990951865, "step": 4510}, {"loss": 1.2496, "grad_norm": 0.6880282163619995, "learning_rate": 0.0002, "epoch": 3.27180600796236, "step": 4520}, {"loss": 1.0947, "grad_norm": 0.8179270029067993, "learning_rate": 0.0002, "epoch": 3.2790445168295332, "step": 4530}, {"loss": 1.1909, "grad_norm": 0.6718448996543884, "learning_rate": 0.0002, "epoch": 3.2862830256967066, "step": 4540}, {"loss": 1.2708, "grad_norm": 0.8300657868385315, "learning_rate": 0.0002, "epoch": 3.29352153456388, "step": 4550}, {"loss": 1.2594, "grad_norm": 0.6433690786361694, "learning_rate": 0.0002, "epoch": 3.3007600434310533, "step": 4560}, {"loss": 1.2479, "grad_norm": 0.690262496471405, "learning_rate": 0.0002, "epoch": 3.3079985522982267, "step": 4570}, {"loss": 1.1342, "grad_norm": 0.7022852301597595, "learning_rate": 0.0002, "epoch": 3.3152370611654, "step": 4580}, {"loss": 1.0844, "grad_norm": 0.6438387632369995, "learning_rate": 0.0002, "epoch": 3.3224755700325734, "step": 4590}, {"loss": 1.17, "grad_norm": 0.6866899132728577, "learning_rate": 0.0002, "epoch": 3.329714078899747, "step": 4600}, {"loss": 1.1289, "grad_norm": 0.8233968019485474, "learning_rate": 0.0002, "epoch": 3.33695258776692, "step": 4610}, {"loss": 1.1855, "grad_norm": 0.7251574993133545, "learning_rate": 0.0002, "epoch": 3.3441910966340935, "step": 4620}, {"loss": 1.3403, "grad_norm": 0.7855110168457031, "learning_rate": 0.0002, "epoch": 3.351429605501267, "step": 4630}, {"loss": 1.2922, "grad_norm": 0.8487356305122375, "learning_rate": 0.0002, "epoch": 3.3586681143684403, "step": 4640}, {"loss": 1.2462, "grad_norm": 0.6429011225700378, "learning_rate": 0.0002, "epoch": 3.3659066232356136, "step": 4650}, {"loss": 1.129, "grad_norm": 0.7095270156860352, "learning_rate": 0.0002, "epoch": 3.373145132102787, "step": 4660}, {"loss": 1.262, "grad_norm": 0.6792303323745728, "learning_rate": 0.0002, "epoch": 3.3803836409699604, "step": 4670}, {"loss": 1.256, "grad_norm": 0.6784825921058655, "learning_rate": 0.0002, "epoch": 3.3876221498371337, "step": 4680}, {"loss": 1.0838, "grad_norm": 0.6362888216972351, "learning_rate": 0.0002, "epoch": 3.394860658704307, "step": 4690}, {"loss": 1.2165, "grad_norm": 0.7794778943061829, "learning_rate": 0.0002, "epoch": 3.4020991675714805, "step": 4700}, {"loss": 1.0644, "grad_norm": 0.7287485003471375, "learning_rate": 0.0002, "epoch": 3.409337676438654, "step": 4710}, {"loss": 1.2925, "grad_norm": 0.6481451392173767, "learning_rate": 0.0002, "epoch": 3.416576185305827, "step": 4720}, {"loss": 1.2121, "grad_norm": 0.9200371503829956, "learning_rate": 0.0002, "epoch": 3.4238146941730006, "step": 4730}, {"loss": 1.072, "grad_norm": 1.074180245399475, "learning_rate": 0.0002, "epoch": 3.431053203040174, "step": 4740}, {"loss": 1.0421, "grad_norm": 0.6722986698150635, "learning_rate": 0.0002, "epoch": 3.438291711907347, "step": 4750}, {"loss": 1.2258, "grad_norm": 0.7945933938026428, "learning_rate": 0.0002, "epoch": 3.44553022077452, "step": 4760}, {"loss": 1.0927, "grad_norm": 0.7624640464782715, "learning_rate": 0.0002, "epoch": 3.4527687296416936, "step": 4770}, {"loss": 1.2428, "grad_norm": 0.7763656377792358, "learning_rate": 0.0002, "epoch": 3.460007238508867, "step": 4780}, {"loss": 1.2584, "grad_norm": 0.7736947536468506, "learning_rate": 0.0002, "epoch": 3.4672457473760403, "step": 4790}, {"loss": 1.1953, "grad_norm": 0.8450354933738708, "learning_rate": 0.0002, "epoch": 3.4744842562432137, "step": 4800}, {"loss": 1.1362, "grad_norm": 0.6480133533477783, "learning_rate": 0.0002, "epoch": 3.481722765110387, "step": 4810}, {"loss": 1.1882, "grad_norm": 0.8437445759773254, "learning_rate": 0.0002, "epoch": 3.4889612739775604, "step": 4820}, {"loss": 1.1519, "grad_norm": 0.7781730890274048, "learning_rate": 0.0002, "epoch": 3.4961997828447338, "step": 4830}, {"loss": 1.1836, "grad_norm": 0.8523228168487549, "learning_rate": 0.0002, "epoch": 3.503438291711907, "step": 4840}, {"loss": 1.1672, "grad_norm": 0.6236732006072998, "learning_rate": 0.0002, "epoch": 3.5106768005790805, "step": 4850}, {"loss": 1.1926, "grad_norm": 0.7500787377357483, "learning_rate": 0.0002, "epoch": 3.517915309446254, "step": 4860}, {"loss": 1.1998, "grad_norm": 0.7665374875068665, "learning_rate": 0.0002, "epoch": 3.5251538183134272, "step": 4870}, {"loss": 1.1551, "grad_norm": 0.787857711315155, "learning_rate": 0.0002, "epoch": 3.5323923271806006, "step": 4880}, {"loss": 1.2758, "grad_norm": 0.970595121383667, "learning_rate": 0.0002, "epoch": 3.539630836047774, "step": 4890}, {"loss": 1.1274, "grad_norm": 0.6409347057342529, "learning_rate": 0.0002, "epoch": 3.5468693449149473, "step": 4900}, {"loss": 1.1596, "grad_norm": 0.888551652431488, "learning_rate": 0.0002, "epoch": 3.5541078537821207, "step": 4910}, {"loss": 1.1644, "grad_norm": 1.0808377265930176, "learning_rate": 0.0002, "epoch": 3.561346362649294, "step": 4920}, {"loss": 1.2564, "grad_norm": 0.7501053214073181, "learning_rate": 0.0002, "epoch": 3.5685848715164674, "step": 4930}, {"loss": 1.2351, "grad_norm": 0.7375240325927734, "learning_rate": 0.0002, "epoch": 3.575823380383641, "step": 4940}, {"loss": 1.3568, "grad_norm": 0.7075039744377136, "learning_rate": 0.0002, "epoch": 3.583061889250814, "step": 4950}, {"loss": 1.3355, "grad_norm": 0.939337432384491, "learning_rate": 0.0002, "epoch": 3.5903003981179875, "step": 4960}, {"loss": 1.1722, "grad_norm": 0.6717396974563599, "learning_rate": 0.0002, "epoch": 3.597538906985161, "step": 4970}, {"loss": 1.1186, "grad_norm": 0.7141643762588501, "learning_rate": 0.0002, "epoch": 3.6047774158523342, "step": 4980}, {"loss": 1.1011, "grad_norm": 0.7109216451644897, "learning_rate": 0.0002, "epoch": 3.6120159247195076, "step": 4990}, {"loss": 1.2178, "grad_norm": 0.7020776867866516, "learning_rate": 0.0002, "epoch": 3.619254433586681, "step": 5000}, {"loss": 1.1939, "grad_norm": 0.7158873677253723, "learning_rate": 0.0002, "epoch": 3.6264929424538543, "step": 5010}, {"loss": 1.2624, "grad_norm": 0.7062035202980042, "learning_rate": 0.0002, "epoch": 3.6337314513210277, "step": 5020}, {"loss": 1.0224, "grad_norm": 0.7081155776977539, "learning_rate": 0.0002, "epoch": 3.640969960188201, "step": 5030}, {"loss": 1.2195, "grad_norm": 1.2210607528686523, "learning_rate": 0.0002, "epoch": 3.6482084690553744, "step": 5040}, {"loss": 1.2596, "grad_norm": 0.6650236248970032, "learning_rate": 0.0002, "epoch": 3.655446977922548, "step": 5050}, {"loss": 1.1072, "grad_norm": 0.6884829998016357, "learning_rate": 0.0002, "epoch": 3.662685486789721, "step": 5060}, {"loss": 1.2292, "grad_norm": 0.7317819595336914, "learning_rate": 0.0002, "epoch": 3.6699239956568945, "step": 5070}, {"loss": 1.1917, "grad_norm": 0.7406691908836365, "learning_rate": 0.0002, "epoch": 3.677162504524068, "step": 5080}, {"loss": 1.2949, "grad_norm": 0.9009454250335693, "learning_rate": 0.0002, "epoch": 3.6844010133912413, "step": 5090}, {"loss": 1.1528, "grad_norm": 0.8189385533332825, "learning_rate": 0.0002, "epoch": 3.6916395222584146, "step": 5100}, {"loss": 1.3408, "grad_norm": 1.0793628692626953, "learning_rate": 0.0002, "epoch": 3.698878031125588, "step": 5110}, {"loss": 1.2417, "grad_norm": 0.8593027591705322, "learning_rate": 0.0002, "epoch": 3.7061165399927614, "step": 5120}, {"loss": 1.2141, "grad_norm": 0.8481812477111816, "learning_rate": 0.0002, "epoch": 3.7133550488599347, "step": 5130}, {"loss": 1.125, "grad_norm": 0.6527451276779175, "learning_rate": 0.0002, "epoch": 3.720593557727108, "step": 5140}, {"loss": 1.1584, "grad_norm": 0.9220114350318909, "learning_rate": 0.0002, "epoch": 3.7278320665942815, "step": 5150}, {"loss": 1.2267, "grad_norm": 1.0842019319534302, "learning_rate": 0.0002, "epoch": 3.735070575461455, "step": 5160}, {"loss": 1.3083, "grad_norm": 0.965453565120697, "learning_rate": 0.0002, "epoch": 3.742309084328628, "step": 5170}, {"loss": 1.1772, "grad_norm": 0.9903319478034973, "learning_rate": 0.0002, "epoch": 3.7495475931958016, "step": 5180}, {"loss": 1.2515, "grad_norm": 0.7434818148612976, "learning_rate": 0.0002, "epoch": 3.756786102062975, "step": 5190}, {"loss": 1.2631, "grad_norm": 0.6717280745506287, "learning_rate": 0.0002, "epoch": 3.7640246109301483, "step": 5200}, {"loss": 1.2012, "grad_norm": 0.7754665613174438, "learning_rate": 0.0002, "epoch": 3.7712631197973217, "step": 5210}, {"loss": 1.305, "grad_norm": 1.028374433517456, "learning_rate": 0.0002, "epoch": 3.778501628664495, "step": 5220}, {"loss": 1.1866, "grad_norm": 0.6026996374130249, "learning_rate": 0.0002, "epoch": 3.7857401375316684, "step": 5230}, {"loss": 1.1901, "grad_norm": 0.6978490948677063, "learning_rate": 0.0002, "epoch": 3.7929786463988417, "step": 5240}, {"loss": 1.2576, "grad_norm": 0.7303446531295776, "learning_rate": 0.0002, "epoch": 3.800217155266015, "step": 5250}, {"loss": 1.3173, "grad_norm": 1.0734210014343262, "learning_rate": 0.0002, "epoch": 3.8074556641331885, "step": 5260}, {"loss": 1.1137, "grad_norm": 0.6383201479911804, "learning_rate": 0.0002, "epoch": 3.814694173000362, "step": 5270}, {"loss": 1.0904, "grad_norm": 0.7742630243301392, "learning_rate": 0.0002, "epoch": 3.821932681867535, "step": 5280}, {"loss": 1.2232, "grad_norm": 0.8477074503898621, "learning_rate": 0.0002, "epoch": 3.8291711907347086, "step": 5290}, {"loss": 1.2047, "grad_norm": 0.6675317883491516, "learning_rate": 0.0002, "epoch": 3.836409699601882, "step": 5300}, {"loss": 1.2275, "grad_norm": 0.7515445351600647, "learning_rate": 0.0002, "epoch": 3.8436482084690553, "step": 5310}, {"loss": 1.2569, "grad_norm": 1.1441220045089722, "learning_rate": 0.0002, "epoch": 3.8508867173362287, "step": 5320}, {"loss": 1.1512, "grad_norm": 0.7968795895576477, "learning_rate": 0.0002, "epoch": 3.858125226203402, "step": 5330}, {"loss": 1.232, "grad_norm": 0.7842824459075928, "learning_rate": 0.0002, "epoch": 3.8653637350705754, "step": 5340}, {"loss": 1.1847, "grad_norm": 0.8272225260734558, "learning_rate": 0.0002, "epoch": 3.8726022439377488, "step": 5350}, {"loss": 1.1381, "grad_norm": 0.8413397669792175, "learning_rate": 0.0002, "epoch": 3.879840752804922, "step": 5360}, {"loss": 1.2349, "grad_norm": 1.141764760017395, "learning_rate": 0.0002, "epoch": 3.8870792616720955, "step": 5370}, {"loss": 1.212, "grad_norm": 0.9826975464820862, "learning_rate": 0.0002, "epoch": 3.894317770539269, "step": 5380}, {"loss": 1.1833, "grad_norm": 0.8598255515098572, "learning_rate": 0.0002, "epoch": 3.9015562794064422, "step": 5390}, {"loss": 1.1247, "grad_norm": 0.6271058320999146, "learning_rate": 0.0002, "epoch": 3.9087947882736156, "step": 5400}, {"loss": 1.2212, "grad_norm": 0.6379870772361755, "learning_rate": 0.0002, "epoch": 3.916033297140789, "step": 5410}, {"loss": 1.2481, "grad_norm": 1.0313376188278198, "learning_rate": 0.0002, "epoch": 3.9232718060079623, "step": 5420}, {"loss": 1.1872, "grad_norm": 0.8220619559288025, "learning_rate": 0.0002, "epoch": 3.9305103148751357, "step": 5430}, {"loss": 1.2006, "grad_norm": 0.7576116919517517, "learning_rate": 0.0002, "epoch": 3.937748823742309, "step": 5440}, {"loss": 1.1969, "grad_norm": 1.226235032081604, "learning_rate": 0.0002, "epoch": 3.9449873326094824, "step": 5450}, {"loss": 1.2945, "grad_norm": 0.7979229688644409, "learning_rate": 0.0002, "epoch": 3.952225841476656, "step": 5460}, {"loss": 1.1922, "grad_norm": 0.9911929965019226, "learning_rate": 0.0002, "epoch": 3.959464350343829, "step": 5470}, {"loss": 1.0924, "grad_norm": 0.643738865852356, "learning_rate": 0.0002, "epoch": 3.9667028592110025, "step": 5480}, {"loss": 1.0607, "grad_norm": 0.682305634021759, "learning_rate": 0.0002, "epoch": 3.973941368078176, "step": 5490}, {"loss": 1.2908, "grad_norm": 1.18373441696167, "learning_rate": 0.0002, "epoch": 3.9811798769453492, "step": 5500}, {"loss": 1.0889, "grad_norm": 0.7190203070640564, "learning_rate": 0.0002, "epoch": 3.9884183858125226, "step": 5510}, {"loss": 1.2745, "grad_norm": 0.7516948580741882, "learning_rate": 0.0002, "epoch": 3.995656894679696, "step": 5520}, {"eval_loss": 1.4252897500991821, "eval_runtime": 27.235, "eval_samples_per_second": 16.009, "eval_steps_per_second": 2.019, "epoch": 4.0, "step": 5526}, {"loss": 1.0088, "grad_norm": 0.6353074312210083, "learning_rate": 0.0002, "epoch": 4.002895403546869, "step": 5530}, {"loss": 1.0326, "grad_norm": 0.7424906492233276, "learning_rate": 0.0002, "epoch": 4.010133912414043, "step": 5540}, {"loss": 1.0667, "grad_norm": 0.8856638073921204, "learning_rate": 0.0002, "epoch": 4.017372421281216, "step": 5550}, {"loss": 1.0905, "grad_norm": 0.9627974033355713, "learning_rate": 0.0002, "epoch": 4.024610930148389, "step": 5560}, {"loss": 1.0965, "grad_norm": 0.9048978686332703, "learning_rate": 0.0002, "epoch": 4.031849439015563, "step": 5570}, {"loss": 1.1108, "grad_norm": 0.921119213104248, "learning_rate": 0.0002, "epoch": 4.039087947882736, "step": 5580}, {"loss": 1.1235, "grad_norm": 0.8654361963272095, "learning_rate": 0.0002, "epoch": 4.0463264567499095, "step": 5590}, {"loss": 1.0794, "grad_norm": 0.7947945594787598, "learning_rate": 0.0002, "epoch": 4.053564965617083, "step": 5600}, {"loss": 1.0674, "grad_norm": 0.8307326436042786, "learning_rate": 0.0002, "epoch": 4.060803474484256, "step": 5610}, {"loss": 1.0076, "grad_norm": 0.793273389339447, "learning_rate": 0.0002, "epoch": 4.06804198335143, "step": 5620}, {"loss": 1.0651, "grad_norm": 0.8748673796653748, "learning_rate": 0.0002, "epoch": 4.075280492218603, "step": 5630}, {"loss": 1.111, "grad_norm": 0.7926856279373169, "learning_rate": 0.0002, "epoch": 4.082519001085776, "step": 5640}, {"loss": 1.044, "grad_norm": 0.922645092010498, "learning_rate": 0.0002, "epoch": 4.08975750995295, "step": 5650}, {"loss": 1.109, "grad_norm": 0.9539641737937927, "learning_rate": 0.0002, "epoch": 4.096996018820123, "step": 5660}, {"loss": 1.0788, "grad_norm": 0.8674443364143372, "learning_rate": 0.0002, "epoch": 4.1042345276872965, "step": 5670}, {"loss": 0.9867, "grad_norm": 0.7097609043121338, "learning_rate": 0.0002, "epoch": 4.11147303655447, "step": 5680}, {"loss": 1.1154, "grad_norm": 0.8875522613525391, "learning_rate": 0.0002, "epoch": 4.118711545421643, "step": 5690}, {"loss": 1.1217, "grad_norm": 0.8583634495735168, "learning_rate": 0.0002, "epoch": 4.125950054288817, "step": 5700}, {"loss": 1.0973, "grad_norm": 0.6736377477645874, "learning_rate": 0.0002, "epoch": 4.13318856315599, "step": 5710}, {"loss": 1.1199, "grad_norm": 0.9349062442779541, "learning_rate": 0.0002, "epoch": 4.140427072023163, "step": 5720}, {"loss": 1.0508, "grad_norm": 1.0610365867614746, "learning_rate": 0.0002, "epoch": 4.147665580890337, "step": 5730}, {"loss": 1.1146, "grad_norm": 1.5838189125061035, "learning_rate": 0.0002, "epoch": 4.15490408975751, "step": 5740}, {"loss": 1.0222, "grad_norm": 0.747522234916687, "learning_rate": 0.0002, "epoch": 4.162142598624683, "step": 5750}, {"loss": 1.1328, "grad_norm": 1.3247915506362915, "learning_rate": 0.0002, "epoch": 4.169381107491857, "step": 5760}, {"loss": 1.1655, "grad_norm": 0.8750247955322266, "learning_rate": 0.0002, "epoch": 4.17661961635903, "step": 5770}, {"loss": 1.199, "grad_norm": 0.7914144992828369, "learning_rate": 0.0002, "epoch": 4.1838581252262035, "step": 5780}, {"loss": 1.1213, "grad_norm": 0.9493299126625061, "learning_rate": 0.0002, "epoch": 4.191096634093377, "step": 5790}, {"loss": 1.1515, "grad_norm": 0.7802295088768005, "learning_rate": 0.0002, "epoch": 4.19833514296055, "step": 5800}, {"loss": 1.0704, "grad_norm": 0.6987314820289612, "learning_rate": 0.0002, "epoch": 4.205573651827724, "step": 5810}, {"loss": 1.1699, "grad_norm": 0.9220341444015503, "learning_rate": 0.0002, "epoch": 4.212812160694897, "step": 5820}, {"loss": 1.1394, "grad_norm": 0.8932939767837524, "learning_rate": 0.0002, "epoch": 4.22005066956207, "step": 5830}, {"loss": 1.0048, "grad_norm": 0.920002818107605, "learning_rate": 0.0002, "epoch": 4.227289178429244, "step": 5840}, {"loss": 0.964, "grad_norm": 0.6662752032279968, "learning_rate": 0.0002, "epoch": 4.234527687296417, "step": 5850}, {"loss": 0.986, "grad_norm": 0.8679718971252441, "learning_rate": 0.0002, "epoch": 4.24176619616359, "step": 5860}, {"loss": 0.8991, "grad_norm": 0.7020887732505798, "learning_rate": 0.0002, "epoch": 4.249004705030764, "step": 5870}, {"loss": 1.1132, "grad_norm": 0.869611382484436, "learning_rate": 0.0002, "epoch": 4.256243213897937, "step": 5880}, {"loss": 1.1026, "grad_norm": 0.7796585559844971, "learning_rate": 0.0002, "epoch": 4.2634817227651105, "step": 5890}, {"loss": 1.0957, "grad_norm": 0.8978819251060486, "learning_rate": 0.0002, "epoch": 4.270720231632284, "step": 5900}, {"loss": 1.1325, "grad_norm": 1.0837205648422241, "learning_rate": 0.0002, "epoch": 4.277958740499457, "step": 5910}, {"loss": 1.1279, "grad_norm": 0.7584353089332581, "learning_rate": 0.0002, "epoch": 4.285197249366631, "step": 5920}, {"loss": 1.0513, "grad_norm": 0.7313185334205627, "learning_rate": 0.0002, "epoch": 4.292435758233804, "step": 5930}, {"loss": 1.1101, "grad_norm": 0.8004671335220337, "learning_rate": 0.0002, "epoch": 4.299674267100977, "step": 5940}, {"loss": 1.14, "grad_norm": 2.154958724975586, "learning_rate": 0.0002, "epoch": 4.306912775968151, "step": 5950}, {"loss": 1.1206, "grad_norm": 0.9163479804992676, "learning_rate": 0.0002, "epoch": 4.314151284835324, "step": 5960}, {"loss": 0.9941, "grad_norm": 0.9151589274406433, "learning_rate": 0.0002, "epoch": 4.321389793702497, "step": 5970}, {"loss": 1.0606, "grad_norm": 0.8624112010002136, "learning_rate": 0.0002, "epoch": 4.328628302569671, "step": 5980}, {"loss": 1.1625, "grad_norm": 0.9357741475105286, "learning_rate": 0.0002, "epoch": 4.335866811436844, "step": 5990}, {"loss": 1.0712, "grad_norm": 1.3482335805892944, "learning_rate": 0.0002, "epoch": 4.3431053203040175, "step": 6000}, {"loss": 1.1224, "grad_norm": 0.7156149744987488, "learning_rate": 0.0002, "epoch": 4.350343829171191, "step": 6010}, {"loss": 1.0753, "grad_norm": 0.8480049967765808, "learning_rate": 0.0002, "epoch": 4.357582338038364, "step": 6020}, {"loss": 1.051, "grad_norm": 0.8262244462966919, "learning_rate": 0.0002, "epoch": 4.364820846905538, "step": 6030}, {"loss": 0.9966, "grad_norm": 0.7733905911445618, "learning_rate": 0.0002, "epoch": 4.372059355772711, "step": 6040}, {"loss": 1.1008, "grad_norm": 0.8553919792175293, "learning_rate": 0.0002, "epoch": 4.379297864639884, "step": 6050}, {"loss": 1.1777, "grad_norm": 0.8666832447052002, "learning_rate": 0.0002, "epoch": 4.386536373507058, "step": 6060}, {"loss": 1.1934, "grad_norm": 0.9168295860290527, "learning_rate": 0.0002, "epoch": 4.393774882374231, "step": 6070}, {"loss": 1.0988, "grad_norm": 0.7315238118171692, "learning_rate": 0.0002, "epoch": 4.4010133912414044, "step": 6080}, {"loss": 1.1599, "grad_norm": 1.020263433456421, "learning_rate": 0.0002, "epoch": 4.408251900108578, "step": 6090}, {"loss": 1.133, "grad_norm": 0.9978243708610535, "learning_rate": 0.0002, "epoch": 4.415490408975751, "step": 6100}, {"loss": 1.1324, "grad_norm": 0.995453953742981, "learning_rate": 0.0002, "epoch": 4.4227289178429245, "step": 6110}, {"loss": 1.0957, "grad_norm": 0.9360884428024292, "learning_rate": 0.0002, "epoch": 4.429967426710098, "step": 6120}, {"loss": 0.9506, "grad_norm": 0.8099448084831238, "learning_rate": 0.0002, "epoch": 4.437205935577271, "step": 6130}, {"loss": 1.0887, "grad_norm": 0.8173841238021851, "learning_rate": 0.0002, "epoch": 4.444444444444445, "step": 6140}, {"loss": 1.1219, "grad_norm": 0.7972666025161743, "learning_rate": 0.0002, "epoch": 4.451682953311618, "step": 6150}, {"loss": 1.0226, "grad_norm": 0.7685779333114624, "learning_rate": 0.0002, "epoch": 4.458921462178791, "step": 6160}, {"loss": 1.0732, "grad_norm": 0.7872623801231384, "learning_rate": 0.0002, "epoch": 4.466159971045965, "step": 6170}, {"loss": 0.9911, "grad_norm": 0.7677070498466492, "learning_rate": 0.0002, "epoch": 4.473398479913138, "step": 6180}, {"loss": 1.0919, "grad_norm": 0.7878316044807434, "learning_rate": 0.0002, "epoch": 4.4806369887803115, "step": 6190}, {"loss": 1.018, "grad_norm": 0.8178079724311829, "learning_rate": 0.0002, "epoch": 4.487875497647485, "step": 6200}, {"loss": 1.0517, "grad_norm": 1.2820082902908325, "learning_rate": 0.0002, "epoch": 4.495114006514658, "step": 6210}, {"loss": 1.3101, "grad_norm": 0.9380832314491272, "learning_rate": 0.0002, "epoch": 4.502352515381832, "step": 6220}, {"loss": 0.9818, "grad_norm": 0.7810422778129578, "learning_rate": 0.0002, "epoch": 4.509591024249005, "step": 6230}, {"loss": 1.1677, "grad_norm": 1.1022917032241821, "learning_rate": 0.0002, "epoch": 4.516829533116178, "step": 6240}, {"loss": 1.1579, "grad_norm": 1.4275553226470947, "learning_rate": 0.0002, "epoch": 4.524068041983352, "step": 6250}, {"loss": 1.3237, "grad_norm": 0.7597777247428894, "learning_rate": 0.0002, "epoch": 4.531306550850525, "step": 6260}, {"loss": 1.1529, "grad_norm": 1.10992431640625, "learning_rate": 0.0002, "epoch": 4.538545059717698, "step": 6270}, {"loss": 1.0732, "grad_norm": 0.8981178998947144, "learning_rate": 0.0002, "epoch": 4.545783568584872, "step": 6280}, {"loss": 1.086, "grad_norm": 0.7863979339599609, "learning_rate": 0.0002, "epoch": 4.553022077452045, "step": 6290}, {"loss": 1.2008, "grad_norm": 0.9071474671363831, "learning_rate": 0.0002, "epoch": 4.5602605863192185, "step": 6300}, {"loss": 1.0916, "grad_norm": 0.7429424524307251, "learning_rate": 0.0002, "epoch": 4.567499095186392, "step": 6310}, {"loss": 1.095, "grad_norm": 1.0767850875854492, "learning_rate": 0.0002, "epoch": 4.574737604053565, "step": 6320}, {"loss": 1.1023, "grad_norm": 0.7885915637016296, "learning_rate": 0.0002, "epoch": 4.581976112920739, "step": 6330}, {"loss": 1.1131, "grad_norm": 0.8350457549095154, "learning_rate": 0.0002, "epoch": 4.589214621787912, "step": 6340}, {"loss": 1.0743, "grad_norm": 0.7853530645370483, "learning_rate": 0.0002, "epoch": 4.596453130655085, "step": 6350}, {"loss": 1.1912, "grad_norm": 1.1220661401748657, "learning_rate": 0.0002, "epoch": 4.603691639522259, "step": 6360}, {"loss": 1.0927, "grad_norm": 0.7959423065185547, "learning_rate": 0.0002, "epoch": 4.610930148389432, "step": 6370}, {"loss": 1.1542, "grad_norm": 0.7782652378082275, "learning_rate": 0.0002, "epoch": 4.618168657256605, "step": 6380}, {"loss": 1.0753, "grad_norm": 0.7882203459739685, "learning_rate": 0.0002, "epoch": 4.625407166123779, "step": 6390}, {"loss": 1.0676, "grad_norm": 0.8841899037361145, "learning_rate": 0.0002, "epoch": 4.632645674990952, "step": 6400}, {"loss": 1.0815, "grad_norm": 0.7936127781867981, "learning_rate": 0.0002, "epoch": 4.6398841838581255, "step": 6410}, {"loss": 1.0198, "grad_norm": 0.9213966131210327, "learning_rate": 0.0002, "epoch": 4.647122692725299, "step": 6420}, {"loss": 0.9872, "grad_norm": 0.9246473908424377, "learning_rate": 0.0002, "epoch": 4.654361201592472, "step": 6430}, {"loss": 1.1309, "grad_norm": 0.766572892665863, "learning_rate": 0.0002, "epoch": 4.661599710459646, "step": 6440}, {"loss": 1.1095, "grad_norm": 0.8596171736717224, "learning_rate": 0.0002, "epoch": 4.668838219326819, "step": 6450}, {"loss": 1.1869, "grad_norm": 0.8482751846313477, "learning_rate": 0.0002, "epoch": 4.676076728193992, "step": 6460}, {"loss": 1.0622, "grad_norm": 1.0826905965805054, "learning_rate": 0.0002, "epoch": 4.683315237061166, "step": 6470}, {"loss": 1.0256, "grad_norm": 1.1048457622528076, "learning_rate": 0.0002, "epoch": 4.690553745928339, "step": 6480}, {"loss": 1.0514, "grad_norm": 0.9429134726524353, "learning_rate": 0.0002, "epoch": 4.697792254795512, "step": 6490}, {"loss": 1.1351, "grad_norm": 0.8587502837181091, "learning_rate": 0.0002, "epoch": 4.705030763662686, "step": 6500}, {"loss": 1.0969, "grad_norm": 1.0387083292007446, "learning_rate": 0.0002, "epoch": 4.712269272529859, "step": 6510}, {"loss": 1.0493, "grad_norm": 0.7471951842308044, "learning_rate": 0.0002, "epoch": 4.7195077813970325, "step": 6520}, {"loss": 1.2632, "grad_norm": 0.8800424933433533, "learning_rate": 0.0002, "epoch": 4.726746290264206, "step": 6530}, {"loss": 1.2126, "grad_norm": 0.8136811852455139, "learning_rate": 0.0002, "epoch": 4.733984799131379, "step": 6540}, {"loss": 1.195, "grad_norm": 0.9910339713096619, "learning_rate": 0.0002, "epoch": 4.741223307998553, "step": 6550}, {"loss": 1.1201, "grad_norm": 1.0679163932800293, "learning_rate": 0.0002, "epoch": 4.748461816865726, "step": 6560}, {"loss": 1.0297, "grad_norm": 0.8468248248100281, "learning_rate": 0.0002, "epoch": 4.755700325732899, "step": 6570}, {"loss": 1.0858, "grad_norm": 0.8771235942840576, "learning_rate": 0.0002, "epoch": 4.762938834600073, "step": 6580}, {"loss": 1.077, "grad_norm": 0.7024846076965332, "learning_rate": 0.0002, "epoch": 4.770177343467246, "step": 6590}, {"loss": 1.0876, "grad_norm": 0.7836683392524719, "learning_rate": 0.0002, "epoch": 4.7774158523344195, "step": 6600}, {"loss": 1.1006, "grad_norm": 0.7717288136482239, "learning_rate": 0.0002, "epoch": 4.784654361201593, "step": 6610}, {"loss": 1.0376, "grad_norm": 0.884183943271637, "learning_rate": 0.0002, "epoch": 4.791892870068766, "step": 6620}, {"loss": 1.1757, "grad_norm": 1.383867621421814, "learning_rate": 0.0002, "epoch": 4.7991313789359396, "step": 6630}, {"loss": 1.0861, "grad_norm": 0.9741523861885071, "learning_rate": 0.0002, "epoch": 4.806369887803113, "step": 6640}, {"loss": 1.0884, "grad_norm": 0.9723693132400513, "learning_rate": 0.0002, "epoch": 4.813608396670286, "step": 6650}, {"loss": 1.2203, "grad_norm": 1.8324809074401855, "learning_rate": 0.0002, "epoch": 4.82084690553746, "step": 6660}, {"loss": 1.0292, "grad_norm": 0.904909074306488, "learning_rate": 0.0002, "epoch": 4.828085414404633, "step": 6670}, {"loss": 1.0349, "grad_norm": 0.7355411648750305, "learning_rate": 0.0002, "epoch": 4.835323923271806, "step": 6680}, {"loss": 1.0793, "grad_norm": 0.8934960961341858, "learning_rate": 0.0002, "epoch": 4.84256243213898, "step": 6690}, {"loss": 1.0375, "grad_norm": 1.4596954584121704, "learning_rate": 0.0002, "epoch": 4.849800941006153, "step": 6700}, {"loss": 1.1065, "grad_norm": 0.8310341238975525, "learning_rate": 0.0002, "epoch": 4.8570394498733265, "step": 6710}, {"loss": 1.1089, "grad_norm": 0.9709894061088562, "learning_rate": 0.0002, "epoch": 4.8642779587405, "step": 6720}, {"loss": 1.0069, "grad_norm": 0.852142333984375, "learning_rate": 0.0002, "epoch": 4.871516467607673, "step": 6730}, {"loss": 1.0507, "grad_norm": 1.0643625259399414, "learning_rate": 0.0002, "epoch": 4.878754976474847, "step": 6740}, {"loss": 1.056, "grad_norm": 0.9419508576393127, "learning_rate": 0.0002, "epoch": 4.88599348534202, "step": 6750}, {"loss": 1.1995, "grad_norm": 1.1818498373031616, "learning_rate": 0.0002, "epoch": 4.893231994209193, "step": 6760}, {"loss": 1.0925, "grad_norm": 0.9369569420814514, "learning_rate": 0.0002, "epoch": 4.900470503076367, "step": 6770}, {"loss": 1.1648, "grad_norm": 0.7012579441070557, "learning_rate": 0.0002, "epoch": 4.90770901194354, "step": 6780}, {"loss": 1.0926, "grad_norm": 0.9109319448471069, "learning_rate": 0.0002, "epoch": 4.914947520810713, "step": 6790}, {"loss": 1.0358, "grad_norm": 0.8077534437179565, "learning_rate": 0.0002, "epoch": 4.922186029677887, "step": 6800}, {"loss": 1.2549, "grad_norm": 0.7571148872375488, "learning_rate": 0.0002, "epoch": 4.92942453854506, "step": 6810}, {"loss": 0.9638, "grad_norm": 0.7325633764266968, "learning_rate": 0.0002, "epoch": 4.9366630474122335, "step": 6820}, {"loss": 1.0128, "grad_norm": 0.8465084433555603, "learning_rate": 0.0002, "epoch": 4.943901556279407, "step": 6830}, {"loss": 1.153, "grad_norm": 0.8753737807273865, "learning_rate": 0.0002, "epoch": 4.95114006514658, "step": 6840}, {"loss": 1.0247, "grad_norm": 0.9421748518943787, "learning_rate": 0.0002, "epoch": 4.958378574013754, "step": 6850}, {"loss": 1.1483, "grad_norm": 0.8245896697044373, "learning_rate": 0.0002, "epoch": 4.965617082880927, "step": 6860}, {"loss": 0.9905, "grad_norm": 0.8823089599609375, "learning_rate": 0.0002, "epoch": 4.9728555917481, "step": 6870}, {"loss": 1.1664, "grad_norm": 0.8406389355659485, "learning_rate": 0.0002, "epoch": 4.980094100615274, "step": 6880}, {"loss": 1.0944, "grad_norm": 0.9732868075370789, "learning_rate": 0.0002, "epoch": 4.987332609482447, "step": 6890}, {"loss": 1.1776, "grad_norm": 2.125141143798828, "learning_rate": 0.0002, "epoch": 4.99457111834962, "step": 6900}]}
+{"epoch": 6.0, "step": 8289, "epoch_duration": 1227.6048402786255, "total_accumulated_duration": 7895.64093542099, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 3020.60888671875}, "peak_memory_usage": {"GPU_0": 15079.2998046875}, "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-0.9-num-8828-sd-1/checkpoint-4144", "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": 4.7061, "grad_norm": 1.2523442506790161, "learning_rate": 0.0002, "epoch": 0.007238508867173362, "step": 10}, {"loss": 3.3493, "grad_norm": 1.8887330293655396, "learning_rate": 0.0002, "epoch": 0.014477017734346724, "step": 20}, {"loss": 2.7585, "grad_norm": 0.9668035507202148, "learning_rate": 0.0002, "epoch": 0.021715526601520086, "step": 30}, {"loss": 2.3699, "grad_norm": 2.9167306423187256, "learning_rate": 0.0002, "epoch": 0.028954035468693448, "step": 40}, {"loss": 2.2679, "grad_norm": 2.649867296218872, "learning_rate": 0.0002, "epoch": 0.036192544335866814, "step": 50}, {"loss": 2.2202, "grad_norm": 1.5120655298233032, "learning_rate": 0.0002, "epoch": 0.04343105320304017, "step": 60}, {"loss": 2.2026, "grad_norm": 0.7879868149757385, "learning_rate": 0.0002, "epoch": 0.05066956207021354, "step": 70}, {"loss": 1.9447, "grad_norm": 0.7616953253746033, "learning_rate": 0.0002, "epoch": 0.057908070937386896, "step": 80}, {"loss": 2.0112, "grad_norm": 1.8809149265289307, "learning_rate": 0.0002, "epoch": 0.06514657980456026, "step": 90}, {"loss": 1.8337, "grad_norm": 0.9294016361236572, "learning_rate": 0.0002, "epoch": 0.07238508867173363, "step": 100}, {"loss": 1.8419, "grad_norm": 0.7145281434059143, "learning_rate": 0.0002, "epoch": 0.07962359753890698, "step": 110}, {"loss": 2.0036, "grad_norm": 0.7564446330070496, "learning_rate": 0.0002, "epoch": 0.08686210640608034, "step": 120}, {"loss": 1.9306, "grad_norm": 1.1681925058364868, "learning_rate": 0.0002, "epoch": 0.09410061527325371, "step": 130}, {"loss": 1.7875, "grad_norm": 0.6708641648292542, "learning_rate": 0.0002, "epoch": 0.10133912414042708, "step": 140}, {"loss": 1.786, "grad_norm": 0.7625647783279419, "learning_rate": 0.0002, "epoch": 0.10857763300760044, "step": 150}, {"loss": 1.6687, "grad_norm": 0.8463464975357056, "learning_rate": 0.0002, "epoch": 0.11581614187477379, "step": 160}, {"loss": 1.6214, "grad_norm": 0.7502335906028748, "learning_rate": 0.0002, "epoch": 0.12305465074194716, "step": 170}, {"loss": 1.7433, "grad_norm": 0.6929958462715149, "learning_rate": 0.0002, "epoch": 0.13029315960912052, "step": 180}, {"loss": 1.6009, "grad_norm": 0.6798707842826843, "learning_rate": 0.0002, "epoch": 0.1375316684762939, "step": 190}, {"loss": 1.6208, "grad_norm": 0.7566508650779724, "learning_rate": 0.0002, "epoch": 0.14477017734346725, "step": 200}, {"loss": 1.5823, "grad_norm": 0.7196869850158691, "learning_rate": 0.0002, "epoch": 0.15200868621064062, "step": 210}, {"loss": 1.738, "grad_norm": 0.8401045799255371, "learning_rate": 0.0002, "epoch": 0.15924719507781396, "step": 220}, {"loss": 1.7574, "grad_norm": 0.8503773212432861, "learning_rate": 0.0002, "epoch": 0.16648570394498732, "step": 230}, {"loss": 1.7861, "grad_norm": 0.7183733582496643, "learning_rate": 0.0002, "epoch": 0.1737242128121607, "step": 240}, {"loss": 1.6693, "grad_norm": 0.7082605957984924, "learning_rate": 0.0002, "epoch": 0.18096272167933405, "step": 250}, {"loss": 1.619, "grad_norm": 0.9386326670646667, "learning_rate": 0.0002, "epoch": 0.18820123054650742, "step": 260}, {"loss": 1.6511, "grad_norm": 0.7332451939582825, "learning_rate": 0.0002, "epoch": 0.19543973941368079, "step": 270}, {"loss": 1.6353, "grad_norm": 0.7092869877815247, "learning_rate": 0.0002, "epoch": 0.20267824828085415, "step": 280}, {"loss": 1.5996, "grad_norm": 0.7256413698196411, "learning_rate": 0.0002, "epoch": 0.20991675714802752, "step": 290}, {"loss": 1.6754, "grad_norm": 0.6398681402206421, "learning_rate": 0.0002, "epoch": 0.21715526601520088, "step": 300}, {"loss": 1.397, "grad_norm": 0.6273287534713745, "learning_rate": 0.0002, "epoch": 0.22439377488237422, "step": 310}, {"loss": 1.5115, "grad_norm": 0.511648416519165, "learning_rate": 0.0002, "epoch": 0.23163228374954759, "step": 320}, {"loss": 1.5424, "grad_norm": 0.8677352070808411, "learning_rate": 0.0002, "epoch": 0.23887079261672095, "step": 330}, {"loss": 1.6779, "grad_norm": 0.6270743012428284, "learning_rate": 0.0002, "epoch": 0.24610930148389432, "step": 340}, {"loss": 1.626, "grad_norm": 0.7980281114578247, "learning_rate": 0.0002, "epoch": 0.2533478103510677, "step": 350}, {"loss": 1.5238, "grad_norm": 0.632486879825592, "learning_rate": 0.0002, "epoch": 0.26058631921824105, "step": 360}, {"loss": 1.5175, "grad_norm": 0.6527034640312195, "learning_rate": 0.0002, "epoch": 0.2678248280854144, "step": 370}, {"loss": 1.627, "grad_norm": 0.7672118544578552, "learning_rate": 0.0002, "epoch": 0.2750633369525878, "step": 380}, {"loss": 1.5605, "grad_norm": 0.6035117506980896, "learning_rate": 0.0002, "epoch": 0.28230184581976114, "step": 390}, {"loss": 1.4603, "grad_norm": 0.5955103039741516, "learning_rate": 0.0002, "epoch": 0.2895403546869345, "step": 400}, {"loss": 1.558, "grad_norm": 0.6015191674232483, "learning_rate": 0.0002, "epoch": 0.2967788635541079, "step": 410}, {"loss": 1.6091, "grad_norm": 0.6380982398986816, "learning_rate": 0.0002, "epoch": 0.30401737242128124, "step": 420}, {"loss": 1.5292, "grad_norm": 0.6707863211631775, "learning_rate": 0.0002, "epoch": 0.3112558812884546, "step": 430}, {"loss": 1.4426, "grad_norm": 0.7010176777839661, "learning_rate": 0.0002, "epoch": 0.3184943901556279, "step": 440}, {"loss": 1.5572, "grad_norm": 0.8263739943504333, "learning_rate": 0.0002, "epoch": 0.3257328990228013, "step": 450}, {"loss": 1.5188, "grad_norm": 0.7253276109695435, "learning_rate": 0.0002, "epoch": 0.33297140788997465, "step": 460}, {"loss": 1.584, "grad_norm": 0.5238934755325317, "learning_rate": 0.0002, "epoch": 0.340209916757148, "step": 470}, {"loss": 1.7035, "grad_norm": 0.7869495749473572, "learning_rate": 0.0002, "epoch": 0.3474484256243214, "step": 480}, {"loss": 1.5776, "grad_norm": 0.7485215663909912, "learning_rate": 0.0002, "epoch": 0.35468693449149474, "step": 490}, {"loss": 1.6274, "grad_norm": 0.5413193106651306, "learning_rate": 0.0002, "epoch": 0.3619254433586681, "step": 500}, {"loss": 1.7323, "grad_norm": 0.7615048885345459, "learning_rate": 0.0002, "epoch": 0.3691639522258415, "step": 510}, {"loss": 1.532, "grad_norm": 0.7685340046882629, "learning_rate": 0.0002, "epoch": 0.37640246109301484, "step": 520}, {"loss": 1.6312, "grad_norm": 0.6379081010818481, "learning_rate": 0.0002, "epoch": 0.3836409699601882, "step": 530}, {"loss": 1.5645, "grad_norm": 0.7946939468383789, "learning_rate": 0.0002, "epoch": 0.39087947882736157, "step": 540}, {"loss": 1.4001, "grad_norm": 0.6287278532981873, "learning_rate": 0.0002, "epoch": 0.39811798769453494, "step": 550}, {"loss": 1.5982, "grad_norm": 0.6811642646789551, "learning_rate": 0.0002, "epoch": 0.4053564965617083, "step": 560}, {"loss": 1.4953, "grad_norm": 0.671073317527771, "learning_rate": 0.0002, "epoch": 0.41259500542888167, "step": 570}, {"loss": 1.6753, "grad_norm": 0.6313900351524353, "learning_rate": 0.0002, "epoch": 0.41983351429605503, "step": 580}, {"loss": 1.546, "grad_norm": 0.5291772484779358, "learning_rate": 0.0002, "epoch": 0.4270720231632284, "step": 590}, {"loss": 1.5441, "grad_norm": 0.62503582239151, "learning_rate": 0.0002, "epoch": 0.43431053203040176, "step": 600}, {"loss": 1.6276, "grad_norm": 0.5777305364608765, "learning_rate": 0.0002, "epoch": 0.4415490408975751, "step": 610}, {"loss": 1.4758, "grad_norm": 0.7013497352600098, "learning_rate": 0.0002, "epoch": 0.44878754976474844, "step": 620}, {"loss": 1.4029, "grad_norm": 0.8044822216033936, "learning_rate": 0.0002, "epoch": 0.4560260586319218, "step": 630}, {"loss": 1.7195, "grad_norm": 0.672531247138977, "learning_rate": 0.0002, "epoch": 0.46326456749909517, "step": 640}, {"loss": 1.614, "grad_norm": 0.6233910322189331, "learning_rate": 0.0002, "epoch": 0.47050307636626854, "step": 650}, {"loss": 1.6041, "grad_norm": 0.651524543762207, "learning_rate": 0.0002, "epoch": 0.4777415852334419, "step": 660}, {"loss": 1.5842, "grad_norm": 0.7213939428329468, "learning_rate": 0.0002, "epoch": 0.48498009410061527, "step": 670}, {"loss": 1.5453, "grad_norm": 0.6541454792022705, "learning_rate": 0.0002, "epoch": 0.49221860296778863, "step": 680}, {"loss": 1.662, "grad_norm": 0.6568936109542847, "learning_rate": 0.0002, "epoch": 0.499457111834962, "step": 690}, {"loss": 1.624, "grad_norm": 0.7176415324211121, "learning_rate": 0.0002, "epoch": 0.5066956207021354, "step": 700}, {"loss": 1.6099, "grad_norm": 0.6553855538368225, "learning_rate": 0.0002, "epoch": 0.5139341295693087, "step": 710}, {"loss": 1.5508, "grad_norm": 0.5654335618019104, "learning_rate": 0.0002, "epoch": 0.5211726384364821, "step": 720}, {"loss": 1.392, "grad_norm": 0.5671001672744751, "learning_rate": 0.0002, "epoch": 0.5284111473036555, "step": 730}, {"loss": 1.388, "grad_norm": 0.7914412021636963, "learning_rate": 0.0002, "epoch": 0.5356496561708288, "step": 740}, {"loss": 1.5931, "grad_norm": 0.6172138452529907, "learning_rate": 0.0002, "epoch": 0.5428881650380022, "step": 750}, {"loss": 1.4018, "grad_norm": 0.6132623553276062, "learning_rate": 0.0002, "epoch": 0.5501266739051756, "step": 760}, {"loss": 1.513, "grad_norm": 0.654000461101532, "learning_rate": 0.0002, "epoch": 0.5573651827723489, "step": 770}, {"loss": 1.5035, "grad_norm": 0.5691370964050293, "learning_rate": 0.0002, "epoch": 0.5646036916395223, "step": 780}, {"loss": 1.65, "grad_norm": 0.7922580242156982, "learning_rate": 0.0002, "epoch": 0.5718422005066957, "step": 790}, {"loss": 1.4521, "grad_norm": 0.6831880211830139, "learning_rate": 0.0002, "epoch": 0.579080709373869, "step": 800}, {"loss": 1.4734, "grad_norm": 0.6740124821662903, "learning_rate": 0.0002, "epoch": 0.5863192182410424, "step": 810}, {"loss": 1.6498, "grad_norm": 1.380016803741455, "learning_rate": 0.0002, "epoch": 0.5935577271082157, "step": 820}, {"loss": 1.4642, "grad_norm": 0.6552878022193909, "learning_rate": 0.0002, "epoch": 0.6007962359753891, "step": 830}, {"loss": 1.6271, "grad_norm": 0.6649535298347473, "learning_rate": 0.0002, "epoch": 0.6080347448425625, "step": 840}, {"loss": 1.5886, "grad_norm": 0.561738133430481, "learning_rate": 0.0002, "epoch": 0.6152732537097358, "step": 850}, {"loss": 1.5364, "grad_norm": 0.6133047938346863, "learning_rate": 0.0002, "epoch": 0.6225117625769092, "step": 860}, {"loss": 1.3489, "grad_norm": 0.559843122959137, "learning_rate": 0.0002, "epoch": 0.6297502714440825, "step": 870}, {"loss": 1.4878, "grad_norm": 0.6117811799049377, "learning_rate": 0.0002, "epoch": 0.6369887803112558, "step": 880}, {"loss": 1.56, "grad_norm": 0.6209776401519775, "learning_rate": 0.0002, "epoch": 0.6442272891784292, "step": 890}, {"loss": 1.6747, "grad_norm": 0.6234082579612732, "learning_rate": 0.0002, "epoch": 0.6514657980456026, "step": 900}, {"loss": 1.6963, "grad_norm": 0.7623258233070374, "learning_rate": 0.0002, "epoch": 0.6587043069127759, "step": 910}, {"loss": 1.2424, "grad_norm": 0.6148061752319336, "learning_rate": 0.0002, "epoch": 0.6659428157799493, "step": 920}, {"loss": 1.4319, "grad_norm": 0.6682973504066467, "learning_rate": 0.0002, "epoch": 0.6731813246471227, "step": 930}, {"loss": 1.5377, "grad_norm": 0.5513041615486145, "learning_rate": 0.0002, "epoch": 0.680419833514296, "step": 940}, {"loss": 1.3991, "grad_norm": 0.5197525024414062, "learning_rate": 0.0002, "epoch": 0.6876583423814694, "step": 950}, {"loss": 1.4398, "grad_norm": 0.6490758061408997, "learning_rate": 0.0002, "epoch": 0.6948968512486428, "step": 960}, {"loss": 1.5251, "grad_norm": 0.6450682878494263, "learning_rate": 0.0002, "epoch": 0.7021353601158161, "step": 970}, {"loss": 1.5417, "grad_norm": 0.6203766465187073, "learning_rate": 0.0002, "epoch": 0.7093738689829895, "step": 980}, {"loss": 1.4575, "grad_norm": 0.6023609638214111, "learning_rate": 0.0002, "epoch": 0.7166123778501629, "step": 990}, {"loss": 1.4973, "grad_norm": 0.5765255093574524, "learning_rate": 0.0002, "epoch": 0.7238508867173362, "step": 1000}, {"loss": 1.483, "grad_norm": 0.6650075316429138, "learning_rate": 0.0002, "epoch": 0.7310893955845096, "step": 1010}, {"loss": 1.5959, "grad_norm": 0.5610854029655457, "learning_rate": 0.0002, "epoch": 0.738327904451683, "step": 1020}, {"loss": 1.5248, "grad_norm": 0.7072813510894775, "learning_rate": 0.0002, "epoch": 0.7455664133188563, "step": 1030}, {"loss": 1.5776, "grad_norm": 0.6815407872200012, "learning_rate": 0.0002, "epoch": 0.7528049221860297, "step": 1040}, {"loss": 1.4577, "grad_norm": 0.7932390570640564, "learning_rate": 0.0002, "epoch": 0.760043431053203, "step": 1050}, {"loss": 1.4515, "grad_norm": 0.5798183083534241, "learning_rate": 0.0002, "epoch": 0.7672819399203764, "step": 1060}, {"loss": 1.5053, "grad_norm": 0.7898504137992859, "learning_rate": 0.0002, "epoch": 0.7745204487875498, "step": 1070}, {"loss": 1.4776, "grad_norm": 0.4983280301094055, "learning_rate": 0.0002, "epoch": 0.7817589576547231, "step": 1080}, {"loss": 1.5007, "grad_norm": 0.691403329372406, "learning_rate": 0.0002, "epoch": 0.7889974665218965, "step": 1090}, {"loss": 1.5153, "grad_norm": 0.5394481420516968, "learning_rate": 0.0002, "epoch": 0.7962359753890699, "step": 1100}, {"loss": 1.6892, "grad_norm": 0.5136822462081909, "learning_rate": 0.0002, "epoch": 0.8034744842562432, "step": 1110}, {"loss": 1.4902, "grad_norm": 0.6828126907348633, "learning_rate": 0.0002, "epoch": 0.8107129931234166, "step": 1120}, {"loss": 1.4346, "grad_norm": 0.6799656748771667, "learning_rate": 0.0002, "epoch": 0.81795150199059, "step": 1130}, {"loss": 1.2678, "grad_norm": 0.5428406000137329, "learning_rate": 0.0002, "epoch": 0.8251900108577633, "step": 1140}, {"loss": 1.4072, "grad_norm": 0.4811290502548218, "learning_rate": 0.0002, "epoch": 0.8324285197249367, "step": 1150}, {"loss": 1.4512, "grad_norm": 0.5519434809684753, "learning_rate": 0.0002, "epoch": 0.8396670285921101, "step": 1160}, {"loss": 1.4072, "grad_norm": 0.9748060703277588, "learning_rate": 0.0002, "epoch": 0.8469055374592834, "step": 1170}, {"loss": 1.4309, "grad_norm": 0.712609589099884, "learning_rate": 0.0002, "epoch": 0.8541440463264568, "step": 1180}, {"loss": 1.434, "grad_norm": 0.6866157054901123, "learning_rate": 0.0002, "epoch": 0.8613825551936302, "step": 1190}, {"loss": 1.3704, "grad_norm": 0.5068854093551636, "learning_rate": 0.0002, "epoch": 0.8686210640608035, "step": 1200}, {"loss": 1.5601, "grad_norm": 0.6333245038986206, "learning_rate": 0.0002, "epoch": 0.8758595729279768, "step": 1210}, {"loss": 1.4636, "grad_norm": 0.6424421072006226, "learning_rate": 0.0002, "epoch": 0.8830980817951501, "step": 1220}, {"loss": 1.4186, "grad_norm": 0.4771921932697296, "learning_rate": 0.0002, "epoch": 0.8903365906623235, "step": 1230}, {"loss": 1.6323, "grad_norm": 0.5191764235496521, "learning_rate": 0.0002, "epoch": 0.8975750995294969, "step": 1240}, {"loss": 1.6105, "grad_norm": 0.756222128868103, "learning_rate": 0.0002, "epoch": 0.9048136083966702, "step": 1250}, {"loss": 1.4396, "grad_norm": 0.623823881149292, "learning_rate": 0.0002, "epoch": 0.9120521172638436, "step": 1260}, {"loss": 1.3097, "grad_norm": 0.8166571259498596, "learning_rate": 0.0002, "epoch": 0.919290626131017, "step": 1270}, {"loss": 1.4625, "grad_norm": 0.6059346795082092, "learning_rate": 0.0002, "epoch": 0.9265291349981903, "step": 1280}, {"loss": 1.3555, "grad_norm": 0.5842690467834473, "learning_rate": 0.0002, "epoch": 0.9337676438653637, "step": 1290}, {"loss": 1.5859, "grad_norm": 0.7649800777435303, "learning_rate": 0.0002, "epoch": 0.9410061527325371, "step": 1300}, {"loss": 1.5915, "grad_norm": 0.6420919895172119, "learning_rate": 0.0002, "epoch": 0.9482446615997104, "step": 1310}, {"loss": 1.453, "grad_norm": 0.7011452913284302, "learning_rate": 0.0002, "epoch": 0.9554831704668838, "step": 1320}, {"loss": 1.6766, "grad_norm": 0.5783746242523193, "learning_rate": 0.0002, "epoch": 0.9627216793340572, "step": 1330}, {"loss": 1.6308, "grad_norm": 0.5973192453384399, "learning_rate": 0.0002, "epoch": 0.9699601882012305, "step": 1340}, {"loss": 1.5901, "grad_norm": 0.6181833744049072, "learning_rate": 0.0002, "epoch": 0.9771986970684039, "step": 1350}, {"loss": 1.5258, "grad_norm": 0.5563396215438843, "learning_rate": 0.0002, "epoch": 0.9844372059355773, "step": 1360}, {"loss": 1.4508, "grad_norm": 0.45723360776901245, "learning_rate": 0.0002, "epoch": 0.9916757148027506, "step": 1370}, {"loss": 1.3291, "grad_norm": 0.5947498679161072, "learning_rate": 0.0002, "epoch": 0.998914223669924, "step": 1380}, {"eval_loss": 1.480796456336975, "eval_runtime": 27.3103, "eval_samples_per_second": 15.965, "eval_steps_per_second": 2.014, "epoch": 0.9996380745566413, "step": 1381}, {"loss": 1.3057, "grad_norm": 0.5599952936172485, "learning_rate": 0.0002, "epoch": 1.0061527325370974, "step": 1390}, {"loss": 1.4991, "grad_norm": 0.5932008028030396, "learning_rate": 0.0002, "epoch": 1.0133912414042707, "step": 1400}, {"loss": 1.4506, "grad_norm": 0.6194121837615967, "learning_rate": 0.0002, "epoch": 1.020629750271444, "step": 1410}, {"loss": 1.5966, "grad_norm": 0.6995621919631958, "learning_rate": 0.0002, "epoch": 1.0278682591386175, "step": 1420}, {"loss": 1.4153, "grad_norm": 0.7905810475349426, "learning_rate": 0.0002, "epoch": 1.0351067680057908, "step": 1430}, {"loss": 1.4414, "grad_norm": 0.7221615314483643, "learning_rate": 0.0002, "epoch": 1.0423452768729642, "step": 1440}, {"loss": 1.3859, "grad_norm": 0.6170642375946045, "learning_rate": 0.0002, "epoch": 1.0495837857401376, "step": 1450}, {"loss": 1.3806, "grad_norm": 0.5844094753265381, "learning_rate": 0.0002, "epoch": 1.056822294607311, "step": 1460}, {"loss": 1.4871, "grad_norm": 0.7731822729110718, "learning_rate": 0.0002, "epoch": 1.0640608034744843, "step": 1470}, {"loss": 1.4286, "grad_norm": 0.4554748237133026, "learning_rate": 0.0002, "epoch": 1.0712993123416577, "step": 1480}, {"loss": 1.3977, "grad_norm": 0.6923259496688843, "learning_rate": 0.0002, "epoch": 1.078537821208831, "step": 1490}, {"loss": 1.3936, "grad_norm": 0.6008219122886658, "learning_rate": 0.0002, "epoch": 1.0857763300760044, "step": 1500}, {"loss": 1.4821, "grad_norm": 0.6450045704841614, "learning_rate": 0.0002, "epoch": 1.0930148389431777, "step": 1510}, {"loss": 1.3295, "grad_norm": 0.7833753824234009, "learning_rate": 0.0002, "epoch": 1.1002533478103511, "step": 1520}, {"loss": 1.3424, "grad_norm": 0.5076758861541748, "learning_rate": 0.0002, "epoch": 1.1074918566775245, "step": 1530}, {"loss": 1.4043, "grad_norm": 0.5661332011222839, "learning_rate": 0.0002, "epoch": 1.1147303655446978, "step": 1540}, {"loss": 1.4963, "grad_norm": 0.6526919603347778, "learning_rate": 0.0002, "epoch": 1.1219688744118712, "step": 1550}, {"loss": 1.3671, "grad_norm": 0.5613082647323608, "learning_rate": 0.0002, "epoch": 1.1292073832790446, "step": 1560}, {"loss": 1.4458, "grad_norm": 0.6113885641098022, "learning_rate": 0.0002, "epoch": 1.136445892146218, "step": 1570}, {"loss": 1.3552, "grad_norm": 0.6732510328292847, "learning_rate": 0.0002, "epoch": 1.1436844010133913, "step": 1580}, {"loss": 1.3114, "grad_norm": 0.6146392226219177, "learning_rate": 0.0002, "epoch": 1.1509229098805647, "step": 1590}, {"loss": 1.411, "grad_norm": 0.6766974329948425, "learning_rate": 0.0002, "epoch": 1.158161418747738, "step": 1600}, {"loss": 1.2401, "grad_norm": 0.7621957659721375, "learning_rate": 0.0002, "epoch": 1.1653999276149114, "step": 1610}, {"loss": 1.3758, "grad_norm": 0.6959581971168518, "learning_rate": 0.0002, "epoch": 1.1726384364820848, "step": 1620}, {"loss": 1.382, "grad_norm": 0.6691278219223022, "learning_rate": 0.0002, "epoch": 1.1798769453492581, "step": 1630}, {"loss": 1.4147, "grad_norm": 0.4927774965763092, "learning_rate": 0.0002, "epoch": 1.1871154542164315, "step": 1640}, {"loss": 1.449, "grad_norm": 0.7724234461784363, "learning_rate": 0.0002, "epoch": 1.1943539630836049, "step": 1650}, {"loss": 1.4778, "grad_norm": 0.6817787885665894, "learning_rate": 0.0002, "epoch": 1.2015924719507782, "step": 1660}, {"loss": 1.3776, "grad_norm": 0.6500699520111084, "learning_rate": 0.0002, "epoch": 1.2088309808179516, "step": 1670}, {"loss": 1.3875, "grad_norm": 0.5703568458557129, "learning_rate": 0.0002, "epoch": 1.216069489685125, "step": 1680}, {"loss": 1.4735, "grad_norm": 0.6261579990386963, "learning_rate": 0.0002, "epoch": 1.2233079985522983, "step": 1690}, {"loss": 1.3898, "grad_norm": 0.651713490486145, "learning_rate": 0.0002, "epoch": 1.2305465074194717, "step": 1700}, {"loss": 1.4002, "grad_norm": 0.684399425983429, "learning_rate": 0.0002, "epoch": 1.237785016286645, "step": 1710}, {"loss": 1.5027, "grad_norm": 0.6996857523918152, "learning_rate": 0.0002, "epoch": 1.2450235251538184, "step": 1720}, {"loss": 1.3326, "grad_norm": 0.7102537751197815, "learning_rate": 0.0002, "epoch": 1.2522620340209918, "step": 1730}, {"loss": 1.3675, "grad_norm": 0.45809897780418396, "learning_rate": 0.0002, "epoch": 1.2595005428881652, "step": 1740}, {"loss": 1.4175, "grad_norm": 0.6377046704292297, "learning_rate": 0.0002, "epoch": 1.2667390517553385, "step": 1750}, {"loss": 1.3479, "grad_norm": 0.6965704560279846, "learning_rate": 0.0002, "epoch": 1.2739775606225119, "step": 1760}, {"loss": 1.5647, "grad_norm": 0.5688214302062988, "learning_rate": 0.0002, "epoch": 1.2812160694896852, "step": 1770}, {"loss": 1.3967, "grad_norm": 0.6384190320968628, "learning_rate": 0.0002, "epoch": 1.2884545783568586, "step": 1780}, {"loss": 1.3671, "grad_norm": 0.5629363656044006, "learning_rate": 0.0002, "epoch": 1.295693087224032, "step": 1790}, {"loss": 1.2292, "grad_norm": 0.6148255467414856, "learning_rate": 0.0002, "epoch": 1.3029315960912053, "step": 1800}, {"loss": 1.5806, "grad_norm": 0.655580997467041, "learning_rate": 0.0002, "epoch": 1.3101701049583787, "step": 1810}, {"loss": 1.2398, "grad_norm": 0.5642657279968262, "learning_rate": 0.0002, "epoch": 1.3174086138255519, "step": 1820}, {"loss": 1.3246, "grad_norm": 0.59607994556427, "learning_rate": 0.0002, "epoch": 1.3246471226927252, "step": 1830}, {"loss": 1.3274, "grad_norm": 0.5564199090003967, "learning_rate": 0.0002, "epoch": 1.3318856315598986, "step": 1840}, {"loss": 1.5834, "grad_norm": 0.6949955821037292, "learning_rate": 0.0002, "epoch": 1.339124140427072, "step": 1850}, {"loss": 1.4722, "grad_norm": 0.7036856412887573, "learning_rate": 0.0002, "epoch": 1.3463626492942453, "step": 1860}, {"loss": 1.333, "grad_norm": 0.722062885761261, "learning_rate": 0.0002, "epoch": 1.3536011581614187, "step": 1870}, {"loss": 1.4044, "grad_norm": 0.6098677515983582, "learning_rate": 0.0002, "epoch": 1.360839667028592, "step": 1880}, {"loss": 1.6217, "grad_norm": 0.5376402735710144, "learning_rate": 0.0002, "epoch": 1.3680781758957654, "step": 1890}, {"loss": 1.5071, "grad_norm": 0.6974610090255737, "learning_rate": 0.0002, "epoch": 1.3753166847629388, "step": 1900}, {"loss": 1.5854, "grad_norm": 0.6520763635635376, "learning_rate": 0.0002, "epoch": 1.3825551936301121, "step": 1910}, {"loss": 1.4271, "grad_norm": 0.6604374647140503, "learning_rate": 0.0002, "epoch": 1.3897937024972855, "step": 1920}, {"loss": 1.419, "grad_norm": 0.7364398241043091, "learning_rate": 0.0002, "epoch": 1.3970322113644589, "step": 1930}, {"loss": 1.4585, "grad_norm": 0.6849475502967834, "learning_rate": 0.0002, "epoch": 1.4042707202316322, "step": 1940}, {"loss": 1.5577, "grad_norm": 0.6562670469284058, "learning_rate": 0.0002, "epoch": 1.4115092290988056, "step": 1950}, {"loss": 1.4725, "grad_norm": 0.5695616006851196, "learning_rate": 0.0002, "epoch": 1.418747737965979, "step": 1960}, {"loss": 1.3088, "grad_norm": 0.5244464874267578, "learning_rate": 0.0002, "epoch": 1.4259862468331523, "step": 1970}, {"loss": 1.5069, "grad_norm": 0.6347293257713318, "learning_rate": 0.0002, "epoch": 1.4332247557003257, "step": 1980}, {"loss": 1.3502, "grad_norm": 0.5528361201286316, "learning_rate": 0.0002, "epoch": 1.440463264567499, "step": 1990}, {"loss": 1.3978, "grad_norm": 0.6987585425376892, "learning_rate": 0.0002, "epoch": 1.4477017734346724, "step": 2000}, {"loss": 1.4262, "grad_norm": 0.6568987369537354, "learning_rate": 0.0002, "epoch": 1.4549402823018458, "step": 2010}, {"loss": 1.4175, "grad_norm": 0.7665994763374329, "learning_rate": 0.0002, "epoch": 1.4621787911690192, "step": 2020}, {"loss": 1.244, "grad_norm": 0.5127707123756409, "learning_rate": 0.0002, "epoch": 1.4694173000361925, "step": 2030}, {"loss": 1.3699, "grad_norm": 0.5406824946403503, "learning_rate": 0.0002, "epoch": 1.476655808903366, "step": 2040}, {"loss": 1.3353, "grad_norm": 0.5990166664123535, "learning_rate": 0.0002, "epoch": 1.4838943177705393, "step": 2050}, {"loss": 1.2454, "grad_norm": 0.6186193823814392, "learning_rate": 0.0002, "epoch": 1.4911328266377126, "step": 2060}, {"loss": 1.428, "grad_norm": 0.6154307126998901, "learning_rate": 0.0002, "epoch": 1.498371335504886, "step": 2070}, {"loss": 1.4528, "grad_norm": 0.5606056451797485, "learning_rate": 0.0002, "epoch": 1.5056098443720594, "step": 2080}, {"loss": 1.2405, "grad_norm": 0.5006417036056519, "learning_rate": 0.0002, "epoch": 1.5128483532392327, "step": 2090}, {"loss": 1.4258, "grad_norm": 0.5968486070632935, "learning_rate": 0.0002, "epoch": 1.520086862106406, "step": 2100}, {"loss": 1.2752, "grad_norm": 0.5835496187210083, "learning_rate": 0.0002, "epoch": 1.5273253709735795, "step": 2110}, {"loss": 1.5443, "grad_norm": 0.6753535270690918, "learning_rate": 0.0002, "epoch": 1.5345638798407528, "step": 2120}, {"loss": 1.2139, "grad_norm": 0.7299720644950867, "learning_rate": 0.0002, "epoch": 1.5418023887079262, "step": 2130}, {"loss": 1.2364, "grad_norm": 0.5105988383293152, "learning_rate": 0.0002, "epoch": 1.5490408975750996, "step": 2140}, {"loss": 1.4528, "grad_norm": 0.5675431489944458, "learning_rate": 0.0002, "epoch": 1.556279406442273, "step": 2150}, {"loss": 1.4563, "grad_norm": 0.6246723532676697, "learning_rate": 0.0002, "epoch": 1.5635179153094463, "step": 2160}, {"loss": 1.5255, "grad_norm": 0.7291720509529114, "learning_rate": 0.0002, "epoch": 1.5707564241766196, "step": 2170}, {"loss": 1.5432, "grad_norm": 0.678114116191864, "learning_rate": 0.0002, "epoch": 1.577994933043793, "step": 2180}, {"loss": 1.5212, "grad_norm": 0.5136260986328125, "learning_rate": 0.0002, "epoch": 1.5852334419109664, "step": 2190}, {"loss": 1.3271, "grad_norm": 0.6359935998916626, "learning_rate": 0.0002, "epoch": 1.5924719507781397, "step": 2200}, {"loss": 1.4038, "grad_norm": 0.7650278806686401, "learning_rate": 0.0002, "epoch": 1.599710459645313, "step": 2210}, {"loss": 1.5478, "grad_norm": 0.7256110906600952, "learning_rate": 0.0002, "epoch": 1.6069489685124865, "step": 2220}, {"loss": 1.4387, "grad_norm": 0.688689649105072, "learning_rate": 0.0002, "epoch": 1.6141874773796598, "step": 2230}, {"loss": 1.4096, "grad_norm": 0.6045311093330383, "learning_rate": 0.0002, "epoch": 1.6214259862468332, "step": 2240}, {"loss": 1.4097, "grad_norm": 0.7064604163169861, "learning_rate": 0.0002, "epoch": 1.6286644951140063, "step": 2250}, {"loss": 1.3477, "grad_norm": 0.5309562087059021, "learning_rate": 0.0002, "epoch": 1.6359030039811797, "step": 2260}, {"loss": 1.4022, "grad_norm": 0.5687053203582764, "learning_rate": 0.0002, "epoch": 1.643141512848353, "step": 2270}, {"loss": 1.2977, "grad_norm": 0.535872757434845, "learning_rate": 0.0002, "epoch": 1.6503800217155264, "step": 2280}, {"loss": 1.3844, "grad_norm": 0.5502381920814514, "learning_rate": 0.0002, "epoch": 1.6576185305826998, "step": 2290}, {"loss": 1.3764, "grad_norm": 0.6158602237701416, "learning_rate": 0.0002, "epoch": 1.6648570394498732, "step": 2300}, {"loss": 1.3515, "grad_norm": 0.5804675817489624, "learning_rate": 0.0002, "epoch": 1.6720955483170465, "step": 2310}, {"loss": 1.2532, "grad_norm": 0.600742757320404, "learning_rate": 0.0002, "epoch": 1.67933405718422, "step": 2320}, {"loss": 1.477, "grad_norm": 0.7101941108703613, "learning_rate": 0.0002, "epoch": 1.6865725660513933, "step": 2330}, {"loss": 1.4849, "grad_norm": 0.7507809996604919, "learning_rate": 0.0002, "epoch": 1.6938110749185666, "step": 2340}, {"loss": 1.2703, "grad_norm": 0.768502414226532, "learning_rate": 0.0002, "epoch": 1.70104958378574, "step": 2350}, {"loss": 1.3332, "grad_norm": 0.4801851212978363, "learning_rate": 0.0002, "epoch": 1.7082880926529134, "step": 2360}, {"loss": 1.4158, "grad_norm": 0.5322122573852539, "learning_rate": 0.0002, "epoch": 1.7155266015200867, "step": 2370}, {"loss": 1.4136, "grad_norm": 0.587661862373352, "learning_rate": 0.0002, "epoch": 1.72276511038726, "step": 2380}, {"loss": 1.3771, "grad_norm": 0.6073525547981262, "learning_rate": 0.0002, "epoch": 1.7300036192544335, "step": 2390}, {"loss": 1.2754, "grad_norm": 0.6950460076332092, "learning_rate": 0.0002, "epoch": 1.7372421281216068, "step": 2400}, {"loss": 1.3858, "grad_norm": 0.5981102585792542, "learning_rate": 0.0002, "epoch": 1.7444806369887802, "step": 2410}, {"loss": 1.4075, "grad_norm": 0.544570803642273, "learning_rate": 0.0002, "epoch": 1.7517191458559536, "step": 2420}, {"loss": 1.3861, "grad_norm": 0.5304399728775024, "learning_rate": 0.0002, "epoch": 1.758957654723127, "step": 2430}, {"loss": 1.4244, "grad_norm": 0.7921594977378845, "learning_rate": 0.0002, "epoch": 1.7661961635903003, "step": 2440}, {"loss": 1.3053, "grad_norm": 0.6084808707237244, "learning_rate": 0.0002, "epoch": 1.7734346724574737, "step": 2450}, {"loss": 1.3781, "grad_norm": 0.8844701051712036, "learning_rate": 0.0002, "epoch": 1.780673181324647, "step": 2460}, {"loss": 1.3227, "grad_norm": 0.5729258060455322, "learning_rate": 0.0002, "epoch": 1.7879116901918204, "step": 2470}, {"loss": 1.3422, "grad_norm": 0.6303611993789673, "learning_rate": 0.0002, "epoch": 1.7951501990589938, "step": 2480}, {"loss": 1.3926, "grad_norm": 0.5627942085266113, "learning_rate": 0.0002, "epoch": 1.8023887079261671, "step": 2490}, {"loss": 1.3816, "grad_norm": 0.6724274158477783, "learning_rate": 0.0002, "epoch": 1.8096272167933405, "step": 2500}, {"loss": 1.2951, "grad_norm": 0.5030826330184937, "learning_rate": 0.0002, "epoch": 1.8168657256605139, "step": 2510}, {"loss": 1.2839, "grad_norm": 0.5504099130630493, "learning_rate": 0.0002, "epoch": 1.8241042345276872, "step": 2520}, {"loss": 1.4264, "grad_norm": 0.6338945627212524, "learning_rate": 0.0002, "epoch": 1.8313427433948606, "step": 2530}, {"loss": 1.563, "grad_norm": 0.5902037620544434, "learning_rate": 0.0002, "epoch": 1.838581252262034, "step": 2540}, {"loss": 1.2961, "grad_norm": 0.48814457654953003, "learning_rate": 0.0002, "epoch": 1.8458197611292073, "step": 2550}, {"loss": 1.466, "grad_norm": 0.6216312646865845, "learning_rate": 0.0002, "epoch": 1.8530582699963807, "step": 2560}, {"loss": 1.5123, "grad_norm": 0.635603666305542, "learning_rate": 0.0002, "epoch": 1.860296778863554, "step": 2570}, {"loss": 1.372, "grad_norm": 0.6938216090202332, "learning_rate": 0.0002, "epoch": 1.8675352877307274, "step": 2580}, {"loss": 1.5011, "grad_norm": 0.599557638168335, "learning_rate": 0.0002, "epoch": 1.8747737965979008, "step": 2590}, {"loss": 1.2714, "grad_norm": 0.564424455165863, "learning_rate": 0.0002, "epoch": 1.8820123054650741, "step": 2600}, {"loss": 1.3403, "grad_norm": 0.5430700182914734, "learning_rate": 0.0002, "epoch": 1.8892508143322475, "step": 2610}, {"loss": 1.4347, "grad_norm": 0.6150169372558594, "learning_rate": 0.0002, "epoch": 1.8964893231994209, "step": 2620}, {"loss": 1.2474, "grad_norm": 0.48159119486808777, "learning_rate": 0.0002, "epoch": 1.9037278320665942, "step": 2630}, {"loss": 1.3716, "grad_norm": 0.5608997941017151, "learning_rate": 0.0002, "epoch": 1.9109663409337676, "step": 2640}, {"loss": 1.5787, "grad_norm": 0.6454501748085022, "learning_rate": 0.0002, "epoch": 1.918204849800941, "step": 2650}, {"loss": 1.3238, "grad_norm": 0.5458073616027832, "learning_rate": 0.0002, "epoch": 1.9254433586681143, "step": 2660}, {"loss": 1.3208, "grad_norm": 0.5328490734100342, "learning_rate": 0.0002, "epoch": 1.9326818675352877, "step": 2670}, {"loss": 1.4971, "grad_norm": 0.6444696187973022, "learning_rate": 0.0002, "epoch": 1.939920376402461, "step": 2680}, {"loss": 1.5387, "grad_norm": 0.7126023769378662, "learning_rate": 0.0002, "epoch": 1.9471588852696344, "step": 2690}, {"loss": 1.3637, "grad_norm": 0.5164045095443726, "learning_rate": 0.0002, "epoch": 1.9543973941368078, "step": 2700}, {"loss": 1.5303, "grad_norm": 0.5347061157226562, "learning_rate": 0.0002, "epoch": 1.9616359030039812, "step": 2710}, {"loss": 1.2815, "grad_norm": 0.5297950506210327, "learning_rate": 0.0002, "epoch": 1.9688744118711545, "step": 2720}, {"loss": 1.3566, "grad_norm": 0.6537790298461914, "learning_rate": 0.0002, "epoch": 1.976112920738328, "step": 2730}, {"loss": 1.332, "grad_norm": 0.5536222457885742, "learning_rate": 0.0002, "epoch": 1.9833514296055013, "step": 2740}, {"loss": 1.3333, "grad_norm": 0.4856105446815491, "learning_rate": 0.0002, "epoch": 1.9905899384726746, "step": 2750}, {"loss": 1.3521, "grad_norm": 0.6642730832099915, "learning_rate": 0.0002, "epoch": 1.997828447339848, "step": 2760}, {"eval_loss": 1.4366681575775146, "eval_runtime": 27.3729, "eval_samples_per_second": 15.928, "eval_steps_per_second": 2.009, "epoch": 2.0, "step": 2763}, {"loss": 1.4322, "grad_norm": 0.740253210067749, "learning_rate": 0.0002, "epoch": 2.0050669562070214, "step": 2770}, {"loss": 1.277, "grad_norm": 0.5826276540756226, "learning_rate": 0.0002, "epoch": 2.0123054650741947, "step": 2780}, {"loss": 1.2424, "grad_norm": 0.607356071472168, "learning_rate": 0.0002, "epoch": 2.019543973941368, "step": 2790}, {"loss": 1.2601, "grad_norm": 0.5918063521385193, "learning_rate": 0.0002, "epoch": 2.0267824828085415, "step": 2800}, {"loss": 1.3715, "grad_norm": 0.5610089898109436, "learning_rate": 0.0002, "epoch": 2.034020991675715, "step": 2810}, {"loss": 1.2092, "grad_norm": 0.5869926810264587, "learning_rate": 0.0002, "epoch": 2.041259500542888, "step": 2820}, {"loss": 1.1929, "grad_norm": 0.5753467679023743, "learning_rate": 0.0002, "epoch": 2.0484980094100615, "step": 2830}, {"loss": 1.333, "grad_norm": 0.7096508145332336, "learning_rate": 0.0002, "epoch": 2.055736518277235, "step": 2840}, {"loss": 1.1766, "grad_norm": 0.7653635144233704, "learning_rate": 0.0002, "epoch": 2.0629750271444083, "step": 2850}, {"loss": 1.2331, "grad_norm": 0.6202841997146606, "learning_rate": 0.0002, "epoch": 2.0702135360115816, "step": 2860}, {"loss": 1.3298, "grad_norm": 0.6810227632522583, "learning_rate": 0.0002, "epoch": 2.077452044878755, "step": 2870}, {"loss": 1.2505, "grad_norm": 0.7481493353843689, "learning_rate": 0.0002, "epoch": 2.0846905537459284, "step": 2880}, {"loss": 1.2484, "grad_norm": 0.7089637517929077, "learning_rate": 0.0002, "epoch": 2.0919290626131017, "step": 2890}, {"loss": 1.3095, "grad_norm": 0.7472923398017883, "learning_rate": 0.0002, "epoch": 2.099167571480275, "step": 2900}, {"loss": 1.304, "grad_norm": 0.8135465979576111, "learning_rate": 0.0002, "epoch": 2.1064060803474485, "step": 2910}, {"loss": 1.273, "grad_norm": 0.6097133159637451, "learning_rate": 0.0002, "epoch": 2.113644589214622, "step": 2920}, {"loss": 1.3384, "grad_norm": 0.5970117449760437, "learning_rate": 0.0002, "epoch": 2.120883098081795, "step": 2930}, {"loss": 1.3233, "grad_norm": 0.6169309616088867, "learning_rate": 0.0002, "epoch": 2.1281216069489686, "step": 2940}, {"loss": 1.4246, "grad_norm": 0.9428738355636597, "learning_rate": 0.0002, "epoch": 2.135360115816142, "step": 2950}, {"loss": 1.3527, "grad_norm": 0.5671679973602295, "learning_rate": 0.0002, "epoch": 2.1425986246833153, "step": 2960}, {"loss": 1.1375, "grad_norm": 0.7007262110710144, "learning_rate": 0.0002, "epoch": 2.1498371335504887, "step": 2970}, {"loss": 1.2015, "grad_norm": 0.6294044256210327, "learning_rate": 0.0002, "epoch": 2.157075642417662, "step": 2980}, {"loss": 1.2167, "grad_norm": 0.6105241775512695, "learning_rate": 0.0002, "epoch": 2.1643141512848354, "step": 2990}, {"loss": 1.2065, "grad_norm": 0.557124137878418, "learning_rate": 0.0002, "epoch": 2.1715526601520088, "step": 3000}, {"loss": 1.2515, "grad_norm": 0.6250392198562622, "learning_rate": 0.0002, "epoch": 2.178791169019182, "step": 3010}, {"loss": 1.385, "grad_norm": 0.645218551158905, "learning_rate": 0.0002, "epoch": 2.1860296778863555, "step": 3020}, {"loss": 1.3928, "grad_norm": 0.9033605456352234, "learning_rate": 0.0002, "epoch": 2.193268186753529, "step": 3030}, {"loss": 1.2458, "grad_norm": 0.5325747132301331, "learning_rate": 0.0002, "epoch": 2.2005066956207022, "step": 3040}, {"loss": 1.261, "grad_norm": 0.6334700584411621, "learning_rate": 0.0002, "epoch": 2.2077452044878756, "step": 3050}, {"loss": 1.2385, "grad_norm": 0.5206325054168701, "learning_rate": 0.0002, "epoch": 2.214983713355049, "step": 3060}, {"loss": 1.3103, "grad_norm": 0.5987200140953064, "learning_rate": 0.0002, "epoch": 2.2222222222222223, "step": 3070}, {"loss": 1.1756, "grad_norm": 0.5893264412879944, "learning_rate": 0.0002, "epoch": 2.2294607310893957, "step": 3080}, {"loss": 1.235, "grad_norm": 0.6869237422943115, "learning_rate": 0.0002, "epoch": 2.236699239956569, "step": 3090}, {"loss": 1.3285, "grad_norm": 0.5040048360824585, "learning_rate": 0.0002, "epoch": 2.2439377488237424, "step": 3100}, {"loss": 1.3316, "grad_norm": 0.6660613417625427, "learning_rate": 0.0002, "epoch": 2.251176257690916, "step": 3110}, {"loss": 1.3108, "grad_norm": 0.5890918970108032, "learning_rate": 0.0002, "epoch": 2.258414766558089, "step": 3120}, {"loss": 1.248, "grad_norm": 0.6458896994590759, "learning_rate": 0.0002, "epoch": 2.2656532754252625, "step": 3130}, {"loss": 1.4151, "grad_norm": 0.6832690834999084, "learning_rate": 0.0002, "epoch": 2.272891784292436, "step": 3140}, {"loss": 1.4458, "grad_norm": 0.833908200263977, "learning_rate": 0.0002, "epoch": 2.2801302931596092, "step": 3150}, {"loss": 1.2931, "grad_norm": 0.4596034586429596, "learning_rate": 0.0002, "epoch": 2.2873688020267826, "step": 3160}, {"loss": 1.449, "grad_norm": 0.9130966067314148, "learning_rate": 0.0002, "epoch": 2.294607310893956, "step": 3170}, {"loss": 1.3806, "grad_norm": 0.7143292427062988, "learning_rate": 0.0002, "epoch": 2.3018458197611293, "step": 3180}, {"loss": 1.2692, "grad_norm": 0.5388900637626648, "learning_rate": 0.0002, "epoch": 2.3090843286283027, "step": 3190}, {"loss": 1.2402, "grad_norm": 0.5607513189315796, "learning_rate": 0.0002, "epoch": 2.316322837495476, "step": 3200}, {"loss": 1.3874, "grad_norm": 0.6795142292976379, "learning_rate": 0.0002, "epoch": 2.3235613463626494, "step": 3210}, {"loss": 1.3042, "grad_norm": 0.6561070680618286, "learning_rate": 0.0002, "epoch": 2.330799855229823, "step": 3220}, {"loss": 1.4636, "grad_norm": 0.8858118057250977, "learning_rate": 0.0002, "epoch": 2.338038364096996, "step": 3230}, {"loss": 1.3214, "grad_norm": 0.6604151725769043, "learning_rate": 0.0002, "epoch": 2.3452768729641695, "step": 3240}, {"loss": 1.4004, "grad_norm": 0.6755785346031189, "learning_rate": 0.0002, "epoch": 2.352515381831343, "step": 3250}, {"loss": 1.2503, "grad_norm": 0.6981677412986755, "learning_rate": 0.0002, "epoch": 2.3597538906985163, "step": 3260}, {"loss": 1.3078, "grad_norm": 0.6338568329811096, "learning_rate": 0.0002, "epoch": 2.3669923995656896, "step": 3270}, {"loss": 1.285, "grad_norm": 0.5754265785217285, "learning_rate": 0.0002, "epoch": 2.374230908432863, "step": 3280}, {"loss": 1.2924, "grad_norm": 0.7533153295516968, "learning_rate": 0.0002, "epoch": 2.3814694173000364, "step": 3290}, {"loss": 1.3711, "grad_norm": 0.675065279006958, "learning_rate": 0.0002, "epoch": 2.3887079261672097, "step": 3300}, {"loss": 1.3548, "grad_norm": 0.5686452984809875, "learning_rate": 0.0002, "epoch": 2.395946435034383, "step": 3310}, {"loss": 1.1998, "grad_norm": 0.8129481673240662, "learning_rate": 0.0002, "epoch": 2.4031849439015565, "step": 3320}, {"loss": 1.2584, "grad_norm": 0.6615934371948242, "learning_rate": 0.0002, "epoch": 2.41042345276873, "step": 3330}, {"loss": 1.3691, "grad_norm": 0.6678834557533264, "learning_rate": 0.0002, "epoch": 2.417661961635903, "step": 3340}, {"loss": 1.2381, "grad_norm": 0.5581308007240295, "learning_rate": 0.0002, "epoch": 2.4249004705030766, "step": 3350}, {"loss": 1.3853, "grad_norm": 0.6098920106887817, "learning_rate": 0.0002, "epoch": 2.43213897937025, "step": 3360}, {"loss": 1.3692, "grad_norm": 0.8101736903190613, "learning_rate": 0.0002, "epoch": 2.4393774882374233, "step": 3370}, {"loss": 1.4418, "grad_norm": 0.6621488928794861, "learning_rate": 0.0002, "epoch": 2.4466159971045967, "step": 3380}, {"loss": 1.4579, "grad_norm": 0.8693289160728455, "learning_rate": 0.0002, "epoch": 2.45385450597177, "step": 3390}, {"loss": 1.3644, "grad_norm": 0.6724580526351929, "learning_rate": 0.0002, "epoch": 2.4610930148389434, "step": 3400}, {"loss": 1.2006, "grad_norm": 0.6776891946792603, "learning_rate": 0.0002, "epoch": 2.4683315237061167, "step": 3410}, {"loss": 1.2937, "grad_norm": 0.7214453816413879, "learning_rate": 0.0002, "epoch": 2.47557003257329, "step": 3420}, {"loss": 1.4051, "grad_norm": 0.8390451073646545, "learning_rate": 0.0002, "epoch": 2.4828085414404635, "step": 3430}, {"loss": 1.25, "grad_norm": 0.7130982279777527, "learning_rate": 0.0002, "epoch": 2.490047050307637, "step": 3440}, {"loss": 1.2231, "grad_norm": 0.8873937129974365, "learning_rate": 0.0002, "epoch": 2.49728555917481, "step": 3450}, {"loss": 1.1429, "grad_norm": 0.725185751914978, "learning_rate": 0.0002, "epoch": 2.5045240680419836, "step": 3460}, {"loss": 1.2699, "grad_norm": 0.6120352149009705, "learning_rate": 0.0002, "epoch": 2.511762576909157, "step": 3470}, {"loss": 1.2552, "grad_norm": 0.7713613510131836, "learning_rate": 0.0002, "epoch": 2.5190010857763303, "step": 3480}, {"loss": 1.4648, "grad_norm": 0.895309567451477, "learning_rate": 0.0002, "epoch": 2.5262395946435037, "step": 3490}, {"loss": 1.3043, "grad_norm": 0.9631021022796631, "learning_rate": 0.0002, "epoch": 2.533478103510677, "step": 3500}, {"loss": 1.3492, "grad_norm": 0.7475683093070984, "learning_rate": 0.0002, "epoch": 2.5407166123778504, "step": 3510}, {"loss": 1.3637, "grad_norm": 0.7271341681480408, "learning_rate": 0.0002, "epoch": 2.5479551212450238, "step": 3520}, {"loss": 1.304, "grad_norm": 0.6979510188102722, "learning_rate": 0.0002, "epoch": 2.555193630112197, "step": 3530}, {"loss": 1.2353, "grad_norm": 0.6504196524620056, "learning_rate": 0.0002, "epoch": 2.5624321389793705, "step": 3540}, {"loss": 1.2699, "grad_norm": 0.7226675748825073, "learning_rate": 0.0002, "epoch": 2.569670647846544, "step": 3550}, {"loss": 1.3002, "grad_norm": 0.6143222451210022, "learning_rate": 0.0002, "epoch": 2.5769091567137172, "step": 3560}, {"loss": 1.1585, "grad_norm": 0.7245154976844788, "learning_rate": 0.0002, "epoch": 2.5841476655808906, "step": 3570}, {"loss": 1.3651, "grad_norm": 0.943540632724762, "learning_rate": 0.0002, "epoch": 2.591386174448064, "step": 3580}, {"loss": 1.3034, "grad_norm": 0.7707241773605347, "learning_rate": 0.0002, "epoch": 2.5986246833152373, "step": 3590}, {"loss": 1.3063, "grad_norm": 0.6705001592636108, "learning_rate": 0.0002, "epoch": 2.6058631921824107, "step": 3600}, {"loss": 1.2437, "grad_norm": 0.6360933780670166, "learning_rate": 0.0002, "epoch": 2.613101701049584, "step": 3610}, {"loss": 1.1844, "grad_norm": 0.5846424698829651, "learning_rate": 0.0002, "epoch": 2.6203402099167574, "step": 3620}, {"loss": 1.3674, "grad_norm": 0.5958625674247742, "learning_rate": 0.0002, "epoch": 2.6275787187839303, "step": 3630}, {"loss": 1.3599, "grad_norm": 0.6819243431091309, "learning_rate": 0.0002, "epoch": 2.6348172276511037, "step": 3640}, {"loss": 1.3884, "grad_norm": 0.7033445835113525, "learning_rate": 0.0002, "epoch": 2.642055736518277, "step": 3650}, {"loss": 1.3392, "grad_norm": 0.6134849786758423, "learning_rate": 0.0002, "epoch": 2.6492942453854504, "step": 3660}, {"loss": 1.2661, "grad_norm": 0.658009946346283, "learning_rate": 0.0002, "epoch": 2.656532754252624, "step": 3670}, {"loss": 1.3987, "grad_norm": 0.6280999779701233, "learning_rate": 0.0002, "epoch": 2.663771263119797, "step": 3680}, {"loss": 1.2995, "grad_norm": 0.5536085963249207, "learning_rate": 0.0002, "epoch": 2.6710097719869705, "step": 3690}, {"loss": 1.2044, "grad_norm": 0.8603981733322144, "learning_rate": 0.0002, "epoch": 2.678248280854144, "step": 3700}, {"loss": 1.3879, "grad_norm": 0.5509994626045227, "learning_rate": 0.0002, "epoch": 2.6854867897213173, "step": 3710}, {"loss": 1.3253, "grad_norm": 0.9093621969223022, "learning_rate": 0.0002, "epoch": 2.6927252985884906, "step": 3720}, {"loss": 1.2668, "grad_norm": 0.7525952458381653, "learning_rate": 0.0002, "epoch": 2.699963807455664, "step": 3730}, {"loss": 1.248, "grad_norm": 0.6737023591995239, "learning_rate": 0.0002, "epoch": 2.7072023163228374, "step": 3740}, {"loss": 1.2981, "grad_norm": 0.8656924962997437, "learning_rate": 0.0002, "epoch": 2.7144408251900107, "step": 3750}, {"loss": 1.2342, "grad_norm": 0.7494133114814758, "learning_rate": 0.0002, "epoch": 2.721679334057184, "step": 3760}, {"loss": 1.2417, "grad_norm": 0.5725520849227905, "learning_rate": 0.0002, "epoch": 2.7289178429243575, "step": 3770}, {"loss": 1.28, "grad_norm": 0.836412787437439, "learning_rate": 0.0002, "epoch": 2.736156351791531, "step": 3780}, {"loss": 1.3784, "grad_norm": 0.6893242597579956, "learning_rate": 0.0002, "epoch": 2.743394860658704, "step": 3790}, {"loss": 1.2929, "grad_norm": 0.6696223020553589, "learning_rate": 0.0002, "epoch": 2.7506333695258776, "step": 3800}, {"loss": 1.2449, "grad_norm": 0.6483015418052673, "learning_rate": 0.0002, "epoch": 2.757871878393051, "step": 3810}, {"loss": 1.3282, "grad_norm": 0.8084456920623779, "learning_rate": 0.0002, "epoch": 2.7651103872602243, "step": 3820}, {"loss": 1.3694, "grad_norm": 0.6601949334144592, "learning_rate": 0.0002, "epoch": 2.7723488961273977, "step": 3830}, {"loss": 1.3568, "grad_norm": 0.6905533671379089, "learning_rate": 0.0002, "epoch": 2.779587404994571, "step": 3840}, {"loss": 1.3854, "grad_norm": 0.619318425655365, "learning_rate": 0.0002, "epoch": 2.7868259138617444, "step": 3850}, {"loss": 1.2551, "grad_norm": 0.5994023084640503, "learning_rate": 0.0002, "epoch": 2.7940644227289178, "step": 3860}, {"loss": 1.2022, "grad_norm": 0.5627168416976929, "learning_rate": 0.0002, "epoch": 2.801302931596091, "step": 3870}, {"loss": 1.3921, "grad_norm": 0.6001605987548828, "learning_rate": 0.0002, "epoch": 2.8085414404632645, "step": 3880}, {"loss": 1.3026, "grad_norm": 0.6022412776947021, "learning_rate": 0.0002, "epoch": 2.815779949330438, "step": 3890}, {"loss": 1.2765, "grad_norm": 0.6832426190376282, "learning_rate": 0.0002, "epoch": 2.823018458197611, "step": 3900}, {"loss": 1.1363, "grad_norm": 0.5936811566352844, "learning_rate": 0.0002, "epoch": 2.8302569670647846, "step": 3910}, {"loss": 1.1707, "grad_norm": 0.6960572600364685, "learning_rate": 0.0002, "epoch": 2.837495475931958, "step": 3920}, {"loss": 1.4063, "grad_norm": 0.5913406610488892, "learning_rate": 0.0002, "epoch": 2.8447339847991313, "step": 3930}, {"loss": 1.3245, "grad_norm": 0.678154706954956, "learning_rate": 0.0002, "epoch": 2.8519724936663047, "step": 3940}, {"loss": 1.366, "grad_norm": 0.7898936867713928, "learning_rate": 0.0002, "epoch": 2.859211002533478, "step": 3950}, {"loss": 1.3948, "grad_norm": 0.9234195351600647, "learning_rate": 0.0002, "epoch": 2.8664495114006514, "step": 3960}, {"loss": 1.2773, "grad_norm": 0.5960825085639954, "learning_rate": 0.0002, "epoch": 2.8736880202678248, "step": 3970}, {"loss": 1.3127, "grad_norm": 0.677118182182312, "learning_rate": 0.0002, "epoch": 2.880926529134998, "step": 3980}, {"loss": 1.2652, "grad_norm": 0.6505142450332642, "learning_rate": 0.0002, "epoch": 2.8881650380021715, "step": 3990}, {"loss": 1.2078, "grad_norm": 0.550826907157898, "learning_rate": 0.0002, "epoch": 2.895403546869345, "step": 4000}, {"loss": 1.1811, "grad_norm": 0.6209215521812439, "learning_rate": 0.0002, "epoch": 2.9026420557365182, "step": 4010}, {"loss": 1.4001, "grad_norm": 0.6549018025398254, "learning_rate": 0.0002, "epoch": 2.9098805646036916, "step": 4020}, {"loss": 1.2285, "grad_norm": 0.570682168006897, "learning_rate": 0.0002, "epoch": 2.917119073470865, "step": 4030}, {"loss": 1.0832, "grad_norm": 1.1807632446289062, "learning_rate": 0.0002, "epoch": 2.9243575823380383, "step": 4040}, {"loss": 1.2693, "grad_norm": 0.7058857679367065, "learning_rate": 0.0002, "epoch": 2.9315960912052117, "step": 4050}, {"loss": 1.2905, "grad_norm": 0.5542812943458557, "learning_rate": 0.0002, "epoch": 2.938834600072385, "step": 4060}, {"loss": 1.33, "grad_norm": 0.63167804479599, "learning_rate": 0.0002, "epoch": 2.9460731089395584, "step": 4070}, {"loss": 1.3075, "grad_norm": 0.5702962279319763, "learning_rate": 0.0002, "epoch": 2.953311617806732, "step": 4080}, {"loss": 1.2007, "grad_norm": 0.620944082736969, "learning_rate": 0.0002, "epoch": 2.960550126673905, "step": 4090}, {"loss": 1.2864, "grad_norm": 0.5866289734840393, "learning_rate": 0.0002, "epoch": 2.9677886355410785, "step": 4100}, {"loss": 1.3293, "grad_norm": 0.560170590877533, "learning_rate": 0.0002, "epoch": 2.975027144408252, "step": 4110}, {"loss": 1.2071, "grad_norm": 0.675082802772522, "learning_rate": 0.0002, "epoch": 2.9822656532754253, "step": 4120}, {"loss": 1.2981, "grad_norm": 0.62708580493927, "learning_rate": 0.0002, "epoch": 2.9895041621425986, "step": 4130}, {"loss": 1.2758, "grad_norm": 0.7893929481506348, "learning_rate": 0.0002, "epoch": 2.996742671009772, "step": 4140}, {"eval_loss": 1.4217946529388428, "eval_runtime": 27.1596, "eval_samples_per_second": 16.053, "eval_steps_per_second": 2.025, "epoch": 2.9996380745566413, "step": 4144}, {"loss": 1.2152, "grad_norm": 0.7043836116790771, "learning_rate": 0.0002, "epoch": 3.0039811798769454, "step": 4150}, {"loss": 1.1664, "grad_norm": 0.6806283593177795, "learning_rate": 0.0002, "epoch": 3.0112196887441187, "step": 4160}, {"loss": 1.292, "grad_norm": 0.7684550285339355, "learning_rate": 0.0002, "epoch": 3.018458197611292, "step": 4170}, {"loss": 1.3467, "grad_norm": 0.7895237803459167, "learning_rate": 0.0002, "epoch": 3.0256967064784654, "step": 4180}, {"loss": 1.1324, "grad_norm": 0.7464531064033508, "learning_rate": 0.0002, "epoch": 3.032935215345639, "step": 4190}, {"loss": 1.1614, "grad_norm": 0.9358500838279724, "learning_rate": 0.0002, "epoch": 3.040173724212812, "step": 4200}, {"loss": 1.1834, "grad_norm": 1.1066628694534302, "learning_rate": 0.0002, "epoch": 3.0474122330799855, "step": 4210}, {"loss": 1.1557, "grad_norm": 0.6663267612457275, "learning_rate": 0.0002, "epoch": 3.054650741947159, "step": 4220}, {"loss": 1.1707, "grad_norm": 0.6669464707374573, "learning_rate": 0.0002, "epoch": 3.0618892508143323, "step": 4230}, {"loss": 1.1841, "grad_norm": 0.7052164077758789, "learning_rate": 0.0002, "epoch": 3.0691277596815056, "step": 4240}, {"loss": 1.2913, "grad_norm": 0.6118432879447937, "learning_rate": 0.0002, "epoch": 3.076366268548679, "step": 4250}, {"loss": 1.1526, "grad_norm": 0.6915903687477112, "learning_rate": 0.0002, "epoch": 3.0836047774158524, "step": 4260}, {"loss": 1.1348, "grad_norm": 0.7441644668579102, "learning_rate": 0.0002, "epoch": 3.0908432862830257, "step": 4270}, {"loss": 1.1672, "grad_norm": 0.823850691318512, "learning_rate": 0.0002, "epoch": 3.098081795150199, "step": 4280}, {"loss": 1.2655, "grad_norm": 0.9677883386611938, "learning_rate": 0.0002, "epoch": 3.1053203040173725, "step": 4290}, {"loss": 1.1794, "grad_norm": 0.7002579569816589, "learning_rate": 0.0002, "epoch": 3.112558812884546, "step": 4300}, {"loss": 1.135, "grad_norm": 0.778789758682251, "learning_rate": 0.0002, "epoch": 3.119797321751719, "step": 4310}, {"loss": 1.0818, "grad_norm": 0.7236007452011108, "learning_rate": 0.0002, "epoch": 3.1270358306188926, "step": 4320}, {"loss": 1.1803, "grad_norm": 0.8809133768081665, "learning_rate": 0.0002, "epoch": 3.134274339486066, "step": 4330}, {"loss": 1.2571, "grad_norm": 0.7924913167953491, "learning_rate": 0.0002, "epoch": 3.1415128483532393, "step": 4340}, {"loss": 1.1413, "grad_norm": 0.7437422275543213, "learning_rate": 0.0002, "epoch": 3.1487513572204127, "step": 4350}, {"loss": 1.2088, "grad_norm": 0.6428450345993042, "learning_rate": 0.0002, "epoch": 3.155989866087586, "step": 4360}, {"loss": 1.3032, "grad_norm": 0.7922873497009277, "learning_rate": 0.0002, "epoch": 3.1632283749547594, "step": 4370}, {"loss": 1.216, "grad_norm": 0.5252506732940674, "learning_rate": 0.0002, "epoch": 3.1704668838219328, "step": 4380}, {"loss": 1.1297, "grad_norm": 0.8570457696914673, "learning_rate": 0.0002, "epoch": 3.177705392689106, "step": 4390}, {"loss": 1.0994, "grad_norm": 0.7218987345695496, "learning_rate": 0.0002, "epoch": 3.1849439015562795, "step": 4400}, {"loss": 1.2891, "grad_norm": 0.6921393275260925, "learning_rate": 0.0002, "epoch": 3.192182410423453, "step": 4410}, {"loss": 1.2668, "grad_norm": 0.7386137843132019, "learning_rate": 0.0002, "epoch": 3.199420919290626, "step": 4420}, {"loss": 1.1654, "grad_norm": 0.6227759122848511, "learning_rate": 0.0002, "epoch": 3.2066594281577996, "step": 4430}, {"loss": 1.1752, "grad_norm": 0.7180278897285461, "learning_rate": 0.0002, "epoch": 3.213897937024973, "step": 4440}, {"loss": 1.1757, "grad_norm": 0.745830774307251, "learning_rate": 0.0002, "epoch": 3.2211364458921463, "step": 4450}, {"loss": 1.234, "grad_norm": 0.6766072511672974, "learning_rate": 0.0002, "epoch": 3.2283749547593197, "step": 4460}, {"loss": 1.1999, "grad_norm": 0.8325067162513733, "learning_rate": 0.0002, "epoch": 3.235613463626493, "step": 4470}, {"loss": 1.1606, "grad_norm": 0.7148305177688599, "learning_rate": 0.0002, "epoch": 3.2428519724936664, "step": 4480}, {"loss": 1.1383, "grad_norm": 0.7752676010131836, "learning_rate": 0.0002, "epoch": 3.25009048136084, "step": 4490}, {"loss": 1.3006, "grad_norm": 0.6776860952377319, "learning_rate": 0.0002, "epoch": 3.257328990228013, "step": 4500}, {"loss": 1.0796, "grad_norm": 0.704359769821167, "learning_rate": 0.0002, "epoch": 3.2645674990951865, "step": 4510}, {"loss": 1.2496, "grad_norm": 0.6880282163619995, "learning_rate": 0.0002, "epoch": 3.27180600796236, "step": 4520}, {"loss": 1.0947, "grad_norm": 0.8179270029067993, "learning_rate": 0.0002, "epoch": 3.2790445168295332, "step": 4530}, {"loss": 1.1909, "grad_norm": 0.6718448996543884, "learning_rate": 0.0002, "epoch": 3.2862830256967066, "step": 4540}, {"loss": 1.2708, "grad_norm": 0.8300657868385315, "learning_rate": 0.0002, "epoch": 3.29352153456388, "step": 4550}, {"loss": 1.2594, "grad_norm": 0.6433690786361694, "learning_rate": 0.0002, "epoch": 3.3007600434310533, "step": 4560}, {"loss": 1.2479, "grad_norm": 0.690262496471405, "learning_rate": 0.0002, "epoch": 3.3079985522982267, "step": 4570}, {"loss": 1.1342, "grad_norm": 0.7022852301597595, "learning_rate": 0.0002, "epoch": 3.3152370611654, "step": 4580}, {"loss": 1.0844, "grad_norm": 0.6438387632369995, "learning_rate": 0.0002, "epoch": 3.3224755700325734, "step": 4590}, {"loss": 1.17, "grad_norm": 0.6866899132728577, "learning_rate": 0.0002, "epoch": 3.329714078899747, "step": 4600}, {"loss": 1.1289, "grad_norm": 0.8233968019485474, "learning_rate": 0.0002, "epoch": 3.33695258776692, "step": 4610}, {"loss": 1.1855, "grad_norm": 0.7251574993133545, "learning_rate": 0.0002, "epoch": 3.3441910966340935, "step": 4620}, {"loss": 1.3403, "grad_norm": 0.7855110168457031, "learning_rate": 0.0002, "epoch": 3.351429605501267, "step": 4630}, {"loss": 1.2922, "grad_norm": 0.8487356305122375, "learning_rate": 0.0002, "epoch": 3.3586681143684403, "step": 4640}, {"loss": 1.2462, "grad_norm": 0.6429011225700378, "learning_rate": 0.0002, "epoch": 3.3659066232356136, "step": 4650}, {"loss": 1.129, "grad_norm": 0.7095270156860352, "learning_rate": 0.0002, "epoch": 3.373145132102787, "step": 4660}, {"loss": 1.262, "grad_norm": 0.6792303323745728, "learning_rate": 0.0002, "epoch": 3.3803836409699604, "step": 4670}, {"loss": 1.256, "grad_norm": 0.6784825921058655, "learning_rate": 0.0002, "epoch": 3.3876221498371337, "step": 4680}, {"loss": 1.0838, "grad_norm": 0.6362888216972351, "learning_rate": 0.0002, "epoch": 3.394860658704307, "step": 4690}, {"loss": 1.2165, "grad_norm": 0.7794778943061829, "learning_rate": 0.0002, "epoch": 3.4020991675714805, "step": 4700}, {"loss": 1.0644, "grad_norm": 0.7287485003471375, "learning_rate": 0.0002, "epoch": 3.409337676438654, "step": 4710}, {"loss": 1.2925, "grad_norm": 0.6481451392173767, "learning_rate": 0.0002, "epoch": 3.416576185305827, "step": 4720}, {"loss": 1.2121, "grad_norm": 0.9200371503829956, "learning_rate": 0.0002, "epoch": 3.4238146941730006, "step": 4730}, {"loss": 1.072, "grad_norm": 1.074180245399475, "learning_rate": 0.0002, "epoch": 3.431053203040174, "step": 4740}, {"loss": 1.0421, "grad_norm": 0.6722986698150635, "learning_rate": 0.0002, "epoch": 3.438291711907347, "step": 4750}, {"loss": 1.2258, "grad_norm": 0.7945933938026428, "learning_rate": 0.0002, "epoch": 3.44553022077452, "step": 4760}, {"loss": 1.0927, "grad_norm": 0.7624640464782715, "learning_rate": 0.0002, "epoch": 3.4527687296416936, "step": 4770}, {"loss": 1.2428, "grad_norm": 0.7763656377792358, "learning_rate": 0.0002, "epoch": 3.460007238508867, "step": 4780}, {"loss": 1.2584, "grad_norm": 0.7736947536468506, "learning_rate": 0.0002, "epoch": 3.4672457473760403, "step": 4790}, {"loss": 1.1953, "grad_norm": 0.8450354933738708, "learning_rate": 0.0002, "epoch": 3.4744842562432137, "step": 4800}, {"loss": 1.1362, "grad_norm": 0.6480133533477783, "learning_rate": 0.0002, "epoch": 3.481722765110387, "step": 4810}, {"loss": 1.1882, "grad_norm": 0.8437445759773254, "learning_rate": 0.0002, "epoch": 3.4889612739775604, "step": 4820}, {"loss": 1.1519, "grad_norm": 0.7781730890274048, "learning_rate": 0.0002, "epoch": 3.4961997828447338, "step": 4830}, {"loss": 1.1836, "grad_norm": 0.8523228168487549, "learning_rate": 0.0002, "epoch": 3.503438291711907, "step": 4840}, {"loss": 1.1672, "grad_norm": 0.6236732006072998, "learning_rate": 0.0002, "epoch": 3.5106768005790805, "step": 4850}, {"loss": 1.1926, "grad_norm": 0.7500787377357483, "learning_rate": 0.0002, "epoch": 3.517915309446254, "step": 4860}, {"loss": 1.1998, "grad_norm": 0.7665374875068665, "learning_rate": 0.0002, "epoch": 3.5251538183134272, "step": 4870}, {"loss": 1.1551, "grad_norm": 0.787857711315155, "learning_rate": 0.0002, "epoch": 3.5323923271806006, "step": 4880}, {"loss": 1.2758, "grad_norm": 0.970595121383667, "learning_rate": 0.0002, "epoch": 3.539630836047774, "step": 4890}, {"loss": 1.1274, "grad_norm": 0.6409347057342529, "learning_rate": 0.0002, "epoch": 3.5468693449149473, "step": 4900}, {"loss": 1.1596, "grad_norm": 0.888551652431488, "learning_rate": 0.0002, "epoch": 3.5541078537821207, "step": 4910}, {"loss": 1.1644, "grad_norm": 1.0808377265930176, "learning_rate": 0.0002, "epoch": 3.561346362649294, "step": 4920}, {"loss": 1.2564, "grad_norm": 0.7501053214073181, "learning_rate": 0.0002, "epoch": 3.5685848715164674, "step": 4930}, {"loss": 1.2351, "grad_norm": 0.7375240325927734, "learning_rate": 0.0002, "epoch": 3.575823380383641, "step": 4940}, {"loss": 1.3568, "grad_norm": 0.7075039744377136, "learning_rate": 0.0002, "epoch": 3.583061889250814, "step": 4950}, {"loss": 1.3355, "grad_norm": 0.939337432384491, "learning_rate": 0.0002, "epoch": 3.5903003981179875, "step": 4960}, {"loss": 1.1722, "grad_norm": 0.6717396974563599, "learning_rate": 0.0002, "epoch": 3.597538906985161, "step": 4970}, {"loss": 1.1186, "grad_norm": 0.7141643762588501, "learning_rate": 0.0002, "epoch": 3.6047774158523342, "step": 4980}, {"loss": 1.1011, "grad_norm": 0.7109216451644897, "learning_rate": 0.0002, "epoch": 3.6120159247195076, "step": 4990}, {"loss": 1.2178, "grad_norm": 0.7020776867866516, "learning_rate": 0.0002, "epoch": 3.619254433586681, "step": 5000}, {"loss": 1.1939, "grad_norm": 0.7158873677253723, "learning_rate": 0.0002, "epoch": 3.6264929424538543, "step": 5010}, {"loss": 1.2624, "grad_norm": 0.7062035202980042, "learning_rate": 0.0002, "epoch": 3.6337314513210277, "step": 5020}, {"loss": 1.0224, "grad_norm": 0.7081155776977539, "learning_rate": 0.0002, "epoch": 3.640969960188201, "step": 5030}, {"loss": 1.2195, "grad_norm": 1.2210607528686523, "learning_rate": 0.0002, "epoch": 3.6482084690553744, "step": 5040}, {"loss": 1.2596, "grad_norm": 0.6650236248970032, "learning_rate": 0.0002, "epoch": 3.655446977922548, "step": 5050}, {"loss": 1.1072, "grad_norm": 0.6884829998016357, "learning_rate": 0.0002, "epoch": 3.662685486789721, "step": 5060}, {"loss": 1.2292, "grad_norm": 0.7317819595336914, "learning_rate": 0.0002, "epoch": 3.6699239956568945, "step": 5070}, {"loss": 1.1917, "grad_norm": 0.7406691908836365, "learning_rate": 0.0002, "epoch": 3.677162504524068, "step": 5080}, {"loss": 1.2949, "grad_norm": 0.9009454250335693, "learning_rate": 0.0002, "epoch": 3.6844010133912413, "step": 5090}, {"loss": 1.1528, "grad_norm": 0.8189385533332825, "learning_rate": 0.0002, "epoch": 3.6916395222584146, "step": 5100}, {"loss": 1.3408, "grad_norm": 1.0793628692626953, "learning_rate": 0.0002, "epoch": 3.698878031125588, "step": 5110}, {"loss": 1.2417, "grad_norm": 0.8593027591705322, "learning_rate": 0.0002, "epoch": 3.7061165399927614, "step": 5120}, {"loss": 1.2141, "grad_norm": 0.8481812477111816, "learning_rate": 0.0002, "epoch": 3.7133550488599347, "step": 5130}, {"loss": 1.125, "grad_norm": 0.6527451276779175, "learning_rate": 0.0002, "epoch": 3.720593557727108, "step": 5140}, {"loss": 1.1584, "grad_norm": 0.9220114350318909, "learning_rate": 0.0002, "epoch": 3.7278320665942815, "step": 5150}, {"loss": 1.2267, "grad_norm": 1.0842019319534302, "learning_rate": 0.0002, "epoch": 3.735070575461455, "step": 5160}, {"loss": 1.3083, "grad_norm": 0.965453565120697, "learning_rate": 0.0002, "epoch": 3.742309084328628, "step": 5170}, {"loss": 1.1772, "grad_norm": 0.9903319478034973, "learning_rate": 0.0002, "epoch": 3.7495475931958016, "step": 5180}, {"loss": 1.2515, "grad_norm": 0.7434818148612976, "learning_rate": 0.0002, "epoch": 3.756786102062975, "step": 5190}, {"loss": 1.2631, "grad_norm": 0.6717280745506287, "learning_rate": 0.0002, "epoch": 3.7640246109301483, "step": 5200}, {"loss": 1.2012, "grad_norm": 0.7754665613174438, "learning_rate": 0.0002, "epoch": 3.7712631197973217, "step": 5210}, {"loss": 1.305, "grad_norm": 1.028374433517456, "learning_rate": 0.0002, "epoch": 3.778501628664495, "step": 5220}, {"loss": 1.1866, "grad_norm": 0.6026996374130249, "learning_rate": 0.0002, "epoch": 3.7857401375316684, "step": 5230}, {"loss": 1.1901, "grad_norm": 0.6978490948677063, "learning_rate": 0.0002, "epoch": 3.7929786463988417, "step": 5240}, {"loss": 1.2576, "grad_norm": 0.7303446531295776, "learning_rate": 0.0002, "epoch": 3.800217155266015, "step": 5250}, {"loss": 1.3173, "grad_norm": 1.0734210014343262, "learning_rate": 0.0002, "epoch": 3.8074556641331885, "step": 5260}, {"loss": 1.1137, "grad_norm": 0.6383201479911804, "learning_rate": 0.0002, "epoch": 3.814694173000362, "step": 5270}, {"loss": 1.0904, "grad_norm": 0.7742630243301392, "learning_rate": 0.0002, "epoch": 3.821932681867535, "step": 5280}, {"loss": 1.2232, "grad_norm": 0.8477074503898621, "learning_rate": 0.0002, "epoch": 3.8291711907347086, "step": 5290}, {"loss": 1.2047, "grad_norm": 0.6675317883491516, "learning_rate": 0.0002, "epoch": 3.836409699601882, "step": 5300}, {"loss": 1.2275, "grad_norm": 0.7515445351600647, "learning_rate": 0.0002, "epoch": 3.8436482084690553, "step": 5310}, {"loss": 1.2569, "grad_norm": 1.1441220045089722, "learning_rate": 0.0002, "epoch": 3.8508867173362287, "step": 5320}, {"loss": 1.1512, "grad_norm": 0.7968795895576477, "learning_rate": 0.0002, "epoch": 3.858125226203402, "step": 5330}, {"loss": 1.232, "grad_norm": 0.7842824459075928, "learning_rate": 0.0002, "epoch": 3.8653637350705754, "step": 5340}, {"loss": 1.1847, "grad_norm": 0.8272225260734558, "learning_rate": 0.0002, "epoch": 3.8726022439377488, "step": 5350}, {"loss": 1.1381, "grad_norm": 0.8413397669792175, "learning_rate": 0.0002, "epoch": 3.879840752804922, "step": 5360}, {"loss": 1.2349, "grad_norm": 1.141764760017395, "learning_rate": 0.0002, "epoch": 3.8870792616720955, "step": 5370}, {"loss": 1.212, "grad_norm": 0.9826975464820862, "learning_rate": 0.0002, "epoch": 3.894317770539269, "step": 5380}, {"loss": 1.1833, "grad_norm": 0.8598255515098572, "learning_rate": 0.0002, "epoch": 3.9015562794064422, "step": 5390}, {"loss": 1.1247, "grad_norm": 0.6271058320999146, "learning_rate": 0.0002, "epoch": 3.9087947882736156, "step": 5400}, {"loss": 1.2212, "grad_norm": 0.6379870772361755, "learning_rate": 0.0002, "epoch": 3.916033297140789, "step": 5410}, {"loss": 1.2481, "grad_norm": 1.0313376188278198, "learning_rate": 0.0002, "epoch": 3.9232718060079623, "step": 5420}, {"loss": 1.1872, "grad_norm": 0.8220619559288025, "learning_rate": 0.0002, "epoch": 3.9305103148751357, "step": 5430}, {"loss": 1.2006, "grad_norm": 0.7576116919517517, "learning_rate": 0.0002, "epoch": 3.937748823742309, "step": 5440}, {"loss": 1.1969, "grad_norm": 1.226235032081604, "learning_rate": 0.0002, "epoch": 3.9449873326094824, "step": 5450}, {"loss": 1.2945, "grad_norm": 0.7979229688644409, "learning_rate": 0.0002, "epoch": 3.952225841476656, "step": 5460}, {"loss": 1.1922, "grad_norm": 0.9911929965019226, "learning_rate": 0.0002, "epoch": 3.959464350343829, "step": 5470}, {"loss": 1.0924, "grad_norm": 0.643738865852356, "learning_rate": 0.0002, "epoch": 3.9667028592110025, "step": 5480}, {"loss": 1.0607, "grad_norm": 0.682305634021759, "learning_rate": 0.0002, "epoch": 3.973941368078176, "step": 5490}, {"loss": 1.2908, "grad_norm": 1.18373441696167, "learning_rate": 0.0002, "epoch": 3.9811798769453492, "step": 5500}, {"loss": 1.0889, "grad_norm": 0.7190203070640564, "learning_rate": 0.0002, "epoch": 3.9884183858125226, "step": 5510}, {"loss": 1.2745, "grad_norm": 0.7516948580741882, "learning_rate": 0.0002, "epoch": 3.995656894679696, "step": 5520}, {"eval_loss": 1.4252897500991821, "eval_runtime": 27.235, "eval_samples_per_second": 16.009, "eval_steps_per_second": 2.019, "epoch": 4.0, "step": 5526}, {"loss": 1.0088, "grad_norm": 0.6353074312210083, "learning_rate": 0.0002, "epoch": 4.002895403546869, "step": 5530}, {"loss": 1.0326, "grad_norm": 0.7424906492233276, "learning_rate": 0.0002, "epoch": 4.010133912414043, "step": 5540}, {"loss": 1.0667, "grad_norm": 0.8856638073921204, "learning_rate": 0.0002, "epoch": 4.017372421281216, "step": 5550}, {"loss": 1.0905, "grad_norm": 0.9627974033355713, "learning_rate": 0.0002, "epoch": 4.024610930148389, "step": 5560}, {"loss": 1.0965, "grad_norm": 0.9048978686332703, "learning_rate": 0.0002, "epoch": 4.031849439015563, "step": 5570}, {"loss": 1.1108, "grad_norm": 0.921119213104248, "learning_rate": 0.0002, "epoch": 4.039087947882736, "step": 5580}, {"loss": 1.1235, "grad_norm": 0.8654361963272095, "learning_rate": 0.0002, "epoch": 4.0463264567499095, "step": 5590}, {"loss": 1.0794, "grad_norm": 0.7947945594787598, "learning_rate": 0.0002, "epoch": 4.053564965617083, "step": 5600}, {"loss": 1.0674, "grad_norm": 0.8307326436042786, "learning_rate": 0.0002, "epoch": 4.060803474484256, "step": 5610}, {"loss": 1.0076, "grad_norm": 0.793273389339447, "learning_rate": 0.0002, "epoch": 4.06804198335143, "step": 5620}, {"loss": 1.0651, "grad_norm": 0.8748673796653748, "learning_rate": 0.0002, "epoch": 4.075280492218603, "step": 5630}, {"loss": 1.111, "grad_norm": 0.7926856279373169, "learning_rate": 0.0002, "epoch": 4.082519001085776, "step": 5640}, {"loss": 1.044, "grad_norm": 0.922645092010498, "learning_rate": 0.0002, "epoch": 4.08975750995295, "step": 5650}, {"loss": 1.109, "grad_norm": 0.9539641737937927, "learning_rate": 0.0002, "epoch": 4.096996018820123, "step": 5660}, {"loss": 1.0788, "grad_norm": 0.8674443364143372, "learning_rate": 0.0002, "epoch": 4.1042345276872965, "step": 5670}, {"loss": 0.9867, "grad_norm": 0.7097609043121338, "learning_rate": 0.0002, "epoch": 4.11147303655447, "step": 5680}, {"loss": 1.1154, "grad_norm": 0.8875522613525391, "learning_rate": 0.0002, "epoch": 4.118711545421643, "step": 5690}, {"loss": 1.1217, "grad_norm": 0.8583634495735168, "learning_rate": 0.0002, "epoch": 4.125950054288817, "step": 5700}, {"loss": 1.0973, "grad_norm": 0.6736377477645874, "learning_rate": 0.0002, "epoch": 4.13318856315599, "step": 5710}, {"loss": 1.1199, "grad_norm": 0.9349062442779541, "learning_rate": 0.0002, "epoch": 4.140427072023163, "step": 5720}, {"loss": 1.0508, "grad_norm": 1.0610365867614746, "learning_rate": 0.0002, "epoch": 4.147665580890337, "step": 5730}, {"loss": 1.1146, "grad_norm": 1.5838189125061035, "learning_rate": 0.0002, "epoch": 4.15490408975751, "step": 5740}, {"loss": 1.0222, "grad_norm": 0.747522234916687, "learning_rate": 0.0002, "epoch": 4.162142598624683, "step": 5750}, {"loss": 1.1328, "grad_norm": 1.3247915506362915, "learning_rate": 0.0002, "epoch": 4.169381107491857, "step": 5760}, {"loss": 1.1655, "grad_norm": 0.8750247955322266, "learning_rate": 0.0002, "epoch": 4.17661961635903, "step": 5770}, {"loss": 1.199, "grad_norm": 0.7914144992828369, "learning_rate": 0.0002, "epoch": 4.1838581252262035, "step": 5780}, {"loss": 1.1213, "grad_norm": 0.9493299126625061, "learning_rate": 0.0002, "epoch": 4.191096634093377, "step": 5790}, {"loss": 1.1515, "grad_norm": 0.7802295088768005, "learning_rate": 0.0002, "epoch": 4.19833514296055, "step": 5800}, {"loss": 1.0704, "grad_norm": 0.6987314820289612, "learning_rate": 0.0002, "epoch": 4.205573651827724, "step": 5810}, {"loss": 1.1699, "grad_norm": 0.9220341444015503, "learning_rate": 0.0002, "epoch": 4.212812160694897, "step": 5820}, {"loss": 1.1394, "grad_norm": 0.8932939767837524, "learning_rate": 0.0002, "epoch": 4.22005066956207, "step": 5830}, {"loss": 1.0048, "grad_norm": 0.920002818107605, "learning_rate": 0.0002, "epoch": 4.227289178429244, "step": 5840}, {"loss": 0.964, "grad_norm": 0.6662752032279968, "learning_rate": 0.0002, "epoch": 4.234527687296417, "step": 5850}, {"loss": 0.986, "grad_norm": 0.8679718971252441, "learning_rate": 0.0002, "epoch": 4.24176619616359, "step": 5860}, {"loss": 0.8991, "grad_norm": 0.7020887732505798, "learning_rate": 0.0002, "epoch": 4.249004705030764, "step": 5870}, {"loss": 1.1132, "grad_norm": 0.869611382484436, "learning_rate": 0.0002, "epoch": 4.256243213897937, "step": 5880}, {"loss": 1.1026, "grad_norm": 0.7796585559844971, "learning_rate": 0.0002, "epoch": 4.2634817227651105, "step": 5890}, {"loss": 1.0957, "grad_norm": 0.8978819251060486, "learning_rate": 0.0002, "epoch": 4.270720231632284, "step": 5900}, {"loss": 1.1325, "grad_norm": 1.0837205648422241, "learning_rate": 0.0002, "epoch": 4.277958740499457, "step": 5910}, {"loss": 1.1279, "grad_norm": 0.7584353089332581, "learning_rate": 0.0002, "epoch": 4.285197249366631, "step": 5920}, {"loss": 1.0513, "grad_norm": 0.7313185334205627, "learning_rate": 0.0002, "epoch": 4.292435758233804, "step": 5930}, {"loss": 1.1101, "grad_norm": 0.8004671335220337, "learning_rate": 0.0002, "epoch": 4.299674267100977, "step": 5940}, {"loss": 1.14, "grad_norm": 2.154958724975586, "learning_rate": 0.0002, "epoch": 4.306912775968151, "step": 5950}, {"loss": 1.1206, "grad_norm": 0.9163479804992676, "learning_rate": 0.0002, "epoch": 4.314151284835324, "step": 5960}, {"loss": 0.9941, "grad_norm": 0.9151589274406433, "learning_rate": 0.0002, "epoch": 4.321389793702497, "step": 5970}, {"loss": 1.0606, "grad_norm": 0.8624112010002136, "learning_rate": 0.0002, "epoch": 4.328628302569671, "step": 5980}, {"loss": 1.1625, "grad_norm": 0.9357741475105286, "learning_rate": 0.0002, "epoch": 4.335866811436844, "step": 5990}, {"loss": 1.0712, "grad_norm": 1.3482335805892944, "learning_rate": 0.0002, "epoch": 4.3431053203040175, "step": 6000}, {"loss": 1.1224, "grad_norm": 0.7156149744987488, "learning_rate": 0.0002, "epoch": 4.350343829171191, "step": 6010}, {"loss": 1.0753, "grad_norm": 0.8480049967765808, "learning_rate": 0.0002, "epoch": 4.357582338038364, "step": 6020}, {"loss": 1.051, "grad_norm": 0.8262244462966919, "learning_rate": 0.0002, "epoch": 4.364820846905538, "step": 6030}, {"loss": 0.9966, "grad_norm": 0.7733905911445618, "learning_rate": 0.0002, "epoch": 4.372059355772711, "step": 6040}, {"loss": 1.1008, "grad_norm": 0.8553919792175293, "learning_rate": 0.0002, "epoch": 4.379297864639884, "step": 6050}, {"loss": 1.1777, "grad_norm": 0.8666832447052002, "learning_rate": 0.0002, "epoch": 4.386536373507058, "step": 6060}, {"loss": 1.1934, "grad_norm": 0.9168295860290527, "learning_rate": 0.0002, "epoch": 4.393774882374231, "step": 6070}, {"loss": 1.0988, "grad_norm": 0.7315238118171692, "learning_rate": 0.0002, "epoch": 4.4010133912414044, "step": 6080}, {"loss": 1.1599, "grad_norm": 1.020263433456421, "learning_rate": 0.0002, "epoch": 4.408251900108578, "step": 6090}, {"loss": 1.133, "grad_norm": 0.9978243708610535, "learning_rate": 0.0002, "epoch": 4.415490408975751, "step": 6100}, {"loss": 1.1324, "grad_norm": 0.995453953742981, "learning_rate": 0.0002, "epoch": 4.4227289178429245, "step": 6110}, {"loss": 1.0957, "grad_norm": 0.9360884428024292, "learning_rate": 0.0002, "epoch": 4.429967426710098, "step": 6120}, {"loss": 0.9506, "grad_norm": 0.8099448084831238, "learning_rate": 0.0002, "epoch": 4.437205935577271, "step": 6130}, {"loss": 1.0887, "grad_norm": 0.8173841238021851, "learning_rate": 0.0002, "epoch": 4.444444444444445, "step": 6140}, {"loss": 1.1219, "grad_norm": 0.7972666025161743, "learning_rate": 0.0002, "epoch": 4.451682953311618, "step": 6150}, {"loss": 1.0226, "grad_norm": 0.7685779333114624, "learning_rate": 0.0002, "epoch": 4.458921462178791, "step": 6160}, {"loss": 1.0732, "grad_norm": 0.7872623801231384, "learning_rate": 0.0002, "epoch": 4.466159971045965, "step": 6170}, {"loss": 0.9911, "grad_norm": 0.7677070498466492, "learning_rate": 0.0002, "epoch": 4.473398479913138, "step": 6180}, {"loss": 1.0919, "grad_norm": 0.7878316044807434, "learning_rate": 0.0002, "epoch": 4.4806369887803115, "step": 6190}, {"loss": 1.018, "grad_norm": 0.8178079724311829, "learning_rate": 0.0002, "epoch": 4.487875497647485, "step": 6200}, {"loss": 1.0517, "grad_norm": 1.2820082902908325, "learning_rate": 0.0002, "epoch": 4.495114006514658, "step": 6210}, {"loss": 1.3101, "grad_norm": 0.9380832314491272, "learning_rate": 0.0002, "epoch": 4.502352515381832, "step": 6220}, {"loss": 0.9818, "grad_norm": 0.7810422778129578, "learning_rate": 0.0002, "epoch": 4.509591024249005, "step": 6230}, {"loss": 1.1677, "grad_norm": 1.1022917032241821, "learning_rate": 0.0002, "epoch": 4.516829533116178, "step": 6240}, {"loss": 1.1579, "grad_norm": 1.4275553226470947, "learning_rate": 0.0002, "epoch": 4.524068041983352, "step": 6250}, {"loss": 1.3237, "grad_norm": 0.7597777247428894, "learning_rate": 0.0002, "epoch": 4.531306550850525, "step": 6260}, {"loss": 1.1529, "grad_norm": 1.10992431640625, "learning_rate": 0.0002, "epoch": 4.538545059717698, "step": 6270}, {"loss": 1.0732, "grad_norm": 0.8981178998947144, "learning_rate": 0.0002, "epoch": 4.545783568584872, "step": 6280}, {"loss": 1.086, "grad_norm": 0.7863979339599609, "learning_rate": 0.0002, "epoch": 4.553022077452045, "step": 6290}, {"loss": 1.2008, "grad_norm": 0.9071474671363831, "learning_rate": 0.0002, "epoch": 4.5602605863192185, "step": 6300}, {"loss": 1.0916, "grad_norm": 0.7429424524307251, "learning_rate": 0.0002, "epoch": 4.567499095186392, "step": 6310}, {"loss": 1.095, "grad_norm": 1.0767850875854492, "learning_rate": 0.0002, "epoch": 4.574737604053565, "step": 6320}, {"loss": 1.1023, "grad_norm": 0.7885915637016296, "learning_rate": 0.0002, "epoch": 4.581976112920739, "step": 6330}, {"loss": 1.1131, "grad_norm": 0.8350457549095154, "learning_rate": 0.0002, "epoch": 4.589214621787912, "step": 6340}, {"loss": 1.0743, "grad_norm": 0.7853530645370483, "learning_rate": 0.0002, "epoch": 4.596453130655085, "step": 6350}, {"loss": 1.1912, "grad_norm": 1.1220661401748657, "learning_rate": 0.0002, "epoch": 4.603691639522259, "step": 6360}, {"loss": 1.0927, "grad_norm": 0.7959423065185547, "learning_rate": 0.0002, "epoch": 4.610930148389432, "step": 6370}, {"loss": 1.1542, "grad_norm": 0.7782652378082275, "learning_rate": 0.0002, "epoch": 4.618168657256605, "step": 6380}, {"loss": 1.0753, "grad_norm": 0.7882203459739685, "learning_rate": 0.0002, "epoch": 4.625407166123779, "step": 6390}, {"loss": 1.0676, "grad_norm": 0.8841899037361145, "learning_rate": 0.0002, "epoch": 4.632645674990952, "step": 6400}, {"loss": 1.0815, "grad_norm": 0.7936127781867981, "learning_rate": 0.0002, "epoch": 4.6398841838581255, "step": 6410}, {"loss": 1.0198, "grad_norm": 0.9213966131210327, "learning_rate": 0.0002, "epoch": 4.647122692725299, "step": 6420}, {"loss": 0.9872, "grad_norm": 0.9246473908424377, "learning_rate": 0.0002, "epoch": 4.654361201592472, "step": 6430}, {"loss": 1.1309, "grad_norm": 0.766572892665863, "learning_rate": 0.0002, "epoch": 4.661599710459646, "step": 6440}, {"loss": 1.1095, "grad_norm": 0.8596171736717224, "learning_rate": 0.0002, "epoch": 4.668838219326819, "step": 6450}, {"loss": 1.1869, "grad_norm": 0.8482751846313477, "learning_rate": 0.0002, "epoch": 4.676076728193992, "step": 6460}, {"loss": 1.0622, "grad_norm": 1.0826905965805054, "learning_rate": 0.0002, "epoch": 4.683315237061166, "step": 6470}, {"loss": 1.0256, "grad_norm": 1.1048457622528076, "learning_rate": 0.0002, "epoch": 4.690553745928339, "step": 6480}, {"loss": 1.0514, "grad_norm": 0.9429134726524353, "learning_rate": 0.0002, "epoch": 4.697792254795512, "step": 6490}, {"loss": 1.1351, "grad_norm": 0.8587502837181091, "learning_rate": 0.0002, "epoch": 4.705030763662686, "step": 6500}, {"loss": 1.0969, "grad_norm": 1.0387083292007446, "learning_rate": 0.0002, "epoch": 4.712269272529859, "step": 6510}, {"loss": 1.0493, "grad_norm": 0.7471951842308044, "learning_rate": 0.0002, "epoch": 4.7195077813970325, "step": 6520}, {"loss": 1.2632, "grad_norm": 0.8800424933433533, "learning_rate": 0.0002, "epoch": 4.726746290264206, "step": 6530}, {"loss": 1.2126, "grad_norm": 0.8136811852455139, "learning_rate": 0.0002, "epoch": 4.733984799131379, "step": 6540}, {"loss": 1.195, "grad_norm": 0.9910339713096619, "learning_rate": 0.0002, "epoch": 4.741223307998553, "step": 6550}, {"loss": 1.1201, "grad_norm": 1.0679163932800293, "learning_rate": 0.0002, "epoch": 4.748461816865726, "step": 6560}, {"loss": 1.0297, "grad_norm": 0.8468248248100281, "learning_rate": 0.0002, "epoch": 4.755700325732899, "step": 6570}, {"loss": 1.0858, "grad_norm": 0.8771235942840576, "learning_rate": 0.0002, "epoch": 4.762938834600073, "step": 6580}, {"loss": 1.077, "grad_norm": 0.7024846076965332, "learning_rate": 0.0002, "epoch": 4.770177343467246, "step": 6590}, {"loss": 1.0876, "grad_norm": 0.7836683392524719, "learning_rate": 0.0002, "epoch": 4.7774158523344195, "step": 6600}, {"loss": 1.1006, "grad_norm": 0.7717288136482239, "learning_rate": 0.0002, "epoch": 4.784654361201593, "step": 6610}, {"loss": 1.0376, "grad_norm": 0.884183943271637, "learning_rate": 0.0002, "epoch": 4.791892870068766, "step": 6620}, {"loss": 1.1757, "grad_norm": 1.383867621421814, "learning_rate": 0.0002, "epoch": 4.7991313789359396, "step": 6630}, {"loss": 1.0861, "grad_norm": 0.9741523861885071, "learning_rate": 0.0002, "epoch": 4.806369887803113, "step": 6640}, {"loss": 1.0884, "grad_norm": 0.9723693132400513, "learning_rate": 0.0002, "epoch": 4.813608396670286, "step": 6650}, {"loss": 1.2203, "grad_norm": 1.8324809074401855, "learning_rate": 0.0002, "epoch": 4.82084690553746, "step": 6660}, {"loss": 1.0292, "grad_norm": 0.904909074306488, "learning_rate": 0.0002, "epoch": 4.828085414404633, "step": 6670}, {"loss": 1.0349, "grad_norm": 0.7355411648750305, "learning_rate": 0.0002, "epoch": 4.835323923271806, "step": 6680}, {"loss": 1.0793, "grad_norm": 0.8934960961341858, "learning_rate": 0.0002, "epoch": 4.84256243213898, "step": 6690}, {"loss": 1.0375, "grad_norm": 1.4596954584121704, "learning_rate": 0.0002, "epoch": 4.849800941006153, "step": 6700}, {"loss": 1.1065, "grad_norm": 0.8310341238975525, "learning_rate": 0.0002, "epoch": 4.8570394498733265, "step": 6710}, {"loss": 1.1089, "grad_norm": 0.9709894061088562, "learning_rate": 0.0002, "epoch": 4.8642779587405, "step": 6720}, {"loss": 1.0069, "grad_norm": 0.852142333984375, "learning_rate": 0.0002, "epoch": 4.871516467607673, "step": 6730}, {"loss": 1.0507, "grad_norm": 1.0643625259399414, "learning_rate": 0.0002, "epoch": 4.878754976474847, "step": 6740}, {"loss": 1.056, "grad_norm": 0.9419508576393127, "learning_rate": 0.0002, "epoch": 4.88599348534202, "step": 6750}, {"loss": 1.1995, "grad_norm": 1.1818498373031616, "learning_rate": 0.0002, "epoch": 4.893231994209193, "step": 6760}, {"loss": 1.0925, "grad_norm": 0.9369569420814514, "learning_rate": 0.0002, "epoch": 4.900470503076367, "step": 6770}, {"loss": 1.1648, "grad_norm": 0.7012579441070557, "learning_rate": 0.0002, "epoch": 4.90770901194354, "step": 6780}, {"loss": 1.0926, "grad_norm": 0.9109319448471069, "learning_rate": 0.0002, "epoch": 4.914947520810713, "step": 6790}, {"loss": 1.0358, "grad_norm": 0.8077534437179565, "learning_rate": 0.0002, "epoch": 4.922186029677887, "step": 6800}, {"loss": 1.2549, "grad_norm": 0.7571148872375488, "learning_rate": 0.0002, "epoch": 4.92942453854506, "step": 6810}, {"loss": 0.9638, "grad_norm": 0.7325633764266968, "learning_rate": 0.0002, "epoch": 4.9366630474122335, "step": 6820}, {"loss": 1.0128, "grad_norm": 0.8465084433555603, "learning_rate": 0.0002, "epoch": 4.943901556279407, "step": 6830}, {"loss": 1.153, "grad_norm": 0.8753737807273865, "learning_rate": 0.0002, "epoch": 4.95114006514658, "step": 6840}, {"loss": 1.0247, "grad_norm": 0.9421748518943787, "learning_rate": 0.0002, "epoch": 4.958378574013754, "step": 6850}, {"loss": 1.1483, "grad_norm": 0.8245896697044373, "learning_rate": 0.0002, "epoch": 4.965617082880927, "step": 6860}, {"loss": 0.9905, "grad_norm": 0.8823089599609375, "learning_rate": 0.0002, "epoch": 4.9728555917481, "step": 6870}, {"loss": 1.1664, "grad_norm": 0.8406389355659485, "learning_rate": 0.0002, "epoch": 4.980094100615274, "step": 6880}, {"loss": 1.0944, "grad_norm": 0.9732868075370789, "learning_rate": 0.0002, "epoch": 4.987332609482447, "step": 6890}, {"loss": 1.1776, "grad_norm": 2.125141143798828, "learning_rate": 0.0002, "epoch": 4.99457111834962, "step": 6900}, {"eval_loss": 1.445176601409912, "eval_runtime": 27.2351, "eval_samples_per_second": 16.009, "eval_steps_per_second": 2.019, "epoch": 4.999638074556641, "step": 6907}, {"loss": 1.1362, "grad_norm": 0.9465792775154114, "learning_rate": 0.0002, "epoch": 5.001809627216793, "step": 6910}, {"loss": 0.982, "grad_norm": 1.2834891080856323, "learning_rate": 0.0002, "epoch": 5.009048136083966, "step": 6920}, {"loss": 0.9803, "grad_norm": 1.0297378301620483, "learning_rate": 0.0002, "epoch": 5.01628664495114, "step": 6930}, {"loss": 1.0447, "grad_norm": 1.1705161333084106, "learning_rate": 0.0002, "epoch": 5.023525153818313, "step": 6940}, {"loss": 1.0113, "grad_norm": 0.8293961882591248, "learning_rate": 0.0002, "epoch": 5.030763662685486, "step": 6950}, {"loss": 0.9203, "grad_norm": 1.0422210693359375, "learning_rate": 0.0002, "epoch": 5.03800217155266, "step": 6960}, {"loss": 1.0553, "grad_norm": 1.116104245185852, "learning_rate": 0.0002, "epoch": 5.045240680419833, "step": 6970}, {"loss": 0.9011, "grad_norm": 1.5118416547775269, "learning_rate": 0.0002, "epoch": 5.0524791892870065, "step": 6980}, {"loss": 0.9969, "grad_norm": 0.8383979797363281, "learning_rate": 0.0002, "epoch": 5.05971769815418, "step": 6990}, {"loss": 0.9659, "grad_norm": 1.3378649950027466, "learning_rate": 0.0002, "epoch": 5.066956207021353, "step": 7000}, {"loss": 1.0212, "grad_norm": 1.1840510368347168, "learning_rate": 0.0002, "epoch": 5.0741947158885266, "step": 7010}, {"loss": 0.9939, "grad_norm": 1.2354751825332642, "learning_rate": 0.0002, "epoch": 5.0814332247557, "step": 7020}, {"loss": 0.9831, "grad_norm": 1.3830451965332031, "learning_rate": 0.0002, "epoch": 5.088671733622873, "step": 7030}, {"loss": 1.1827, "grad_norm": 0.8101674318313599, "learning_rate": 0.0002, "epoch": 5.095910242490047, "step": 7040}, {"loss": 0.9255, "grad_norm": 0.897982656955719, "learning_rate": 0.0002, "epoch": 5.10314875135722, "step": 7050}, {"loss": 0.8784, "grad_norm": 1.2049678564071655, "learning_rate": 0.0002, "epoch": 5.110387260224393, "step": 7060}, {"loss": 1.0182, "grad_norm": 1.5912116765975952, "learning_rate": 0.0002, "epoch": 5.117625769091567, "step": 7070}, {"loss": 1.0909, "grad_norm": 0.9261530041694641, "learning_rate": 0.0002, "epoch": 5.12486427795874, "step": 7080}, {"loss": 0.9603, "grad_norm": 1.1454812288284302, "learning_rate": 0.0002, "epoch": 5.1321027868259135, "step": 7090}, {"loss": 0.9149, "grad_norm": 1.0049978494644165, "learning_rate": 0.0002, "epoch": 5.139341295693087, "step": 7100}, {"loss": 0.9463, "grad_norm": 1.4513251781463623, "learning_rate": 0.0002, "epoch": 5.14657980456026, "step": 7110}, {"loss": 0.8995, "grad_norm": 0.9800849556922913, "learning_rate": 0.0002, "epoch": 5.153818313427434, "step": 7120}, {"loss": 0.9835, "grad_norm": 0.9698708653450012, "learning_rate": 0.0002, "epoch": 5.161056822294607, "step": 7130}, {"loss": 0.9672, "grad_norm": 1.1126646995544434, "learning_rate": 0.0002, "epoch": 5.16829533116178, "step": 7140}, {"loss": 0.9384, "grad_norm": 0.9248330593109131, "learning_rate": 0.0002, "epoch": 5.175533840028954, "step": 7150}, {"loss": 0.826, "grad_norm": 0.7967255711555481, "learning_rate": 0.0002, "epoch": 5.182772348896127, "step": 7160}, {"loss": 1.0078, "grad_norm": 0.9933333992958069, "learning_rate": 0.0002, "epoch": 5.1900108577633, "step": 7170}, {"loss": 1.0276, "grad_norm": 1.0080649852752686, "learning_rate": 0.0002, "epoch": 5.197249366630474, "step": 7180}, {"loss": 1.0201, "grad_norm": 1.3954921960830688, "learning_rate": 0.0002, "epoch": 5.204487875497647, "step": 7190}, {"loss": 1.0863, "grad_norm": 1.2386271953582764, "learning_rate": 0.0002, "epoch": 5.2117263843648205, "step": 7200}, {"loss": 0.8863, "grad_norm": 1.2379488945007324, "learning_rate": 0.0002, "epoch": 5.218964893231994, "step": 7210}, {"loss": 1.0518, "grad_norm": 0.9882503747940063, "learning_rate": 0.0002, "epoch": 5.226203402099167, "step": 7220}, {"loss": 0.9834, "grad_norm": 1.1728729009628296, "learning_rate": 0.0002, "epoch": 5.233441910966341, "step": 7230}, {"loss": 0.9269, "grad_norm": 0.9849673509597778, "learning_rate": 0.0002, "epoch": 5.240680419833514, "step": 7240}, {"loss": 0.9935, "grad_norm": 1.177639365196228, "learning_rate": 0.0002, "epoch": 5.247918928700687, "step": 7250}, {"loss": 1.0639, "grad_norm": 1.2395055294036865, "learning_rate": 0.0002, "epoch": 5.255157437567861, "step": 7260}, {"loss": 1.0138, "grad_norm": 1.3999171257019043, "learning_rate": 0.0002, "epoch": 5.262395946435034, "step": 7270}, {"loss": 0.9745, "grad_norm": 0.7698732018470764, "learning_rate": 0.0002, "epoch": 5.269634455302207, "step": 7280}, {"loss": 1.0389, "grad_norm": 0.9167453646659851, "learning_rate": 0.0002, "epoch": 5.276872964169381, "step": 7290}, {"loss": 0.9858, "grad_norm": 1.113830804824829, "learning_rate": 0.0002, "epoch": 5.284111473036554, "step": 7300}, {"loss": 0.9577, "grad_norm": 0.9644396901130676, "learning_rate": 0.0002, "epoch": 5.2913499819037275, "step": 7310}, {"loss": 1.0556, "grad_norm": 1.462435007095337, "learning_rate": 0.0002, "epoch": 5.298588490770901, "step": 7320}, {"loss": 0.871, "grad_norm": 0.9406287670135498, "learning_rate": 0.0002, "epoch": 5.305826999638074, "step": 7330}, {"loss": 1.0022, "grad_norm": 0.9698247909545898, "learning_rate": 0.0002, "epoch": 5.313065508505248, "step": 7340}, {"loss": 0.915, "grad_norm": 1.12003755569458, "learning_rate": 0.0002, "epoch": 5.320304017372421, "step": 7350}, {"loss": 0.9838, "grad_norm": 1.598681926727295, "learning_rate": 0.0002, "epoch": 5.327542526239594, "step": 7360}, {"loss": 1.0, "grad_norm": 1.0450010299682617, "learning_rate": 0.0002, "epoch": 5.334781035106768, "step": 7370}, {"loss": 0.9983, "grad_norm": 0.8680008053779602, "learning_rate": 0.0002, "epoch": 5.342019543973941, "step": 7380}, {"loss": 0.9851, "grad_norm": 1.0115476846694946, "learning_rate": 0.0002, "epoch": 5.349258052841114, "step": 7390}, {"loss": 1.0702, "grad_norm": 0.9589748382568359, "learning_rate": 0.0002, "epoch": 5.356496561708288, "step": 7400}, {"loss": 0.9366, "grad_norm": 0.6729998588562012, "learning_rate": 0.0002, "epoch": 5.363735070575461, "step": 7410}, {"loss": 1.0126, "grad_norm": 0.9246699213981628, "learning_rate": 0.0002, "epoch": 5.3709735794426345, "step": 7420}, {"loss": 0.9815, "grad_norm": 1.1266791820526123, "learning_rate": 0.0002, "epoch": 5.378212088309808, "step": 7430}, {"loss": 1.1166, "grad_norm": 1.8056942224502563, "learning_rate": 0.0002, "epoch": 5.385450597176981, "step": 7440}, {"loss": 0.9604, "grad_norm": 0.9802932739257812, "learning_rate": 0.0002, "epoch": 5.392689106044155, "step": 7450}, {"loss": 0.9656, "grad_norm": 1.0504707098007202, "learning_rate": 0.0002, "epoch": 5.399927614911328, "step": 7460}, {"loss": 1.0132, "grad_norm": 1.1915022134780884, "learning_rate": 0.0002, "epoch": 5.407166123778501, "step": 7470}, {"loss": 1.0041, "grad_norm": 1.1856611967086792, "learning_rate": 0.0002, "epoch": 5.414404632645675, "step": 7480}, {"loss": 0.9747, "grad_norm": 1.292152762413025, "learning_rate": 0.0002, "epoch": 5.421643141512848, "step": 7490}, {"loss": 0.9659, "grad_norm": 1.2675740718841553, "learning_rate": 0.0002, "epoch": 5.4288816503800215, "step": 7500}, {"loss": 1.0271, "grad_norm": 1.4034695625305176, "learning_rate": 0.0002, "epoch": 5.436120159247195, "step": 7510}, {"loss": 1.0318, "grad_norm": 0.984588623046875, "learning_rate": 0.0002, "epoch": 5.443358668114368, "step": 7520}, {"loss": 1.0726, "grad_norm": 0.8419108390808105, "learning_rate": 0.0002, "epoch": 5.450597176981542, "step": 7530}, {"loss": 1.0499, "grad_norm": 1.0270143747329712, "learning_rate": 0.0002, "epoch": 5.457835685848715, "step": 7540}, {"loss": 0.9804, "grad_norm": 2.2158689498901367, "learning_rate": 0.0002, "epoch": 5.465074194715888, "step": 7550}, {"loss": 0.9856, "grad_norm": 1.0740524530410767, "learning_rate": 0.0002, "epoch": 5.472312703583062, "step": 7560}, {"loss": 1.0522, "grad_norm": 1.3804482221603394, "learning_rate": 0.0002, "epoch": 5.479551212450235, "step": 7570}, {"loss": 1.0297, "grad_norm": 0.9428979754447937, "learning_rate": 0.0002, "epoch": 5.486789721317408, "step": 7580}, {"loss": 1.0906, "grad_norm": 0.9548295736312866, "learning_rate": 0.0002, "epoch": 5.494028230184582, "step": 7590}, {"loss": 0.8853, "grad_norm": 1.0691065788269043, "learning_rate": 0.0002, "epoch": 5.501266739051755, "step": 7600}, {"loss": 1.0375, "grad_norm": 1.0987380743026733, "learning_rate": 0.0002, "epoch": 5.5085052479189285, "step": 7610}, {"loss": 1.0162, "grad_norm": 0.9483979344367981, "learning_rate": 0.0002, "epoch": 5.515743756786102, "step": 7620}, {"loss": 1.105, "grad_norm": 1.16624915599823, "learning_rate": 0.0002, "epoch": 5.522982265653275, "step": 7630}, {"loss": 0.8695, "grad_norm": 0.8563777208328247, "learning_rate": 0.0002, "epoch": 5.530220774520449, "step": 7640}, {"loss": 0.9297, "grad_norm": 1.268186092376709, "learning_rate": 0.0002, "epoch": 5.537459283387622, "step": 7650}, {"loss": 1.1152, "grad_norm": 1.0752092599868774, "learning_rate": 0.0002, "epoch": 5.544697792254795, "step": 7660}, {"loss": 0.9344, "grad_norm": 1.210389256477356, "learning_rate": 0.0002, "epoch": 5.551936301121969, "step": 7670}, {"loss": 1.0349, "grad_norm": 1.669063925743103, "learning_rate": 0.0002, "epoch": 5.559174809989142, "step": 7680}, {"loss": 0.9833, "grad_norm": 1.038020133972168, "learning_rate": 0.0002, "epoch": 5.566413318856315, "step": 7690}, {"loss": 0.8907, "grad_norm": 1.316673994064331, "learning_rate": 0.0002, "epoch": 5.573651827723489, "step": 7700}, {"loss": 0.9614, "grad_norm": 1.029935359954834, "learning_rate": 0.0002, "epoch": 5.580890336590662, "step": 7710}, {"loss": 1.0409, "grad_norm": 0.9401940703392029, "learning_rate": 0.0002, "epoch": 5.5881288454578355, "step": 7720}, {"loss": 0.9272, "grad_norm": 2.4811816215515137, "learning_rate": 0.0002, "epoch": 5.595367354325009, "step": 7730}, {"loss": 0.992, "grad_norm": 1.0329105854034424, "learning_rate": 0.0002, "epoch": 5.602605863192182, "step": 7740}, {"loss": 0.9493, "grad_norm": 1.479629635810852, "learning_rate": 0.0002, "epoch": 5.609844372059356, "step": 7750}, {"loss": 1.0727, "grad_norm": 1.9232319593429565, "learning_rate": 0.0002, "epoch": 5.617082880926529, "step": 7760}, {"loss": 1.0741, "grad_norm": 1.0055509805679321, "learning_rate": 0.0002, "epoch": 5.624321389793702, "step": 7770}, {"loss": 1.0731, "grad_norm": 1.0037437677383423, "learning_rate": 0.0002, "epoch": 5.631559898660876, "step": 7780}, {"loss": 1.0913, "grad_norm": 1.4245030879974365, "learning_rate": 0.0002, "epoch": 5.638798407528049, "step": 7790}, {"loss": 0.9711, "grad_norm": 1.080687403678894, "learning_rate": 0.0002, "epoch": 5.646036916395222, "step": 7800}, {"loss": 1.0276, "grad_norm": 1.354953408241272, "learning_rate": 0.0002, "epoch": 5.653275425262396, "step": 7810}, {"loss": 1.0534, "grad_norm": 0.8966761231422424, "learning_rate": 0.0002, "epoch": 5.660513934129569, "step": 7820}, {"loss": 1.0662, "grad_norm": 1.0675480365753174, "learning_rate": 0.0002, "epoch": 5.6677524429967425, "step": 7830}, {"loss": 1.1077, "grad_norm": 1.2104216814041138, "learning_rate": 0.0002, "epoch": 5.674990951863916, "step": 7840}, {"loss": 0.9627, "grad_norm": 1.105790376663208, "learning_rate": 0.0002, "epoch": 5.682229460731089, "step": 7850}, {"loss": 1.0483, "grad_norm": 1.0915391445159912, "learning_rate": 0.0002, "epoch": 5.689467969598263, "step": 7860}, {"loss": 1.0291, "grad_norm": 0.8957812786102295, "learning_rate": 0.0002, "epoch": 5.696706478465436, "step": 7870}, {"loss": 0.9785, "grad_norm": 1.9189311265945435, "learning_rate": 0.0002, "epoch": 5.703944987332609, "step": 7880}, {"loss": 1.0076, "grad_norm": 1.0867321491241455, "learning_rate": 0.0002, "epoch": 5.711183496199783, "step": 7890}, {"loss": 1.0236, "grad_norm": 1.0233147144317627, "learning_rate": 0.0002, "epoch": 5.718422005066956, "step": 7900}, {"loss": 0.9872, "grad_norm": 1.16460382938385, "learning_rate": 0.0002, "epoch": 5.7256605139341294, "step": 7910}, {"loss": 1.0762, "grad_norm": 1.1098358631134033, "learning_rate": 0.0002, "epoch": 5.732899022801303, "step": 7920}, {"loss": 0.9937, "grad_norm": 0.8555701375007629, "learning_rate": 0.0002, "epoch": 5.740137531668476, "step": 7930}, {"loss": 1.0081, "grad_norm": 0.9885705709457397, "learning_rate": 0.0002, "epoch": 5.7473760405356495, "step": 7940}, {"loss": 0.9909, "grad_norm": 0.9184203147888184, "learning_rate": 0.0002, "epoch": 5.754614549402823, "step": 7950}, {"loss": 1.0767, "grad_norm": 0.9653698205947876, "learning_rate": 0.0002, "epoch": 5.761853058269996, "step": 7960}, {"loss": 0.9317, "grad_norm": 1.0014251470565796, "learning_rate": 0.0002, "epoch": 5.76909156713717, "step": 7970}, {"loss": 1.0271, "grad_norm": 1.004701018333435, "learning_rate": 0.0002, "epoch": 5.776330076004343, "step": 7980}, {"loss": 1.0397, "grad_norm": 0.950577974319458, "learning_rate": 0.0002, "epoch": 5.783568584871516, "step": 7990}, {"loss": 0.9725, "grad_norm": 1.2986834049224854, "learning_rate": 0.0002, "epoch": 5.79080709373869, "step": 8000}, {"loss": 1.039, "grad_norm": 1.3353424072265625, "learning_rate": 0.0002, "epoch": 5.798045602605863, "step": 8010}, {"loss": 1.0626, "grad_norm": 0.7650562524795532, "learning_rate": 0.0002, "epoch": 5.8052841114730365, "step": 8020}, {"loss": 1.0802, "grad_norm": 1.0156235694885254, "learning_rate": 0.0002, "epoch": 5.81252262034021, "step": 8030}, {"loss": 1.0185, "grad_norm": 1.3092900514602661, "learning_rate": 0.0002, "epoch": 5.819761129207383, "step": 8040}, {"loss": 0.9905, "grad_norm": 1.184428095817566, "learning_rate": 0.0002, "epoch": 5.826999638074557, "step": 8050}, {"loss": 1.0548, "grad_norm": 0.979401707649231, "learning_rate": 0.0002, "epoch": 5.83423814694173, "step": 8060}, {"loss": 0.9721, "grad_norm": 1.3557400703430176, "learning_rate": 0.0002, "epoch": 5.841476655808903, "step": 8070}, {"loss": 1.0235, "grad_norm": 0.8429333567619324, "learning_rate": 0.0002, "epoch": 5.848715164676077, "step": 8080}, {"loss": 0.952, "grad_norm": 1.3167692422866821, "learning_rate": 0.0002, "epoch": 5.85595367354325, "step": 8090}, {"loss": 0.9609, "grad_norm": 0.9750998020172119, "learning_rate": 0.0002, "epoch": 5.863192182410423, "step": 8100}, {"loss": 1.0789, "grad_norm": 1.1869813203811646, "learning_rate": 0.0002, "epoch": 5.870430691277597, "step": 8110}, {"loss": 1.0331, "grad_norm": 1.508615255355835, "learning_rate": 0.0002, "epoch": 5.87766920014477, "step": 8120}, {"loss": 1.0171, "grad_norm": 0.9439908266067505, "learning_rate": 0.0002, "epoch": 5.8849077090119435, "step": 8130}, {"loss": 0.9682, "grad_norm": 0.910508930683136, "learning_rate": 0.0002, "epoch": 5.892146217879117, "step": 8140}, {"loss": 1.0032, "grad_norm": 1.111501932144165, "learning_rate": 0.0002, "epoch": 5.89938472674629, "step": 8150}, {"loss": 1.0266, "grad_norm": 0.726554274559021, "learning_rate": 0.0002, "epoch": 5.906623235613464, "step": 8160}, {"loss": 1.0681, "grad_norm": 1.1084556579589844, "learning_rate": 0.0002, "epoch": 5.913861744480637, "step": 8170}, {"loss": 0.969, "grad_norm": 0.9695167541503906, "learning_rate": 0.0002, "epoch": 5.92110025334781, "step": 8180}, {"loss": 0.9858, "grad_norm": 1.1169592142105103, "learning_rate": 0.0002, "epoch": 5.928338762214984, "step": 8190}, {"loss": 1.0924, "grad_norm": 1.5116780996322632, "learning_rate": 0.0002, "epoch": 5.935577271082157, "step": 8200}, {"loss": 0.878, "grad_norm": 1.0073388814926147, "learning_rate": 0.0002, "epoch": 5.94281577994933, "step": 8210}, {"loss": 1.0462, "grad_norm": 0.9323263168334961, "learning_rate": 0.0002, "epoch": 5.950054288816504, "step": 8220}, {"loss": 1.0291, "grad_norm": 0.9422887563705444, "learning_rate": 0.0002, "epoch": 5.957292797683677, "step": 8230}, {"loss": 0.953, "grad_norm": 0.9691047668457031, "learning_rate": 0.0002, "epoch": 5.9645313065508505, "step": 8240}, {"loss": 0.9842, "grad_norm": 0.9650622606277466, "learning_rate": 0.0002, "epoch": 5.971769815418024, "step": 8250}, {"loss": 0.907, "grad_norm": 1.077958345413208, "learning_rate": 0.0002, "epoch": 5.979008324285197, "step": 8260}, {"loss": 0.9162, "grad_norm": 0.8946306109428406, "learning_rate": 0.0002, "epoch": 5.986246833152371, "step": 8270}, {"loss": 1.0439, "grad_norm": 1.34098219871521, "learning_rate": 0.0002, "epoch": 5.993485342019544, "step": 8280}]}
+{"epoch": 6.999638074556641, "step": 9670, "epoch_duration": 1214.7583508491516, "total_accumulated_duration": 9110.399286270142, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 3048.73388671875}, "peak_memory_usage": {"GPU_0": 15079.2998046875}, "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-0.9-num-8828-sd-1/checkpoint-4144", "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": 4.7061, "grad_norm": 1.2523442506790161, "learning_rate": 0.0002, "epoch": 0.007238508867173362, "step": 10}, {"loss": 3.3493, "grad_norm": 1.8887330293655396, "learning_rate": 0.0002, "epoch": 0.014477017734346724, "step": 20}, {"loss": 2.7585, "grad_norm": 0.9668035507202148, "learning_rate": 0.0002, "epoch": 0.021715526601520086, "step": 30}, {"loss": 2.3699, "grad_norm": 2.9167306423187256, "learning_rate": 0.0002, "epoch": 0.028954035468693448, "step": 40}, {"loss": 2.2679, "grad_norm": 2.649867296218872, "learning_rate": 0.0002, "epoch": 0.036192544335866814, "step": 50}, {"loss": 2.2202, "grad_norm": 1.5120655298233032, "learning_rate": 0.0002, "epoch": 0.04343105320304017, "step": 60}, {"loss": 2.2026, "grad_norm": 0.7879868149757385, "learning_rate": 0.0002, "epoch": 0.05066956207021354, "step": 70}, {"loss": 1.9447, "grad_norm": 0.7616953253746033, "learning_rate": 0.0002, "epoch": 0.057908070937386896, "step": 80}, {"loss": 2.0112, "grad_norm": 1.8809149265289307, "learning_rate": 0.0002, "epoch": 0.06514657980456026, "step": 90}, {"loss": 1.8337, "grad_norm": 0.9294016361236572, "learning_rate": 0.0002, "epoch": 0.07238508867173363, "step": 100}, {"loss": 1.8419, "grad_norm": 0.7145281434059143, "learning_rate": 0.0002, "epoch": 0.07962359753890698, "step": 110}, {"loss": 2.0036, "grad_norm": 0.7564446330070496, "learning_rate": 0.0002, "epoch": 0.08686210640608034, "step": 120}, {"loss": 1.9306, "grad_norm": 1.1681925058364868, "learning_rate": 0.0002, "epoch": 0.09410061527325371, "step": 130}, {"loss": 1.7875, "grad_norm": 0.6708641648292542, "learning_rate": 0.0002, "epoch": 0.10133912414042708, "step": 140}, {"loss": 1.786, "grad_norm": 0.7625647783279419, "learning_rate": 0.0002, "epoch": 0.10857763300760044, "step": 150}, {"loss": 1.6687, "grad_norm": 0.8463464975357056, "learning_rate": 0.0002, "epoch": 0.11581614187477379, "step": 160}, {"loss": 1.6214, "grad_norm": 0.7502335906028748, "learning_rate": 0.0002, "epoch": 0.12305465074194716, "step": 170}, {"loss": 1.7433, "grad_norm": 0.6929958462715149, "learning_rate": 0.0002, "epoch": 0.13029315960912052, "step": 180}, {"loss": 1.6009, "grad_norm": 0.6798707842826843, "learning_rate": 0.0002, "epoch": 0.1375316684762939, "step": 190}, {"loss": 1.6208, "grad_norm": 0.7566508650779724, "learning_rate": 0.0002, "epoch": 0.14477017734346725, "step": 200}, {"loss": 1.5823, "grad_norm": 0.7196869850158691, "learning_rate": 0.0002, "epoch": 0.15200868621064062, "step": 210}, {"loss": 1.738, "grad_norm": 0.8401045799255371, "learning_rate": 0.0002, "epoch": 0.15924719507781396, "step": 220}, {"loss": 1.7574, "grad_norm": 0.8503773212432861, "learning_rate": 0.0002, "epoch": 0.16648570394498732, "step": 230}, {"loss": 1.7861, "grad_norm": 0.7183733582496643, "learning_rate": 0.0002, "epoch": 0.1737242128121607, "step": 240}, {"loss": 1.6693, "grad_norm": 0.7082605957984924, "learning_rate": 0.0002, "epoch": 0.18096272167933405, "step": 250}, {"loss": 1.619, "grad_norm": 0.9386326670646667, "learning_rate": 0.0002, "epoch": 0.18820123054650742, "step": 260}, {"loss": 1.6511, "grad_norm": 0.7332451939582825, "learning_rate": 0.0002, "epoch": 0.19543973941368079, "step": 270}, {"loss": 1.6353, "grad_norm": 0.7092869877815247, "learning_rate": 0.0002, "epoch": 0.20267824828085415, "step": 280}, {"loss": 1.5996, "grad_norm": 0.7256413698196411, "learning_rate": 0.0002, "epoch": 0.20991675714802752, "step": 290}, {"loss": 1.6754, "grad_norm": 0.6398681402206421, "learning_rate": 0.0002, "epoch": 0.21715526601520088, "step": 300}, {"loss": 1.397, "grad_norm": 0.6273287534713745, "learning_rate": 0.0002, "epoch": 0.22439377488237422, "step": 310}, {"loss": 1.5115, "grad_norm": 0.511648416519165, "learning_rate": 0.0002, "epoch": 0.23163228374954759, "step": 320}, {"loss": 1.5424, "grad_norm": 0.8677352070808411, "learning_rate": 0.0002, "epoch": 0.23887079261672095, "step": 330}, {"loss": 1.6779, "grad_norm": 0.6270743012428284, "learning_rate": 0.0002, "epoch": 0.24610930148389432, "step": 340}, {"loss": 1.626, "grad_norm": 0.7980281114578247, "learning_rate": 0.0002, "epoch": 0.2533478103510677, "step": 350}, {"loss": 1.5238, "grad_norm": 0.632486879825592, "learning_rate": 0.0002, "epoch": 0.26058631921824105, "step": 360}, {"loss": 1.5175, "grad_norm": 0.6527034640312195, "learning_rate": 0.0002, "epoch": 0.2678248280854144, "step": 370}, {"loss": 1.627, "grad_norm": 0.7672118544578552, "learning_rate": 0.0002, "epoch": 0.2750633369525878, "step": 380}, {"loss": 1.5605, "grad_norm": 0.6035117506980896, "learning_rate": 0.0002, "epoch": 0.28230184581976114, "step": 390}, {"loss": 1.4603, "grad_norm": 0.5955103039741516, "learning_rate": 0.0002, "epoch": 0.2895403546869345, "step": 400}, {"loss": 1.558, "grad_norm": 0.6015191674232483, "learning_rate": 0.0002, "epoch": 0.2967788635541079, "step": 410}, {"loss": 1.6091, "grad_norm": 0.6380982398986816, "learning_rate": 0.0002, "epoch": 0.30401737242128124, "step": 420}, {"loss": 1.5292, "grad_norm": 0.6707863211631775, "learning_rate": 0.0002, "epoch": 0.3112558812884546, "step": 430}, {"loss": 1.4426, "grad_norm": 0.7010176777839661, "learning_rate": 0.0002, "epoch": 0.3184943901556279, "step": 440}, {"loss": 1.5572, "grad_norm": 0.8263739943504333, "learning_rate": 0.0002, "epoch": 0.3257328990228013, "step": 450}, {"loss": 1.5188, "grad_norm": 0.7253276109695435, "learning_rate": 0.0002, "epoch": 0.33297140788997465, "step": 460}, {"loss": 1.584, "grad_norm": 0.5238934755325317, "learning_rate": 0.0002, "epoch": 0.340209916757148, "step": 470}, {"loss": 1.7035, "grad_norm": 0.7869495749473572, "learning_rate": 0.0002, "epoch": 0.3474484256243214, "step": 480}, {"loss": 1.5776, "grad_norm": 0.7485215663909912, "learning_rate": 0.0002, "epoch": 0.35468693449149474, "step": 490}, {"loss": 1.6274, "grad_norm": 0.5413193106651306, "learning_rate": 0.0002, "epoch": 0.3619254433586681, "step": 500}, {"loss": 1.7323, "grad_norm": 0.7615048885345459, "learning_rate": 0.0002, "epoch": 0.3691639522258415, "step": 510}, {"loss": 1.532, "grad_norm": 0.7685340046882629, "learning_rate": 0.0002, "epoch": 0.37640246109301484, "step": 520}, {"loss": 1.6312, "grad_norm": 0.6379081010818481, "learning_rate": 0.0002, "epoch": 0.3836409699601882, "step": 530}, {"loss": 1.5645, "grad_norm": 0.7946939468383789, "learning_rate": 0.0002, "epoch": 0.39087947882736157, "step": 540}, {"loss": 1.4001, "grad_norm": 0.6287278532981873, "learning_rate": 0.0002, "epoch": 0.39811798769453494, "step": 550}, {"loss": 1.5982, "grad_norm": 0.6811642646789551, "learning_rate": 0.0002, "epoch": 0.4053564965617083, "step": 560}, {"loss": 1.4953, "grad_norm": 0.671073317527771, "learning_rate": 0.0002, "epoch": 0.41259500542888167, "step": 570}, {"loss": 1.6753, "grad_norm": 0.6313900351524353, "learning_rate": 0.0002, "epoch": 0.41983351429605503, "step": 580}, {"loss": 1.546, "grad_norm": 0.5291772484779358, "learning_rate": 0.0002, "epoch": 0.4270720231632284, "step": 590}, {"loss": 1.5441, "grad_norm": 0.62503582239151, "learning_rate": 0.0002, "epoch": 0.43431053203040176, "step": 600}, {"loss": 1.6276, "grad_norm": 0.5777305364608765, "learning_rate": 0.0002, "epoch": 0.4415490408975751, "step": 610}, {"loss": 1.4758, "grad_norm": 0.7013497352600098, "learning_rate": 0.0002, "epoch": 0.44878754976474844, "step": 620}, {"loss": 1.4029, "grad_norm": 0.8044822216033936, "learning_rate": 0.0002, "epoch": 0.4560260586319218, "step": 630}, {"loss": 1.7195, "grad_norm": 0.672531247138977, "learning_rate": 0.0002, "epoch": 0.46326456749909517, "step": 640}, {"loss": 1.614, "grad_norm": 0.6233910322189331, "learning_rate": 0.0002, "epoch": 0.47050307636626854, "step": 650}, {"loss": 1.6041, "grad_norm": 0.651524543762207, "learning_rate": 0.0002, "epoch": 0.4777415852334419, "step": 660}, {"loss": 1.5842, "grad_norm": 0.7213939428329468, "learning_rate": 0.0002, "epoch": 0.48498009410061527, "step": 670}, {"loss": 1.5453, "grad_norm": 0.6541454792022705, "learning_rate": 0.0002, "epoch": 0.49221860296778863, "step": 680}, {"loss": 1.662, "grad_norm": 0.6568936109542847, "learning_rate": 0.0002, "epoch": 0.499457111834962, "step": 690}, {"loss": 1.624, "grad_norm": 0.7176415324211121, "learning_rate": 0.0002, "epoch": 0.5066956207021354, "step": 700}, {"loss": 1.6099, "grad_norm": 0.6553855538368225, "learning_rate": 0.0002, "epoch": 0.5139341295693087, "step": 710}, {"loss": 1.5508, "grad_norm": 0.5654335618019104, "learning_rate": 0.0002, "epoch": 0.5211726384364821, "step": 720}, {"loss": 1.392, "grad_norm": 0.5671001672744751, "learning_rate": 0.0002, "epoch": 0.5284111473036555, "step": 730}, {"loss": 1.388, "grad_norm": 0.7914412021636963, "learning_rate": 0.0002, "epoch": 0.5356496561708288, "step": 740}, {"loss": 1.5931, "grad_norm": 0.6172138452529907, "learning_rate": 0.0002, "epoch": 0.5428881650380022, "step": 750}, {"loss": 1.4018, "grad_norm": 0.6132623553276062, "learning_rate": 0.0002, "epoch": 0.5501266739051756, "step": 760}, {"loss": 1.513, "grad_norm": 0.654000461101532, "learning_rate": 0.0002, "epoch": 0.5573651827723489, "step": 770}, {"loss": 1.5035, "grad_norm": 0.5691370964050293, "learning_rate": 0.0002, "epoch": 0.5646036916395223, "step": 780}, {"loss": 1.65, "grad_norm": 0.7922580242156982, "learning_rate": 0.0002, "epoch": 0.5718422005066957, "step": 790}, {"loss": 1.4521, "grad_norm": 0.6831880211830139, "learning_rate": 0.0002, "epoch": 0.579080709373869, "step": 800}, {"loss": 1.4734, "grad_norm": 0.6740124821662903, "learning_rate": 0.0002, "epoch": 0.5863192182410424, "step": 810}, {"loss": 1.6498, "grad_norm": 1.380016803741455, "learning_rate": 0.0002, "epoch": 0.5935577271082157, "step": 820}, {"loss": 1.4642, "grad_norm": 0.6552878022193909, "learning_rate": 0.0002, "epoch": 0.6007962359753891, "step": 830}, {"loss": 1.6271, "grad_norm": 0.6649535298347473, "learning_rate": 0.0002, "epoch": 0.6080347448425625, "step": 840}, {"loss": 1.5886, "grad_norm": 0.561738133430481, "learning_rate": 0.0002, "epoch": 0.6152732537097358, "step": 850}, {"loss": 1.5364, "grad_norm": 0.6133047938346863, "learning_rate": 0.0002, "epoch": 0.6225117625769092, "step": 860}, {"loss": 1.3489, "grad_norm": 0.559843122959137, "learning_rate": 0.0002, "epoch": 0.6297502714440825, "step": 870}, {"loss": 1.4878, "grad_norm": 0.6117811799049377, "learning_rate": 0.0002, "epoch": 0.6369887803112558, "step": 880}, {"loss": 1.56, "grad_norm": 0.6209776401519775, "learning_rate": 0.0002, "epoch": 0.6442272891784292, "step": 890}, {"loss": 1.6747, "grad_norm": 0.6234082579612732, "learning_rate": 0.0002, "epoch": 0.6514657980456026, "step": 900}, {"loss": 1.6963, "grad_norm": 0.7623258233070374, "learning_rate": 0.0002, "epoch": 0.6587043069127759, "step": 910}, {"loss": 1.2424, "grad_norm": 0.6148061752319336, "learning_rate": 0.0002, "epoch": 0.6659428157799493, "step": 920}, {"loss": 1.4319, "grad_norm": 0.6682973504066467, "learning_rate": 0.0002, "epoch": 0.6731813246471227, "step": 930}, {"loss": 1.5377, "grad_norm": 0.5513041615486145, "learning_rate": 0.0002, "epoch": 0.680419833514296, "step": 940}, {"loss": 1.3991, "grad_norm": 0.5197525024414062, "learning_rate": 0.0002, "epoch": 0.6876583423814694, "step": 950}, {"loss": 1.4398, "grad_norm": 0.6490758061408997, "learning_rate": 0.0002, "epoch": 0.6948968512486428, "step": 960}, {"loss": 1.5251, "grad_norm": 0.6450682878494263, "learning_rate": 0.0002, "epoch": 0.7021353601158161, "step": 970}, {"loss": 1.5417, "grad_norm": 0.6203766465187073, "learning_rate": 0.0002, "epoch": 0.7093738689829895, "step": 980}, {"loss": 1.4575, "grad_norm": 0.6023609638214111, "learning_rate": 0.0002, "epoch": 0.7166123778501629, "step": 990}, {"loss": 1.4973, "grad_norm": 0.5765255093574524, "learning_rate": 0.0002, "epoch": 0.7238508867173362, "step": 1000}, {"loss": 1.483, "grad_norm": 0.6650075316429138, "learning_rate": 0.0002, "epoch": 0.7310893955845096, "step": 1010}, {"loss": 1.5959, "grad_norm": 0.5610854029655457, "learning_rate": 0.0002, "epoch": 0.738327904451683, "step": 1020}, {"loss": 1.5248, "grad_norm": 0.7072813510894775, "learning_rate": 0.0002, "epoch": 0.7455664133188563, "step": 1030}, {"loss": 1.5776, "grad_norm": 0.6815407872200012, "learning_rate": 0.0002, "epoch": 0.7528049221860297, "step": 1040}, {"loss": 1.4577, "grad_norm": 0.7932390570640564, "learning_rate": 0.0002, "epoch": 0.760043431053203, "step": 1050}, {"loss": 1.4515, "grad_norm": 0.5798183083534241, "learning_rate": 0.0002, "epoch": 0.7672819399203764, "step": 1060}, {"loss": 1.5053, "grad_norm": 0.7898504137992859, "learning_rate": 0.0002, "epoch": 0.7745204487875498, "step": 1070}, {"loss": 1.4776, "grad_norm": 0.4983280301094055, "learning_rate": 0.0002, "epoch": 0.7817589576547231, "step": 1080}, {"loss": 1.5007, "grad_norm": 0.691403329372406, "learning_rate": 0.0002, "epoch": 0.7889974665218965, "step": 1090}, {"loss": 1.5153, "grad_norm": 0.5394481420516968, "learning_rate": 0.0002, "epoch": 0.7962359753890699, "step": 1100}, {"loss": 1.6892, "grad_norm": 0.5136822462081909, "learning_rate": 0.0002, "epoch": 0.8034744842562432, "step": 1110}, {"loss": 1.4902, "grad_norm": 0.6828126907348633, "learning_rate": 0.0002, "epoch": 0.8107129931234166, "step": 1120}, {"loss": 1.4346, "grad_norm": 0.6799656748771667, "learning_rate": 0.0002, "epoch": 0.81795150199059, "step": 1130}, {"loss": 1.2678, "grad_norm": 0.5428406000137329, "learning_rate": 0.0002, "epoch": 0.8251900108577633, "step": 1140}, {"loss": 1.4072, "grad_norm": 0.4811290502548218, "learning_rate": 0.0002, "epoch": 0.8324285197249367, "step": 1150}, {"loss": 1.4512, "grad_norm": 0.5519434809684753, "learning_rate": 0.0002, "epoch": 0.8396670285921101, "step": 1160}, {"loss": 1.4072, "grad_norm": 0.9748060703277588, "learning_rate": 0.0002, "epoch": 0.8469055374592834, "step": 1170}, {"loss": 1.4309, "grad_norm": 0.712609589099884, "learning_rate": 0.0002, "epoch": 0.8541440463264568, "step": 1180}, {"loss": 1.434, "grad_norm": 0.6866157054901123, "learning_rate": 0.0002, "epoch": 0.8613825551936302, "step": 1190}, {"loss": 1.3704, "grad_norm": 0.5068854093551636, "learning_rate": 0.0002, "epoch": 0.8686210640608035, "step": 1200}, {"loss": 1.5601, "grad_norm": 0.6333245038986206, "learning_rate": 0.0002, "epoch": 0.8758595729279768, "step": 1210}, {"loss": 1.4636, "grad_norm": 0.6424421072006226, "learning_rate": 0.0002, "epoch": 0.8830980817951501, "step": 1220}, {"loss": 1.4186, "grad_norm": 0.4771921932697296, "learning_rate": 0.0002, "epoch": 0.8903365906623235, "step": 1230}, {"loss": 1.6323, "grad_norm": 0.5191764235496521, "learning_rate": 0.0002, "epoch": 0.8975750995294969, "step": 1240}, {"loss": 1.6105, "grad_norm": 0.756222128868103, "learning_rate": 0.0002, "epoch": 0.9048136083966702, "step": 1250}, {"loss": 1.4396, "grad_norm": 0.623823881149292, "learning_rate": 0.0002, "epoch": 0.9120521172638436, "step": 1260}, {"loss": 1.3097, "grad_norm": 0.8166571259498596, "learning_rate": 0.0002, "epoch": 0.919290626131017, "step": 1270}, {"loss": 1.4625, "grad_norm": 0.6059346795082092, "learning_rate": 0.0002, "epoch": 0.9265291349981903, "step": 1280}, {"loss": 1.3555, "grad_norm": 0.5842690467834473, "learning_rate": 0.0002, "epoch": 0.9337676438653637, "step": 1290}, {"loss": 1.5859, "grad_norm": 0.7649800777435303, "learning_rate": 0.0002, "epoch": 0.9410061527325371, "step": 1300}, {"loss": 1.5915, "grad_norm": 0.6420919895172119, "learning_rate": 0.0002, "epoch": 0.9482446615997104, "step": 1310}, {"loss": 1.453, "grad_norm": 0.7011452913284302, "learning_rate": 0.0002, "epoch": 0.9554831704668838, "step": 1320}, {"loss": 1.6766, "grad_norm": 0.5783746242523193, "learning_rate": 0.0002, "epoch": 0.9627216793340572, "step": 1330}, {"loss": 1.6308, "grad_norm": 0.5973192453384399, "learning_rate": 0.0002, "epoch": 0.9699601882012305, "step": 1340}, {"loss": 1.5901, "grad_norm": 0.6181833744049072, "learning_rate": 0.0002, "epoch": 0.9771986970684039, "step": 1350}, {"loss": 1.5258, "grad_norm": 0.5563396215438843, "learning_rate": 0.0002, "epoch": 0.9844372059355773, "step": 1360}, {"loss": 1.4508, "grad_norm": 0.45723360776901245, "learning_rate": 0.0002, "epoch": 0.9916757148027506, "step": 1370}, {"loss": 1.3291, "grad_norm": 0.5947498679161072, "learning_rate": 0.0002, "epoch": 0.998914223669924, "step": 1380}, {"eval_loss": 1.480796456336975, "eval_runtime": 27.3103, "eval_samples_per_second": 15.965, "eval_steps_per_second": 2.014, "epoch": 0.9996380745566413, "step": 1381}, {"loss": 1.3057, "grad_norm": 0.5599952936172485, "learning_rate": 0.0002, "epoch": 1.0061527325370974, "step": 1390}, {"loss": 1.4991, "grad_norm": 0.5932008028030396, "learning_rate": 0.0002, "epoch": 1.0133912414042707, "step": 1400}, {"loss": 1.4506, "grad_norm": 0.6194121837615967, "learning_rate": 0.0002, "epoch": 1.020629750271444, "step": 1410}, {"loss": 1.5966, "grad_norm": 0.6995621919631958, "learning_rate": 0.0002, "epoch": 1.0278682591386175, "step": 1420}, {"loss": 1.4153, "grad_norm": 0.7905810475349426, "learning_rate": 0.0002, "epoch": 1.0351067680057908, "step": 1430}, {"loss": 1.4414, "grad_norm": 0.7221615314483643, "learning_rate": 0.0002, "epoch": 1.0423452768729642, "step": 1440}, {"loss": 1.3859, "grad_norm": 0.6170642375946045, "learning_rate": 0.0002, "epoch": 1.0495837857401376, "step": 1450}, {"loss": 1.3806, "grad_norm": 0.5844094753265381, "learning_rate": 0.0002, "epoch": 1.056822294607311, "step": 1460}, {"loss": 1.4871, "grad_norm": 0.7731822729110718, "learning_rate": 0.0002, "epoch": 1.0640608034744843, "step": 1470}, {"loss": 1.4286, "grad_norm": 0.4554748237133026, "learning_rate": 0.0002, "epoch": 1.0712993123416577, "step": 1480}, {"loss": 1.3977, "grad_norm": 0.6923259496688843, "learning_rate": 0.0002, "epoch": 1.078537821208831, "step": 1490}, {"loss": 1.3936, "grad_norm": 0.6008219122886658, "learning_rate": 0.0002, "epoch": 1.0857763300760044, "step": 1500}, {"loss": 1.4821, "grad_norm": 0.6450045704841614, "learning_rate": 0.0002, "epoch": 1.0930148389431777, "step": 1510}, {"loss": 1.3295, "grad_norm": 0.7833753824234009, "learning_rate": 0.0002, "epoch": 1.1002533478103511, "step": 1520}, {"loss": 1.3424, "grad_norm": 0.5076758861541748, "learning_rate": 0.0002, "epoch": 1.1074918566775245, "step": 1530}, {"loss": 1.4043, "grad_norm": 0.5661332011222839, "learning_rate": 0.0002, "epoch": 1.1147303655446978, "step": 1540}, {"loss": 1.4963, "grad_norm": 0.6526919603347778, "learning_rate": 0.0002, "epoch": 1.1219688744118712, "step": 1550}, {"loss": 1.3671, "grad_norm": 0.5613082647323608, "learning_rate": 0.0002, "epoch": 1.1292073832790446, "step": 1560}, {"loss": 1.4458, "grad_norm": 0.6113885641098022, "learning_rate": 0.0002, "epoch": 1.136445892146218, "step": 1570}, {"loss": 1.3552, "grad_norm": 0.6732510328292847, "learning_rate": 0.0002, "epoch": 1.1436844010133913, "step": 1580}, {"loss": 1.3114, "grad_norm": 0.6146392226219177, "learning_rate": 0.0002, "epoch": 1.1509229098805647, "step": 1590}, {"loss": 1.411, "grad_norm": 0.6766974329948425, "learning_rate": 0.0002, "epoch": 1.158161418747738, "step": 1600}, {"loss": 1.2401, "grad_norm": 0.7621957659721375, "learning_rate": 0.0002, "epoch": 1.1653999276149114, "step": 1610}, {"loss": 1.3758, "grad_norm": 0.6959581971168518, "learning_rate": 0.0002, "epoch": 1.1726384364820848, "step": 1620}, {"loss": 1.382, "grad_norm": 0.6691278219223022, "learning_rate": 0.0002, "epoch": 1.1798769453492581, "step": 1630}, {"loss": 1.4147, "grad_norm": 0.4927774965763092, "learning_rate": 0.0002, "epoch": 1.1871154542164315, "step": 1640}, {"loss": 1.449, "grad_norm": 0.7724234461784363, "learning_rate": 0.0002, "epoch": 1.1943539630836049, "step": 1650}, {"loss": 1.4778, "grad_norm": 0.6817787885665894, "learning_rate": 0.0002, "epoch": 1.2015924719507782, "step": 1660}, {"loss": 1.3776, "grad_norm": 0.6500699520111084, "learning_rate": 0.0002, "epoch": 1.2088309808179516, "step": 1670}, {"loss": 1.3875, "grad_norm": 0.5703568458557129, "learning_rate": 0.0002, "epoch": 1.216069489685125, "step": 1680}, {"loss": 1.4735, "grad_norm": 0.6261579990386963, "learning_rate": 0.0002, "epoch": 1.2233079985522983, "step": 1690}, {"loss": 1.3898, "grad_norm": 0.651713490486145, "learning_rate": 0.0002, "epoch": 1.2305465074194717, "step": 1700}, {"loss": 1.4002, "grad_norm": 0.684399425983429, "learning_rate": 0.0002, "epoch": 1.237785016286645, "step": 1710}, {"loss": 1.5027, "grad_norm": 0.6996857523918152, "learning_rate": 0.0002, "epoch": 1.2450235251538184, "step": 1720}, {"loss": 1.3326, "grad_norm": 0.7102537751197815, "learning_rate": 0.0002, "epoch": 1.2522620340209918, "step": 1730}, {"loss": 1.3675, "grad_norm": 0.45809897780418396, "learning_rate": 0.0002, "epoch": 1.2595005428881652, "step": 1740}, {"loss": 1.4175, "grad_norm": 0.6377046704292297, "learning_rate": 0.0002, "epoch": 1.2667390517553385, "step": 1750}, {"loss": 1.3479, "grad_norm": 0.6965704560279846, "learning_rate": 0.0002, "epoch": 1.2739775606225119, "step": 1760}, {"loss": 1.5647, "grad_norm": 0.5688214302062988, "learning_rate": 0.0002, "epoch": 1.2812160694896852, "step": 1770}, {"loss": 1.3967, "grad_norm": 0.6384190320968628, "learning_rate": 0.0002, "epoch": 1.2884545783568586, "step": 1780}, {"loss": 1.3671, "grad_norm": 0.5629363656044006, "learning_rate": 0.0002, "epoch": 1.295693087224032, "step": 1790}, {"loss": 1.2292, "grad_norm": 0.6148255467414856, "learning_rate": 0.0002, "epoch": 1.3029315960912053, "step": 1800}, {"loss": 1.5806, "grad_norm": 0.655580997467041, "learning_rate": 0.0002, "epoch": 1.3101701049583787, "step": 1810}, {"loss": 1.2398, "grad_norm": 0.5642657279968262, "learning_rate": 0.0002, "epoch": 1.3174086138255519, "step": 1820}, {"loss": 1.3246, "grad_norm": 0.59607994556427, "learning_rate": 0.0002, "epoch": 1.3246471226927252, "step": 1830}, {"loss": 1.3274, "grad_norm": 0.5564199090003967, "learning_rate": 0.0002, "epoch": 1.3318856315598986, "step": 1840}, {"loss": 1.5834, "grad_norm": 0.6949955821037292, "learning_rate": 0.0002, "epoch": 1.339124140427072, "step": 1850}, {"loss": 1.4722, "grad_norm": 0.7036856412887573, "learning_rate": 0.0002, "epoch": 1.3463626492942453, "step": 1860}, {"loss": 1.333, "grad_norm": 0.722062885761261, "learning_rate": 0.0002, "epoch": 1.3536011581614187, "step": 1870}, {"loss": 1.4044, "grad_norm": 0.6098677515983582, "learning_rate": 0.0002, "epoch": 1.360839667028592, "step": 1880}, {"loss": 1.6217, "grad_norm": 0.5376402735710144, "learning_rate": 0.0002, "epoch": 1.3680781758957654, "step": 1890}, {"loss": 1.5071, "grad_norm": 0.6974610090255737, "learning_rate": 0.0002, "epoch": 1.3753166847629388, "step": 1900}, {"loss": 1.5854, "grad_norm": 0.6520763635635376, "learning_rate": 0.0002, "epoch": 1.3825551936301121, "step": 1910}, {"loss": 1.4271, "grad_norm": 0.6604374647140503, "learning_rate": 0.0002, "epoch": 1.3897937024972855, "step": 1920}, {"loss": 1.419, "grad_norm": 0.7364398241043091, "learning_rate": 0.0002, "epoch": 1.3970322113644589, "step": 1930}, {"loss": 1.4585, "grad_norm": 0.6849475502967834, "learning_rate": 0.0002, "epoch": 1.4042707202316322, "step": 1940}, {"loss": 1.5577, "grad_norm": 0.6562670469284058, "learning_rate": 0.0002, "epoch": 1.4115092290988056, "step": 1950}, {"loss": 1.4725, "grad_norm": 0.5695616006851196, "learning_rate": 0.0002, "epoch": 1.418747737965979, "step": 1960}, {"loss": 1.3088, "grad_norm": 0.5244464874267578, "learning_rate": 0.0002, "epoch": 1.4259862468331523, "step": 1970}, {"loss": 1.5069, "grad_norm": 0.6347293257713318, "learning_rate": 0.0002, "epoch": 1.4332247557003257, "step": 1980}, {"loss": 1.3502, "grad_norm": 0.5528361201286316, "learning_rate": 0.0002, "epoch": 1.440463264567499, "step": 1990}, {"loss": 1.3978, "grad_norm": 0.6987585425376892, "learning_rate": 0.0002, "epoch": 1.4477017734346724, "step": 2000}, {"loss": 1.4262, "grad_norm": 0.6568987369537354, "learning_rate": 0.0002, "epoch": 1.4549402823018458, "step": 2010}, {"loss": 1.4175, "grad_norm": 0.7665994763374329, "learning_rate": 0.0002, "epoch": 1.4621787911690192, "step": 2020}, {"loss": 1.244, "grad_norm": 0.5127707123756409, "learning_rate": 0.0002, "epoch": 1.4694173000361925, "step": 2030}, {"loss": 1.3699, "grad_norm": 0.5406824946403503, "learning_rate": 0.0002, "epoch": 1.476655808903366, "step": 2040}, {"loss": 1.3353, "grad_norm": 0.5990166664123535, "learning_rate": 0.0002, "epoch": 1.4838943177705393, "step": 2050}, {"loss": 1.2454, "grad_norm": 0.6186193823814392, "learning_rate": 0.0002, "epoch": 1.4911328266377126, "step": 2060}, {"loss": 1.428, "grad_norm": 0.6154307126998901, "learning_rate": 0.0002, "epoch": 1.498371335504886, "step": 2070}, {"loss": 1.4528, "grad_norm": 0.5606056451797485, "learning_rate": 0.0002, "epoch": 1.5056098443720594, "step": 2080}, {"loss": 1.2405, "grad_norm": 0.5006417036056519, "learning_rate": 0.0002, "epoch": 1.5128483532392327, "step": 2090}, {"loss": 1.4258, "grad_norm": 0.5968486070632935, "learning_rate": 0.0002, "epoch": 1.520086862106406, "step": 2100}, {"loss": 1.2752, "grad_norm": 0.5835496187210083, "learning_rate": 0.0002, "epoch": 1.5273253709735795, "step": 2110}, {"loss": 1.5443, "grad_norm": 0.6753535270690918, "learning_rate": 0.0002, "epoch": 1.5345638798407528, "step": 2120}, {"loss": 1.2139, "grad_norm": 0.7299720644950867, "learning_rate": 0.0002, "epoch": 1.5418023887079262, "step": 2130}, {"loss": 1.2364, "grad_norm": 0.5105988383293152, "learning_rate": 0.0002, "epoch": 1.5490408975750996, "step": 2140}, {"loss": 1.4528, "grad_norm": 0.5675431489944458, "learning_rate": 0.0002, "epoch": 1.556279406442273, "step": 2150}, {"loss": 1.4563, "grad_norm": 0.6246723532676697, "learning_rate": 0.0002, "epoch": 1.5635179153094463, "step": 2160}, {"loss": 1.5255, "grad_norm": 0.7291720509529114, "learning_rate": 0.0002, "epoch": 1.5707564241766196, "step": 2170}, {"loss": 1.5432, "grad_norm": 0.678114116191864, "learning_rate": 0.0002, "epoch": 1.577994933043793, "step": 2180}, {"loss": 1.5212, "grad_norm": 0.5136260986328125, "learning_rate": 0.0002, "epoch": 1.5852334419109664, "step": 2190}, {"loss": 1.3271, "grad_norm": 0.6359935998916626, "learning_rate": 0.0002, "epoch": 1.5924719507781397, "step": 2200}, {"loss": 1.4038, "grad_norm": 0.7650278806686401, "learning_rate": 0.0002, "epoch": 1.599710459645313, "step": 2210}, {"loss": 1.5478, "grad_norm": 0.7256110906600952, "learning_rate": 0.0002, "epoch": 1.6069489685124865, "step": 2220}, {"loss": 1.4387, "grad_norm": 0.688689649105072, "learning_rate": 0.0002, "epoch": 1.6141874773796598, "step": 2230}, {"loss": 1.4096, "grad_norm": 0.6045311093330383, "learning_rate": 0.0002, "epoch": 1.6214259862468332, "step": 2240}, {"loss": 1.4097, "grad_norm": 0.7064604163169861, "learning_rate": 0.0002, "epoch": 1.6286644951140063, "step": 2250}, {"loss": 1.3477, "grad_norm": 0.5309562087059021, "learning_rate": 0.0002, "epoch": 1.6359030039811797, "step": 2260}, {"loss": 1.4022, "grad_norm": 0.5687053203582764, "learning_rate": 0.0002, "epoch": 1.643141512848353, "step": 2270}, {"loss": 1.2977, "grad_norm": 0.535872757434845, "learning_rate": 0.0002, "epoch": 1.6503800217155264, "step": 2280}, {"loss": 1.3844, "grad_norm": 0.5502381920814514, "learning_rate": 0.0002, "epoch": 1.6576185305826998, "step": 2290}, {"loss": 1.3764, "grad_norm": 0.6158602237701416, "learning_rate": 0.0002, "epoch": 1.6648570394498732, "step": 2300}, {"loss": 1.3515, "grad_norm": 0.5804675817489624, "learning_rate": 0.0002, "epoch": 1.6720955483170465, "step": 2310}, {"loss": 1.2532, "grad_norm": 0.600742757320404, "learning_rate": 0.0002, "epoch": 1.67933405718422, "step": 2320}, {"loss": 1.477, "grad_norm": 0.7101941108703613, "learning_rate": 0.0002, "epoch": 1.6865725660513933, "step": 2330}, {"loss": 1.4849, "grad_norm": 0.7507809996604919, "learning_rate": 0.0002, "epoch": 1.6938110749185666, "step": 2340}, {"loss": 1.2703, "grad_norm": 0.768502414226532, "learning_rate": 0.0002, "epoch": 1.70104958378574, "step": 2350}, {"loss": 1.3332, "grad_norm": 0.4801851212978363, "learning_rate": 0.0002, "epoch": 1.7082880926529134, "step": 2360}, {"loss": 1.4158, "grad_norm": 0.5322122573852539, "learning_rate": 0.0002, "epoch": 1.7155266015200867, "step": 2370}, {"loss": 1.4136, "grad_norm": 0.587661862373352, "learning_rate": 0.0002, "epoch": 1.72276511038726, "step": 2380}, {"loss": 1.3771, "grad_norm": 0.6073525547981262, "learning_rate": 0.0002, "epoch": 1.7300036192544335, "step": 2390}, {"loss": 1.2754, "grad_norm": 0.6950460076332092, "learning_rate": 0.0002, "epoch": 1.7372421281216068, "step": 2400}, {"loss": 1.3858, "grad_norm": 0.5981102585792542, "learning_rate": 0.0002, "epoch": 1.7444806369887802, "step": 2410}, {"loss": 1.4075, "grad_norm": 0.544570803642273, "learning_rate": 0.0002, "epoch": 1.7517191458559536, "step": 2420}, {"loss": 1.3861, "grad_norm": 0.5304399728775024, "learning_rate": 0.0002, "epoch": 1.758957654723127, "step": 2430}, {"loss": 1.4244, "grad_norm": 0.7921594977378845, "learning_rate": 0.0002, "epoch": 1.7661961635903003, "step": 2440}, {"loss": 1.3053, "grad_norm": 0.6084808707237244, "learning_rate": 0.0002, "epoch": 1.7734346724574737, "step": 2450}, {"loss": 1.3781, "grad_norm": 0.8844701051712036, "learning_rate": 0.0002, "epoch": 1.780673181324647, "step": 2460}, {"loss": 1.3227, "grad_norm": 0.5729258060455322, "learning_rate": 0.0002, "epoch": 1.7879116901918204, "step": 2470}, {"loss": 1.3422, "grad_norm": 0.6303611993789673, "learning_rate": 0.0002, "epoch": 1.7951501990589938, "step": 2480}, {"loss": 1.3926, "grad_norm": 0.5627942085266113, "learning_rate": 0.0002, "epoch": 1.8023887079261671, "step": 2490}, {"loss": 1.3816, "grad_norm": 0.6724274158477783, "learning_rate": 0.0002, "epoch": 1.8096272167933405, "step": 2500}, {"loss": 1.2951, "grad_norm": 0.5030826330184937, "learning_rate": 0.0002, "epoch": 1.8168657256605139, "step": 2510}, {"loss": 1.2839, "grad_norm": 0.5504099130630493, "learning_rate": 0.0002, "epoch": 1.8241042345276872, "step": 2520}, {"loss": 1.4264, "grad_norm": 0.6338945627212524, "learning_rate": 0.0002, "epoch": 1.8313427433948606, "step": 2530}, {"loss": 1.563, "grad_norm": 0.5902037620544434, "learning_rate": 0.0002, "epoch": 1.838581252262034, "step": 2540}, {"loss": 1.2961, "grad_norm": 0.48814457654953003, "learning_rate": 0.0002, "epoch": 1.8458197611292073, "step": 2550}, {"loss": 1.466, "grad_norm": 0.6216312646865845, "learning_rate": 0.0002, "epoch": 1.8530582699963807, "step": 2560}, {"loss": 1.5123, "grad_norm": 0.635603666305542, "learning_rate": 0.0002, "epoch": 1.860296778863554, "step": 2570}, {"loss": 1.372, "grad_norm": 0.6938216090202332, "learning_rate": 0.0002, "epoch": 1.8675352877307274, "step": 2580}, {"loss": 1.5011, "grad_norm": 0.599557638168335, "learning_rate": 0.0002, "epoch": 1.8747737965979008, "step": 2590}, {"loss": 1.2714, "grad_norm": 0.564424455165863, "learning_rate": 0.0002, "epoch": 1.8820123054650741, "step": 2600}, {"loss": 1.3403, "grad_norm": 0.5430700182914734, "learning_rate": 0.0002, "epoch": 1.8892508143322475, "step": 2610}, {"loss": 1.4347, "grad_norm": 0.6150169372558594, "learning_rate": 0.0002, "epoch": 1.8964893231994209, "step": 2620}, {"loss": 1.2474, "grad_norm": 0.48159119486808777, "learning_rate": 0.0002, "epoch": 1.9037278320665942, "step": 2630}, {"loss": 1.3716, "grad_norm": 0.5608997941017151, "learning_rate": 0.0002, "epoch": 1.9109663409337676, "step": 2640}, {"loss": 1.5787, "grad_norm": 0.6454501748085022, "learning_rate": 0.0002, "epoch": 1.918204849800941, "step": 2650}, {"loss": 1.3238, "grad_norm": 0.5458073616027832, "learning_rate": 0.0002, "epoch": 1.9254433586681143, "step": 2660}, {"loss": 1.3208, "grad_norm": 0.5328490734100342, "learning_rate": 0.0002, "epoch": 1.9326818675352877, "step": 2670}, {"loss": 1.4971, "grad_norm": 0.6444696187973022, "learning_rate": 0.0002, "epoch": 1.939920376402461, "step": 2680}, {"loss": 1.5387, "grad_norm": 0.7126023769378662, "learning_rate": 0.0002, "epoch": 1.9471588852696344, "step": 2690}, {"loss": 1.3637, "grad_norm": 0.5164045095443726, "learning_rate": 0.0002, "epoch": 1.9543973941368078, "step": 2700}, {"loss": 1.5303, "grad_norm": 0.5347061157226562, "learning_rate": 0.0002, "epoch": 1.9616359030039812, "step": 2710}, {"loss": 1.2815, "grad_norm": 0.5297950506210327, "learning_rate": 0.0002, "epoch": 1.9688744118711545, "step": 2720}, {"loss": 1.3566, "grad_norm": 0.6537790298461914, "learning_rate": 0.0002, "epoch": 1.976112920738328, "step": 2730}, {"loss": 1.332, "grad_norm": 0.5536222457885742, "learning_rate": 0.0002, "epoch": 1.9833514296055013, "step": 2740}, {"loss": 1.3333, "grad_norm": 0.4856105446815491, "learning_rate": 0.0002, "epoch": 1.9905899384726746, "step": 2750}, {"loss": 1.3521, "grad_norm": 0.6642730832099915, "learning_rate": 0.0002, "epoch": 1.997828447339848, "step": 2760}, {"eval_loss": 1.4366681575775146, "eval_runtime": 27.3729, "eval_samples_per_second": 15.928, "eval_steps_per_second": 2.009, "epoch": 2.0, "step": 2763}, {"loss": 1.4322, "grad_norm": 0.740253210067749, "learning_rate": 0.0002, "epoch": 2.0050669562070214, "step": 2770}, {"loss": 1.277, "grad_norm": 0.5826276540756226, "learning_rate": 0.0002, "epoch": 2.0123054650741947, "step": 2780}, {"loss": 1.2424, "grad_norm": 0.607356071472168, "learning_rate": 0.0002, "epoch": 2.019543973941368, "step": 2790}, {"loss": 1.2601, "grad_norm": 0.5918063521385193, "learning_rate": 0.0002, "epoch": 2.0267824828085415, "step": 2800}, {"loss": 1.3715, "grad_norm": 0.5610089898109436, "learning_rate": 0.0002, "epoch": 2.034020991675715, "step": 2810}, {"loss": 1.2092, "grad_norm": 0.5869926810264587, "learning_rate": 0.0002, "epoch": 2.041259500542888, "step": 2820}, {"loss": 1.1929, "grad_norm": 0.5753467679023743, "learning_rate": 0.0002, "epoch": 2.0484980094100615, "step": 2830}, {"loss": 1.333, "grad_norm": 0.7096508145332336, "learning_rate": 0.0002, "epoch": 2.055736518277235, "step": 2840}, {"loss": 1.1766, "grad_norm": 0.7653635144233704, "learning_rate": 0.0002, "epoch": 2.0629750271444083, "step": 2850}, {"loss": 1.2331, "grad_norm": 0.6202841997146606, "learning_rate": 0.0002, "epoch": 2.0702135360115816, "step": 2860}, {"loss": 1.3298, "grad_norm": 0.6810227632522583, "learning_rate": 0.0002, "epoch": 2.077452044878755, "step": 2870}, {"loss": 1.2505, "grad_norm": 0.7481493353843689, "learning_rate": 0.0002, "epoch": 2.0846905537459284, "step": 2880}, {"loss": 1.2484, "grad_norm": 0.7089637517929077, "learning_rate": 0.0002, "epoch": 2.0919290626131017, "step": 2890}, {"loss": 1.3095, "grad_norm": 0.7472923398017883, "learning_rate": 0.0002, "epoch": 2.099167571480275, "step": 2900}, {"loss": 1.304, "grad_norm": 0.8135465979576111, "learning_rate": 0.0002, "epoch": 2.1064060803474485, "step": 2910}, {"loss": 1.273, "grad_norm": 0.6097133159637451, "learning_rate": 0.0002, "epoch": 2.113644589214622, "step": 2920}, {"loss": 1.3384, "grad_norm": 0.5970117449760437, "learning_rate": 0.0002, "epoch": 2.120883098081795, "step": 2930}, {"loss": 1.3233, "grad_norm": 0.6169309616088867, "learning_rate": 0.0002, "epoch": 2.1281216069489686, "step": 2940}, {"loss": 1.4246, "grad_norm": 0.9428738355636597, "learning_rate": 0.0002, "epoch": 2.135360115816142, "step": 2950}, {"loss": 1.3527, "grad_norm": 0.5671679973602295, "learning_rate": 0.0002, "epoch": 2.1425986246833153, "step": 2960}, {"loss": 1.1375, "grad_norm": 0.7007262110710144, "learning_rate": 0.0002, "epoch": 2.1498371335504887, "step": 2970}, {"loss": 1.2015, "grad_norm": 0.6294044256210327, "learning_rate": 0.0002, "epoch": 2.157075642417662, "step": 2980}, {"loss": 1.2167, "grad_norm": 0.6105241775512695, "learning_rate": 0.0002, "epoch": 2.1643141512848354, "step": 2990}, {"loss": 1.2065, "grad_norm": 0.557124137878418, "learning_rate": 0.0002, "epoch": 2.1715526601520088, "step": 3000}, {"loss": 1.2515, "grad_norm": 0.6250392198562622, "learning_rate": 0.0002, "epoch": 2.178791169019182, "step": 3010}, {"loss": 1.385, "grad_norm": 0.645218551158905, "learning_rate": 0.0002, "epoch": 2.1860296778863555, "step": 3020}, {"loss": 1.3928, "grad_norm": 0.9033605456352234, "learning_rate": 0.0002, "epoch": 2.193268186753529, "step": 3030}, {"loss": 1.2458, "grad_norm": 0.5325747132301331, "learning_rate": 0.0002, "epoch": 2.2005066956207022, "step": 3040}, {"loss": 1.261, "grad_norm": 0.6334700584411621, "learning_rate": 0.0002, "epoch": 2.2077452044878756, "step": 3050}, {"loss": 1.2385, "grad_norm": 0.5206325054168701, "learning_rate": 0.0002, "epoch": 2.214983713355049, "step": 3060}, {"loss": 1.3103, "grad_norm": 0.5987200140953064, "learning_rate": 0.0002, "epoch": 2.2222222222222223, "step": 3070}, {"loss": 1.1756, "grad_norm": 0.5893264412879944, "learning_rate": 0.0002, "epoch": 2.2294607310893957, "step": 3080}, {"loss": 1.235, "grad_norm": 0.6869237422943115, "learning_rate": 0.0002, "epoch": 2.236699239956569, "step": 3090}, {"loss": 1.3285, "grad_norm": 0.5040048360824585, "learning_rate": 0.0002, "epoch": 2.2439377488237424, "step": 3100}, {"loss": 1.3316, "grad_norm": 0.6660613417625427, "learning_rate": 0.0002, "epoch": 2.251176257690916, "step": 3110}, {"loss": 1.3108, "grad_norm": 0.5890918970108032, "learning_rate": 0.0002, "epoch": 2.258414766558089, "step": 3120}, {"loss": 1.248, "grad_norm": 0.6458896994590759, "learning_rate": 0.0002, "epoch": 2.2656532754252625, "step": 3130}, {"loss": 1.4151, "grad_norm": 0.6832690834999084, "learning_rate": 0.0002, "epoch": 2.272891784292436, "step": 3140}, {"loss": 1.4458, "grad_norm": 0.833908200263977, "learning_rate": 0.0002, "epoch": 2.2801302931596092, "step": 3150}, {"loss": 1.2931, "grad_norm": 0.4596034586429596, "learning_rate": 0.0002, "epoch": 2.2873688020267826, "step": 3160}, {"loss": 1.449, "grad_norm": 0.9130966067314148, "learning_rate": 0.0002, "epoch": 2.294607310893956, "step": 3170}, {"loss": 1.3806, "grad_norm": 0.7143292427062988, "learning_rate": 0.0002, "epoch": 2.3018458197611293, "step": 3180}, {"loss": 1.2692, "grad_norm": 0.5388900637626648, "learning_rate": 0.0002, "epoch": 2.3090843286283027, "step": 3190}, {"loss": 1.2402, "grad_norm": 0.5607513189315796, "learning_rate": 0.0002, "epoch": 2.316322837495476, "step": 3200}, {"loss": 1.3874, "grad_norm": 0.6795142292976379, "learning_rate": 0.0002, "epoch": 2.3235613463626494, "step": 3210}, {"loss": 1.3042, "grad_norm": 0.6561070680618286, "learning_rate": 0.0002, "epoch": 2.330799855229823, "step": 3220}, {"loss": 1.4636, "grad_norm": 0.8858118057250977, "learning_rate": 0.0002, "epoch": 2.338038364096996, "step": 3230}, {"loss": 1.3214, "grad_norm": 0.6604151725769043, "learning_rate": 0.0002, "epoch": 2.3452768729641695, "step": 3240}, {"loss": 1.4004, "grad_norm": 0.6755785346031189, "learning_rate": 0.0002, "epoch": 2.352515381831343, "step": 3250}, {"loss": 1.2503, "grad_norm": 0.6981677412986755, "learning_rate": 0.0002, "epoch": 2.3597538906985163, "step": 3260}, {"loss": 1.3078, "grad_norm": 0.6338568329811096, "learning_rate": 0.0002, "epoch": 2.3669923995656896, "step": 3270}, {"loss": 1.285, "grad_norm": 0.5754265785217285, "learning_rate": 0.0002, "epoch": 2.374230908432863, "step": 3280}, {"loss": 1.2924, "grad_norm": 0.7533153295516968, "learning_rate": 0.0002, "epoch": 2.3814694173000364, "step": 3290}, {"loss": 1.3711, "grad_norm": 0.675065279006958, "learning_rate": 0.0002, "epoch": 2.3887079261672097, "step": 3300}, {"loss": 1.3548, "grad_norm": 0.5686452984809875, "learning_rate": 0.0002, "epoch": 2.395946435034383, "step": 3310}, {"loss": 1.1998, "grad_norm": 0.8129481673240662, "learning_rate": 0.0002, "epoch": 2.4031849439015565, "step": 3320}, {"loss": 1.2584, "grad_norm": 0.6615934371948242, "learning_rate": 0.0002, "epoch": 2.41042345276873, "step": 3330}, {"loss": 1.3691, "grad_norm": 0.6678834557533264, "learning_rate": 0.0002, "epoch": 2.417661961635903, "step": 3340}, {"loss": 1.2381, "grad_norm": 0.5581308007240295, "learning_rate": 0.0002, "epoch": 2.4249004705030766, "step": 3350}, {"loss": 1.3853, "grad_norm": 0.6098920106887817, "learning_rate": 0.0002, "epoch": 2.43213897937025, "step": 3360}, {"loss": 1.3692, "grad_norm": 0.8101736903190613, "learning_rate": 0.0002, "epoch": 2.4393774882374233, "step": 3370}, {"loss": 1.4418, "grad_norm": 0.6621488928794861, "learning_rate": 0.0002, "epoch": 2.4466159971045967, "step": 3380}, {"loss": 1.4579, "grad_norm": 0.8693289160728455, "learning_rate": 0.0002, "epoch": 2.45385450597177, "step": 3390}, {"loss": 1.3644, "grad_norm": 0.6724580526351929, "learning_rate": 0.0002, "epoch": 2.4610930148389434, "step": 3400}, {"loss": 1.2006, "grad_norm": 0.6776891946792603, "learning_rate": 0.0002, "epoch": 2.4683315237061167, "step": 3410}, {"loss": 1.2937, "grad_norm": 0.7214453816413879, "learning_rate": 0.0002, "epoch": 2.47557003257329, "step": 3420}, {"loss": 1.4051, "grad_norm": 0.8390451073646545, "learning_rate": 0.0002, "epoch": 2.4828085414404635, "step": 3430}, {"loss": 1.25, "grad_norm": 0.7130982279777527, "learning_rate": 0.0002, "epoch": 2.490047050307637, "step": 3440}, {"loss": 1.2231, "grad_norm": 0.8873937129974365, "learning_rate": 0.0002, "epoch": 2.49728555917481, "step": 3450}, {"loss": 1.1429, "grad_norm": 0.725185751914978, "learning_rate": 0.0002, "epoch": 2.5045240680419836, "step": 3460}, {"loss": 1.2699, "grad_norm": 0.6120352149009705, "learning_rate": 0.0002, "epoch": 2.511762576909157, "step": 3470}, {"loss": 1.2552, "grad_norm": 0.7713613510131836, "learning_rate": 0.0002, "epoch": 2.5190010857763303, "step": 3480}, {"loss": 1.4648, "grad_norm": 0.895309567451477, "learning_rate": 0.0002, "epoch": 2.5262395946435037, "step": 3490}, {"loss": 1.3043, "grad_norm": 0.9631021022796631, "learning_rate": 0.0002, "epoch": 2.533478103510677, "step": 3500}, {"loss": 1.3492, "grad_norm": 0.7475683093070984, "learning_rate": 0.0002, "epoch": 2.5407166123778504, "step": 3510}, {"loss": 1.3637, "grad_norm": 0.7271341681480408, "learning_rate": 0.0002, "epoch": 2.5479551212450238, "step": 3520}, {"loss": 1.304, "grad_norm": 0.6979510188102722, "learning_rate": 0.0002, "epoch": 2.555193630112197, "step": 3530}, {"loss": 1.2353, "grad_norm": 0.6504196524620056, "learning_rate": 0.0002, "epoch": 2.5624321389793705, "step": 3540}, {"loss": 1.2699, "grad_norm": 0.7226675748825073, "learning_rate": 0.0002, "epoch": 2.569670647846544, "step": 3550}, {"loss": 1.3002, "grad_norm": 0.6143222451210022, "learning_rate": 0.0002, "epoch": 2.5769091567137172, "step": 3560}, {"loss": 1.1585, "grad_norm": 0.7245154976844788, "learning_rate": 0.0002, "epoch": 2.5841476655808906, "step": 3570}, {"loss": 1.3651, "grad_norm": 0.943540632724762, "learning_rate": 0.0002, "epoch": 2.591386174448064, "step": 3580}, {"loss": 1.3034, "grad_norm": 0.7707241773605347, "learning_rate": 0.0002, "epoch": 2.5986246833152373, "step": 3590}, {"loss": 1.3063, "grad_norm": 0.6705001592636108, "learning_rate": 0.0002, "epoch": 2.6058631921824107, "step": 3600}, {"loss": 1.2437, "grad_norm": 0.6360933780670166, "learning_rate": 0.0002, "epoch": 2.613101701049584, "step": 3610}, {"loss": 1.1844, "grad_norm": 0.5846424698829651, "learning_rate": 0.0002, "epoch": 2.6203402099167574, "step": 3620}, {"loss": 1.3674, "grad_norm": 0.5958625674247742, "learning_rate": 0.0002, "epoch": 2.6275787187839303, "step": 3630}, {"loss": 1.3599, "grad_norm": 0.6819243431091309, "learning_rate": 0.0002, "epoch": 2.6348172276511037, "step": 3640}, {"loss": 1.3884, "grad_norm": 0.7033445835113525, "learning_rate": 0.0002, "epoch": 2.642055736518277, "step": 3650}, {"loss": 1.3392, "grad_norm": 0.6134849786758423, "learning_rate": 0.0002, "epoch": 2.6492942453854504, "step": 3660}, {"loss": 1.2661, "grad_norm": 0.658009946346283, "learning_rate": 0.0002, "epoch": 2.656532754252624, "step": 3670}, {"loss": 1.3987, "grad_norm": 0.6280999779701233, "learning_rate": 0.0002, "epoch": 2.663771263119797, "step": 3680}, {"loss": 1.2995, "grad_norm": 0.5536085963249207, "learning_rate": 0.0002, "epoch": 2.6710097719869705, "step": 3690}, {"loss": 1.2044, "grad_norm": 0.8603981733322144, "learning_rate": 0.0002, "epoch": 2.678248280854144, "step": 3700}, {"loss": 1.3879, "grad_norm": 0.5509994626045227, "learning_rate": 0.0002, "epoch": 2.6854867897213173, "step": 3710}, {"loss": 1.3253, "grad_norm": 0.9093621969223022, "learning_rate": 0.0002, "epoch": 2.6927252985884906, "step": 3720}, {"loss": 1.2668, "grad_norm": 0.7525952458381653, "learning_rate": 0.0002, "epoch": 2.699963807455664, "step": 3730}, {"loss": 1.248, "grad_norm": 0.6737023591995239, "learning_rate": 0.0002, "epoch": 2.7072023163228374, "step": 3740}, {"loss": 1.2981, "grad_norm": 0.8656924962997437, "learning_rate": 0.0002, "epoch": 2.7144408251900107, "step": 3750}, {"loss": 1.2342, "grad_norm": 0.7494133114814758, "learning_rate": 0.0002, "epoch": 2.721679334057184, "step": 3760}, {"loss": 1.2417, "grad_norm": 0.5725520849227905, "learning_rate": 0.0002, "epoch": 2.7289178429243575, "step": 3770}, {"loss": 1.28, "grad_norm": 0.836412787437439, "learning_rate": 0.0002, "epoch": 2.736156351791531, "step": 3780}, {"loss": 1.3784, "grad_norm": 0.6893242597579956, "learning_rate": 0.0002, "epoch": 2.743394860658704, "step": 3790}, {"loss": 1.2929, "grad_norm": 0.6696223020553589, "learning_rate": 0.0002, "epoch": 2.7506333695258776, "step": 3800}, {"loss": 1.2449, "grad_norm": 0.6483015418052673, "learning_rate": 0.0002, "epoch": 2.757871878393051, "step": 3810}, {"loss": 1.3282, "grad_norm": 0.8084456920623779, "learning_rate": 0.0002, "epoch": 2.7651103872602243, "step": 3820}, {"loss": 1.3694, "grad_norm": 0.6601949334144592, "learning_rate": 0.0002, "epoch": 2.7723488961273977, "step": 3830}, {"loss": 1.3568, "grad_norm": 0.6905533671379089, "learning_rate": 0.0002, "epoch": 2.779587404994571, "step": 3840}, {"loss": 1.3854, "grad_norm": 0.619318425655365, "learning_rate": 0.0002, "epoch": 2.7868259138617444, "step": 3850}, {"loss": 1.2551, "grad_norm": 0.5994023084640503, "learning_rate": 0.0002, "epoch": 2.7940644227289178, "step": 3860}, {"loss": 1.2022, "grad_norm": 0.5627168416976929, "learning_rate": 0.0002, "epoch": 2.801302931596091, "step": 3870}, {"loss": 1.3921, "grad_norm": 0.6001605987548828, "learning_rate": 0.0002, "epoch": 2.8085414404632645, "step": 3880}, {"loss": 1.3026, "grad_norm": 0.6022412776947021, "learning_rate": 0.0002, "epoch": 2.815779949330438, "step": 3890}, {"loss": 1.2765, "grad_norm": 0.6832426190376282, "learning_rate": 0.0002, "epoch": 2.823018458197611, "step": 3900}, {"loss": 1.1363, "grad_norm": 0.5936811566352844, "learning_rate": 0.0002, "epoch": 2.8302569670647846, "step": 3910}, {"loss": 1.1707, "grad_norm": 0.6960572600364685, "learning_rate": 0.0002, "epoch": 2.837495475931958, "step": 3920}, {"loss": 1.4063, "grad_norm": 0.5913406610488892, "learning_rate": 0.0002, "epoch": 2.8447339847991313, "step": 3930}, {"loss": 1.3245, "grad_norm": 0.678154706954956, "learning_rate": 0.0002, "epoch": 2.8519724936663047, "step": 3940}, {"loss": 1.366, "grad_norm": 0.7898936867713928, "learning_rate": 0.0002, "epoch": 2.859211002533478, "step": 3950}, {"loss": 1.3948, "grad_norm": 0.9234195351600647, "learning_rate": 0.0002, "epoch": 2.8664495114006514, "step": 3960}, {"loss": 1.2773, "grad_norm": 0.5960825085639954, "learning_rate": 0.0002, "epoch": 2.8736880202678248, "step": 3970}, {"loss": 1.3127, "grad_norm": 0.677118182182312, "learning_rate": 0.0002, "epoch": 2.880926529134998, "step": 3980}, {"loss": 1.2652, "grad_norm": 0.6505142450332642, "learning_rate": 0.0002, "epoch": 2.8881650380021715, "step": 3990}, {"loss": 1.2078, "grad_norm": 0.550826907157898, "learning_rate": 0.0002, "epoch": 2.895403546869345, "step": 4000}, {"loss": 1.1811, "grad_norm": 0.6209215521812439, "learning_rate": 0.0002, "epoch": 2.9026420557365182, "step": 4010}, {"loss": 1.4001, "grad_norm": 0.6549018025398254, "learning_rate": 0.0002, "epoch": 2.9098805646036916, "step": 4020}, {"loss": 1.2285, "grad_norm": 0.570682168006897, "learning_rate": 0.0002, "epoch": 2.917119073470865, "step": 4030}, {"loss": 1.0832, "grad_norm": 1.1807632446289062, "learning_rate": 0.0002, "epoch": 2.9243575823380383, "step": 4040}, {"loss": 1.2693, "grad_norm": 0.7058857679367065, "learning_rate": 0.0002, "epoch": 2.9315960912052117, "step": 4050}, {"loss": 1.2905, "grad_norm": 0.5542812943458557, "learning_rate": 0.0002, "epoch": 2.938834600072385, "step": 4060}, {"loss": 1.33, "grad_norm": 0.63167804479599, "learning_rate": 0.0002, "epoch": 2.9460731089395584, "step": 4070}, {"loss": 1.3075, "grad_norm": 0.5702962279319763, "learning_rate": 0.0002, "epoch": 2.953311617806732, "step": 4080}, {"loss": 1.2007, "grad_norm": 0.620944082736969, "learning_rate": 0.0002, "epoch": 2.960550126673905, "step": 4090}, {"loss": 1.2864, "grad_norm": 0.5866289734840393, "learning_rate": 0.0002, "epoch": 2.9677886355410785, "step": 4100}, {"loss": 1.3293, "grad_norm": 0.560170590877533, "learning_rate": 0.0002, "epoch": 2.975027144408252, "step": 4110}, {"loss": 1.2071, "grad_norm": 0.675082802772522, "learning_rate": 0.0002, "epoch": 2.9822656532754253, "step": 4120}, {"loss": 1.2981, "grad_norm": 0.62708580493927, "learning_rate": 0.0002, "epoch": 2.9895041621425986, "step": 4130}, {"loss": 1.2758, "grad_norm": 0.7893929481506348, "learning_rate": 0.0002, "epoch": 2.996742671009772, "step": 4140}, {"eval_loss": 1.4217946529388428, "eval_runtime": 27.1596, "eval_samples_per_second": 16.053, "eval_steps_per_second": 2.025, "epoch": 2.9996380745566413, "step": 4144}, {"loss": 1.2152, "grad_norm": 0.7043836116790771, "learning_rate": 0.0002, "epoch": 3.0039811798769454, "step": 4150}, {"loss": 1.1664, "grad_norm": 0.6806283593177795, "learning_rate": 0.0002, "epoch": 3.0112196887441187, "step": 4160}, {"loss": 1.292, "grad_norm": 0.7684550285339355, "learning_rate": 0.0002, "epoch": 3.018458197611292, "step": 4170}, {"loss": 1.3467, "grad_norm": 0.7895237803459167, "learning_rate": 0.0002, "epoch": 3.0256967064784654, "step": 4180}, {"loss": 1.1324, "grad_norm": 0.7464531064033508, "learning_rate": 0.0002, "epoch": 3.032935215345639, "step": 4190}, {"loss": 1.1614, "grad_norm": 0.9358500838279724, "learning_rate": 0.0002, "epoch": 3.040173724212812, "step": 4200}, {"loss": 1.1834, "grad_norm": 1.1066628694534302, "learning_rate": 0.0002, "epoch": 3.0474122330799855, "step": 4210}, {"loss": 1.1557, "grad_norm": 0.6663267612457275, "learning_rate": 0.0002, "epoch": 3.054650741947159, "step": 4220}, {"loss": 1.1707, "grad_norm": 0.6669464707374573, "learning_rate": 0.0002, "epoch": 3.0618892508143323, "step": 4230}, {"loss": 1.1841, "grad_norm": 0.7052164077758789, "learning_rate": 0.0002, "epoch": 3.0691277596815056, "step": 4240}, {"loss": 1.2913, "grad_norm": 0.6118432879447937, "learning_rate": 0.0002, "epoch": 3.076366268548679, "step": 4250}, {"loss": 1.1526, "grad_norm": 0.6915903687477112, "learning_rate": 0.0002, "epoch": 3.0836047774158524, "step": 4260}, {"loss": 1.1348, "grad_norm": 0.7441644668579102, "learning_rate": 0.0002, "epoch": 3.0908432862830257, "step": 4270}, {"loss": 1.1672, "grad_norm": 0.823850691318512, "learning_rate": 0.0002, "epoch": 3.098081795150199, "step": 4280}, {"loss": 1.2655, "grad_norm": 0.9677883386611938, "learning_rate": 0.0002, "epoch": 3.1053203040173725, "step": 4290}, {"loss": 1.1794, "grad_norm": 0.7002579569816589, "learning_rate": 0.0002, "epoch": 3.112558812884546, "step": 4300}, {"loss": 1.135, "grad_norm": 0.778789758682251, "learning_rate": 0.0002, "epoch": 3.119797321751719, "step": 4310}, {"loss": 1.0818, "grad_norm": 0.7236007452011108, "learning_rate": 0.0002, "epoch": 3.1270358306188926, "step": 4320}, {"loss": 1.1803, "grad_norm": 0.8809133768081665, "learning_rate": 0.0002, "epoch": 3.134274339486066, "step": 4330}, {"loss": 1.2571, "grad_norm": 0.7924913167953491, "learning_rate": 0.0002, "epoch": 3.1415128483532393, "step": 4340}, {"loss": 1.1413, "grad_norm": 0.7437422275543213, "learning_rate": 0.0002, "epoch": 3.1487513572204127, "step": 4350}, {"loss": 1.2088, "grad_norm": 0.6428450345993042, "learning_rate": 0.0002, "epoch": 3.155989866087586, "step": 4360}, {"loss": 1.3032, "grad_norm": 0.7922873497009277, "learning_rate": 0.0002, "epoch": 3.1632283749547594, "step": 4370}, {"loss": 1.216, "grad_norm": 0.5252506732940674, "learning_rate": 0.0002, "epoch": 3.1704668838219328, "step": 4380}, {"loss": 1.1297, "grad_norm": 0.8570457696914673, "learning_rate": 0.0002, "epoch": 3.177705392689106, "step": 4390}, {"loss": 1.0994, "grad_norm": 0.7218987345695496, "learning_rate": 0.0002, "epoch": 3.1849439015562795, "step": 4400}, {"loss": 1.2891, "grad_norm": 0.6921393275260925, "learning_rate": 0.0002, "epoch": 3.192182410423453, "step": 4410}, {"loss": 1.2668, "grad_norm": 0.7386137843132019, "learning_rate": 0.0002, "epoch": 3.199420919290626, "step": 4420}, {"loss": 1.1654, "grad_norm": 0.6227759122848511, "learning_rate": 0.0002, "epoch": 3.2066594281577996, "step": 4430}, {"loss": 1.1752, "grad_norm": 0.7180278897285461, "learning_rate": 0.0002, "epoch": 3.213897937024973, "step": 4440}, {"loss": 1.1757, "grad_norm": 0.745830774307251, "learning_rate": 0.0002, "epoch": 3.2211364458921463, "step": 4450}, {"loss": 1.234, "grad_norm": 0.6766072511672974, "learning_rate": 0.0002, "epoch": 3.2283749547593197, "step": 4460}, {"loss": 1.1999, "grad_norm": 0.8325067162513733, "learning_rate": 0.0002, "epoch": 3.235613463626493, "step": 4470}, {"loss": 1.1606, "grad_norm": 0.7148305177688599, "learning_rate": 0.0002, "epoch": 3.2428519724936664, "step": 4480}, {"loss": 1.1383, "grad_norm": 0.7752676010131836, "learning_rate": 0.0002, "epoch": 3.25009048136084, "step": 4490}, {"loss": 1.3006, "grad_norm": 0.6776860952377319, "learning_rate": 0.0002, "epoch": 3.257328990228013, "step": 4500}, {"loss": 1.0796, "grad_norm": 0.704359769821167, "learning_rate": 0.0002, "epoch": 3.2645674990951865, "step": 4510}, {"loss": 1.2496, "grad_norm": 0.6880282163619995, "learning_rate": 0.0002, "epoch": 3.27180600796236, "step": 4520}, {"loss": 1.0947, "grad_norm": 0.8179270029067993, "learning_rate": 0.0002, "epoch": 3.2790445168295332, "step": 4530}, {"loss": 1.1909, "grad_norm": 0.6718448996543884, "learning_rate": 0.0002, "epoch": 3.2862830256967066, "step": 4540}, {"loss": 1.2708, "grad_norm": 0.8300657868385315, "learning_rate": 0.0002, "epoch": 3.29352153456388, "step": 4550}, {"loss": 1.2594, "grad_norm": 0.6433690786361694, "learning_rate": 0.0002, "epoch": 3.3007600434310533, "step": 4560}, {"loss": 1.2479, "grad_norm": 0.690262496471405, "learning_rate": 0.0002, "epoch": 3.3079985522982267, "step": 4570}, {"loss": 1.1342, "grad_norm": 0.7022852301597595, "learning_rate": 0.0002, "epoch": 3.3152370611654, "step": 4580}, {"loss": 1.0844, "grad_norm": 0.6438387632369995, "learning_rate": 0.0002, "epoch": 3.3224755700325734, "step": 4590}, {"loss": 1.17, "grad_norm": 0.6866899132728577, "learning_rate": 0.0002, "epoch": 3.329714078899747, "step": 4600}, {"loss": 1.1289, "grad_norm": 0.8233968019485474, "learning_rate": 0.0002, "epoch": 3.33695258776692, "step": 4610}, {"loss": 1.1855, "grad_norm": 0.7251574993133545, "learning_rate": 0.0002, "epoch": 3.3441910966340935, "step": 4620}, {"loss": 1.3403, "grad_norm": 0.7855110168457031, "learning_rate": 0.0002, "epoch": 3.351429605501267, "step": 4630}, {"loss": 1.2922, "grad_norm": 0.8487356305122375, "learning_rate": 0.0002, "epoch": 3.3586681143684403, "step": 4640}, {"loss": 1.2462, "grad_norm": 0.6429011225700378, "learning_rate": 0.0002, "epoch": 3.3659066232356136, "step": 4650}, {"loss": 1.129, "grad_norm": 0.7095270156860352, "learning_rate": 0.0002, "epoch": 3.373145132102787, "step": 4660}, {"loss": 1.262, "grad_norm": 0.6792303323745728, "learning_rate": 0.0002, "epoch": 3.3803836409699604, "step": 4670}, {"loss": 1.256, "grad_norm": 0.6784825921058655, "learning_rate": 0.0002, "epoch": 3.3876221498371337, "step": 4680}, {"loss": 1.0838, "grad_norm": 0.6362888216972351, "learning_rate": 0.0002, "epoch": 3.394860658704307, "step": 4690}, {"loss": 1.2165, "grad_norm": 0.7794778943061829, "learning_rate": 0.0002, "epoch": 3.4020991675714805, "step": 4700}, {"loss": 1.0644, "grad_norm": 0.7287485003471375, "learning_rate": 0.0002, "epoch": 3.409337676438654, "step": 4710}, {"loss": 1.2925, "grad_norm": 0.6481451392173767, "learning_rate": 0.0002, "epoch": 3.416576185305827, "step": 4720}, {"loss": 1.2121, "grad_norm": 0.9200371503829956, "learning_rate": 0.0002, "epoch": 3.4238146941730006, "step": 4730}, {"loss": 1.072, "grad_norm": 1.074180245399475, "learning_rate": 0.0002, "epoch": 3.431053203040174, "step": 4740}, {"loss": 1.0421, "grad_norm": 0.6722986698150635, "learning_rate": 0.0002, "epoch": 3.438291711907347, "step": 4750}, {"loss": 1.2258, "grad_norm": 0.7945933938026428, "learning_rate": 0.0002, "epoch": 3.44553022077452, "step": 4760}, {"loss": 1.0927, "grad_norm": 0.7624640464782715, "learning_rate": 0.0002, "epoch": 3.4527687296416936, "step": 4770}, {"loss": 1.2428, "grad_norm": 0.7763656377792358, "learning_rate": 0.0002, "epoch": 3.460007238508867, "step": 4780}, {"loss": 1.2584, "grad_norm": 0.7736947536468506, "learning_rate": 0.0002, "epoch": 3.4672457473760403, "step": 4790}, {"loss": 1.1953, "grad_norm": 0.8450354933738708, "learning_rate": 0.0002, "epoch": 3.4744842562432137, "step": 4800}, {"loss": 1.1362, "grad_norm": 0.6480133533477783, "learning_rate": 0.0002, "epoch": 3.481722765110387, "step": 4810}, {"loss": 1.1882, "grad_norm": 0.8437445759773254, "learning_rate": 0.0002, "epoch": 3.4889612739775604, "step": 4820}, {"loss": 1.1519, "grad_norm": 0.7781730890274048, "learning_rate": 0.0002, "epoch": 3.4961997828447338, "step": 4830}, {"loss": 1.1836, "grad_norm": 0.8523228168487549, "learning_rate": 0.0002, "epoch": 3.503438291711907, "step": 4840}, {"loss": 1.1672, "grad_norm": 0.6236732006072998, "learning_rate": 0.0002, "epoch": 3.5106768005790805, "step": 4850}, {"loss": 1.1926, "grad_norm": 0.7500787377357483, "learning_rate": 0.0002, "epoch": 3.517915309446254, "step": 4860}, {"loss": 1.1998, "grad_norm": 0.7665374875068665, "learning_rate": 0.0002, "epoch": 3.5251538183134272, "step": 4870}, {"loss": 1.1551, "grad_norm": 0.787857711315155, "learning_rate": 0.0002, "epoch": 3.5323923271806006, "step": 4880}, {"loss": 1.2758, "grad_norm": 0.970595121383667, "learning_rate": 0.0002, "epoch": 3.539630836047774, "step": 4890}, {"loss": 1.1274, "grad_norm": 0.6409347057342529, "learning_rate": 0.0002, "epoch": 3.5468693449149473, "step": 4900}, {"loss": 1.1596, "grad_norm": 0.888551652431488, "learning_rate": 0.0002, "epoch": 3.5541078537821207, "step": 4910}, {"loss": 1.1644, "grad_norm": 1.0808377265930176, "learning_rate": 0.0002, "epoch": 3.561346362649294, "step": 4920}, {"loss": 1.2564, "grad_norm": 0.7501053214073181, "learning_rate": 0.0002, "epoch": 3.5685848715164674, "step": 4930}, {"loss": 1.2351, "grad_norm": 0.7375240325927734, "learning_rate": 0.0002, "epoch": 3.575823380383641, "step": 4940}, {"loss": 1.3568, "grad_norm": 0.7075039744377136, "learning_rate": 0.0002, "epoch": 3.583061889250814, "step": 4950}, {"loss": 1.3355, "grad_norm": 0.939337432384491, "learning_rate": 0.0002, "epoch": 3.5903003981179875, "step": 4960}, {"loss": 1.1722, "grad_norm": 0.6717396974563599, "learning_rate": 0.0002, "epoch": 3.597538906985161, "step": 4970}, {"loss": 1.1186, "grad_norm": 0.7141643762588501, "learning_rate": 0.0002, "epoch": 3.6047774158523342, "step": 4980}, {"loss": 1.1011, "grad_norm": 0.7109216451644897, "learning_rate": 0.0002, "epoch": 3.6120159247195076, "step": 4990}, {"loss": 1.2178, "grad_norm": 0.7020776867866516, "learning_rate": 0.0002, "epoch": 3.619254433586681, "step": 5000}, {"loss": 1.1939, "grad_norm": 0.7158873677253723, "learning_rate": 0.0002, "epoch": 3.6264929424538543, "step": 5010}, {"loss": 1.2624, "grad_norm": 0.7062035202980042, "learning_rate": 0.0002, "epoch": 3.6337314513210277, "step": 5020}, {"loss": 1.0224, "grad_norm": 0.7081155776977539, "learning_rate": 0.0002, "epoch": 3.640969960188201, "step": 5030}, {"loss": 1.2195, "grad_norm": 1.2210607528686523, "learning_rate": 0.0002, "epoch": 3.6482084690553744, "step": 5040}, {"loss": 1.2596, "grad_norm": 0.6650236248970032, "learning_rate": 0.0002, "epoch": 3.655446977922548, "step": 5050}, {"loss": 1.1072, "grad_norm": 0.6884829998016357, "learning_rate": 0.0002, "epoch": 3.662685486789721, "step": 5060}, {"loss": 1.2292, "grad_norm": 0.7317819595336914, "learning_rate": 0.0002, "epoch": 3.6699239956568945, "step": 5070}, {"loss": 1.1917, "grad_norm": 0.7406691908836365, "learning_rate": 0.0002, "epoch": 3.677162504524068, "step": 5080}, {"loss": 1.2949, "grad_norm": 0.9009454250335693, "learning_rate": 0.0002, "epoch": 3.6844010133912413, "step": 5090}, {"loss": 1.1528, "grad_norm": 0.8189385533332825, "learning_rate": 0.0002, "epoch": 3.6916395222584146, "step": 5100}, {"loss": 1.3408, "grad_norm": 1.0793628692626953, "learning_rate": 0.0002, "epoch": 3.698878031125588, "step": 5110}, {"loss": 1.2417, "grad_norm": 0.8593027591705322, "learning_rate": 0.0002, "epoch": 3.7061165399927614, "step": 5120}, {"loss": 1.2141, "grad_norm": 0.8481812477111816, "learning_rate": 0.0002, "epoch": 3.7133550488599347, "step": 5130}, {"loss": 1.125, "grad_norm": 0.6527451276779175, "learning_rate": 0.0002, "epoch": 3.720593557727108, "step": 5140}, {"loss": 1.1584, "grad_norm": 0.9220114350318909, "learning_rate": 0.0002, "epoch": 3.7278320665942815, "step": 5150}, {"loss": 1.2267, "grad_norm": 1.0842019319534302, "learning_rate": 0.0002, "epoch": 3.735070575461455, "step": 5160}, {"loss": 1.3083, "grad_norm": 0.965453565120697, "learning_rate": 0.0002, "epoch": 3.742309084328628, "step": 5170}, {"loss": 1.1772, "grad_norm": 0.9903319478034973, "learning_rate": 0.0002, "epoch": 3.7495475931958016, "step": 5180}, {"loss": 1.2515, "grad_norm": 0.7434818148612976, "learning_rate": 0.0002, "epoch": 3.756786102062975, "step": 5190}, {"loss": 1.2631, "grad_norm": 0.6717280745506287, "learning_rate": 0.0002, "epoch": 3.7640246109301483, "step": 5200}, {"loss": 1.2012, "grad_norm": 0.7754665613174438, "learning_rate": 0.0002, "epoch": 3.7712631197973217, "step": 5210}, {"loss": 1.305, "grad_norm": 1.028374433517456, "learning_rate": 0.0002, "epoch": 3.778501628664495, "step": 5220}, {"loss": 1.1866, "grad_norm": 0.6026996374130249, "learning_rate": 0.0002, "epoch": 3.7857401375316684, "step": 5230}, {"loss": 1.1901, "grad_norm": 0.6978490948677063, "learning_rate": 0.0002, "epoch": 3.7929786463988417, "step": 5240}, {"loss": 1.2576, "grad_norm": 0.7303446531295776, "learning_rate": 0.0002, "epoch": 3.800217155266015, "step": 5250}, {"loss": 1.3173, "grad_norm": 1.0734210014343262, "learning_rate": 0.0002, "epoch": 3.8074556641331885, "step": 5260}, {"loss": 1.1137, "grad_norm": 0.6383201479911804, "learning_rate": 0.0002, "epoch": 3.814694173000362, "step": 5270}, {"loss": 1.0904, "grad_norm": 0.7742630243301392, "learning_rate": 0.0002, "epoch": 3.821932681867535, "step": 5280}, {"loss": 1.2232, "grad_norm": 0.8477074503898621, "learning_rate": 0.0002, "epoch": 3.8291711907347086, "step": 5290}, {"loss": 1.2047, "grad_norm": 0.6675317883491516, "learning_rate": 0.0002, "epoch": 3.836409699601882, "step": 5300}, {"loss": 1.2275, "grad_norm": 0.7515445351600647, "learning_rate": 0.0002, "epoch": 3.8436482084690553, "step": 5310}, {"loss": 1.2569, "grad_norm": 1.1441220045089722, "learning_rate": 0.0002, "epoch": 3.8508867173362287, "step": 5320}, {"loss": 1.1512, "grad_norm": 0.7968795895576477, "learning_rate": 0.0002, "epoch": 3.858125226203402, "step": 5330}, {"loss": 1.232, "grad_norm": 0.7842824459075928, "learning_rate": 0.0002, "epoch": 3.8653637350705754, "step": 5340}, {"loss": 1.1847, "grad_norm": 0.8272225260734558, "learning_rate": 0.0002, "epoch": 3.8726022439377488, "step": 5350}, {"loss": 1.1381, "grad_norm": 0.8413397669792175, "learning_rate": 0.0002, "epoch": 3.879840752804922, "step": 5360}, {"loss": 1.2349, "grad_norm": 1.141764760017395, "learning_rate": 0.0002, "epoch": 3.8870792616720955, "step": 5370}, {"loss": 1.212, "grad_norm": 0.9826975464820862, "learning_rate": 0.0002, "epoch": 3.894317770539269, "step": 5380}, {"loss": 1.1833, "grad_norm": 0.8598255515098572, "learning_rate": 0.0002, "epoch": 3.9015562794064422, "step": 5390}, {"loss": 1.1247, "grad_norm": 0.6271058320999146, "learning_rate": 0.0002, "epoch": 3.9087947882736156, "step": 5400}, {"loss": 1.2212, "grad_norm": 0.6379870772361755, "learning_rate": 0.0002, "epoch": 3.916033297140789, "step": 5410}, {"loss": 1.2481, "grad_norm": 1.0313376188278198, "learning_rate": 0.0002, "epoch": 3.9232718060079623, "step": 5420}, {"loss": 1.1872, "grad_norm": 0.8220619559288025, "learning_rate": 0.0002, "epoch": 3.9305103148751357, "step": 5430}, {"loss": 1.2006, "grad_norm": 0.7576116919517517, "learning_rate": 0.0002, "epoch": 3.937748823742309, "step": 5440}, {"loss": 1.1969, "grad_norm": 1.226235032081604, "learning_rate": 0.0002, "epoch": 3.9449873326094824, "step": 5450}, {"loss": 1.2945, "grad_norm": 0.7979229688644409, "learning_rate": 0.0002, "epoch": 3.952225841476656, "step": 5460}, {"loss": 1.1922, "grad_norm": 0.9911929965019226, "learning_rate": 0.0002, "epoch": 3.959464350343829, "step": 5470}, {"loss": 1.0924, "grad_norm": 0.643738865852356, "learning_rate": 0.0002, "epoch": 3.9667028592110025, "step": 5480}, {"loss": 1.0607, "grad_norm": 0.682305634021759, "learning_rate": 0.0002, "epoch": 3.973941368078176, "step": 5490}, {"loss": 1.2908, "grad_norm": 1.18373441696167, "learning_rate": 0.0002, "epoch": 3.9811798769453492, "step": 5500}, {"loss": 1.0889, "grad_norm": 0.7190203070640564, "learning_rate": 0.0002, "epoch": 3.9884183858125226, "step": 5510}, {"loss": 1.2745, "grad_norm": 0.7516948580741882, "learning_rate": 0.0002, "epoch": 3.995656894679696, "step": 5520}, {"eval_loss": 1.4252897500991821, "eval_runtime": 27.235, "eval_samples_per_second": 16.009, "eval_steps_per_second": 2.019, "epoch": 4.0, "step": 5526}, {"loss": 1.0088, "grad_norm": 0.6353074312210083, "learning_rate": 0.0002, "epoch": 4.002895403546869, "step": 5530}, {"loss": 1.0326, "grad_norm": 0.7424906492233276, "learning_rate": 0.0002, "epoch": 4.010133912414043, "step": 5540}, {"loss": 1.0667, "grad_norm": 0.8856638073921204, "learning_rate": 0.0002, "epoch": 4.017372421281216, "step": 5550}, {"loss": 1.0905, "grad_norm": 0.9627974033355713, "learning_rate": 0.0002, "epoch": 4.024610930148389, "step": 5560}, {"loss": 1.0965, "grad_norm": 0.9048978686332703, "learning_rate": 0.0002, "epoch": 4.031849439015563, "step": 5570}, {"loss": 1.1108, "grad_norm": 0.921119213104248, "learning_rate": 0.0002, "epoch": 4.039087947882736, "step": 5580}, {"loss": 1.1235, "grad_norm": 0.8654361963272095, "learning_rate": 0.0002, "epoch": 4.0463264567499095, "step": 5590}, {"loss": 1.0794, "grad_norm": 0.7947945594787598, "learning_rate": 0.0002, "epoch": 4.053564965617083, "step": 5600}, {"loss": 1.0674, "grad_norm": 0.8307326436042786, "learning_rate": 0.0002, "epoch": 4.060803474484256, "step": 5610}, {"loss": 1.0076, "grad_norm": 0.793273389339447, "learning_rate": 0.0002, "epoch": 4.06804198335143, "step": 5620}, {"loss": 1.0651, "grad_norm": 0.8748673796653748, "learning_rate": 0.0002, "epoch": 4.075280492218603, "step": 5630}, {"loss": 1.111, "grad_norm": 0.7926856279373169, "learning_rate": 0.0002, "epoch": 4.082519001085776, "step": 5640}, {"loss": 1.044, "grad_norm": 0.922645092010498, "learning_rate": 0.0002, "epoch": 4.08975750995295, "step": 5650}, {"loss": 1.109, "grad_norm": 0.9539641737937927, "learning_rate": 0.0002, "epoch": 4.096996018820123, "step": 5660}, {"loss": 1.0788, "grad_norm": 0.8674443364143372, "learning_rate": 0.0002, "epoch": 4.1042345276872965, "step": 5670}, {"loss": 0.9867, "grad_norm": 0.7097609043121338, "learning_rate": 0.0002, "epoch": 4.11147303655447, "step": 5680}, {"loss": 1.1154, "grad_norm": 0.8875522613525391, "learning_rate": 0.0002, "epoch": 4.118711545421643, "step": 5690}, {"loss": 1.1217, "grad_norm": 0.8583634495735168, "learning_rate": 0.0002, "epoch": 4.125950054288817, "step": 5700}, {"loss": 1.0973, "grad_norm": 0.6736377477645874, "learning_rate": 0.0002, "epoch": 4.13318856315599, "step": 5710}, {"loss": 1.1199, "grad_norm": 0.9349062442779541, "learning_rate": 0.0002, "epoch": 4.140427072023163, "step": 5720}, {"loss": 1.0508, "grad_norm": 1.0610365867614746, "learning_rate": 0.0002, "epoch": 4.147665580890337, "step": 5730}, {"loss": 1.1146, "grad_norm": 1.5838189125061035, "learning_rate": 0.0002, "epoch": 4.15490408975751, "step": 5740}, {"loss": 1.0222, "grad_norm": 0.747522234916687, "learning_rate": 0.0002, "epoch": 4.162142598624683, "step": 5750}, {"loss": 1.1328, "grad_norm": 1.3247915506362915, "learning_rate": 0.0002, "epoch": 4.169381107491857, "step": 5760}, {"loss": 1.1655, "grad_norm": 0.8750247955322266, "learning_rate": 0.0002, "epoch": 4.17661961635903, "step": 5770}, {"loss": 1.199, "grad_norm": 0.7914144992828369, "learning_rate": 0.0002, "epoch": 4.1838581252262035, "step": 5780}, {"loss": 1.1213, "grad_norm": 0.9493299126625061, "learning_rate": 0.0002, "epoch": 4.191096634093377, "step": 5790}, {"loss": 1.1515, "grad_norm": 0.7802295088768005, "learning_rate": 0.0002, "epoch": 4.19833514296055, "step": 5800}, {"loss": 1.0704, "grad_norm": 0.6987314820289612, "learning_rate": 0.0002, "epoch": 4.205573651827724, "step": 5810}, {"loss": 1.1699, "grad_norm": 0.9220341444015503, "learning_rate": 0.0002, "epoch": 4.212812160694897, "step": 5820}, {"loss": 1.1394, "grad_norm": 0.8932939767837524, "learning_rate": 0.0002, "epoch": 4.22005066956207, "step": 5830}, {"loss": 1.0048, "grad_norm": 0.920002818107605, "learning_rate": 0.0002, "epoch": 4.227289178429244, "step": 5840}, {"loss": 0.964, "grad_norm": 0.6662752032279968, "learning_rate": 0.0002, "epoch": 4.234527687296417, "step": 5850}, {"loss": 0.986, "grad_norm": 0.8679718971252441, "learning_rate": 0.0002, "epoch": 4.24176619616359, "step": 5860}, {"loss": 0.8991, "grad_norm": 0.7020887732505798, "learning_rate": 0.0002, "epoch": 4.249004705030764, "step": 5870}, {"loss": 1.1132, "grad_norm": 0.869611382484436, "learning_rate": 0.0002, "epoch": 4.256243213897937, "step": 5880}, {"loss": 1.1026, "grad_norm": 0.7796585559844971, "learning_rate": 0.0002, "epoch": 4.2634817227651105, "step": 5890}, {"loss": 1.0957, "grad_norm": 0.8978819251060486, "learning_rate": 0.0002, "epoch": 4.270720231632284, "step": 5900}, {"loss": 1.1325, "grad_norm": 1.0837205648422241, "learning_rate": 0.0002, "epoch": 4.277958740499457, "step": 5910}, {"loss": 1.1279, "grad_norm": 0.7584353089332581, "learning_rate": 0.0002, "epoch": 4.285197249366631, "step": 5920}, {"loss": 1.0513, "grad_norm": 0.7313185334205627, "learning_rate": 0.0002, "epoch": 4.292435758233804, "step": 5930}, {"loss": 1.1101, "grad_norm": 0.8004671335220337, "learning_rate": 0.0002, "epoch": 4.299674267100977, "step": 5940}, {"loss": 1.14, "grad_norm": 2.154958724975586, "learning_rate": 0.0002, "epoch": 4.306912775968151, "step": 5950}, {"loss": 1.1206, "grad_norm": 0.9163479804992676, "learning_rate": 0.0002, "epoch": 4.314151284835324, "step": 5960}, {"loss": 0.9941, "grad_norm": 0.9151589274406433, "learning_rate": 0.0002, "epoch": 4.321389793702497, "step": 5970}, {"loss": 1.0606, "grad_norm": 0.8624112010002136, "learning_rate": 0.0002, "epoch": 4.328628302569671, "step": 5980}, {"loss": 1.1625, "grad_norm": 0.9357741475105286, "learning_rate": 0.0002, "epoch": 4.335866811436844, "step": 5990}, {"loss": 1.0712, "grad_norm": 1.3482335805892944, "learning_rate": 0.0002, "epoch": 4.3431053203040175, "step": 6000}, {"loss": 1.1224, "grad_norm": 0.7156149744987488, "learning_rate": 0.0002, "epoch": 4.350343829171191, "step": 6010}, {"loss": 1.0753, "grad_norm": 0.8480049967765808, "learning_rate": 0.0002, "epoch": 4.357582338038364, "step": 6020}, {"loss": 1.051, "grad_norm": 0.8262244462966919, "learning_rate": 0.0002, "epoch": 4.364820846905538, "step": 6030}, {"loss": 0.9966, "grad_norm": 0.7733905911445618, "learning_rate": 0.0002, "epoch": 4.372059355772711, "step": 6040}, {"loss": 1.1008, "grad_norm": 0.8553919792175293, "learning_rate": 0.0002, "epoch": 4.379297864639884, "step": 6050}, {"loss": 1.1777, "grad_norm": 0.8666832447052002, "learning_rate": 0.0002, "epoch": 4.386536373507058, "step": 6060}, {"loss": 1.1934, "grad_norm": 0.9168295860290527, "learning_rate": 0.0002, "epoch": 4.393774882374231, "step": 6070}, {"loss": 1.0988, "grad_norm": 0.7315238118171692, "learning_rate": 0.0002, "epoch": 4.4010133912414044, "step": 6080}, {"loss": 1.1599, "grad_norm": 1.020263433456421, "learning_rate": 0.0002, "epoch": 4.408251900108578, "step": 6090}, {"loss": 1.133, "grad_norm": 0.9978243708610535, "learning_rate": 0.0002, "epoch": 4.415490408975751, "step": 6100}, {"loss": 1.1324, "grad_norm": 0.995453953742981, "learning_rate": 0.0002, "epoch": 4.4227289178429245, "step": 6110}, {"loss": 1.0957, "grad_norm": 0.9360884428024292, "learning_rate": 0.0002, "epoch": 4.429967426710098, "step": 6120}, {"loss": 0.9506, "grad_norm": 0.8099448084831238, "learning_rate": 0.0002, "epoch": 4.437205935577271, "step": 6130}, {"loss": 1.0887, "grad_norm": 0.8173841238021851, "learning_rate": 0.0002, "epoch": 4.444444444444445, "step": 6140}, {"loss": 1.1219, "grad_norm": 0.7972666025161743, "learning_rate": 0.0002, "epoch": 4.451682953311618, "step": 6150}, {"loss": 1.0226, "grad_norm": 0.7685779333114624, "learning_rate": 0.0002, "epoch": 4.458921462178791, "step": 6160}, {"loss": 1.0732, "grad_norm": 0.7872623801231384, "learning_rate": 0.0002, "epoch": 4.466159971045965, "step": 6170}, {"loss": 0.9911, "grad_norm": 0.7677070498466492, "learning_rate": 0.0002, "epoch": 4.473398479913138, "step": 6180}, {"loss": 1.0919, "grad_norm": 0.7878316044807434, "learning_rate": 0.0002, "epoch": 4.4806369887803115, "step": 6190}, {"loss": 1.018, "grad_norm": 0.8178079724311829, "learning_rate": 0.0002, "epoch": 4.487875497647485, "step": 6200}, {"loss": 1.0517, "grad_norm": 1.2820082902908325, "learning_rate": 0.0002, "epoch": 4.495114006514658, "step": 6210}, {"loss": 1.3101, "grad_norm": 0.9380832314491272, "learning_rate": 0.0002, "epoch": 4.502352515381832, "step": 6220}, {"loss": 0.9818, "grad_norm": 0.7810422778129578, "learning_rate": 0.0002, "epoch": 4.509591024249005, "step": 6230}, {"loss": 1.1677, "grad_norm": 1.1022917032241821, "learning_rate": 0.0002, "epoch": 4.516829533116178, "step": 6240}, {"loss": 1.1579, "grad_norm": 1.4275553226470947, "learning_rate": 0.0002, "epoch": 4.524068041983352, "step": 6250}, {"loss": 1.3237, "grad_norm": 0.7597777247428894, "learning_rate": 0.0002, "epoch": 4.531306550850525, "step": 6260}, {"loss": 1.1529, "grad_norm": 1.10992431640625, "learning_rate": 0.0002, "epoch": 4.538545059717698, "step": 6270}, {"loss": 1.0732, "grad_norm": 0.8981178998947144, "learning_rate": 0.0002, "epoch": 4.545783568584872, "step": 6280}, {"loss": 1.086, "grad_norm": 0.7863979339599609, "learning_rate": 0.0002, "epoch": 4.553022077452045, "step": 6290}, {"loss": 1.2008, "grad_norm": 0.9071474671363831, "learning_rate": 0.0002, "epoch": 4.5602605863192185, "step": 6300}, {"loss": 1.0916, "grad_norm": 0.7429424524307251, "learning_rate": 0.0002, "epoch": 4.567499095186392, "step": 6310}, {"loss": 1.095, "grad_norm": 1.0767850875854492, "learning_rate": 0.0002, "epoch": 4.574737604053565, "step": 6320}, {"loss": 1.1023, "grad_norm": 0.7885915637016296, "learning_rate": 0.0002, "epoch": 4.581976112920739, "step": 6330}, {"loss": 1.1131, "grad_norm": 0.8350457549095154, "learning_rate": 0.0002, "epoch": 4.589214621787912, "step": 6340}, {"loss": 1.0743, "grad_norm": 0.7853530645370483, "learning_rate": 0.0002, "epoch": 4.596453130655085, "step": 6350}, {"loss": 1.1912, "grad_norm": 1.1220661401748657, "learning_rate": 0.0002, "epoch": 4.603691639522259, "step": 6360}, {"loss": 1.0927, "grad_norm": 0.7959423065185547, "learning_rate": 0.0002, "epoch": 4.610930148389432, "step": 6370}, {"loss": 1.1542, "grad_norm": 0.7782652378082275, "learning_rate": 0.0002, "epoch": 4.618168657256605, "step": 6380}, {"loss": 1.0753, "grad_norm": 0.7882203459739685, "learning_rate": 0.0002, "epoch": 4.625407166123779, "step": 6390}, {"loss": 1.0676, "grad_norm": 0.8841899037361145, "learning_rate": 0.0002, "epoch": 4.632645674990952, "step": 6400}, {"loss": 1.0815, "grad_norm": 0.7936127781867981, "learning_rate": 0.0002, "epoch": 4.6398841838581255, "step": 6410}, {"loss": 1.0198, "grad_norm": 0.9213966131210327, "learning_rate": 0.0002, "epoch": 4.647122692725299, "step": 6420}, {"loss": 0.9872, "grad_norm": 0.9246473908424377, "learning_rate": 0.0002, "epoch": 4.654361201592472, "step": 6430}, {"loss": 1.1309, "grad_norm": 0.766572892665863, "learning_rate": 0.0002, "epoch": 4.661599710459646, "step": 6440}, {"loss": 1.1095, "grad_norm": 0.8596171736717224, "learning_rate": 0.0002, "epoch": 4.668838219326819, "step": 6450}, {"loss": 1.1869, "grad_norm": 0.8482751846313477, "learning_rate": 0.0002, "epoch": 4.676076728193992, "step": 6460}, {"loss": 1.0622, "grad_norm": 1.0826905965805054, "learning_rate": 0.0002, "epoch": 4.683315237061166, "step": 6470}, {"loss": 1.0256, "grad_norm": 1.1048457622528076, "learning_rate": 0.0002, "epoch": 4.690553745928339, "step": 6480}, {"loss": 1.0514, "grad_norm": 0.9429134726524353, "learning_rate": 0.0002, "epoch": 4.697792254795512, "step": 6490}, {"loss": 1.1351, "grad_norm": 0.8587502837181091, "learning_rate": 0.0002, "epoch": 4.705030763662686, "step": 6500}, {"loss": 1.0969, "grad_norm": 1.0387083292007446, "learning_rate": 0.0002, "epoch": 4.712269272529859, "step": 6510}, {"loss": 1.0493, "grad_norm": 0.7471951842308044, "learning_rate": 0.0002, "epoch": 4.7195077813970325, "step": 6520}, {"loss": 1.2632, "grad_norm": 0.8800424933433533, "learning_rate": 0.0002, "epoch": 4.726746290264206, "step": 6530}, {"loss": 1.2126, "grad_norm": 0.8136811852455139, "learning_rate": 0.0002, "epoch": 4.733984799131379, "step": 6540}, {"loss": 1.195, "grad_norm": 0.9910339713096619, "learning_rate": 0.0002, "epoch": 4.741223307998553, "step": 6550}, {"loss": 1.1201, "grad_norm": 1.0679163932800293, "learning_rate": 0.0002, "epoch": 4.748461816865726, "step": 6560}, {"loss": 1.0297, "grad_norm": 0.8468248248100281, "learning_rate": 0.0002, "epoch": 4.755700325732899, "step": 6570}, {"loss": 1.0858, "grad_norm": 0.8771235942840576, "learning_rate": 0.0002, "epoch": 4.762938834600073, "step": 6580}, {"loss": 1.077, "grad_norm": 0.7024846076965332, "learning_rate": 0.0002, "epoch": 4.770177343467246, "step": 6590}, {"loss": 1.0876, "grad_norm": 0.7836683392524719, "learning_rate": 0.0002, "epoch": 4.7774158523344195, "step": 6600}, {"loss": 1.1006, "grad_norm": 0.7717288136482239, "learning_rate": 0.0002, "epoch": 4.784654361201593, "step": 6610}, {"loss": 1.0376, "grad_norm": 0.884183943271637, "learning_rate": 0.0002, "epoch": 4.791892870068766, "step": 6620}, {"loss": 1.1757, "grad_norm": 1.383867621421814, "learning_rate": 0.0002, "epoch": 4.7991313789359396, "step": 6630}, {"loss": 1.0861, "grad_norm": 0.9741523861885071, "learning_rate": 0.0002, "epoch": 4.806369887803113, "step": 6640}, {"loss": 1.0884, "grad_norm": 0.9723693132400513, "learning_rate": 0.0002, "epoch": 4.813608396670286, "step": 6650}, {"loss": 1.2203, "grad_norm": 1.8324809074401855, "learning_rate": 0.0002, "epoch": 4.82084690553746, "step": 6660}, {"loss": 1.0292, "grad_norm": 0.904909074306488, "learning_rate": 0.0002, "epoch": 4.828085414404633, "step": 6670}, {"loss": 1.0349, "grad_norm": 0.7355411648750305, "learning_rate": 0.0002, "epoch": 4.835323923271806, "step": 6680}, {"loss": 1.0793, "grad_norm": 0.8934960961341858, "learning_rate": 0.0002, "epoch": 4.84256243213898, "step": 6690}, {"loss": 1.0375, "grad_norm": 1.4596954584121704, "learning_rate": 0.0002, "epoch": 4.849800941006153, "step": 6700}, {"loss": 1.1065, "grad_norm": 0.8310341238975525, "learning_rate": 0.0002, "epoch": 4.8570394498733265, "step": 6710}, {"loss": 1.1089, "grad_norm": 0.9709894061088562, "learning_rate": 0.0002, "epoch": 4.8642779587405, "step": 6720}, {"loss": 1.0069, "grad_norm": 0.852142333984375, "learning_rate": 0.0002, "epoch": 4.871516467607673, "step": 6730}, {"loss": 1.0507, "grad_norm": 1.0643625259399414, "learning_rate": 0.0002, "epoch": 4.878754976474847, "step": 6740}, {"loss": 1.056, "grad_norm": 0.9419508576393127, "learning_rate": 0.0002, "epoch": 4.88599348534202, "step": 6750}, {"loss": 1.1995, "grad_norm": 1.1818498373031616, "learning_rate": 0.0002, "epoch": 4.893231994209193, "step": 6760}, {"loss": 1.0925, "grad_norm": 0.9369569420814514, "learning_rate": 0.0002, "epoch": 4.900470503076367, "step": 6770}, {"loss": 1.1648, "grad_norm": 0.7012579441070557, "learning_rate": 0.0002, "epoch": 4.90770901194354, "step": 6780}, {"loss": 1.0926, "grad_norm": 0.9109319448471069, "learning_rate": 0.0002, "epoch": 4.914947520810713, "step": 6790}, {"loss": 1.0358, "grad_norm": 0.8077534437179565, "learning_rate": 0.0002, "epoch": 4.922186029677887, "step": 6800}, {"loss": 1.2549, "grad_norm": 0.7571148872375488, "learning_rate": 0.0002, "epoch": 4.92942453854506, "step": 6810}, {"loss": 0.9638, "grad_norm": 0.7325633764266968, "learning_rate": 0.0002, "epoch": 4.9366630474122335, "step": 6820}, {"loss": 1.0128, "grad_norm": 0.8465084433555603, "learning_rate": 0.0002, "epoch": 4.943901556279407, "step": 6830}, {"loss": 1.153, "grad_norm": 0.8753737807273865, "learning_rate": 0.0002, "epoch": 4.95114006514658, "step": 6840}, {"loss": 1.0247, "grad_norm": 0.9421748518943787, "learning_rate": 0.0002, "epoch": 4.958378574013754, "step": 6850}, {"loss": 1.1483, "grad_norm": 0.8245896697044373, "learning_rate": 0.0002, "epoch": 4.965617082880927, "step": 6860}, {"loss": 0.9905, "grad_norm": 0.8823089599609375, "learning_rate": 0.0002, "epoch": 4.9728555917481, "step": 6870}, {"loss": 1.1664, "grad_norm": 0.8406389355659485, "learning_rate": 0.0002, "epoch": 4.980094100615274, "step": 6880}, {"loss": 1.0944, "grad_norm": 0.9732868075370789, "learning_rate": 0.0002, "epoch": 4.987332609482447, "step": 6890}, {"loss": 1.1776, "grad_norm": 2.125141143798828, "learning_rate": 0.0002, "epoch": 4.99457111834962, "step": 6900}, {"eval_loss": 1.445176601409912, "eval_runtime": 27.2351, "eval_samples_per_second": 16.009, "eval_steps_per_second": 2.019, "epoch": 4.999638074556641, "step": 6907}, {"loss": 1.1362, "grad_norm": 0.9465792775154114, "learning_rate": 0.0002, "epoch": 5.001809627216793, "step": 6910}, {"loss": 0.982, "grad_norm": 1.2834891080856323, "learning_rate": 0.0002, "epoch": 5.009048136083966, "step": 6920}, {"loss": 0.9803, "grad_norm": 1.0297378301620483, "learning_rate": 0.0002, "epoch": 5.01628664495114, "step": 6930}, {"loss": 1.0447, "grad_norm": 1.1705161333084106, "learning_rate": 0.0002, "epoch": 5.023525153818313, "step": 6940}, {"loss": 1.0113, "grad_norm": 0.8293961882591248, "learning_rate": 0.0002, "epoch": 5.030763662685486, "step": 6950}, {"loss": 0.9203, "grad_norm": 1.0422210693359375, "learning_rate": 0.0002, "epoch": 5.03800217155266, "step": 6960}, {"loss": 1.0553, "grad_norm": 1.116104245185852, "learning_rate": 0.0002, "epoch": 5.045240680419833, "step": 6970}, {"loss": 0.9011, "grad_norm": 1.5118416547775269, "learning_rate": 0.0002, "epoch": 5.0524791892870065, "step": 6980}, {"loss": 0.9969, "grad_norm": 0.8383979797363281, "learning_rate": 0.0002, "epoch": 5.05971769815418, "step": 6990}, {"loss": 0.9659, "grad_norm": 1.3378649950027466, "learning_rate": 0.0002, "epoch": 5.066956207021353, "step": 7000}, {"loss": 1.0212, "grad_norm": 1.1840510368347168, "learning_rate": 0.0002, "epoch": 5.0741947158885266, "step": 7010}, {"loss": 0.9939, "grad_norm": 1.2354751825332642, "learning_rate": 0.0002, "epoch": 5.0814332247557, "step": 7020}, {"loss": 0.9831, "grad_norm": 1.3830451965332031, "learning_rate": 0.0002, "epoch": 5.088671733622873, "step": 7030}, {"loss": 1.1827, "grad_norm": 0.8101674318313599, "learning_rate": 0.0002, "epoch": 5.095910242490047, "step": 7040}, {"loss": 0.9255, "grad_norm": 0.897982656955719, "learning_rate": 0.0002, "epoch": 5.10314875135722, "step": 7050}, {"loss": 0.8784, "grad_norm": 1.2049678564071655, "learning_rate": 0.0002, "epoch": 5.110387260224393, "step": 7060}, {"loss": 1.0182, "grad_norm": 1.5912116765975952, "learning_rate": 0.0002, "epoch": 5.117625769091567, "step": 7070}, {"loss": 1.0909, "grad_norm": 0.9261530041694641, "learning_rate": 0.0002, "epoch": 5.12486427795874, "step": 7080}, {"loss": 0.9603, "grad_norm": 1.1454812288284302, "learning_rate": 0.0002, "epoch": 5.1321027868259135, "step": 7090}, {"loss": 0.9149, "grad_norm": 1.0049978494644165, "learning_rate": 0.0002, "epoch": 5.139341295693087, "step": 7100}, {"loss": 0.9463, "grad_norm": 1.4513251781463623, "learning_rate": 0.0002, "epoch": 5.14657980456026, "step": 7110}, {"loss": 0.8995, "grad_norm": 0.9800849556922913, "learning_rate": 0.0002, "epoch": 5.153818313427434, "step": 7120}, {"loss": 0.9835, "grad_norm": 0.9698708653450012, "learning_rate": 0.0002, "epoch": 5.161056822294607, "step": 7130}, {"loss": 0.9672, "grad_norm": 1.1126646995544434, "learning_rate": 0.0002, "epoch": 5.16829533116178, "step": 7140}, {"loss": 0.9384, "grad_norm": 0.9248330593109131, "learning_rate": 0.0002, "epoch": 5.175533840028954, "step": 7150}, {"loss": 0.826, "grad_norm": 0.7967255711555481, "learning_rate": 0.0002, "epoch": 5.182772348896127, "step": 7160}, {"loss": 1.0078, "grad_norm": 0.9933333992958069, "learning_rate": 0.0002, "epoch": 5.1900108577633, "step": 7170}, {"loss": 1.0276, "grad_norm": 1.0080649852752686, "learning_rate": 0.0002, "epoch": 5.197249366630474, "step": 7180}, {"loss": 1.0201, "grad_norm": 1.3954921960830688, "learning_rate": 0.0002, "epoch": 5.204487875497647, "step": 7190}, {"loss": 1.0863, "grad_norm": 1.2386271953582764, "learning_rate": 0.0002, "epoch": 5.2117263843648205, "step": 7200}, {"loss": 0.8863, "grad_norm": 1.2379488945007324, "learning_rate": 0.0002, "epoch": 5.218964893231994, "step": 7210}, {"loss": 1.0518, "grad_norm": 0.9882503747940063, "learning_rate": 0.0002, "epoch": 5.226203402099167, "step": 7220}, {"loss": 0.9834, "grad_norm": 1.1728729009628296, "learning_rate": 0.0002, "epoch": 5.233441910966341, "step": 7230}, {"loss": 0.9269, "grad_norm": 0.9849673509597778, "learning_rate": 0.0002, "epoch": 5.240680419833514, "step": 7240}, {"loss": 0.9935, "grad_norm": 1.177639365196228, "learning_rate": 0.0002, "epoch": 5.247918928700687, "step": 7250}, {"loss": 1.0639, "grad_norm": 1.2395055294036865, "learning_rate": 0.0002, "epoch": 5.255157437567861, "step": 7260}, {"loss": 1.0138, "grad_norm": 1.3999171257019043, "learning_rate": 0.0002, "epoch": 5.262395946435034, "step": 7270}, {"loss": 0.9745, "grad_norm": 0.7698732018470764, "learning_rate": 0.0002, "epoch": 5.269634455302207, "step": 7280}, {"loss": 1.0389, "grad_norm": 0.9167453646659851, "learning_rate": 0.0002, "epoch": 5.276872964169381, "step": 7290}, {"loss": 0.9858, "grad_norm": 1.113830804824829, "learning_rate": 0.0002, "epoch": 5.284111473036554, "step": 7300}, {"loss": 0.9577, "grad_norm": 0.9644396901130676, "learning_rate": 0.0002, "epoch": 5.2913499819037275, "step": 7310}, {"loss": 1.0556, "grad_norm": 1.462435007095337, "learning_rate": 0.0002, "epoch": 5.298588490770901, "step": 7320}, {"loss": 0.871, "grad_norm": 0.9406287670135498, "learning_rate": 0.0002, "epoch": 5.305826999638074, "step": 7330}, {"loss": 1.0022, "grad_norm": 0.9698247909545898, "learning_rate": 0.0002, "epoch": 5.313065508505248, "step": 7340}, {"loss": 0.915, "grad_norm": 1.12003755569458, "learning_rate": 0.0002, "epoch": 5.320304017372421, "step": 7350}, {"loss": 0.9838, "grad_norm": 1.598681926727295, "learning_rate": 0.0002, "epoch": 5.327542526239594, "step": 7360}, {"loss": 1.0, "grad_norm": 1.0450010299682617, "learning_rate": 0.0002, "epoch": 5.334781035106768, "step": 7370}, {"loss": 0.9983, "grad_norm": 0.8680008053779602, "learning_rate": 0.0002, "epoch": 5.342019543973941, "step": 7380}, {"loss": 0.9851, "grad_norm": 1.0115476846694946, "learning_rate": 0.0002, "epoch": 5.349258052841114, "step": 7390}, {"loss": 1.0702, "grad_norm": 0.9589748382568359, "learning_rate": 0.0002, "epoch": 5.356496561708288, "step": 7400}, {"loss": 0.9366, "grad_norm": 0.6729998588562012, "learning_rate": 0.0002, "epoch": 5.363735070575461, "step": 7410}, {"loss": 1.0126, "grad_norm": 0.9246699213981628, "learning_rate": 0.0002, "epoch": 5.3709735794426345, "step": 7420}, {"loss": 0.9815, "grad_norm": 1.1266791820526123, "learning_rate": 0.0002, "epoch": 5.378212088309808, "step": 7430}, {"loss": 1.1166, "grad_norm": 1.8056942224502563, "learning_rate": 0.0002, "epoch": 5.385450597176981, "step": 7440}, {"loss": 0.9604, "grad_norm": 0.9802932739257812, "learning_rate": 0.0002, "epoch": 5.392689106044155, "step": 7450}, {"loss": 0.9656, "grad_norm": 1.0504707098007202, "learning_rate": 0.0002, "epoch": 5.399927614911328, "step": 7460}, {"loss": 1.0132, "grad_norm": 1.1915022134780884, "learning_rate": 0.0002, "epoch": 5.407166123778501, "step": 7470}, {"loss": 1.0041, "grad_norm": 1.1856611967086792, "learning_rate": 0.0002, "epoch": 5.414404632645675, "step": 7480}, {"loss": 0.9747, "grad_norm": 1.292152762413025, "learning_rate": 0.0002, "epoch": 5.421643141512848, "step": 7490}, {"loss": 0.9659, "grad_norm": 1.2675740718841553, "learning_rate": 0.0002, "epoch": 5.4288816503800215, "step": 7500}, {"loss": 1.0271, "grad_norm": 1.4034695625305176, "learning_rate": 0.0002, "epoch": 5.436120159247195, "step": 7510}, {"loss": 1.0318, "grad_norm": 0.984588623046875, "learning_rate": 0.0002, "epoch": 5.443358668114368, "step": 7520}, {"loss": 1.0726, "grad_norm": 0.8419108390808105, "learning_rate": 0.0002, "epoch": 5.450597176981542, "step": 7530}, {"loss": 1.0499, "grad_norm": 1.0270143747329712, "learning_rate": 0.0002, "epoch": 5.457835685848715, "step": 7540}, {"loss": 0.9804, "grad_norm": 2.2158689498901367, "learning_rate": 0.0002, "epoch": 5.465074194715888, "step": 7550}, {"loss": 0.9856, "grad_norm": 1.0740524530410767, "learning_rate": 0.0002, "epoch": 5.472312703583062, "step": 7560}, {"loss": 1.0522, "grad_norm": 1.3804482221603394, "learning_rate": 0.0002, "epoch": 5.479551212450235, "step": 7570}, {"loss": 1.0297, "grad_norm": 0.9428979754447937, "learning_rate": 0.0002, "epoch": 5.486789721317408, "step": 7580}, {"loss": 1.0906, "grad_norm": 0.9548295736312866, "learning_rate": 0.0002, "epoch": 5.494028230184582, "step": 7590}, {"loss": 0.8853, "grad_norm": 1.0691065788269043, "learning_rate": 0.0002, "epoch": 5.501266739051755, "step": 7600}, {"loss": 1.0375, "grad_norm": 1.0987380743026733, "learning_rate": 0.0002, "epoch": 5.5085052479189285, "step": 7610}, {"loss": 1.0162, "grad_norm": 0.9483979344367981, "learning_rate": 0.0002, "epoch": 5.515743756786102, "step": 7620}, {"loss": 1.105, "grad_norm": 1.16624915599823, "learning_rate": 0.0002, "epoch": 5.522982265653275, "step": 7630}, {"loss": 0.8695, "grad_norm": 0.8563777208328247, "learning_rate": 0.0002, "epoch": 5.530220774520449, "step": 7640}, {"loss": 0.9297, "grad_norm": 1.268186092376709, "learning_rate": 0.0002, "epoch": 5.537459283387622, "step": 7650}, {"loss": 1.1152, "grad_norm": 1.0752092599868774, "learning_rate": 0.0002, "epoch": 5.544697792254795, "step": 7660}, {"loss": 0.9344, "grad_norm": 1.210389256477356, "learning_rate": 0.0002, "epoch": 5.551936301121969, "step": 7670}, {"loss": 1.0349, "grad_norm": 1.669063925743103, "learning_rate": 0.0002, "epoch": 5.559174809989142, "step": 7680}, {"loss": 0.9833, "grad_norm": 1.038020133972168, "learning_rate": 0.0002, "epoch": 5.566413318856315, "step": 7690}, {"loss": 0.8907, "grad_norm": 1.316673994064331, "learning_rate": 0.0002, "epoch": 5.573651827723489, "step": 7700}, {"loss": 0.9614, "grad_norm": 1.029935359954834, "learning_rate": 0.0002, "epoch": 5.580890336590662, "step": 7710}, {"loss": 1.0409, "grad_norm": 0.9401940703392029, "learning_rate": 0.0002, "epoch": 5.5881288454578355, "step": 7720}, {"loss": 0.9272, "grad_norm": 2.4811816215515137, "learning_rate": 0.0002, "epoch": 5.595367354325009, "step": 7730}, {"loss": 0.992, "grad_norm": 1.0329105854034424, "learning_rate": 0.0002, "epoch": 5.602605863192182, "step": 7740}, {"loss": 0.9493, "grad_norm": 1.479629635810852, "learning_rate": 0.0002, "epoch": 5.609844372059356, "step": 7750}, {"loss": 1.0727, "grad_norm": 1.9232319593429565, "learning_rate": 0.0002, "epoch": 5.617082880926529, "step": 7760}, {"loss": 1.0741, "grad_norm": 1.0055509805679321, "learning_rate": 0.0002, "epoch": 5.624321389793702, "step": 7770}, {"loss": 1.0731, "grad_norm": 1.0037437677383423, "learning_rate": 0.0002, "epoch": 5.631559898660876, "step": 7780}, {"loss": 1.0913, "grad_norm": 1.4245030879974365, "learning_rate": 0.0002, "epoch": 5.638798407528049, "step": 7790}, {"loss": 0.9711, "grad_norm": 1.080687403678894, "learning_rate": 0.0002, "epoch": 5.646036916395222, "step": 7800}, {"loss": 1.0276, "grad_norm": 1.354953408241272, "learning_rate": 0.0002, "epoch": 5.653275425262396, "step": 7810}, {"loss": 1.0534, "grad_norm": 0.8966761231422424, "learning_rate": 0.0002, "epoch": 5.660513934129569, "step": 7820}, {"loss": 1.0662, "grad_norm": 1.0675480365753174, "learning_rate": 0.0002, "epoch": 5.6677524429967425, "step": 7830}, {"loss": 1.1077, "grad_norm": 1.2104216814041138, "learning_rate": 0.0002, "epoch": 5.674990951863916, "step": 7840}, {"loss": 0.9627, "grad_norm": 1.105790376663208, "learning_rate": 0.0002, "epoch": 5.682229460731089, "step": 7850}, {"loss": 1.0483, "grad_norm": 1.0915391445159912, "learning_rate": 0.0002, "epoch": 5.689467969598263, "step": 7860}, {"loss": 1.0291, "grad_norm": 0.8957812786102295, "learning_rate": 0.0002, "epoch": 5.696706478465436, "step": 7870}, {"loss": 0.9785, "grad_norm": 1.9189311265945435, "learning_rate": 0.0002, "epoch": 5.703944987332609, "step": 7880}, {"loss": 1.0076, "grad_norm": 1.0867321491241455, "learning_rate": 0.0002, "epoch": 5.711183496199783, "step": 7890}, {"loss": 1.0236, "grad_norm": 1.0233147144317627, "learning_rate": 0.0002, "epoch": 5.718422005066956, "step": 7900}, {"loss": 0.9872, "grad_norm": 1.16460382938385, "learning_rate": 0.0002, "epoch": 5.7256605139341294, "step": 7910}, {"loss": 1.0762, "grad_norm": 1.1098358631134033, "learning_rate": 0.0002, "epoch": 5.732899022801303, "step": 7920}, {"loss": 0.9937, "grad_norm": 0.8555701375007629, "learning_rate": 0.0002, "epoch": 5.740137531668476, "step": 7930}, {"loss": 1.0081, "grad_norm": 0.9885705709457397, "learning_rate": 0.0002, "epoch": 5.7473760405356495, "step": 7940}, {"loss": 0.9909, "grad_norm": 0.9184203147888184, "learning_rate": 0.0002, "epoch": 5.754614549402823, "step": 7950}, {"loss": 1.0767, "grad_norm": 0.9653698205947876, "learning_rate": 0.0002, "epoch": 5.761853058269996, "step": 7960}, {"loss": 0.9317, "grad_norm": 1.0014251470565796, "learning_rate": 0.0002, "epoch": 5.76909156713717, "step": 7970}, {"loss": 1.0271, "grad_norm": 1.004701018333435, "learning_rate": 0.0002, "epoch": 5.776330076004343, "step": 7980}, {"loss": 1.0397, "grad_norm": 0.950577974319458, "learning_rate": 0.0002, "epoch": 5.783568584871516, "step": 7990}, {"loss": 0.9725, "grad_norm": 1.2986834049224854, "learning_rate": 0.0002, "epoch": 5.79080709373869, "step": 8000}, {"loss": 1.039, "grad_norm": 1.3353424072265625, "learning_rate": 0.0002, "epoch": 5.798045602605863, "step": 8010}, {"loss": 1.0626, "grad_norm": 0.7650562524795532, "learning_rate": 0.0002, "epoch": 5.8052841114730365, "step": 8020}, {"loss": 1.0802, "grad_norm": 1.0156235694885254, "learning_rate": 0.0002, "epoch": 5.81252262034021, "step": 8030}, {"loss": 1.0185, "grad_norm": 1.3092900514602661, "learning_rate": 0.0002, "epoch": 5.819761129207383, "step": 8040}, {"loss": 0.9905, "grad_norm": 1.184428095817566, "learning_rate": 0.0002, "epoch": 5.826999638074557, "step": 8050}, {"loss": 1.0548, "grad_norm": 0.979401707649231, "learning_rate": 0.0002, "epoch": 5.83423814694173, "step": 8060}, {"loss": 0.9721, "grad_norm": 1.3557400703430176, "learning_rate": 0.0002, "epoch": 5.841476655808903, "step": 8070}, {"loss": 1.0235, "grad_norm": 0.8429333567619324, "learning_rate": 0.0002, "epoch": 5.848715164676077, "step": 8080}, {"loss": 0.952, "grad_norm": 1.3167692422866821, "learning_rate": 0.0002, "epoch": 5.85595367354325, "step": 8090}, {"loss": 0.9609, "grad_norm": 0.9750998020172119, "learning_rate": 0.0002, "epoch": 5.863192182410423, "step": 8100}, {"loss": 1.0789, "grad_norm": 1.1869813203811646, "learning_rate": 0.0002, "epoch": 5.870430691277597, "step": 8110}, {"loss": 1.0331, "grad_norm": 1.508615255355835, "learning_rate": 0.0002, "epoch": 5.87766920014477, "step": 8120}, {"loss": 1.0171, "grad_norm": 0.9439908266067505, "learning_rate": 0.0002, "epoch": 5.8849077090119435, "step": 8130}, {"loss": 0.9682, "grad_norm": 0.910508930683136, "learning_rate": 0.0002, "epoch": 5.892146217879117, "step": 8140}, {"loss": 1.0032, "grad_norm": 1.111501932144165, "learning_rate": 0.0002, "epoch": 5.89938472674629, "step": 8150}, {"loss": 1.0266, "grad_norm": 0.726554274559021, "learning_rate": 0.0002, "epoch": 5.906623235613464, "step": 8160}, {"loss": 1.0681, "grad_norm": 1.1084556579589844, "learning_rate": 0.0002, "epoch": 5.913861744480637, "step": 8170}, {"loss": 0.969, "grad_norm": 0.9695167541503906, "learning_rate": 0.0002, "epoch": 5.92110025334781, "step": 8180}, {"loss": 0.9858, "grad_norm": 1.1169592142105103, "learning_rate": 0.0002, "epoch": 5.928338762214984, "step": 8190}, {"loss": 1.0924, "grad_norm": 1.5116780996322632, "learning_rate": 0.0002, "epoch": 5.935577271082157, "step": 8200}, {"loss": 0.878, "grad_norm": 1.0073388814926147, "learning_rate": 0.0002, "epoch": 5.94281577994933, "step": 8210}, {"loss": 1.0462, "grad_norm": 0.9323263168334961, "learning_rate": 0.0002, "epoch": 5.950054288816504, "step": 8220}, {"loss": 1.0291, "grad_norm": 0.9422887563705444, "learning_rate": 0.0002, "epoch": 5.957292797683677, "step": 8230}, {"loss": 0.953, "grad_norm": 0.9691047668457031, "learning_rate": 0.0002, "epoch": 5.9645313065508505, "step": 8240}, {"loss": 0.9842, "grad_norm": 0.9650622606277466, "learning_rate": 0.0002, "epoch": 5.971769815418024, "step": 8250}, {"loss": 0.907, "grad_norm": 1.077958345413208, "learning_rate": 0.0002, "epoch": 5.979008324285197, "step": 8260}, {"loss": 0.9162, "grad_norm": 0.8946306109428406, "learning_rate": 0.0002, "epoch": 5.986246833152371, "step": 8270}, {"loss": 1.0439, "grad_norm": 1.34098219871521, "learning_rate": 0.0002, "epoch": 5.993485342019544, "step": 8280}, {"eval_loss": 1.4714229106903076, "eval_runtime": 26.301, "eval_samples_per_second": 16.577, "eval_steps_per_second": 2.091, "epoch": 6.0, "step": 8289}, {"loss": 1.1403, "grad_norm": 0.9737564325332642, "learning_rate": 0.0002, "epoch": 6.000723850886717, "step": 8290}, {"loss": 0.8875, "grad_norm": 1.2205945253372192, "learning_rate": 0.0002, "epoch": 6.007962359753891, "step": 8300}, {"loss": 0.8623, "grad_norm": 1.3529434204101562, "learning_rate": 0.0002, "epoch": 6.015200868621064, "step": 8310}, {"loss": 0.9427, "grad_norm": 1.2300174236297607, "learning_rate": 0.0002, "epoch": 6.022439377488237, "step": 8320}, {"loss": 0.9322, "grad_norm": 0.9248194098472595, "learning_rate": 0.0002, "epoch": 6.029677886355411, "step": 8330}, {"loss": 0.9302, "grad_norm": 1.1140035390853882, "learning_rate": 0.0002, "epoch": 6.036916395222584, "step": 8340}, {"loss": 0.8255, "grad_norm": 1.2097352743148804, "learning_rate": 0.0002, "epoch": 6.0441549040897575, "step": 8350}, {"loss": 0.8792, "grad_norm": 0.9472483396530151, "learning_rate": 0.0002, "epoch": 6.051393412956931, "step": 8360}, {"loss": 0.8865, "grad_norm": 1.0195368528366089, "learning_rate": 0.0002, "epoch": 6.058631921824104, "step": 8370}, {"loss": 0.8858, "grad_norm": 1.182735562324524, "learning_rate": 0.0002, "epoch": 6.065870430691278, "step": 8380}, {"loss": 0.9455, "grad_norm": 1.1042858362197876, "learning_rate": 0.0002, "epoch": 6.073108939558451, "step": 8390}, {"loss": 0.9723, "grad_norm": 0.8606401085853577, "learning_rate": 0.0002, "epoch": 6.080347448425624, "step": 8400}, {"loss": 0.8436, "grad_norm": 1.1015676259994507, "learning_rate": 0.0002, "epoch": 6.087585957292798, "step": 8410}, {"loss": 0.8845, "grad_norm": 1.690224289894104, "learning_rate": 0.0002, "epoch": 6.094824466159971, "step": 8420}, {"loss": 0.8484, "grad_norm": 1.1928749084472656, "learning_rate": 0.0002, "epoch": 6.1020629750271445, "step": 8430}, {"loss": 0.9546, "grad_norm": 1.0816864967346191, "learning_rate": 0.0002, "epoch": 6.109301483894318, "step": 8440}, {"loss": 0.8286, "grad_norm": 1.1638226509094238, "learning_rate": 0.0002, "epoch": 6.116539992761491, "step": 8450}, {"loss": 0.8749, "grad_norm": 1.3782968521118164, "learning_rate": 0.0002, "epoch": 6.1237785016286646, "step": 8460}, {"loss": 0.7956, "grad_norm": 1.2030094861984253, "learning_rate": 0.0002, "epoch": 6.131017010495838, "step": 8470}, {"loss": 0.8393, "grad_norm": 1.3227659463882446, "learning_rate": 0.0002, "epoch": 6.138255519363011, "step": 8480}, {"loss": 0.9175, "grad_norm": 1.104384422302246, "learning_rate": 0.0002, "epoch": 6.145494028230185, "step": 8490}, {"loss": 0.861, "grad_norm": 1.518805980682373, "learning_rate": 0.0002, "epoch": 6.152732537097358, "step": 8500}, {"loss": 0.9169, "grad_norm": 1.2029093503952026, "learning_rate": 0.0002, "epoch": 6.159971045964531, "step": 8510}, {"loss": 0.8701, "grad_norm": 1.2991217374801636, "learning_rate": 0.0002, "epoch": 6.167209554831705, "step": 8520}, {"loss": 0.9748, "grad_norm": 1.7002956867218018, "learning_rate": 0.0002, "epoch": 6.174448063698878, "step": 8530}, {"loss": 0.8881, "grad_norm": 1.6653581857681274, "learning_rate": 0.0002, "epoch": 6.1816865725660515, "step": 8540}, {"loss": 0.817, "grad_norm": 1.0493303537368774, "learning_rate": 0.0002, "epoch": 6.188925081433225, "step": 8550}, {"loss": 0.8726, "grad_norm": 1.539345622062683, "learning_rate": 0.0002, "epoch": 6.196163590300398, "step": 8560}, {"loss": 0.9452, "grad_norm": 1.2757070064544678, "learning_rate": 0.0002, "epoch": 6.203402099167572, "step": 8570}, {"loss": 0.8773, "grad_norm": 1.2416890859603882, "learning_rate": 0.0002, "epoch": 6.210640608034745, "step": 8580}, {"loss": 0.815, "grad_norm": 1.617621898651123, "learning_rate": 0.0002, "epoch": 6.217879116901918, "step": 8590}, {"loss": 0.9137, "grad_norm": 1.058962106704712, "learning_rate": 0.0002, "epoch": 6.225117625769092, "step": 8600}, {"loss": 0.8164, "grad_norm": 1.1489088535308838, "learning_rate": 0.0002, "epoch": 6.232356134636265, "step": 8610}, {"loss": 0.9476, "grad_norm": 0.9391577243804932, "learning_rate": 0.0002, "epoch": 6.239594643503438, "step": 8620}, {"loss": 0.932, "grad_norm": 1.363706111907959, "learning_rate": 0.0002, "epoch": 6.246833152370612, "step": 8630}, {"loss": 0.8917, "grad_norm": 0.779502809047699, "learning_rate": 0.0002, "epoch": 6.254071661237785, "step": 8640}, {"loss": 0.9196, "grad_norm": 2.000821590423584, "learning_rate": 0.0002, "epoch": 6.2613101701049585, "step": 8650}, {"loss": 0.9794, "grad_norm": 1.1521023511886597, "learning_rate": 0.0002, "epoch": 6.268548678972132, "step": 8660}, {"loss": 0.9147, "grad_norm": 1.3734570741653442, "learning_rate": 0.0002, "epoch": 6.275787187839305, "step": 8670}, {"loss": 0.795, "grad_norm": 0.9550670385360718, "learning_rate": 0.0002, "epoch": 6.283025696706479, "step": 8680}, {"loss": 0.9049, "grad_norm": 0.8937032222747803, "learning_rate": 0.0002, "epoch": 6.290264205573652, "step": 8690}, {"loss": 0.8526, "grad_norm": 1.3352779150009155, "learning_rate": 0.0002, "epoch": 6.297502714440825, "step": 8700}, {"loss": 0.8572, "grad_norm": 1.3057222366333008, "learning_rate": 0.0002, "epoch": 6.304741223307999, "step": 8710}, {"loss": 0.8825, "grad_norm": 0.9078314304351807, "learning_rate": 0.0002, "epoch": 6.311979732175172, "step": 8720}, {"loss": 0.8666, "grad_norm": 1.6663457155227661, "learning_rate": 0.0002, "epoch": 6.319218241042345, "step": 8730}, {"loss": 0.927, "grad_norm": 1.2043739557266235, "learning_rate": 0.0002, "epoch": 6.326456749909519, "step": 8740}, {"loss": 0.8014, "grad_norm": 0.9165967702865601, "learning_rate": 0.0002, "epoch": 6.333695258776692, "step": 8750}, {"loss": 0.9761, "grad_norm": 1.016452670097351, "learning_rate": 0.0002, "epoch": 6.3409337676438655, "step": 8760}, {"loss": 1.022, "grad_norm": 1.2209261655807495, "learning_rate": 0.0002, "epoch": 6.348172276511039, "step": 8770}, {"loss": 0.8012, "grad_norm": 1.3380663394927979, "learning_rate": 0.0002, "epoch": 6.355410785378212, "step": 8780}, {"loss": 0.9553, "grad_norm": 2.3311562538146973, "learning_rate": 0.0002, "epoch": 6.362649294245386, "step": 8790}, {"loss": 0.8676, "grad_norm": 1.0330604314804077, "learning_rate": 0.0002, "epoch": 6.369887803112559, "step": 8800}, {"loss": 0.98, "grad_norm": 0.9655511975288391, "learning_rate": 0.0002, "epoch": 6.377126311979732, "step": 8810}, {"loss": 1.0324, "grad_norm": 1.1065765619277954, "learning_rate": 0.0002, "epoch": 6.384364820846906, "step": 8820}, {"loss": 1.0078, "grad_norm": 1.2631285190582275, "learning_rate": 0.0002, "epoch": 6.391603329714079, "step": 8830}, {"loss": 0.8989, "grad_norm": 0.92459636926651, "learning_rate": 0.0002, "epoch": 6.398841838581252, "step": 8840}, {"loss": 0.8536, "grad_norm": 0.9982633590698242, "learning_rate": 0.0002, "epoch": 6.406080347448426, "step": 8850}, {"loss": 0.8949, "grad_norm": 1.0746768712997437, "learning_rate": 0.0002, "epoch": 6.413318856315599, "step": 8860}, {"loss": 0.8547, "grad_norm": 1.3024073839187622, "learning_rate": 0.0002, "epoch": 6.4205573651827725, "step": 8870}, {"loss": 0.9618, "grad_norm": 1.2764527797698975, "learning_rate": 0.0002, "epoch": 6.427795874049946, "step": 8880}, {"loss": 0.8905, "grad_norm": 0.8318809270858765, "learning_rate": 0.0002, "epoch": 6.435034382917119, "step": 8890}, {"loss": 0.917, "grad_norm": 1.7350783348083496, "learning_rate": 0.0002, "epoch": 6.442272891784293, "step": 8900}, {"loss": 1.0229, "grad_norm": 1.3430488109588623, "learning_rate": 0.0002, "epoch": 6.449511400651466, "step": 8910}, {"loss": 0.9678, "grad_norm": 1.5907495021820068, "learning_rate": 0.0002, "epoch": 6.456749909518639, "step": 8920}, {"loss": 0.9639, "grad_norm": 1.8579202890396118, "learning_rate": 0.0002, "epoch": 6.463988418385813, "step": 8930}, {"loss": 0.9302, "grad_norm": 1.2233413457870483, "learning_rate": 0.0002, "epoch": 6.471226927252986, "step": 8940}, {"loss": 0.9169, "grad_norm": 1.009103775024414, "learning_rate": 0.0002, "epoch": 6.4784654361201595, "step": 8950}, {"loss": 0.8969, "grad_norm": 1.1265181303024292, "learning_rate": 0.0002, "epoch": 6.485703944987333, "step": 8960}, {"loss": 0.8374, "grad_norm": 1.1733338832855225, "learning_rate": 0.0002, "epoch": 6.492942453854506, "step": 8970}, {"loss": 0.8764, "grad_norm": 1.0444518327713013, "learning_rate": 0.0002, "epoch": 6.50018096272168, "step": 8980}, {"loss": 0.9582, "grad_norm": 1.2296479940414429, "learning_rate": 0.0002, "epoch": 6.507419471588853, "step": 8990}, {"loss": 0.8557, "grad_norm": 1.370417833328247, "learning_rate": 0.0002, "epoch": 6.514657980456026, "step": 9000}, {"loss": 0.9787, "grad_norm": 1.4787620306015015, "learning_rate": 0.0002, "epoch": 6.5218964893232, "step": 9010}, {"loss": 0.967, "grad_norm": 0.8550514578819275, "learning_rate": 0.0002, "epoch": 6.529134998190373, "step": 9020}, {"loss": 0.9755, "grad_norm": 1.2327991724014282, "learning_rate": 0.0002, "epoch": 6.536373507057546, "step": 9030}, {"loss": 0.9248, "grad_norm": 1.0915621519088745, "learning_rate": 0.0002, "epoch": 6.54361201592472, "step": 9040}, {"loss": 1.0024, "grad_norm": 1.7243309020996094, "learning_rate": 0.0002, "epoch": 6.550850524791893, "step": 9050}, {"loss": 1.0123, "grad_norm": 0.954359769821167, "learning_rate": 0.0002, "epoch": 6.5580890336590665, "step": 9060}, {"loss": 0.8261, "grad_norm": 1.066051959991455, "learning_rate": 0.0002, "epoch": 6.56532754252624, "step": 9070}, {"loss": 0.944, "grad_norm": 1.200271487236023, "learning_rate": 0.0002, "epoch": 6.572566051393413, "step": 9080}, {"loss": 0.9788, "grad_norm": 1.4331457614898682, "learning_rate": 0.0002, "epoch": 6.579804560260587, "step": 9090}, {"loss": 1.0216, "grad_norm": 1.0892444849014282, "learning_rate": 0.0002, "epoch": 6.58704306912776, "step": 9100}, {"loss": 0.8557, "grad_norm": 1.849726915359497, "learning_rate": 0.0002, "epoch": 6.594281577994933, "step": 9110}, {"loss": 0.9495, "grad_norm": 1.1228708028793335, "learning_rate": 0.0002, "epoch": 6.601520086862107, "step": 9120}, {"loss": 1.0169, "grad_norm": 1.0928595066070557, "learning_rate": 0.0002, "epoch": 6.60875859572928, "step": 9130}, {"loss": 0.9342, "grad_norm": 1.2138155698776245, "learning_rate": 0.0002, "epoch": 6.615997104596453, "step": 9140}, {"loss": 0.8715, "grad_norm": 1.5155235528945923, "learning_rate": 0.0002, "epoch": 6.623235613463627, "step": 9150}, {"loss": 0.9806, "grad_norm": 1.3194212913513184, "learning_rate": 0.0002, "epoch": 6.6304741223308, "step": 9160}, {"loss": 0.8958, "grad_norm": 1.045623779296875, "learning_rate": 0.0002, "epoch": 6.6377126311979735, "step": 9170}, {"loss": 0.8698, "grad_norm": 0.9647570252418518, "learning_rate": 0.0002, "epoch": 6.644951140065147, "step": 9180}, {"loss": 0.8829, "grad_norm": 1.0818220376968384, "learning_rate": 0.0002, "epoch": 6.65218964893232, "step": 9190}, {"loss": 0.9745, "grad_norm": 1.2792822122573853, "learning_rate": 0.0002, "epoch": 6.659428157799494, "step": 9200}, {"loss": 0.8854, "grad_norm": 1.2764191627502441, "learning_rate": 0.0002, "epoch": 6.666666666666667, "step": 9210}, {"loss": 0.9709, "grad_norm": 1.0552066564559937, "learning_rate": 0.0002, "epoch": 6.67390517553384, "step": 9220}, {"loss": 0.8855, "grad_norm": 1.082476019859314, "learning_rate": 0.0002, "epoch": 6.681143684401014, "step": 9230}, {"loss": 0.9779, "grad_norm": 1.3313323259353638, "learning_rate": 0.0002, "epoch": 6.688382193268187, "step": 9240}, {"loss": 1.005, "grad_norm": 1.130048394203186, "learning_rate": 0.0002, "epoch": 6.69562070213536, "step": 9250}, {"loss": 0.9969, "grad_norm": 1.1997296810150146, "learning_rate": 0.0002, "epoch": 6.702859211002534, "step": 9260}, {"loss": 0.8691, "grad_norm": 1.0591834783554077, "learning_rate": 0.0002, "epoch": 6.710097719869707, "step": 9270}, {"loss": 0.9603, "grad_norm": 1.2722901105880737, "learning_rate": 0.0002, "epoch": 6.7173362287368805, "step": 9280}, {"loss": 0.9227, "grad_norm": 1.1150950193405151, "learning_rate": 0.0002, "epoch": 6.724574737604054, "step": 9290}, {"loss": 0.95, "grad_norm": 1.1575992107391357, "learning_rate": 0.0002, "epoch": 6.731813246471227, "step": 9300}, {"loss": 0.9822, "grad_norm": 0.9371691346168518, "learning_rate": 0.0002, "epoch": 6.739051755338401, "step": 9310}, {"loss": 0.9773, "grad_norm": 1.4924226999282837, "learning_rate": 0.0002, "epoch": 6.746290264205574, "step": 9320}, {"loss": 0.969, "grad_norm": 1.1524218320846558, "learning_rate": 0.0002, "epoch": 6.753528773072747, "step": 9330}, {"loss": 0.9271, "grad_norm": 0.9500471949577332, "learning_rate": 0.0002, "epoch": 6.760767281939921, "step": 9340}, {"loss": 0.9029, "grad_norm": 1.2062290906906128, "learning_rate": 0.0002, "epoch": 6.768005790807094, "step": 9350}, {"loss": 0.9121, "grad_norm": 1.212631106376648, "learning_rate": 0.0002, "epoch": 6.7752442996742674, "step": 9360}, {"loss": 0.8486, "grad_norm": 1.9135472774505615, "learning_rate": 0.0002, "epoch": 6.782482808541441, "step": 9370}, {"loss": 0.9332, "grad_norm": 0.9682775139808655, "learning_rate": 0.0002, "epoch": 6.789721317408614, "step": 9380}, {"loss": 0.8548, "grad_norm": 1.1405237913131714, "learning_rate": 0.0002, "epoch": 6.7969598262757875, "step": 9390}, {"loss": 0.8922, "grad_norm": 1.6855751276016235, "learning_rate": 0.0002, "epoch": 6.804198335142961, "step": 9400}, {"loss": 0.9417, "grad_norm": 1.6590169668197632, "learning_rate": 0.0002, "epoch": 6.811436844010134, "step": 9410}, {"loss": 0.868, "grad_norm": 1.8795170783996582, "learning_rate": 0.0002, "epoch": 6.818675352877308, "step": 9420}, {"loss": 0.9142, "grad_norm": 1.1087183952331543, "learning_rate": 0.0002, "epoch": 6.825913861744481, "step": 9430}, {"loss": 1.1427, "grad_norm": 1.4178446531295776, "learning_rate": 0.0002, "epoch": 6.833152370611654, "step": 9440}, {"loss": 0.8325, "grad_norm": 1.0792350769042969, "learning_rate": 0.0002, "epoch": 6.840390879478828, "step": 9450}, {"loss": 1.0078, "grad_norm": 1.2159196138381958, "learning_rate": 0.0002, "epoch": 6.847629388346001, "step": 9460}, {"loss": 0.9536, "grad_norm": 0.9998821020126343, "learning_rate": 0.0002, "epoch": 6.8548678972131745, "step": 9470}, {"loss": 0.9277, "grad_norm": 0.7940687537193298, "learning_rate": 0.0002, "epoch": 6.862106406080348, "step": 9480}, {"loss": 0.8612, "grad_norm": 0.9572826027870178, "learning_rate": 0.0002, "epoch": 6.869344914947521, "step": 9490}, {"loss": 0.9611, "grad_norm": 1.1086537837982178, "learning_rate": 0.0002, "epoch": 6.876583423814694, "step": 9500}, {"loss": 0.9276, "grad_norm": 1.1934887170791626, "learning_rate": 0.0002, "epoch": 6.883821932681867, "step": 9510}, {"loss": 0.8416, "grad_norm": 1.207324504852295, "learning_rate": 0.0002, "epoch": 6.89106044154904, "step": 9520}, {"loss": 0.9378, "grad_norm": 1.1303677558898926, "learning_rate": 0.0002, "epoch": 6.898298950416214, "step": 9530}, {"loss": 0.9599, "grad_norm": 1.4958926439285278, "learning_rate": 0.0002, "epoch": 6.905537459283387, "step": 9540}, {"loss": 0.9365, "grad_norm": 1.2141553163528442, "learning_rate": 0.0002, "epoch": 6.9127759681505605, "step": 9550}, {"loss": 1.0291, "grad_norm": 1.6544346809387207, "learning_rate": 0.0002, "epoch": 6.920014477017734, "step": 9560}, {"loss": 0.8439, "grad_norm": 1.0540320873260498, "learning_rate": 0.0002, "epoch": 6.927252985884907, "step": 9570}, {"loss": 0.9831, "grad_norm": 1.3095581531524658, "learning_rate": 0.0002, "epoch": 6.934491494752081, "step": 9580}, {"loss": 0.8694, "grad_norm": 1.4509341716766357, "learning_rate": 0.0002, "epoch": 6.941730003619254, "step": 9590}, {"loss": 0.983, "grad_norm": 1.1091740131378174, "learning_rate": 0.0002, "epoch": 6.948968512486427, "step": 9600}, {"loss": 0.9126, "grad_norm": 1.102929949760437, "learning_rate": 0.0002, "epoch": 6.956207021353601, "step": 9610}, {"loss": 0.9622, "grad_norm": 1.1377743482589722, "learning_rate": 0.0002, "epoch": 6.963445530220774, "step": 9620}, {"loss": 0.9045, "grad_norm": 1.2070361375808716, "learning_rate": 0.0002, "epoch": 6.970684039087947, "step": 9630}, {"loss": 0.9714, "grad_norm": 1.30153489112854, "learning_rate": 0.0002, "epoch": 6.977922547955121, "step": 9640}, {"loss": 0.9555, "grad_norm": 1.4641543626785278, "learning_rate": 0.0002, "epoch": 6.985161056822294, "step": 9650}, {"loss": 0.9177, "grad_norm": 1.0497819185256958, "learning_rate": 0.0002, "epoch": 6.9923995656894675, "step": 9660}, {"loss": 0.8369, "grad_norm": 1.2500354051589966, "learning_rate": 0.0002, "epoch": 6.999638074556641, "step": 9670}]}
+{"epoch": 7.997104596453131, "step": 11048, "epoch_duration": 1212.7061741352081, "total_accumulated_duration": 10323.10546040535, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 3020.60888671875}, "peak_memory_usage": {"GPU_0": 15079.2998046875}, "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-0.9-num-8828-sd-1/checkpoint-4144", "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": 4.7061, "grad_norm": 1.2523442506790161, "learning_rate": 0.0002, "epoch": 0.007238508867173362, "step": 10}, {"loss": 3.3493, "grad_norm": 1.8887330293655396, "learning_rate": 0.0002, "epoch": 0.014477017734346724, "step": 20}, {"loss": 2.7585, "grad_norm": 0.9668035507202148, "learning_rate": 0.0002, "epoch": 0.021715526601520086, "step": 30}, {"loss": 2.3699, "grad_norm": 2.9167306423187256, "learning_rate": 0.0002, "epoch": 0.028954035468693448, "step": 40}, {"loss": 2.2679, "grad_norm": 2.649867296218872, "learning_rate": 0.0002, "epoch": 0.036192544335866814, "step": 50}, {"loss": 2.2202, "grad_norm": 1.5120655298233032, "learning_rate": 0.0002, "epoch": 0.04343105320304017, "step": 60}, {"loss": 2.2026, "grad_norm": 0.7879868149757385, "learning_rate": 0.0002, "epoch": 0.05066956207021354, "step": 70}, {"loss": 1.9447, "grad_norm": 0.7616953253746033, "learning_rate": 0.0002, "epoch": 0.057908070937386896, "step": 80}, {"loss": 2.0112, "grad_norm": 1.8809149265289307, "learning_rate": 0.0002, "epoch": 0.06514657980456026, "step": 90}, {"loss": 1.8337, "grad_norm": 0.9294016361236572, "learning_rate": 0.0002, "epoch": 0.07238508867173363, "step": 100}, {"loss": 1.8419, "grad_norm": 0.7145281434059143, "learning_rate": 0.0002, "epoch": 0.07962359753890698, "step": 110}, {"loss": 2.0036, "grad_norm": 0.7564446330070496, "learning_rate": 0.0002, "epoch": 0.08686210640608034, "step": 120}, {"loss": 1.9306, "grad_norm": 1.1681925058364868, "learning_rate": 0.0002, "epoch": 0.09410061527325371, "step": 130}, {"loss": 1.7875, "grad_norm": 0.6708641648292542, "learning_rate": 0.0002, "epoch": 0.10133912414042708, "step": 140}, {"loss": 1.786, "grad_norm": 0.7625647783279419, "learning_rate": 0.0002, "epoch": 0.10857763300760044, "step": 150}, {"loss": 1.6687, "grad_norm": 0.8463464975357056, "learning_rate": 0.0002, "epoch": 0.11581614187477379, "step": 160}, {"loss": 1.6214, "grad_norm": 0.7502335906028748, "learning_rate": 0.0002, "epoch": 0.12305465074194716, "step": 170}, {"loss": 1.7433, "grad_norm": 0.6929958462715149, "learning_rate": 0.0002, "epoch": 0.13029315960912052, "step": 180}, {"loss": 1.6009, "grad_norm": 0.6798707842826843, "learning_rate": 0.0002, "epoch": 0.1375316684762939, "step": 190}, {"loss": 1.6208, "grad_norm": 0.7566508650779724, "learning_rate": 0.0002, "epoch": 0.14477017734346725, "step": 200}, {"loss": 1.5823, "grad_norm": 0.7196869850158691, "learning_rate": 0.0002, "epoch": 0.15200868621064062, "step": 210}, {"loss": 1.738, "grad_norm": 0.8401045799255371, "learning_rate": 0.0002, "epoch": 0.15924719507781396, "step": 220}, {"loss": 1.7574, "grad_norm": 0.8503773212432861, "learning_rate": 0.0002, "epoch": 0.16648570394498732, "step": 230}, {"loss": 1.7861, "grad_norm": 0.7183733582496643, "learning_rate": 0.0002, "epoch": 0.1737242128121607, "step": 240}, {"loss": 1.6693, "grad_norm": 0.7082605957984924, "learning_rate": 0.0002, "epoch": 0.18096272167933405, "step": 250}, {"loss": 1.619, "grad_norm": 0.9386326670646667, "learning_rate": 0.0002, "epoch": 0.18820123054650742, "step": 260}, {"loss": 1.6511, "grad_norm": 0.7332451939582825, "learning_rate": 0.0002, "epoch": 0.19543973941368079, "step": 270}, {"loss": 1.6353, "grad_norm": 0.7092869877815247, "learning_rate": 0.0002, "epoch": 0.20267824828085415, "step": 280}, {"loss": 1.5996, "grad_norm": 0.7256413698196411, "learning_rate": 0.0002, "epoch": 0.20991675714802752, "step": 290}, {"loss": 1.6754, "grad_norm": 0.6398681402206421, "learning_rate": 0.0002, "epoch": 0.21715526601520088, "step": 300}, {"loss": 1.397, "grad_norm": 0.6273287534713745, "learning_rate": 0.0002, "epoch": 0.22439377488237422, "step": 310}, {"loss": 1.5115, "grad_norm": 0.511648416519165, "learning_rate": 0.0002, "epoch": 0.23163228374954759, "step": 320}, {"loss": 1.5424, "grad_norm": 0.8677352070808411, "learning_rate": 0.0002, "epoch": 0.23887079261672095, "step": 330}, {"loss": 1.6779, "grad_norm": 0.6270743012428284, "learning_rate": 0.0002, "epoch": 0.24610930148389432, "step": 340}, {"loss": 1.626, "grad_norm": 0.7980281114578247, "learning_rate": 0.0002, "epoch": 0.2533478103510677, "step": 350}, {"loss": 1.5238, "grad_norm": 0.632486879825592, "learning_rate": 0.0002, "epoch": 0.26058631921824105, "step": 360}, {"loss": 1.5175, "grad_norm": 0.6527034640312195, "learning_rate": 0.0002, "epoch": 0.2678248280854144, "step": 370}, {"loss": 1.627, "grad_norm": 0.7672118544578552, "learning_rate": 0.0002, "epoch": 0.2750633369525878, "step": 380}, {"loss": 1.5605, "grad_norm": 0.6035117506980896, "learning_rate": 0.0002, "epoch": 0.28230184581976114, "step": 390}, {"loss": 1.4603, "grad_norm": 0.5955103039741516, "learning_rate": 0.0002, "epoch": 0.2895403546869345, "step": 400}, {"loss": 1.558, "grad_norm": 0.6015191674232483, "learning_rate": 0.0002, "epoch": 0.2967788635541079, "step": 410}, {"loss": 1.6091, "grad_norm": 0.6380982398986816, "learning_rate": 0.0002, "epoch": 0.30401737242128124, "step": 420}, {"loss": 1.5292, "grad_norm": 0.6707863211631775, "learning_rate": 0.0002, "epoch": 0.3112558812884546, "step": 430}, {"loss": 1.4426, "grad_norm": 0.7010176777839661, "learning_rate": 0.0002, "epoch": 0.3184943901556279, "step": 440}, {"loss": 1.5572, "grad_norm": 0.8263739943504333, "learning_rate": 0.0002, "epoch": 0.3257328990228013, "step": 450}, {"loss": 1.5188, "grad_norm": 0.7253276109695435, "learning_rate": 0.0002, "epoch": 0.33297140788997465, "step": 460}, {"loss": 1.584, "grad_norm": 0.5238934755325317, "learning_rate": 0.0002, "epoch": 0.340209916757148, "step": 470}, {"loss": 1.7035, "grad_norm": 0.7869495749473572, "learning_rate": 0.0002, "epoch": 0.3474484256243214, "step": 480}, {"loss": 1.5776, "grad_norm": 0.7485215663909912, "learning_rate": 0.0002, "epoch": 0.35468693449149474, "step": 490}, {"loss": 1.6274, "grad_norm": 0.5413193106651306, "learning_rate": 0.0002, "epoch": 0.3619254433586681, "step": 500}, {"loss": 1.7323, "grad_norm": 0.7615048885345459, "learning_rate": 0.0002, "epoch": 0.3691639522258415, "step": 510}, {"loss": 1.532, "grad_norm": 0.7685340046882629, "learning_rate": 0.0002, "epoch": 0.37640246109301484, "step": 520}, {"loss": 1.6312, "grad_norm": 0.6379081010818481, "learning_rate": 0.0002, "epoch": 0.3836409699601882, "step": 530}, {"loss": 1.5645, "grad_norm": 0.7946939468383789, "learning_rate": 0.0002, "epoch": 0.39087947882736157, "step": 540}, {"loss": 1.4001, "grad_norm": 0.6287278532981873, "learning_rate": 0.0002, "epoch": 0.39811798769453494, "step": 550}, {"loss": 1.5982, "grad_norm": 0.6811642646789551, "learning_rate": 0.0002, "epoch": 0.4053564965617083, "step": 560}, {"loss": 1.4953, "grad_norm": 0.671073317527771, "learning_rate": 0.0002, "epoch": 0.41259500542888167, "step": 570}, {"loss": 1.6753, "grad_norm": 0.6313900351524353, "learning_rate": 0.0002, "epoch": 0.41983351429605503, "step": 580}, {"loss": 1.546, "grad_norm": 0.5291772484779358, "learning_rate": 0.0002, "epoch": 0.4270720231632284, "step": 590}, {"loss": 1.5441, "grad_norm": 0.62503582239151, "learning_rate": 0.0002, "epoch": 0.43431053203040176, "step": 600}, {"loss": 1.6276, "grad_norm": 0.5777305364608765, "learning_rate": 0.0002, "epoch": 0.4415490408975751, "step": 610}, {"loss": 1.4758, "grad_norm": 0.7013497352600098, "learning_rate": 0.0002, "epoch": 0.44878754976474844, "step": 620}, {"loss": 1.4029, "grad_norm": 0.8044822216033936, "learning_rate": 0.0002, "epoch": 0.4560260586319218, "step": 630}, {"loss": 1.7195, "grad_norm": 0.672531247138977, "learning_rate": 0.0002, "epoch": 0.46326456749909517, "step": 640}, {"loss": 1.614, "grad_norm": 0.6233910322189331, "learning_rate": 0.0002, "epoch": 0.47050307636626854, "step": 650}, {"loss": 1.6041, "grad_norm": 0.651524543762207, "learning_rate": 0.0002, "epoch": 0.4777415852334419, "step": 660}, {"loss": 1.5842, "grad_norm": 0.7213939428329468, "learning_rate": 0.0002, "epoch": 0.48498009410061527, "step": 670}, {"loss": 1.5453, "grad_norm": 0.6541454792022705, "learning_rate": 0.0002, "epoch": 0.49221860296778863, "step": 680}, {"loss": 1.662, "grad_norm": 0.6568936109542847, "learning_rate": 0.0002, "epoch": 0.499457111834962, "step": 690}, {"loss": 1.624, "grad_norm": 0.7176415324211121, "learning_rate": 0.0002, "epoch": 0.5066956207021354, "step": 700}, {"loss": 1.6099, "grad_norm": 0.6553855538368225, "learning_rate": 0.0002, "epoch": 0.5139341295693087, "step": 710}, {"loss": 1.5508, "grad_norm": 0.5654335618019104, "learning_rate": 0.0002, "epoch": 0.5211726384364821, "step": 720}, {"loss": 1.392, "grad_norm": 0.5671001672744751, "learning_rate": 0.0002, "epoch": 0.5284111473036555, "step": 730}, {"loss": 1.388, "grad_norm": 0.7914412021636963, "learning_rate": 0.0002, "epoch": 0.5356496561708288, "step": 740}, {"loss": 1.5931, "grad_norm": 0.6172138452529907, "learning_rate": 0.0002, "epoch": 0.5428881650380022, "step": 750}, {"loss": 1.4018, "grad_norm": 0.6132623553276062, "learning_rate": 0.0002, "epoch": 0.5501266739051756, "step": 760}, {"loss": 1.513, "grad_norm": 0.654000461101532, "learning_rate": 0.0002, "epoch": 0.5573651827723489, "step": 770}, {"loss": 1.5035, "grad_norm": 0.5691370964050293, "learning_rate": 0.0002, "epoch": 0.5646036916395223, "step": 780}, {"loss": 1.65, "grad_norm": 0.7922580242156982, "learning_rate": 0.0002, "epoch": 0.5718422005066957, "step": 790}, {"loss": 1.4521, "grad_norm": 0.6831880211830139, "learning_rate": 0.0002, "epoch": 0.579080709373869, "step": 800}, {"loss": 1.4734, "grad_norm": 0.6740124821662903, "learning_rate": 0.0002, "epoch": 0.5863192182410424, "step": 810}, {"loss": 1.6498, "grad_norm": 1.380016803741455, "learning_rate": 0.0002, "epoch": 0.5935577271082157, "step": 820}, {"loss": 1.4642, "grad_norm": 0.6552878022193909, "learning_rate": 0.0002, "epoch": 0.6007962359753891, "step": 830}, {"loss": 1.6271, "grad_norm": 0.6649535298347473, "learning_rate": 0.0002, "epoch": 0.6080347448425625, "step": 840}, {"loss": 1.5886, "grad_norm": 0.561738133430481, "learning_rate": 0.0002, "epoch": 0.6152732537097358, "step": 850}, {"loss": 1.5364, "grad_norm": 0.6133047938346863, "learning_rate": 0.0002, "epoch": 0.6225117625769092, "step": 860}, {"loss": 1.3489, "grad_norm": 0.559843122959137, "learning_rate": 0.0002, "epoch": 0.6297502714440825, "step": 870}, {"loss": 1.4878, "grad_norm": 0.6117811799049377, "learning_rate": 0.0002, "epoch": 0.6369887803112558, "step": 880}, {"loss": 1.56, "grad_norm": 0.6209776401519775, "learning_rate": 0.0002, "epoch": 0.6442272891784292, "step": 890}, {"loss": 1.6747, "grad_norm": 0.6234082579612732, "learning_rate": 0.0002, "epoch": 0.6514657980456026, "step": 900}, {"loss": 1.6963, "grad_norm": 0.7623258233070374, "learning_rate": 0.0002, "epoch": 0.6587043069127759, "step": 910}, {"loss": 1.2424, "grad_norm": 0.6148061752319336, "learning_rate": 0.0002, "epoch": 0.6659428157799493, "step": 920}, {"loss": 1.4319, "grad_norm": 0.6682973504066467, "learning_rate": 0.0002, "epoch": 0.6731813246471227, "step": 930}, {"loss": 1.5377, "grad_norm": 0.5513041615486145, "learning_rate": 0.0002, "epoch": 0.680419833514296, "step": 940}, {"loss": 1.3991, "grad_norm": 0.5197525024414062, "learning_rate": 0.0002, "epoch": 0.6876583423814694, "step": 950}, {"loss": 1.4398, "grad_norm": 0.6490758061408997, "learning_rate": 0.0002, "epoch": 0.6948968512486428, "step": 960}, {"loss": 1.5251, "grad_norm": 0.6450682878494263, "learning_rate": 0.0002, "epoch": 0.7021353601158161, "step": 970}, {"loss": 1.5417, "grad_norm": 0.6203766465187073, "learning_rate": 0.0002, "epoch": 0.7093738689829895, "step": 980}, {"loss": 1.4575, "grad_norm": 0.6023609638214111, "learning_rate": 0.0002, "epoch": 0.7166123778501629, "step": 990}, {"loss": 1.4973, "grad_norm": 0.5765255093574524, "learning_rate": 0.0002, "epoch": 0.7238508867173362, "step": 1000}, {"loss": 1.483, "grad_norm": 0.6650075316429138, "learning_rate": 0.0002, "epoch": 0.7310893955845096, "step": 1010}, {"loss": 1.5959, "grad_norm": 0.5610854029655457, "learning_rate": 0.0002, "epoch": 0.738327904451683, "step": 1020}, {"loss": 1.5248, "grad_norm": 0.7072813510894775, "learning_rate": 0.0002, "epoch": 0.7455664133188563, "step": 1030}, {"loss": 1.5776, "grad_norm": 0.6815407872200012, "learning_rate": 0.0002, "epoch": 0.7528049221860297, "step": 1040}, {"loss": 1.4577, "grad_norm": 0.7932390570640564, "learning_rate": 0.0002, "epoch": 0.760043431053203, "step": 1050}, {"loss": 1.4515, "grad_norm": 0.5798183083534241, "learning_rate": 0.0002, "epoch": 0.7672819399203764, "step": 1060}, {"loss": 1.5053, "grad_norm": 0.7898504137992859, "learning_rate": 0.0002, "epoch": 0.7745204487875498, "step": 1070}, {"loss": 1.4776, "grad_norm": 0.4983280301094055, "learning_rate": 0.0002, "epoch": 0.7817589576547231, "step": 1080}, {"loss": 1.5007, "grad_norm": 0.691403329372406, "learning_rate": 0.0002, "epoch": 0.7889974665218965, "step": 1090}, {"loss": 1.5153, "grad_norm": 0.5394481420516968, "learning_rate": 0.0002, "epoch": 0.7962359753890699, "step": 1100}, {"loss": 1.6892, "grad_norm": 0.5136822462081909, "learning_rate": 0.0002, "epoch": 0.8034744842562432, "step": 1110}, {"loss": 1.4902, "grad_norm": 0.6828126907348633, "learning_rate": 0.0002, "epoch": 0.8107129931234166, "step": 1120}, {"loss": 1.4346, "grad_norm": 0.6799656748771667, "learning_rate": 0.0002, "epoch": 0.81795150199059, "step": 1130}, {"loss": 1.2678, "grad_norm": 0.5428406000137329, "learning_rate": 0.0002, "epoch": 0.8251900108577633, "step": 1140}, {"loss": 1.4072, "grad_norm": 0.4811290502548218, "learning_rate": 0.0002, "epoch": 0.8324285197249367, "step": 1150}, {"loss": 1.4512, "grad_norm": 0.5519434809684753, "learning_rate": 0.0002, "epoch": 0.8396670285921101, "step": 1160}, {"loss": 1.4072, "grad_norm": 0.9748060703277588, "learning_rate": 0.0002, "epoch": 0.8469055374592834, "step": 1170}, {"loss": 1.4309, "grad_norm": 0.712609589099884, "learning_rate": 0.0002, "epoch": 0.8541440463264568, "step": 1180}, {"loss": 1.434, "grad_norm": 0.6866157054901123, "learning_rate": 0.0002, "epoch": 0.8613825551936302, "step": 1190}, {"loss": 1.3704, "grad_norm": 0.5068854093551636, "learning_rate": 0.0002, "epoch": 0.8686210640608035, "step": 1200}, {"loss": 1.5601, "grad_norm": 0.6333245038986206, "learning_rate": 0.0002, "epoch": 0.8758595729279768, "step": 1210}, {"loss": 1.4636, "grad_norm": 0.6424421072006226, "learning_rate": 0.0002, "epoch": 0.8830980817951501, "step": 1220}, {"loss": 1.4186, "grad_norm": 0.4771921932697296, "learning_rate": 0.0002, "epoch": 0.8903365906623235, "step": 1230}, {"loss": 1.6323, "grad_norm": 0.5191764235496521, "learning_rate": 0.0002, "epoch": 0.8975750995294969, "step": 1240}, {"loss": 1.6105, "grad_norm": 0.756222128868103, "learning_rate": 0.0002, "epoch": 0.9048136083966702, "step": 1250}, {"loss": 1.4396, "grad_norm": 0.623823881149292, "learning_rate": 0.0002, "epoch": 0.9120521172638436, "step": 1260}, {"loss": 1.3097, "grad_norm": 0.8166571259498596, "learning_rate": 0.0002, "epoch": 0.919290626131017, "step": 1270}, {"loss": 1.4625, "grad_norm": 0.6059346795082092, "learning_rate": 0.0002, "epoch": 0.9265291349981903, "step": 1280}, {"loss": 1.3555, "grad_norm": 0.5842690467834473, "learning_rate": 0.0002, "epoch": 0.9337676438653637, "step": 1290}, {"loss": 1.5859, "grad_norm": 0.7649800777435303, "learning_rate": 0.0002, "epoch": 0.9410061527325371, "step": 1300}, {"loss": 1.5915, "grad_norm": 0.6420919895172119, "learning_rate": 0.0002, "epoch": 0.9482446615997104, "step": 1310}, {"loss": 1.453, "grad_norm": 0.7011452913284302, "learning_rate": 0.0002, "epoch": 0.9554831704668838, "step": 1320}, {"loss": 1.6766, "grad_norm": 0.5783746242523193, "learning_rate": 0.0002, "epoch": 0.9627216793340572, "step": 1330}, {"loss": 1.6308, "grad_norm": 0.5973192453384399, "learning_rate": 0.0002, "epoch": 0.9699601882012305, "step": 1340}, {"loss": 1.5901, "grad_norm": 0.6181833744049072, "learning_rate": 0.0002, "epoch": 0.9771986970684039, "step": 1350}, {"loss": 1.5258, "grad_norm": 0.5563396215438843, "learning_rate": 0.0002, "epoch": 0.9844372059355773, "step": 1360}, {"loss": 1.4508, "grad_norm": 0.45723360776901245, "learning_rate": 0.0002, "epoch": 0.9916757148027506, "step": 1370}, {"loss": 1.3291, "grad_norm": 0.5947498679161072, "learning_rate": 0.0002, "epoch": 0.998914223669924, "step": 1380}, {"eval_loss": 1.480796456336975, "eval_runtime": 27.3103, "eval_samples_per_second": 15.965, "eval_steps_per_second": 2.014, "epoch": 0.9996380745566413, "step": 1381}, {"loss": 1.3057, "grad_norm": 0.5599952936172485, "learning_rate": 0.0002, "epoch": 1.0061527325370974, "step": 1390}, {"loss": 1.4991, "grad_norm": 0.5932008028030396, "learning_rate": 0.0002, "epoch": 1.0133912414042707, "step": 1400}, {"loss": 1.4506, "grad_norm": 0.6194121837615967, "learning_rate": 0.0002, "epoch": 1.020629750271444, "step": 1410}, {"loss": 1.5966, "grad_norm": 0.6995621919631958, "learning_rate": 0.0002, "epoch": 1.0278682591386175, "step": 1420}, {"loss": 1.4153, "grad_norm": 0.7905810475349426, "learning_rate": 0.0002, "epoch": 1.0351067680057908, "step": 1430}, {"loss": 1.4414, "grad_norm": 0.7221615314483643, "learning_rate": 0.0002, "epoch": 1.0423452768729642, "step": 1440}, {"loss": 1.3859, "grad_norm": 0.6170642375946045, "learning_rate": 0.0002, "epoch": 1.0495837857401376, "step": 1450}, {"loss": 1.3806, "grad_norm": 0.5844094753265381, "learning_rate": 0.0002, "epoch": 1.056822294607311, "step": 1460}, {"loss": 1.4871, "grad_norm": 0.7731822729110718, "learning_rate": 0.0002, "epoch": 1.0640608034744843, "step": 1470}, {"loss": 1.4286, "grad_norm": 0.4554748237133026, "learning_rate": 0.0002, "epoch": 1.0712993123416577, "step": 1480}, {"loss": 1.3977, "grad_norm": 0.6923259496688843, "learning_rate": 0.0002, "epoch": 1.078537821208831, "step": 1490}, {"loss": 1.3936, "grad_norm": 0.6008219122886658, "learning_rate": 0.0002, "epoch": 1.0857763300760044, "step": 1500}, {"loss": 1.4821, "grad_norm": 0.6450045704841614, "learning_rate": 0.0002, "epoch": 1.0930148389431777, "step": 1510}, {"loss": 1.3295, "grad_norm": 0.7833753824234009, "learning_rate": 0.0002, "epoch": 1.1002533478103511, "step": 1520}, {"loss": 1.3424, "grad_norm": 0.5076758861541748, "learning_rate": 0.0002, "epoch": 1.1074918566775245, "step": 1530}, {"loss": 1.4043, "grad_norm": 0.5661332011222839, "learning_rate": 0.0002, "epoch": 1.1147303655446978, "step": 1540}, {"loss": 1.4963, "grad_norm": 0.6526919603347778, "learning_rate": 0.0002, "epoch": 1.1219688744118712, "step": 1550}, {"loss": 1.3671, "grad_norm": 0.5613082647323608, "learning_rate": 0.0002, "epoch": 1.1292073832790446, "step": 1560}, {"loss": 1.4458, "grad_norm": 0.6113885641098022, "learning_rate": 0.0002, "epoch": 1.136445892146218, "step": 1570}, {"loss": 1.3552, "grad_norm": 0.6732510328292847, "learning_rate": 0.0002, "epoch": 1.1436844010133913, "step": 1580}, {"loss": 1.3114, "grad_norm": 0.6146392226219177, "learning_rate": 0.0002, "epoch": 1.1509229098805647, "step": 1590}, {"loss": 1.411, "grad_norm": 0.6766974329948425, "learning_rate": 0.0002, "epoch": 1.158161418747738, "step": 1600}, {"loss": 1.2401, "grad_norm": 0.7621957659721375, "learning_rate": 0.0002, "epoch": 1.1653999276149114, "step": 1610}, {"loss": 1.3758, "grad_norm": 0.6959581971168518, "learning_rate": 0.0002, "epoch": 1.1726384364820848, "step": 1620}, {"loss": 1.382, "grad_norm": 0.6691278219223022, "learning_rate": 0.0002, "epoch": 1.1798769453492581, "step": 1630}, {"loss": 1.4147, "grad_norm": 0.4927774965763092, "learning_rate": 0.0002, "epoch": 1.1871154542164315, "step": 1640}, {"loss": 1.449, "grad_norm": 0.7724234461784363, "learning_rate": 0.0002, "epoch": 1.1943539630836049, "step": 1650}, {"loss": 1.4778, "grad_norm": 0.6817787885665894, "learning_rate": 0.0002, "epoch": 1.2015924719507782, "step": 1660}, {"loss": 1.3776, "grad_norm": 0.6500699520111084, "learning_rate": 0.0002, "epoch": 1.2088309808179516, "step": 1670}, {"loss": 1.3875, "grad_norm": 0.5703568458557129, "learning_rate": 0.0002, "epoch": 1.216069489685125, "step": 1680}, {"loss": 1.4735, "grad_norm": 0.6261579990386963, "learning_rate": 0.0002, "epoch": 1.2233079985522983, "step": 1690}, {"loss": 1.3898, "grad_norm": 0.651713490486145, "learning_rate": 0.0002, "epoch": 1.2305465074194717, "step": 1700}, {"loss": 1.4002, "grad_norm": 0.684399425983429, "learning_rate": 0.0002, "epoch": 1.237785016286645, "step": 1710}, {"loss": 1.5027, "grad_norm": 0.6996857523918152, "learning_rate": 0.0002, "epoch": 1.2450235251538184, "step": 1720}, {"loss": 1.3326, "grad_norm": 0.7102537751197815, "learning_rate": 0.0002, "epoch": 1.2522620340209918, "step": 1730}, {"loss": 1.3675, "grad_norm": 0.45809897780418396, "learning_rate": 0.0002, "epoch": 1.2595005428881652, "step": 1740}, {"loss": 1.4175, "grad_norm": 0.6377046704292297, "learning_rate": 0.0002, "epoch": 1.2667390517553385, "step": 1750}, {"loss": 1.3479, "grad_norm": 0.6965704560279846, "learning_rate": 0.0002, "epoch": 1.2739775606225119, "step": 1760}, {"loss": 1.5647, "grad_norm": 0.5688214302062988, "learning_rate": 0.0002, "epoch": 1.2812160694896852, "step": 1770}, {"loss": 1.3967, "grad_norm": 0.6384190320968628, "learning_rate": 0.0002, "epoch": 1.2884545783568586, "step": 1780}, {"loss": 1.3671, "grad_norm": 0.5629363656044006, "learning_rate": 0.0002, "epoch": 1.295693087224032, "step": 1790}, {"loss": 1.2292, "grad_norm": 0.6148255467414856, "learning_rate": 0.0002, "epoch": 1.3029315960912053, "step": 1800}, {"loss": 1.5806, "grad_norm": 0.655580997467041, "learning_rate": 0.0002, "epoch": 1.3101701049583787, "step": 1810}, {"loss": 1.2398, "grad_norm": 0.5642657279968262, "learning_rate": 0.0002, "epoch": 1.3174086138255519, "step": 1820}, {"loss": 1.3246, "grad_norm": 0.59607994556427, "learning_rate": 0.0002, "epoch": 1.3246471226927252, "step": 1830}, {"loss": 1.3274, "grad_norm": 0.5564199090003967, "learning_rate": 0.0002, "epoch": 1.3318856315598986, "step": 1840}, {"loss": 1.5834, "grad_norm": 0.6949955821037292, "learning_rate": 0.0002, "epoch": 1.339124140427072, "step": 1850}, {"loss": 1.4722, "grad_norm": 0.7036856412887573, "learning_rate": 0.0002, "epoch": 1.3463626492942453, "step": 1860}, {"loss": 1.333, "grad_norm": 0.722062885761261, "learning_rate": 0.0002, "epoch": 1.3536011581614187, "step": 1870}, {"loss": 1.4044, "grad_norm": 0.6098677515983582, "learning_rate": 0.0002, "epoch": 1.360839667028592, "step": 1880}, {"loss": 1.6217, "grad_norm": 0.5376402735710144, "learning_rate": 0.0002, "epoch": 1.3680781758957654, "step": 1890}, {"loss": 1.5071, "grad_norm": 0.6974610090255737, "learning_rate": 0.0002, "epoch": 1.3753166847629388, "step": 1900}, {"loss": 1.5854, "grad_norm": 0.6520763635635376, "learning_rate": 0.0002, "epoch": 1.3825551936301121, "step": 1910}, {"loss": 1.4271, "grad_norm": 0.6604374647140503, "learning_rate": 0.0002, "epoch": 1.3897937024972855, "step": 1920}, {"loss": 1.419, "grad_norm": 0.7364398241043091, "learning_rate": 0.0002, "epoch": 1.3970322113644589, "step": 1930}, {"loss": 1.4585, "grad_norm": 0.6849475502967834, "learning_rate": 0.0002, "epoch": 1.4042707202316322, "step": 1940}, {"loss": 1.5577, "grad_norm": 0.6562670469284058, "learning_rate": 0.0002, "epoch": 1.4115092290988056, "step": 1950}, {"loss": 1.4725, "grad_norm": 0.5695616006851196, "learning_rate": 0.0002, "epoch": 1.418747737965979, "step": 1960}, {"loss": 1.3088, "grad_norm": 0.5244464874267578, "learning_rate": 0.0002, "epoch": 1.4259862468331523, "step": 1970}, {"loss": 1.5069, "grad_norm": 0.6347293257713318, "learning_rate": 0.0002, "epoch": 1.4332247557003257, "step": 1980}, {"loss": 1.3502, "grad_norm": 0.5528361201286316, "learning_rate": 0.0002, "epoch": 1.440463264567499, "step": 1990}, {"loss": 1.3978, "grad_norm": 0.6987585425376892, "learning_rate": 0.0002, "epoch": 1.4477017734346724, "step": 2000}, {"loss": 1.4262, "grad_norm": 0.6568987369537354, "learning_rate": 0.0002, "epoch": 1.4549402823018458, "step": 2010}, {"loss": 1.4175, "grad_norm": 0.7665994763374329, "learning_rate": 0.0002, "epoch": 1.4621787911690192, "step": 2020}, {"loss": 1.244, "grad_norm": 0.5127707123756409, "learning_rate": 0.0002, "epoch": 1.4694173000361925, "step": 2030}, {"loss": 1.3699, "grad_norm": 0.5406824946403503, "learning_rate": 0.0002, "epoch": 1.476655808903366, "step": 2040}, {"loss": 1.3353, "grad_norm": 0.5990166664123535, "learning_rate": 0.0002, "epoch": 1.4838943177705393, "step": 2050}, {"loss": 1.2454, "grad_norm": 0.6186193823814392, "learning_rate": 0.0002, "epoch": 1.4911328266377126, "step": 2060}, {"loss": 1.428, "grad_norm": 0.6154307126998901, "learning_rate": 0.0002, "epoch": 1.498371335504886, "step": 2070}, {"loss": 1.4528, "grad_norm": 0.5606056451797485, "learning_rate": 0.0002, "epoch": 1.5056098443720594, "step": 2080}, {"loss": 1.2405, "grad_norm": 0.5006417036056519, "learning_rate": 0.0002, "epoch": 1.5128483532392327, "step": 2090}, {"loss": 1.4258, "grad_norm": 0.5968486070632935, "learning_rate": 0.0002, "epoch": 1.520086862106406, "step": 2100}, {"loss": 1.2752, "grad_norm": 0.5835496187210083, "learning_rate": 0.0002, "epoch": 1.5273253709735795, "step": 2110}, {"loss": 1.5443, "grad_norm": 0.6753535270690918, "learning_rate": 0.0002, "epoch": 1.5345638798407528, "step": 2120}, {"loss": 1.2139, "grad_norm": 0.7299720644950867, "learning_rate": 0.0002, "epoch": 1.5418023887079262, "step": 2130}, {"loss": 1.2364, "grad_norm": 0.5105988383293152, "learning_rate": 0.0002, "epoch": 1.5490408975750996, "step": 2140}, {"loss": 1.4528, "grad_norm": 0.5675431489944458, "learning_rate": 0.0002, "epoch": 1.556279406442273, "step": 2150}, {"loss": 1.4563, "grad_norm": 0.6246723532676697, "learning_rate": 0.0002, "epoch": 1.5635179153094463, "step": 2160}, {"loss": 1.5255, "grad_norm": 0.7291720509529114, "learning_rate": 0.0002, "epoch": 1.5707564241766196, "step": 2170}, {"loss": 1.5432, "grad_norm": 0.678114116191864, "learning_rate": 0.0002, "epoch": 1.577994933043793, "step": 2180}, {"loss": 1.5212, "grad_norm": 0.5136260986328125, "learning_rate": 0.0002, "epoch": 1.5852334419109664, "step": 2190}, {"loss": 1.3271, "grad_norm": 0.6359935998916626, "learning_rate": 0.0002, "epoch": 1.5924719507781397, "step": 2200}, {"loss": 1.4038, "grad_norm": 0.7650278806686401, "learning_rate": 0.0002, "epoch": 1.599710459645313, "step": 2210}, {"loss": 1.5478, "grad_norm": 0.7256110906600952, "learning_rate": 0.0002, "epoch": 1.6069489685124865, "step": 2220}, {"loss": 1.4387, "grad_norm": 0.688689649105072, "learning_rate": 0.0002, "epoch": 1.6141874773796598, "step": 2230}, {"loss": 1.4096, "grad_norm": 0.6045311093330383, "learning_rate": 0.0002, "epoch": 1.6214259862468332, "step": 2240}, {"loss": 1.4097, "grad_norm": 0.7064604163169861, "learning_rate": 0.0002, "epoch": 1.6286644951140063, "step": 2250}, {"loss": 1.3477, "grad_norm": 0.5309562087059021, "learning_rate": 0.0002, "epoch": 1.6359030039811797, "step": 2260}, {"loss": 1.4022, "grad_norm": 0.5687053203582764, "learning_rate": 0.0002, "epoch": 1.643141512848353, "step": 2270}, {"loss": 1.2977, "grad_norm": 0.535872757434845, "learning_rate": 0.0002, "epoch": 1.6503800217155264, "step": 2280}, {"loss": 1.3844, "grad_norm": 0.5502381920814514, "learning_rate": 0.0002, "epoch": 1.6576185305826998, "step": 2290}, {"loss": 1.3764, "grad_norm": 0.6158602237701416, "learning_rate": 0.0002, "epoch": 1.6648570394498732, "step": 2300}, {"loss": 1.3515, "grad_norm": 0.5804675817489624, "learning_rate": 0.0002, "epoch": 1.6720955483170465, "step": 2310}, {"loss": 1.2532, "grad_norm": 0.600742757320404, "learning_rate": 0.0002, "epoch": 1.67933405718422, "step": 2320}, {"loss": 1.477, "grad_norm": 0.7101941108703613, "learning_rate": 0.0002, "epoch": 1.6865725660513933, "step": 2330}, {"loss": 1.4849, "grad_norm": 0.7507809996604919, "learning_rate": 0.0002, "epoch": 1.6938110749185666, "step": 2340}, {"loss": 1.2703, "grad_norm": 0.768502414226532, "learning_rate": 0.0002, "epoch": 1.70104958378574, "step": 2350}, {"loss": 1.3332, "grad_norm": 0.4801851212978363, "learning_rate": 0.0002, "epoch": 1.7082880926529134, "step": 2360}, {"loss": 1.4158, "grad_norm": 0.5322122573852539, "learning_rate": 0.0002, "epoch": 1.7155266015200867, "step": 2370}, {"loss": 1.4136, "grad_norm": 0.587661862373352, "learning_rate": 0.0002, "epoch": 1.72276511038726, "step": 2380}, {"loss": 1.3771, "grad_norm": 0.6073525547981262, "learning_rate": 0.0002, "epoch": 1.7300036192544335, "step": 2390}, {"loss": 1.2754, "grad_norm": 0.6950460076332092, "learning_rate": 0.0002, "epoch": 1.7372421281216068, "step": 2400}, {"loss": 1.3858, "grad_norm": 0.5981102585792542, "learning_rate": 0.0002, "epoch": 1.7444806369887802, "step": 2410}, {"loss": 1.4075, "grad_norm": 0.544570803642273, "learning_rate": 0.0002, "epoch": 1.7517191458559536, "step": 2420}, {"loss": 1.3861, "grad_norm": 0.5304399728775024, "learning_rate": 0.0002, "epoch": 1.758957654723127, "step": 2430}, {"loss": 1.4244, "grad_norm": 0.7921594977378845, "learning_rate": 0.0002, "epoch": 1.7661961635903003, "step": 2440}, {"loss": 1.3053, "grad_norm": 0.6084808707237244, "learning_rate": 0.0002, "epoch": 1.7734346724574737, "step": 2450}, {"loss": 1.3781, "grad_norm": 0.8844701051712036, "learning_rate": 0.0002, "epoch": 1.780673181324647, "step": 2460}, {"loss": 1.3227, "grad_norm": 0.5729258060455322, "learning_rate": 0.0002, "epoch": 1.7879116901918204, "step": 2470}, {"loss": 1.3422, "grad_norm": 0.6303611993789673, "learning_rate": 0.0002, "epoch": 1.7951501990589938, "step": 2480}, {"loss": 1.3926, "grad_norm": 0.5627942085266113, "learning_rate": 0.0002, "epoch": 1.8023887079261671, "step": 2490}, {"loss": 1.3816, "grad_norm": 0.6724274158477783, "learning_rate": 0.0002, "epoch": 1.8096272167933405, "step": 2500}, {"loss": 1.2951, "grad_norm": 0.5030826330184937, "learning_rate": 0.0002, "epoch": 1.8168657256605139, "step": 2510}, {"loss": 1.2839, "grad_norm": 0.5504099130630493, "learning_rate": 0.0002, "epoch": 1.8241042345276872, "step": 2520}, {"loss": 1.4264, "grad_norm": 0.6338945627212524, "learning_rate": 0.0002, "epoch": 1.8313427433948606, "step": 2530}, {"loss": 1.563, "grad_norm": 0.5902037620544434, "learning_rate": 0.0002, "epoch": 1.838581252262034, "step": 2540}, {"loss": 1.2961, "grad_norm": 0.48814457654953003, "learning_rate": 0.0002, "epoch": 1.8458197611292073, "step": 2550}, {"loss": 1.466, "grad_norm": 0.6216312646865845, "learning_rate": 0.0002, "epoch": 1.8530582699963807, "step": 2560}, {"loss": 1.5123, "grad_norm": 0.635603666305542, "learning_rate": 0.0002, "epoch": 1.860296778863554, "step": 2570}, {"loss": 1.372, "grad_norm": 0.6938216090202332, "learning_rate": 0.0002, "epoch": 1.8675352877307274, "step": 2580}, {"loss": 1.5011, "grad_norm": 0.599557638168335, "learning_rate": 0.0002, "epoch": 1.8747737965979008, "step": 2590}, {"loss": 1.2714, "grad_norm": 0.564424455165863, "learning_rate": 0.0002, "epoch": 1.8820123054650741, "step": 2600}, {"loss": 1.3403, "grad_norm": 0.5430700182914734, "learning_rate": 0.0002, "epoch": 1.8892508143322475, "step": 2610}, {"loss": 1.4347, "grad_norm": 0.6150169372558594, "learning_rate": 0.0002, "epoch": 1.8964893231994209, "step": 2620}, {"loss": 1.2474, "grad_norm": 0.48159119486808777, "learning_rate": 0.0002, "epoch": 1.9037278320665942, "step": 2630}, {"loss": 1.3716, "grad_norm": 0.5608997941017151, "learning_rate": 0.0002, "epoch": 1.9109663409337676, "step": 2640}, {"loss": 1.5787, "grad_norm": 0.6454501748085022, "learning_rate": 0.0002, "epoch": 1.918204849800941, "step": 2650}, {"loss": 1.3238, "grad_norm": 0.5458073616027832, "learning_rate": 0.0002, "epoch": 1.9254433586681143, "step": 2660}, {"loss": 1.3208, "grad_norm": 0.5328490734100342, "learning_rate": 0.0002, "epoch": 1.9326818675352877, "step": 2670}, {"loss": 1.4971, "grad_norm": 0.6444696187973022, "learning_rate": 0.0002, "epoch": 1.939920376402461, "step": 2680}, {"loss": 1.5387, "grad_norm": 0.7126023769378662, "learning_rate": 0.0002, "epoch": 1.9471588852696344, "step": 2690}, {"loss": 1.3637, "grad_norm": 0.5164045095443726, "learning_rate": 0.0002, "epoch": 1.9543973941368078, "step": 2700}, {"loss": 1.5303, "grad_norm": 0.5347061157226562, "learning_rate": 0.0002, "epoch": 1.9616359030039812, "step": 2710}, {"loss": 1.2815, "grad_norm": 0.5297950506210327, "learning_rate": 0.0002, "epoch": 1.9688744118711545, "step": 2720}, {"loss": 1.3566, "grad_norm": 0.6537790298461914, "learning_rate": 0.0002, "epoch": 1.976112920738328, "step": 2730}, {"loss": 1.332, "grad_norm": 0.5536222457885742, "learning_rate": 0.0002, "epoch": 1.9833514296055013, "step": 2740}, {"loss": 1.3333, "grad_norm": 0.4856105446815491, "learning_rate": 0.0002, "epoch": 1.9905899384726746, "step": 2750}, {"loss": 1.3521, "grad_norm": 0.6642730832099915, "learning_rate": 0.0002, "epoch": 1.997828447339848, "step": 2760}, {"eval_loss": 1.4366681575775146, "eval_runtime": 27.3729, "eval_samples_per_second": 15.928, "eval_steps_per_second": 2.009, "epoch": 2.0, "step": 2763}, {"loss": 1.4322, "grad_norm": 0.740253210067749, "learning_rate": 0.0002, "epoch": 2.0050669562070214, "step": 2770}, {"loss": 1.277, "grad_norm": 0.5826276540756226, "learning_rate": 0.0002, "epoch": 2.0123054650741947, "step": 2780}, {"loss": 1.2424, "grad_norm": 0.607356071472168, "learning_rate": 0.0002, "epoch": 2.019543973941368, "step": 2790}, {"loss": 1.2601, "grad_norm": 0.5918063521385193, "learning_rate": 0.0002, "epoch": 2.0267824828085415, "step": 2800}, {"loss": 1.3715, "grad_norm": 0.5610089898109436, "learning_rate": 0.0002, "epoch": 2.034020991675715, "step": 2810}, {"loss": 1.2092, "grad_norm": 0.5869926810264587, "learning_rate": 0.0002, "epoch": 2.041259500542888, "step": 2820}, {"loss": 1.1929, "grad_norm": 0.5753467679023743, "learning_rate": 0.0002, "epoch": 2.0484980094100615, "step": 2830}, {"loss": 1.333, "grad_norm": 0.7096508145332336, "learning_rate": 0.0002, "epoch": 2.055736518277235, "step": 2840}, {"loss": 1.1766, "grad_norm": 0.7653635144233704, "learning_rate": 0.0002, "epoch": 2.0629750271444083, "step": 2850}, {"loss": 1.2331, "grad_norm": 0.6202841997146606, "learning_rate": 0.0002, "epoch": 2.0702135360115816, "step": 2860}, {"loss": 1.3298, "grad_norm": 0.6810227632522583, "learning_rate": 0.0002, "epoch": 2.077452044878755, "step": 2870}, {"loss": 1.2505, "grad_norm": 0.7481493353843689, "learning_rate": 0.0002, "epoch": 2.0846905537459284, "step": 2880}, {"loss": 1.2484, "grad_norm": 0.7089637517929077, "learning_rate": 0.0002, "epoch": 2.0919290626131017, "step": 2890}, {"loss": 1.3095, "grad_norm": 0.7472923398017883, "learning_rate": 0.0002, "epoch": 2.099167571480275, "step": 2900}, {"loss": 1.304, "grad_norm": 0.8135465979576111, "learning_rate": 0.0002, "epoch": 2.1064060803474485, "step": 2910}, {"loss": 1.273, "grad_norm": 0.6097133159637451, "learning_rate": 0.0002, "epoch": 2.113644589214622, "step": 2920}, {"loss": 1.3384, "grad_norm": 0.5970117449760437, "learning_rate": 0.0002, "epoch": 2.120883098081795, "step": 2930}, {"loss": 1.3233, "grad_norm": 0.6169309616088867, "learning_rate": 0.0002, "epoch": 2.1281216069489686, "step": 2940}, {"loss": 1.4246, "grad_norm": 0.9428738355636597, "learning_rate": 0.0002, "epoch": 2.135360115816142, "step": 2950}, {"loss": 1.3527, "grad_norm": 0.5671679973602295, "learning_rate": 0.0002, "epoch": 2.1425986246833153, "step": 2960}, {"loss": 1.1375, "grad_norm": 0.7007262110710144, "learning_rate": 0.0002, "epoch": 2.1498371335504887, "step": 2970}, {"loss": 1.2015, "grad_norm": 0.6294044256210327, "learning_rate": 0.0002, "epoch": 2.157075642417662, "step": 2980}, {"loss": 1.2167, "grad_norm": 0.6105241775512695, "learning_rate": 0.0002, "epoch": 2.1643141512848354, "step": 2990}, {"loss": 1.2065, "grad_norm": 0.557124137878418, "learning_rate": 0.0002, "epoch": 2.1715526601520088, "step": 3000}, {"loss": 1.2515, "grad_norm": 0.6250392198562622, "learning_rate": 0.0002, "epoch": 2.178791169019182, "step": 3010}, {"loss": 1.385, "grad_norm": 0.645218551158905, "learning_rate": 0.0002, "epoch": 2.1860296778863555, "step": 3020}, {"loss": 1.3928, "grad_norm": 0.9033605456352234, "learning_rate": 0.0002, "epoch": 2.193268186753529, "step": 3030}, {"loss": 1.2458, "grad_norm": 0.5325747132301331, "learning_rate": 0.0002, "epoch": 2.2005066956207022, "step": 3040}, {"loss": 1.261, "grad_norm": 0.6334700584411621, "learning_rate": 0.0002, "epoch": 2.2077452044878756, "step": 3050}, {"loss": 1.2385, "grad_norm": 0.5206325054168701, "learning_rate": 0.0002, "epoch": 2.214983713355049, "step": 3060}, {"loss": 1.3103, "grad_norm": 0.5987200140953064, "learning_rate": 0.0002, "epoch": 2.2222222222222223, "step": 3070}, {"loss": 1.1756, "grad_norm": 0.5893264412879944, "learning_rate": 0.0002, "epoch": 2.2294607310893957, "step": 3080}, {"loss": 1.235, "grad_norm": 0.6869237422943115, "learning_rate": 0.0002, "epoch": 2.236699239956569, "step": 3090}, {"loss": 1.3285, "grad_norm": 0.5040048360824585, "learning_rate": 0.0002, "epoch": 2.2439377488237424, "step": 3100}, {"loss": 1.3316, "grad_norm": 0.6660613417625427, "learning_rate": 0.0002, "epoch": 2.251176257690916, "step": 3110}, {"loss": 1.3108, "grad_norm": 0.5890918970108032, "learning_rate": 0.0002, "epoch": 2.258414766558089, "step": 3120}, {"loss": 1.248, "grad_norm": 0.6458896994590759, "learning_rate": 0.0002, "epoch": 2.2656532754252625, "step": 3130}, {"loss": 1.4151, "grad_norm": 0.6832690834999084, "learning_rate": 0.0002, "epoch": 2.272891784292436, "step": 3140}, {"loss": 1.4458, "grad_norm": 0.833908200263977, "learning_rate": 0.0002, "epoch": 2.2801302931596092, "step": 3150}, {"loss": 1.2931, "grad_norm": 0.4596034586429596, "learning_rate": 0.0002, "epoch": 2.2873688020267826, "step": 3160}, {"loss": 1.449, "grad_norm": 0.9130966067314148, "learning_rate": 0.0002, "epoch": 2.294607310893956, "step": 3170}, {"loss": 1.3806, "grad_norm": 0.7143292427062988, "learning_rate": 0.0002, "epoch": 2.3018458197611293, "step": 3180}, {"loss": 1.2692, "grad_norm": 0.5388900637626648, "learning_rate": 0.0002, "epoch": 2.3090843286283027, "step": 3190}, {"loss": 1.2402, "grad_norm": 0.5607513189315796, "learning_rate": 0.0002, "epoch": 2.316322837495476, "step": 3200}, {"loss": 1.3874, "grad_norm": 0.6795142292976379, "learning_rate": 0.0002, "epoch": 2.3235613463626494, "step": 3210}, {"loss": 1.3042, "grad_norm": 0.6561070680618286, "learning_rate": 0.0002, "epoch": 2.330799855229823, "step": 3220}, {"loss": 1.4636, "grad_norm": 0.8858118057250977, "learning_rate": 0.0002, "epoch": 2.338038364096996, "step": 3230}, {"loss": 1.3214, "grad_norm": 0.6604151725769043, "learning_rate": 0.0002, "epoch": 2.3452768729641695, "step": 3240}, {"loss": 1.4004, "grad_norm": 0.6755785346031189, "learning_rate": 0.0002, "epoch": 2.352515381831343, "step": 3250}, {"loss": 1.2503, "grad_norm": 0.6981677412986755, "learning_rate": 0.0002, "epoch": 2.3597538906985163, "step": 3260}, {"loss": 1.3078, "grad_norm": 0.6338568329811096, "learning_rate": 0.0002, "epoch": 2.3669923995656896, "step": 3270}, {"loss": 1.285, "grad_norm": 0.5754265785217285, "learning_rate": 0.0002, "epoch": 2.374230908432863, "step": 3280}, {"loss": 1.2924, "grad_norm": 0.7533153295516968, "learning_rate": 0.0002, "epoch": 2.3814694173000364, "step": 3290}, {"loss": 1.3711, "grad_norm": 0.675065279006958, "learning_rate": 0.0002, "epoch": 2.3887079261672097, "step": 3300}, {"loss": 1.3548, "grad_norm": 0.5686452984809875, "learning_rate": 0.0002, "epoch": 2.395946435034383, "step": 3310}, {"loss": 1.1998, "grad_norm": 0.8129481673240662, "learning_rate": 0.0002, "epoch": 2.4031849439015565, "step": 3320}, {"loss": 1.2584, "grad_norm": 0.6615934371948242, "learning_rate": 0.0002, "epoch": 2.41042345276873, "step": 3330}, {"loss": 1.3691, "grad_norm": 0.6678834557533264, "learning_rate": 0.0002, "epoch": 2.417661961635903, "step": 3340}, {"loss": 1.2381, "grad_norm": 0.5581308007240295, "learning_rate": 0.0002, "epoch": 2.4249004705030766, "step": 3350}, {"loss": 1.3853, "grad_norm": 0.6098920106887817, "learning_rate": 0.0002, "epoch": 2.43213897937025, "step": 3360}, {"loss": 1.3692, "grad_norm": 0.8101736903190613, "learning_rate": 0.0002, "epoch": 2.4393774882374233, "step": 3370}, {"loss": 1.4418, "grad_norm": 0.6621488928794861, "learning_rate": 0.0002, "epoch": 2.4466159971045967, "step": 3380}, {"loss": 1.4579, "grad_norm": 0.8693289160728455, "learning_rate": 0.0002, "epoch": 2.45385450597177, "step": 3390}, {"loss": 1.3644, "grad_norm": 0.6724580526351929, "learning_rate": 0.0002, "epoch": 2.4610930148389434, "step": 3400}, {"loss": 1.2006, "grad_norm": 0.6776891946792603, "learning_rate": 0.0002, "epoch": 2.4683315237061167, "step": 3410}, {"loss": 1.2937, "grad_norm": 0.7214453816413879, "learning_rate": 0.0002, "epoch": 2.47557003257329, "step": 3420}, {"loss": 1.4051, "grad_norm": 0.8390451073646545, "learning_rate": 0.0002, "epoch": 2.4828085414404635, "step": 3430}, {"loss": 1.25, "grad_norm": 0.7130982279777527, "learning_rate": 0.0002, "epoch": 2.490047050307637, "step": 3440}, {"loss": 1.2231, "grad_norm": 0.8873937129974365, "learning_rate": 0.0002, "epoch": 2.49728555917481, "step": 3450}, {"loss": 1.1429, "grad_norm": 0.725185751914978, "learning_rate": 0.0002, "epoch": 2.5045240680419836, "step": 3460}, {"loss": 1.2699, "grad_norm": 0.6120352149009705, "learning_rate": 0.0002, "epoch": 2.511762576909157, "step": 3470}, {"loss": 1.2552, "grad_norm": 0.7713613510131836, "learning_rate": 0.0002, "epoch": 2.5190010857763303, "step": 3480}, {"loss": 1.4648, "grad_norm": 0.895309567451477, "learning_rate": 0.0002, "epoch": 2.5262395946435037, "step": 3490}, {"loss": 1.3043, "grad_norm": 0.9631021022796631, "learning_rate": 0.0002, "epoch": 2.533478103510677, "step": 3500}, {"loss": 1.3492, "grad_norm": 0.7475683093070984, "learning_rate": 0.0002, "epoch": 2.5407166123778504, "step": 3510}, {"loss": 1.3637, "grad_norm": 0.7271341681480408, "learning_rate": 0.0002, "epoch": 2.5479551212450238, "step": 3520}, {"loss": 1.304, "grad_norm": 0.6979510188102722, "learning_rate": 0.0002, "epoch": 2.555193630112197, "step": 3530}, {"loss": 1.2353, "grad_norm": 0.6504196524620056, "learning_rate": 0.0002, "epoch": 2.5624321389793705, "step": 3540}, {"loss": 1.2699, "grad_norm": 0.7226675748825073, "learning_rate": 0.0002, "epoch": 2.569670647846544, "step": 3550}, {"loss": 1.3002, "grad_norm": 0.6143222451210022, "learning_rate": 0.0002, "epoch": 2.5769091567137172, "step": 3560}, {"loss": 1.1585, "grad_norm": 0.7245154976844788, "learning_rate": 0.0002, "epoch": 2.5841476655808906, "step": 3570}, {"loss": 1.3651, "grad_norm": 0.943540632724762, "learning_rate": 0.0002, "epoch": 2.591386174448064, "step": 3580}, {"loss": 1.3034, "grad_norm": 0.7707241773605347, "learning_rate": 0.0002, "epoch": 2.5986246833152373, "step": 3590}, {"loss": 1.3063, "grad_norm": 0.6705001592636108, "learning_rate": 0.0002, "epoch": 2.6058631921824107, "step": 3600}, {"loss": 1.2437, "grad_norm": 0.6360933780670166, "learning_rate": 0.0002, "epoch": 2.613101701049584, "step": 3610}, {"loss": 1.1844, "grad_norm": 0.5846424698829651, "learning_rate": 0.0002, "epoch": 2.6203402099167574, "step": 3620}, {"loss": 1.3674, "grad_norm": 0.5958625674247742, "learning_rate": 0.0002, "epoch": 2.6275787187839303, "step": 3630}, {"loss": 1.3599, "grad_norm": 0.6819243431091309, "learning_rate": 0.0002, "epoch": 2.6348172276511037, "step": 3640}, {"loss": 1.3884, "grad_norm": 0.7033445835113525, "learning_rate": 0.0002, "epoch": 2.642055736518277, "step": 3650}, {"loss": 1.3392, "grad_norm": 0.6134849786758423, "learning_rate": 0.0002, "epoch": 2.6492942453854504, "step": 3660}, {"loss": 1.2661, "grad_norm": 0.658009946346283, "learning_rate": 0.0002, "epoch": 2.656532754252624, "step": 3670}, {"loss": 1.3987, "grad_norm": 0.6280999779701233, "learning_rate": 0.0002, "epoch": 2.663771263119797, "step": 3680}, {"loss": 1.2995, "grad_norm": 0.5536085963249207, "learning_rate": 0.0002, "epoch": 2.6710097719869705, "step": 3690}, {"loss": 1.2044, "grad_norm": 0.8603981733322144, "learning_rate": 0.0002, "epoch": 2.678248280854144, "step": 3700}, {"loss": 1.3879, "grad_norm": 0.5509994626045227, "learning_rate": 0.0002, "epoch": 2.6854867897213173, "step": 3710}, {"loss": 1.3253, "grad_norm": 0.9093621969223022, "learning_rate": 0.0002, "epoch": 2.6927252985884906, "step": 3720}, {"loss": 1.2668, "grad_norm": 0.7525952458381653, "learning_rate": 0.0002, "epoch": 2.699963807455664, "step": 3730}, {"loss": 1.248, "grad_norm": 0.6737023591995239, "learning_rate": 0.0002, "epoch": 2.7072023163228374, "step": 3740}, {"loss": 1.2981, "grad_norm": 0.8656924962997437, "learning_rate": 0.0002, "epoch": 2.7144408251900107, "step": 3750}, {"loss": 1.2342, "grad_norm": 0.7494133114814758, "learning_rate": 0.0002, "epoch": 2.721679334057184, "step": 3760}, {"loss": 1.2417, "grad_norm": 0.5725520849227905, "learning_rate": 0.0002, "epoch": 2.7289178429243575, "step": 3770}, {"loss": 1.28, "grad_norm": 0.836412787437439, "learning_rate": 0.0002, "epoch": 2.736156351791531, "step": 3780}, {"loss": 1.3784, "grad_norm": 0.6893242597579956, "learning_rate": 0.0002, "epoch": 2.743394860658704, "step": 3790}, {"loss": 1.2929, "grad_norm": 0.6696223020553589, "learning_rate": 0.0002, "epoch": 2.7506333695258776, "step": 3800}, {"loss": 1.2449, "grad_norm": 0.6483015418052673, "learning_rate": 0.0002, "epoch": 2.757871878393051, "step": 3810}, {"loss": 1.3282, "grad_norm": 0.8084456920623779, "learning_rate": 0.0002, "epoch": 2.7651103872602243, "step": 3820}, {"loss": 1.3694, "grad_norm": 0.6601949334144592, "learning_rate": 0.0002, "epoch": 2.7723488961273977, "step": 3830}, {"loss": 1.3568, "grad_norm": 0.6905533671379089, "learning_rate": 0.0002, "epoch": 2.779587404994571, "step": 3840}, {"loss": 1.3854, "grad_norm": 0.619318425655365, "learning_rate": 0.0002, "epoch": 2.7868259138617444, "step": 3850}, {"loss": 1.2551, "grad_norm": 0.5994023084640503, "learning_rate": 0.0002, "epoch": 2.7940644227289178, "step": 3860}, {"loss": 1.2022, "grad_norm": 0.5627168416976929, "learning_rate": 0.0002, "epoch": 2.801302931596091, "step": 3870}, {"loss": 1.3921, "grad_norm": 0.6001605987548828, "learning_rate": 0.0002, "epoch": 2.8085414404632645, "step": 3880}, {"loss": 1.3026, "grad_norm": 0.6022412776947021, "learning_rate": 0.0002, "epoch": 2.815779949330438, "step": 3890}, {"loss": 1.2765, "grad_norm": 0.6832426190376282, "learning_rate": 0.0002, "epoch": 2.823018458197611, "step": 3900}, {"loss": 1.1363, "grad_norm": 0.5936811566352844, "learning_rate": 0.0002, "epoch": 2.8302569670647846, "step": 3910}, {"loss": 1.1707, "grad_norm": 0.6960572600364685, "learning_rate": 0.0002, "epoch": 2.837495475931958, "step": 3920}, {"loss": 1.4063, "grad_norm": 0.5913406610488892, "learning_rate": 0.0002, "epoch": 2.8447339847991313, "step": 3930}, {"loss": 1.3245, "grad_norm": 0.678154706954956, "learning_rate": 0.0002, "epoch": 2.8519724936663047, "step": 3940}, {"loss": 1.366, "grad_norm": 0.7898936867713928, "learning_rate": 0.0002, "epoch": 2.859211002533478, "step": 3950}, {"loss": 1.3948, "grad_norm": 0.9234195351600647, "learning_rate": 0.0002, "epoch": 2.8664495114006514, "step": 3960}, {"loss": 1.2773, "grad_norm": 0.5960825085639954, "learning_rate": 0.0002, "epoch": 2.8736880202678248, "step": 3970}, {"loss": 1.3127, "grad_norm": 0.677118182182312, "learning_rate": 0.0002, "epoch": 2.880926529134998, "step": 3980}, {"loss": 1.2652, "grad_norm": 0.6505142450332642, "learning_rate": 0.0002, "epoch": 2.8881650380021715, "step": 3990}, {"loss": 1.2078, "grad_norm": 0.550826907157898, "learning_rate": 0.0002, "epoch": 2.895403546869345, "step": 4000}, {"loss": 1.1811, "grad_norm": 0.6209215521812439, "learning_rate": 0.0002, "epoch": 2.9026420557365182, "step": 4010}, {"loss": 1.4001, "grad_norm": 0.6549018025398254, "learning_rate": 0.0002, "epoch": 2.9098805646036916, "step": 4020}, {"loss": 1.2285, "grad_norm": 0.570682168006897, "learning_rate": 0.0002, "epoch": 2.917119073470865, "step": 4030}, {"loss": 1.0832, "grad_norm": 1.1807632446289062, "learning_rate": 0.0002, "epoch": 2.9243575823380383, "step": 4040}, {"loss": 1.2693, "grad_norm": 0.7058857679367065, "learning_rate": 0.0002, "epoch": 2.9315960912052117, "step": 4050}, {"loss": 1.2905, "grad_norm": 0.5542812943458557, "learning_rate": 0.0002, "epoch": 2.938834600072385, "step": 4060}, {"loss": 1.33, "grad_norm": 0.63167804479599, "learning_rate": 0.0002, "epoch": 2.9460731089395584, "step": 4070}, {"loss": 1.3075, "grad_norm": 0.5702962279319763, "learning_rate": 0.0002, "epoch": 2.953311617806732, "step": 4080}, {"loss": 1.2007, "grad_norm": 0.620944082736969, "learning_rate": 0.0002, "epoch": 2.960550126673905, "step": 4090}, {"loss": 1.2864, "grad_norm": 0.5866289734840393, "learning_rate": 0.0002, "epoch": 2.9677886355410785, "step": 4100}, {"loss": 1.3293, "grad_norm": 0.560170590877533, "learning_rate": 0.0002, "epoch": 2.975027144408252, "step": 4110}, {"loss": 1.2071, "grad_norm": 0.675082802772522, "learning_rate": 0.0002, "epoch": 2.9822656532754253, "step": 4120}, {"loss": 1.2981, "grad_norm": 0.62708580493927, "learning_rate": 0.0002, "epoch": 2.9895041621425986, "step": 4130}, {"loss": 1.2758, "grad_norm": 0.7893929481506348, "learning_rate": 0.0002, "epoch": 2.996742671009772, "step": 4140}, {"eval_loss": 1.4217946529388428, "eval_runtime": 27.1596, "eval_samples_per_second": 16.053, "eval_steps_per_second": 2.025, "epoch": 2.9996380745566413, "step": 4144}, {"loss": 1.2152, "grad_norm": 0.7043836116790771, "learning_rate": 0.0002, "epoch": 3.0039811798769454, "step": 4150}, {"loss": 1.1664, "grad_norm": 0.6806283593177795, "learning_rate": 0.0002, "epoch": 3.0112196887441187, "step": 4160}, {"loss": 1.292, "grad_norm": 0.7684550285339355, "learning_rate": 0.0002, "epoch": 3.018458197611292, "step": 4170}, {"loss": 1.3467, "grad_norm": 0.7895237803459167, "learning_rate": 0.0002, "epoch": 3.0256967064784654, "step": 4180}, {"loss": 1.1324, "grad_norm": 0.7464531064033508, "learning_rate": 0.0002, "epoch": 3.032935215345639, "step": 4190}, {"loss": 1.1614, "grad_norm": 0.9358500838279724, "learning_rate": 0.0002, "epoch": 3.040173724212812, "step": 4200}, {"loss": 1.1834, "grad_norm": 1.1066628694534302, "learning_rate": 0.0002, "epoch": 3.0474122330799855, "step": 4210}, {"loss": 1.1557, "grad_norm": 0.6663267612457275, "learning_rate": 0.0002, "epoch": 3.054650741947159, "step": 4220}, {"loss": 1.1707, "grad_norm": 0.6669464707374573, "learning_rate": 0.0002, "epoch": 3.0618892508143323, "step": 4230}, {"loss": 1.1841, "grad_norm": 0.7052164077758789, "learning_rate": 0.0002, "epoch": 3.0691277596815056, "step": 4240}, {"loss": 1.2913, "grad_norm": 0.6118432879447937, "learning_rate": 0.0002, "epoch": 3.076366268548679, "step": 4250}, {"loss": 1.1526, "grad_norm": 0.6915903687477112, "learning_rate": 0.0002, "epoch": 3.0836047774158524, "step": 4260}, {"loss": 1.1348, "grad_norm": 0.7441644668579102, "learning_rate": 0.0002, "epoch": 3.0908432862830257, "step": 4270}, {"loss": 1.1672, "grad_norm": 0.823850691318512, "learning_rate": 0.0002, "epoch": 3.098081795150199, "step": 4280}, {"loss": 1.2655, "grad_norm": 0.9677883386611938, "learning_rate": 0.0002, "epoch": 3.1053203040173725, "step": 4290}, {"loss": 1.1794, "grad_norm": 0.7002579569816589, "learning_rate": 0.0002, "epoch": 3.112558812884546, "step": 4300}, {"loss": 1.135, "grad_norm": 0.778789758682251, "learning_rate": 0.0002, "epoch": 3.119797321751719, "step": 4310}, {"loss": 1.0818, "grad_norm": 0.7236007452011108, "learning_rate": 0.0002, "epoch": 3.1270358306188926, "step": 4320}, {"loss": 1.1803, "grad_norm": 0.8809133768081665, "learning_rate": 0.0002, "epoch": 3.134274339486066, "step": 4330}, {"loss": 1.2571, "grad_norm": 0.7924913167953491, "learning_rate": 0.0002, "epoch": 3.1415128483532393, "step": 4340}, {"loss": 1.1413, "grad_norm": 0.7437422275543213, "learning_rate": 0.0002, "epoch": 3.1487513572204127, "step": 4350}, {"loss": 1.2088, "grad_norm": 0.6428450345993042, "learning_rate": 0.0002, "epoch": 3.155989866087586, "step": 4360}, {"loss": 1.3032, "grad_norm": 0.7922873497009277, "learning_rate": 0.0002, "epoch": 3.1632283749547594, "step": 4370}, {"loss": 1.216, "grad_norm": 0.5252506732940674, "learning_rate": 0.0002, "epoch": 3.1704668838219328, "step": 4380}, {"loss": 1.1297, "grad_norm": 0.8570457696914673, "learning_rate": 0.0002, "epoch": 3.177705392689106, "step": 4390}, {"loss": 1.0994, "grad_norm": 0.7218987345695496, "learning_rate": 0.0002, "epoch": 3.1849439015562795, "step": 4400}, {"loss": 1.2891, "grad_norm": 0.6921393275260925, "learning_rate": 0.0002, "epoch": 3.192182410423453, "step": 4410}, {"loss": 1.2668, "grad_norm": 0.7386137843132019, "learning_rate": 0.0002, "epoch": 3.199420919290626, "step": 4420}, {"loss": 1.1654, "grad_norm": 0.6227759122848511, "learning_rate": 0.0002, "epoch": 3.2066594281577996, "step": 4430}, {"loss": 1.1752, "grad_norm": 0.7180278897285461, "learning_rate": 0.0002, "epoch": 3.213897937024973, "step": 4440}, {"loss": 1.1757, "grad_norm": 0.745830774307251, "learning_rate": 0.0002, "epoch": 3.2211364458921463, "step": 4450}, {"loss": 1.234, "grad_norm": 0.6766072511672974, "learning_rate": 0.0002, "epoch": 3.2283749547593197, "step": 4460}, {"loss": 1.1999, "grad_norm": 0.8325067162513733, "learning_rate": 0.0002, "epoch": 3.235613463626493, "step": 4470}, {"loss": 1.1606, "grad_norm": 0.7148305177688599, "learning_rate": 0.0002, "epoch": 3.2428519724936664, "step": 4480}, {"loss": 1.1383, "grad_norm": 0.7752676010131836, "learning_rate": 0.0002, "epoch": 3.25009048136084, "step": 4490}, {"loss": 1.3006, "grad_norm": 0.6776860952377319, "learning_rate": 0.0002, "epoch": 3.257328990228013, "step": 4500}, {"loss": 1.0796, "grad_norm": 0.704359769821167, "learning_rate": 0.0002, "epoch": 3.2645674990951865, "step": 4510}, {"loss": 1.2496, "grad_norm": 0.6880282163619995, "learning_rate": 0.0002, "epoch": 3.27180600796236, "step": 4520}, {"loss": 1.0947, "grad_norm": 0.8179270029067993, "learning_rate": 0.0002, "epoch": 3.2790445168295332, "step": 4530}, {"loss": 1.1909, "grad_norm": 0.6718448996543884, "learning_rate": 0.0002, "epoch": 3.2862830256967066, "step": 4540}, {"loss": 1.2708, "grad_norm": 0.8300657868385315, "learning_rate": 0.0002, "epoch": 3.29352153456388, "step": 4550}, {"loss": 1.2594, "grad_norm": 0.6433690786361694, "learning_rate": 0.0002, "epoch": 3.3007600434310533, "step": 4560}, {"loss": 1.2479, "grad_norm": 0.690262496471405, "learning_rate": 0.0002, "epoch": 3.3079985522982267, "step": 4570}, {"loss": 1.1342, "grad_norm": 0.7022852301597595, "learning_rate": 0.0002, "epoch": 3.3152370611654, "step": 4580}, {"loss": 1.0844, "grad_norm": 0.6438387632369995, "learning_rate": 0.0002, "epoch": 3.3224755700325734, "step": 4590}, {"loss": 1.17, "grad_norm": 0.6866899132728577, "learning_rate": 0.0002, "epoch": 3.329714078899747, "step": 4600}, {"loss": 1.1289, "grad_norm": 0.8233968019485474, "learning_rate": 0.0002, "epoch": 3.33695258776692, "step": 4610}, {"loss": 1.1855, "grad_norm": 0.7251574993133545, "learning_rate": 0.0002, "epoch": 3.3441910966340935, "step": 4620}, {"loss": 1.3403, "grad_norm": 0.7855110168457031, "learning_rate": 0.0002, "epoch": 3.351429605501267, "step": 4630}, {"loss": 1.2922, "grad_norm": 0.8487356305122375, "learning_rate": 0.0002, "epoch": 3.3586681143684403, "step": 4640}, {"loss": 1.2462, "grad_norm": 0.6429011225700378, "learning_rate": 0.0002, "epoch": 3.3659066232356136, "step": 4650}, {"loss": 1.129, "grad_norm": 0.7095270156860352, "learning_rate": 0.0002, "epoch": 3.373145132102787, "step": 4660}, {"loss": 1.262, "grad_norm": 0.6792303323745728, "learning_rate": 0.0002, "epoch": 3.3803836409699604, "step": 4670}, {"loss": 1.256, "grad_norm": 0.6784825921058655, "learning_rate": 0.0002, "epoch": 3.3876221498371337, "step": 4680}, {"loss": 1.0838, "grad_norm": 0.6362888216972351, "learning_rate": 0.0002, "epoch": 3.394860658704307, "step": 4690}, {"loss": 1.2165, "grad_norm": 0.7794778943061829, "learning_rate": 0.0002, "epoch": 3.4020991675714805, "step": 4700}, {"loss": 1.0644, "grad_norm": 0.7287485003471375, "learning_rate": 0.0002, "epoch": 3.409337676438654, "step": 4710}, {"loss": 1.2925, "grad_norm": 0.6481451392173767, "learning_rate": 0.0002, "epoch": 3.416576185305827, "step": 4720}, {"loss": 1.2121, "grad_norm": 0.9200371503829956, "learning_rate": 0.0002, "epoch": 3.4238146941730006, "step": 4730}, {"loss": 1.072, "grad_norm": 1.074180245399475, "learning_rate": 0.0002, "epoch": 3.431053203040174, "step": 4740}, {"loss": 1.0421, "grad_norm": 0.6722986698150635, "learning_rate": 0.0002, "epoch": 3.438291711907347, "step": 4750}, {"loss": 1.2258, "grad_norm": 0.7945933938026428, "learning_rate": 0.0002, "epoch": 3.44553022077452, "step": 4760}, {"loss": 1.0927, "grad_norm": 0.7624640464782715, "learning_rate": 0.0002, "epoch": 3.4527687296416936, "step": 4770}, {"loss": 1.2428, "grad_norm": 0.7763656377792358, "learning_rate": 0.0002, "epoch": 3.460007238508867, "step": 4780}, {"loss": 1.2584, "grad_norm": 0.7736947536468506, "learning_rate": 0.0002, "epoch": 3.4672457473760403, "step": 4790}, {"loss": 1.1953, "grad_norm": 0.8450354933738708, "learning_rate": 0.0002, "epoch": 3.4744842562432137, "step": 4800}, {"loss": 1.1362, "grad_norm": 0.6480133533477783, "learning_rate": 0.0002, "epoch": 3.481722765110387, "step": 4810}, {"loss": 1.1882, "grad_norm": 0.8437445759773254, "learning_rate": 0.0002, "epoch": 3.4889612739775604, "step": 4820}, {"loss": 1.1519, "grad_norm": 0.7781730890274048, "learning_rate": 0.0002, "epoch": 3.4961997828447338, "step": 4830}, {"loss": 1.1836, "grad_norm": 0.8523228168487549, "learning_rate": 0.0002, "epoch": 3.503438291711907, "step": 4840}, {"loss": 1.1672, "grad_norm": 0.6236732006072998, "learning_rate": 0.0002, "epoch": 3.5106768005790805, "step": 4850}, {"loss": 1.1926, "grad_norm": 0.7500787377357483, "learning_rate": 0.0002, "epoch": 3.517915309446254, "step": 4860}, {"loss": 1.1998, "grad_norm": 0.7665374875068665, "learning_rate": 0.0002, "epoch": 3.5251538183134272, "step": 4870}, {"loss": 1.1551, "grad_norm": 0.787857711315155, "learning_rate": 0.0002, "epoch": 3.5323923271806006, "step": 4880}, {"loss": 1.2758, "grad_norm": 0.970595121383667, "learning_rate": 0.0002, "epoch": 3.539630836047774, "step": 4890}, {"loss": 1.1274, "grad_norm": 0.6409347057342529, "learning_rate": 0.0002, "epoch": 3.5468693449149473, "step": 4900}, {"loss": 1.1596, "grad_norm": 0.888551652431488, "learning_rate": 0.0002, "epoch": 3.5541078537821207, "step": 4910}, {"loss": 1.1644, "grad_norm": 1.0808377265930176, "learning_rate": 0.0002, "epoch": 3.561346362649294, "step": 4920}, {"loss": 1.2564, "grad_norm": 0.7501053214073181, "learning_rate": 0.0002, "epoch": 3.5685848715164674, "step": 4930}, {"loss": 1.2351, "grad_norm": 0.7375240325927734, "learning_rate": 0.0002, "epoch": 3.575823380383641, "step": 4940}, {"loss": 1.3568, "grad_norm": 0.7075039744377136, "learning_rate": 0.0002, "epoch": 3.583061889250814, "step": 4950}, {"loss": 1.3355, "grad_norm": 0.939337432384491, "learning_rate": 0.0002, "epoch": 3.5903003981179875, "step": 4960}, {"loss": 1.1722, "grad_norm": 0.6717396974563599, "learning_rate": 0.0002, "epoch": 3.597538906985161, "step": 4970}, {"loss": 1.1186, "grad_norm": 0.7141643762588501, "learning_rate": 0.0002, "epoch": 3.6047774158523342, "step": 4980}, {"loss": 1.1011, "grad_norm": 0.7109216451644897, "learning_rate": 0.0002, "epoch": 3.6120159247195076, "step": 4990}, {"loss": 1.2178, "grad_norm": 0.7020776867866516, "learning_rate": 0.0002, "epoch": 3.619254433586681, "step": 5000}, {"loss": 1.1939, "grad_norm": 0.7158873677253723, "learning_rate": 0.0002, "epoch": 3.6264929424538543, "step": 5010}, {"loss": 1.2624, "grad_norm": 0.7062035202980042, "learning_rate": 0.0002, "epoch": 3.6337314513210277, "step": 5020}, {"loss": 1.0224, "grad_norm": 0.7081155776977539, "learning_rate": 0.0002, "epoch": 3.640969960188201, "step": 5030}, {"loss": 1.2195, "grad_norm": 1.2210607528686523, "learning_rate": 0.0002, "epoch": 3.6482084690553744, "step": 5040}, {"loss": 1.2596, "grad_norm": 0.6650236248970032, "learning_rate": 0.0002, "epoch": 3.655446977922548, "step": 5050}, {"loss": 1.1072, "grad_norm": 0.6884829998016357, "learning_rate": 0.0002, "epoch": 3.662685486789721, "step": 5060}, {"loss": 1.2292, "grad_norm": 0.7317819595336914, "learning_rate": 0.0002, "epoch": 3.6699239956568945, "step": 5070}, {"loss": 1.1917, "grad_norm": 0.7406691908836365, "learning_rate": 0.0002, "epoch": 3.677162504524068, "step": 5080}, {"loss": 1.2949, "grad_norm": 0.9009454250335693, "learning_rate": 0.0002, "epoch": 3.6844010133912413, "step": 5090}, {"loss": 1.1528, "grad_norm": 0.8189385533332825, "learning_rate": 0.0002, "epoch": 3.6916395222584146, "step": 5100}, {"loss": 1.3408, "grad_norm": 1.0793628692626953, "learning_rate": 0.0002, "epoch": 3.698878031125588, "step": 5110}, {"loss": 1.2417, "grad_norm": 0.8593027591705322, "learning_rate": 0.0002, "epoch": 3.7061165399927614, "step": 5120}, {"loss": 1.2141, "grad_norm": 0.8481812477111816, "learning_rate": 0.0002, "epoch": 3.7133550488599347, "step": 5130}, {"loss": 1.125, "grad_norm": 0.6527451276779175, "learning_rate": 0.0002, "epoch": 3.720593557727108, "step": 5140}, {"loss": 1.1584, "grad_norm": 0.9220114350318909, "learning_rate": 0.0002, "epoch": 3.7278320665942815, "step": 5150}, {"loss": 1.2267, "grad_norm": 1.0842019319534302, "learning_rate": 0.0002, "epoch": 3.735070575461455, "step": 5160}, {"loss": 1.3083, "grad_norm": 0.965453565120697, "learning_rate": 0.0002, "epoch": 3.742309084328628, "step": 5170}, {"loss": 1.1772, "grad_norm": 0.9903319478034973, "learning_rate": 0.0002, "epoch": 3.7495475931958016, "step": 5180}, {"loss": 1.2515, "grad_norm": 0.7434818148612976, "learning_rate": 0.0002, "epoch": 3.756786102062975, "step": 5190}, {"loss": 1.2631, "grad_norm": 0.6717280745506287, "learning_rate": 0.0002, "epoch": 3.7640246109301483, "step": 5200}, {"loss": 1.2012, "grad_norm": 0.7754665613174438, "learning_rate": 0.0002, "epoch": 3.7712631197973217, "step": 5210}, {"loss": 1.305, "grad_norm": 1.028374433517456, "learning_rate": 0.0002, "epoch": 3.778501628664495, "step": 5220}, {"loss": 1.1866, "grad_norm": 0.6026996374130249, "learning_rate": 0.0002, "epoch": 3.7857401375316684, "step": 5230}, {"loss": 1.1901, "grad_norm": 0.6978490948677063, "learning_rate": 0.0002, "epoch": 3.7929786463988417, "step": 5240}, {"loss": 1.2576, "grad_norm": 0.7303446531295776, "learning_rate": 0.0002, "epoch": 3.800217155266015, "step": 5250}, {"loss": 1.3173, "grad_norm": 1.0734210014343262, "learning_rate": 0.0002, "epoch": 3.8074556641331885, "step": 5260}, {"loss": 1.1137, "grad_norm": 0.6383201479911804, "learning_rate": 0.0002, "epoch": 3.814694173000362, "step": 5270}, {"loss": 1.0904, "grad_norm": 0.7742630243301392, "learning_rate": 0.0002, "epoch": 3.821932681867535, "step": 5280}, {"loss": 1.2232, "grad_norm": 0.8477074503898621, "learning_rate": 0.0002, "epoch": 3.8291711907347086, "step": 5290}, {"loss": 1.2047, "grad_norm": 0.6675317883491516, "learning_rate": 0.0002, "epoch": 3.836409699601882, "step": 5300}, {"loss": 1.2275, "grad_norm": 0.7515445351600647, "learning_rate": 0.0002, "epoch": 3.8436482084690553, "step": 5310}, {"loss": 1.2569, "grad_norm": 1.1441220045089722, "learning_rate": 0.0002, "epoch": 3.8508867173362287, "step": 5320}, {"loss": 1.1512, "grad_norm": 0.7968795895576477, "learning_rate": 0.0002, "epoch": 3.858125226203402, "step": 5330}, {"loss": 1.232, "grad_norm": 0.7842824459075928, "learning_rate": 0.0002, "epoch": 3.8653637350705754, "step": 5340}, {"loss": 1.1847, "grad_norm": 0.8272225260734558, "learning_rate": 0.0002, "epoch": 3.8726022439377488, "step": 5350}, {"loss": 1.1381, "grad_norm": 0.8413397669792175, "learning_rate": 0.0002, "epoch": 3.879840752804922, "step": 5360}, {"loss": 1.2349, "grad_norm": 1.141764760017395, "learning_rate": 0.0002, "epoch": 3.8870792616720955, "step": 5370}, {"loss": 1.212, "grad_norm": 0.9826975464820862, "learning_rate": 0.0002, "epoch": 3.894317770539269, "step": 5380}, {"loss": 1.1833, "grad_norm": 0.8598255515098572, "learning_rate": 0.0002, "epoch": 3.9015562794064422, "step": 5390}, {"loss": 1.1247, "grad_norm": 0.6271058320999146, "learning_rate": 0.0002, "epoch": 3.9087947882736156, "step": 5400}, {"loss": 1.2212, "grad_norm": 0.6379870772361755, "learning_rate": 0.0002, "epoch": 3.916033297140789, "step": 5410}, {"loss": 1.2481, "grad_norm": 1.0313376188278198, "learning_rate": 0.0002, "epoch": 3.9232718060079623, "step": 5420}, {"loss": 1.1872, "grad_norm": 0.8220619559288025, "learning_rate": 0.0002, "epoch": 3.9305103148751357, "step": 5430}, {"loss": 1.2006, "grad_norm": 0.7576116919517517, "learning_rate": 0.0002, "epoch": 3.937748823742309, "step": 5440}, {"loss": 1.1969, "grad_norm": 1.226235032081604, "learning_rate": 0.0002, "epoch": 3.9449873326094824, "step": 5450}, {"loss": 1.2945, "grad_norm": 0.7979229688644409, "learning_rate": 0.0002, "epoch": 3.952225841476656, "step": 5460}, {"loss": 1.1922, "grad_norm": 0.9911929965019226, "learning_rate": 0.0002, "epoch": 3.959464350343829, "step": 5470}, {"loss": 1.0924, "grad_norm": 0.643738865852356, "learning_rate": 0.0002, "epoch": 3.9667028592110025, "step": 5480}, {"loss": 1.0607, "grad_norm": 0.682305634021759, "learning_rate": 0.0002, "epoch": 3.973941368078176, "step": 5490}, {"loss": 1.2908, "grad_norm": 1.18373441696167, "learning_rate": 0.0002, "epoch": 3.9811798769453492, "step": 5500}, {"loss": 1.0889, "grad_norm": 0.7190203070640564, "learning_rate": 0.0002, "epoch": 3.9884183858125226, "step": 5510}, {"loss": 1.2745, "grad_norm": 0.7516948580741882, "learning_rate": 0.0002, "epoch": 3.995656894679696, "step": 5520}, {"eval_loss": 1.4252897500991821, "eval_runtime": 27.235, "eval_samples_per_second": 16.009, "eval_steps_per_second": 2.019, "epoch": 4.0, "step": 5526}, {"loss": 1.0088, "grad_norm": 0.6353074312210083, "learning_rate": 0.0002, "epoch": 4.002895403546869, "step": 5530}, {"loss": 1.0326, "grad_norm": 0.7424906492233276, "learning_rate": 0.0002, "epoch": 4.010133912414043, "step": 5540}, {"loss": 1.0667, "grad_norm": 0.8856638073921204, "learning_rate": 0.0002, "epoch": 4.017372421281216, "step": 5550}, {"loss": 1.0905, "grad_norm": 0.9627974033355713, "learning_rate": 0.0002, "epoch": 4.024610930148389, "step": 5560}, {"loss": 1.0965, "grad_norm": 0.9048978686332703, "learning_rate": 0.0002, "epoch": 4.031849439015563, "step": 5570}, {"loss": 1.1108, "grad_norm": 0.921119213104248, "learning_rate": 0.0002, "epoch": 4.039087947882736, "step": 5580}, {"loss": 1.1235, "grad_norm": 0.8654361963272095, "learning_rate": 0.0002, "epoch": 4.0463264567499095, "step": 5590}, {"loss": 1.0794, "grad_norm": 0.7947945594787598, "learning_rate": 0.0002, "epoch": 4.053564965617083, "step": 5600}, {"loss": 1.0674, "grad_norm": 0.8307326436042786, "learning_rate": 0.0002, "epoch": 4.060803474484256, "step": 5610}, {"loss": 1.0076, "grad_norm": 0.793273389339447, "learning_rate": 0.0002, "epoch": 4.06804198335143, "step": 5620}, {"loss": 1.0651, "grad_norm": 0.8748673796653748, "learning_rate": 0.0002, "epoch": 4.075280492218603, "step": 5630}, {"loss": 1.111, "grad_norm": 0.7926856279373169, "learning_rate": 0.0002, "epoch": 4.082519001085776, "step": 5640}, {"loss": 1.044, "grad_norm": 0.922645092010498, "learning_rate": 0.0002, "epoch": 4.08975750995295, "step": 5650}, {"loss": 1.109, "grad_norm": 0.9539641737937927, "learning_rate": 0.0002, "epoch": 4.096996018820123, "step": 5660}, {"loss": 1.0788, "grad_norm": 0.8674443364143372, "learning_rate": 0.0002, "epoch": 4.1042345276872965, "step": 5670}, {"loss": 0.9867, "grad_norm": 0.7097609043121338, "learning_rate": 0.0002, "epoch": 4.11147303655447, "step": 5680}, {"loss": 1.1154, "grad_norm": 0.8875522613525391, "learning_rate": 0.0002, "epoch": 4.118711545421643, "step": 5690}, {"loss": 1.1217, "grad_norm": 0.8583634495735168, "learning_rate": 0.0002, "epoch": 4.125950054288817, "step": 5700}, {"loss": 1.0973, "grad_norm": 0.6736377477645874, "learning_rate": 0.0002, "epoch": 4.13318856315599, "step": 5710}, {"loss": 1.1199, "grad_norm": 0.9349062442779541, "learning_rate": 0.0002, "epoch": 4.140427072023163, "step": 5720}, {"loss": 1.0508, "grad_norm": 1.0610365867614746, "learning_rate": 0.0002, "epoch": 4.147665580890337, "step": 5730}, {"loss": 1.1146, "grad_norm": 1.5838189125061035, "learning_rate": 0.0002, "epoch": 4.15490408975751, "step": 5740}, {"loss": 1.0222, "grad_norm": 0.747522234916687, "learning_rate": 0.0002, "epoch": 4.162142598624683, "step": 5750}, {"loss": 1.1328, "grad_norm": 1.3247915506362915, "learning_rate": 0.0002, "epoch": 4.169381107491857, "step": 5760}, {"loss": 1.1655, "grad_norm": 0.8750247955322266, "learning_rate": 0.0002, "epoch": 4.17661961635903, "step": 5770}, {"loss": 1.199, "grad_norm": 0.7914144992828369, "learning_rate": 0.0002, "epoch": 4.1838581252262035, "step": 5780}, {"loss": 1.1213, "grad_norm": 0.9493299126625061, "learning_rate": 0.0002, "epoch": 4.191096634093377, "step": 5790}, {"loss": 1.1515, "grad_norm": 0.7802295088768005, "learning_rate": 0.0002, "epoch": 4.19833514296055, "step": 5800}, {"loss": 1.0704, "grad_norm": 0.6987314820289612, "learning_rate": 0.0002, "epoch": 4.205573651827724, "step": 5810}, {"loss": 1.1699, "grad_norm": 0.9220341444015503, "learning_rate": 0.0002, "epoch": 4.212812160694897, "step": 5820}, {"loss": 1.1394, "grad_norm": 0.8932939767837524, "learning_rate": 0.0002, "epoch": 4.22005066956207, "step": 5830}, {"loss": 1.0048, "grad_norm": 0.920002818107605, "learning_rate": 0.0002, "epoch": 4.227289178429244, "step": 5840}, {"loss": 0.964, "grad_norm": 0.6662752032279968, "learning_rate": 0.0002, "epoch": 4.234527687296417, "step": 5850}, {"loss": 0.986, "grad_norm": 0.8679718971252441, "learning_rate": 0.0002, "epoch": 4.24176619616359, "step": 5860}, {"loss": 0.8991, "grad_norm": 0.7020887732505798, "learning_rate": 0.0002, "epoch": 4.249004705030764, "step": 5870}, {"loss": 1.1132, "grad_norm": 0.869611382484436, "learning_rate": 0.0002, "epoch": 4.256243213897937, "step": 5880}, {"loss": 1.1026, "grad_norm": 0.7796585559844971, "learning_rate": 0.0002, "epoch": 4.2634817227651105, "step": 5890}, {"loss": 1.0957, "grad_norm": 0.8978819251060486, "learning_rate": 0.0002, "epoch": 4.270720231632284, "step": 5900}, {"loss": 1.1325, "grad_norm": 1.0837205648422241, "learning_rate": 0.0002, "epoch": 4.277958740499457, "step": 5910}, {"loss": 1.1279, "grad_norm": 0.7584353089332581, "learning_rate": 0.0002, "epoch": 4.285197249366631, "step": 5920}, {"loss": 1.0513, "grad_norm": 0.7313185334205627, "learning_rate": 0.0002, "epoch": 4.292435758233804, "step": 5930}, {"loss": 1.1101, "grad_norm": 0.8004671335220337, "learning_rate": 0.0002, "epoch": 4.299674267100977, "step": 5940}, {"loss": 1.14, "grad_norm": 2.154958724975586, "learning_rate": 0.0002, "epoch": 4.306912775968151, "step": 5950}, {"loss": 1.1206, "grad_norm": 0.9163479804992676, "learning_rate": 0.0002, "epoch": 4.314151284835324, "step": 5960}, {"loss": 0.9941, "grad_norm": 0.9151589274406433, "learning_rate": 0.0002, "epoch": 4.321389793702497, "step": 5970}, {"loss": 1.0606, "grad_norm": 0.8624112010002136, "learning_rate": 0.0002, "epoch": 4.328628302569671, "step": 5980}, {"loss": 1.1625, "grad_norm": 0.9357741475105286, "learning_rate": 0.0002, "epoch": 4.335866811436844, "step": 5990}, {"loss": 1.0712, "grad_norm": 1.3482335805892944, "learning_rate": 0.0002, "epoch": 4.3431053203040175, "step": 6000}, {"loss": 1.1224, "grad_norm": 0.7156149744987488, "learning_rate": 0.0002, "epoch": 4.350343829171191, "step": 6010}, {"loss": 1.0753, "grad_norm": 0.8480049967765808, "learning_rate": 0.0002, "epoch": 4.357582338038364, "step": 6020}, {"loss": 1.051, "grad_norm": 0.8262244462966919, "learning_rate": 0.0002, "epoch": 4.364820846905538, "step": 6030}, {"loss": 0.9966, "grad_norm": 0.7733905911445618, "learning_rate": 0.0002, "epoch": 4.372059355772711, "step": 6040}, {"loss": 1.1008, "grad_norm": 0.8553919792175293, "learning_rate": 0.0002, "epoch": 4.379297864639884, "step": 6050}, {"loss": 1.1777, "grad_norm": 0.8666832447052002, "learning_rate": 0.0002, "epoch": 4.386536373507058, "step": 6060}, {"loss": 1.1934, "grad_norm": 0.9168295860290527, "learning_rate": 0.0002, "epoch": 4.393774882374231, "step": 6070}, {"loss": 1.0988, "grad_norm": 0.7315238118171692, "learning_rate": 0.0002, "epoch": 4.4010133912414044, "step": 6080}, {"loss": 1.1599, "grad_norm": 1.020263433456421, "learning_rate": 0.0002, "epoch": 4.408251900108578, "step": 6090}, {"loss": 1.133, "grad_norm": 0.9978243708610535, "learning_rate": 0.0002, "epoch": 4.415490408975751, "step": 6100}, {"loss": 1.1324, "grad_norm": 0.995453953742981, "learning_rate": 0.0002, "epoch": 4.4227289178429245, "step": 6110}, {"loss": 1.0957, "grad_norm": 0.9360884428024292, "learning_rate": 0.0002, "epoch": 4.429967426710098, "step": 6120}, {"loss": 0.9506, "grad_norm": 0.8099448084831238, "learning_rate": 0.0002, "epoch": 4.437205935577271, "step": 6130}, {"loss": 1.0887, "grad_norm": 0.8173841238021851, "learning_rate": 0.0002, "epoch": 4.444444444444445, "step": 6140}, {"loss": 1.1219, "grad_norm": 0.7972666025161743, "learning_rate": 0.0002, "epoch": 4.451682953311618, "step": 6150}, {"loss": 1.0226, "grad_norm": 0.7685779333114624, "learning_rate": 0.0002, "epoch": 4.458921462178791, "step": 6160}, {"loss": 1.0732, "grad_norm": 0.7872623801231384, "learning_rate": 0.0002, "epoch": 4.466159971045965, "step": 6170}, {"loss": 0.9911, "grad_norm": 0.7677070498466492, "learning_rate": 0.0002, "epoch": 4.473398479913138, "step": 6180}, {"loss": 1.0919, "grad_norm": 0.7878316044807434, "learning_rate": 0.0002, "epoch": 4.4806369887803115, "step": 6190}, {"loss": 1.018, "grad_norm": 0.8178079724311829, "learning_rate": 0.0002, "epoch": 4.487875497647485, "step": 6200}, {"loss": 1.0517, "grad_norm": 1.2820082902908325, "learning_rate": 0.0002, "epoch": 4.495114006514658, "step": 6210}, {"loss": 1.3101, "grad_norm": 0.9380832314491272, "learning_rate": 0.0002, "epoch": 4.502352515381832, "step": 6220}, {"loss": 0.9818, "grad_norm": 0.7810422778129578, "learning_rate": 0.0002, "epoch": 4.509591024249005, "step": 6230}, {"loss": 1.1677, "grad_norm": 1.1022917032241821, "learning_rate": 0.0002, "epoch": 4.516829533116178, "step": 6240}, {"loss": 1.1579, "grad_norm": 1.4275553226470947, "learning_rate": 0.0002, "epoch": 4.524068041983352, "step": 6250}, {"loss": 1.3237, "grad_norm": 0.7597777247428894, "learning_rate": 0.0002, "epoch": 4.531306550850525, "step": 6260}, {"loss": 1.1529, "grad_norm": 1.10992431640625, "learning_rate": 0.0002, "epoch": 4.538545059717698, "step": 6270}, {"loss": 1.0732, "grad_norm": 0.8981178998947144, "learning_rate": 0.0002, "epoch": 4.545783568584872, "step": 6280}, {"loss": 1.086, "grad_norm": 0.7863979339599609, "learning_rate": 0.0002, "epoch": 4.553022077452045, "step": 6290}, {"loss": 1.2008, "grad_norm": 0.9071474671363831, "learning_rate": 0.0002, "epoch": 4.5602605863192185, "step": 6300}, {"loss": 1.0916, "grad_norm": 0.7429424524307251, "learning_rate": 0.0002, "epoch": 4.567499095186392, "step": 6310}, {"loss": 1.095, "grad_norm": 1.0767850875854492, "learning_rate": 0.0002, "epoch": 4.574737604053565, "step": 6320}, {"loss": 1.1023, "grad_norm": 0.7885915637016296, "learning_rate": 0.0002, "epoch": 4.581976112920739, "step": 6330}, {"loss": 1.1131, "grad_norm": 0.8350457549095154, "learning_rate": 0.0002, "epoch": 4.589214621787912, "step": 6340}, {"loss": 1.0743, "grad_norm": 0.7853530645370483, "learning_rate": 0.0002, "epoch": 4.596453130655085, "step": 6350}, {"loss": 1.1912, "grad_norm": 1.1220661401748657, "learning_rate": 0.0002, "epoch": 4.603691639522259, "step": 6360}, {"loss": 1.0927, "grad_norm": 0.7959423065185547, "learning_rate": 0.0002, "epoch": 4.610930148389432, "step": 6370}, {"loss": 1.1542, "grad_norm": 0.7782652378082275, "learning_rate": 0.0002, "epoch": 4.618168657256605, "step": 6380}, {"loss": 1.0753, "grad_norm": 0.7882203459739685, "learning_rate": 0.0002, "epoch": 4.625407166123779, "step": 6390}, {"loss": 1.0676, "grad_norm": 0.8841899037361145, "learning_rate": 0.0002, "epoch": 4.632645674990952, "step": 6400}, {"loss": 1.0815, "grad_norm": 0.7936127781867981, "learning_rate": 0.0002, "epoch": 4.6398841838581255, "step": 6410}, {"loss": 1.0198, "grad_norm": 0.9213966131210327, "learning_rate": 0.0002, "epoch": 4.647122692725299, "step": 6420}, {"loss": 0.9872, "grad_norm": 0.9246473908424377, "learning_rate": 0.0002, "epoch": 4.654361201592472, "step": 6430}, {"loss": 1.1309, "grad_norm": 0.766572892665863, "learning_rate": 0.0002, "epoch": 4.661599710459646, "step": 6440}, {"loss": 1.1095, "grad_norm": 0.8596171736717224, "learning_rate": 0.0002, "epoch": 4.668838219326819, "step": 6450}, {"loss": 1.1869, "grad_norm": 0.8482751846313477, "learning_rate": 0.0002, "epoch": 4.676076728193992, "step": 6460}, {"loss": 1.0622, "grad_norm": 1.0826905965805054, "learning_rate": 0.0002, "epoch": 4.683315237061166, "step": 6470}, {"loss": 1.0256, "grad_norm": 1.1048457622528076, "learning_rate": 0.0002, "epoch": 4.690553745928339, "step": 6480}, {"loss": 1.0514, "grad_norm": 0.9429134726524353, "learning_rate": 0.0002, "epoch": 4.697792254795512, "step": 6490}, {"loss": 1.1351, "grad_norm": 0.8587502837181091, "learning_rate": 0.0002, "epoch": 4.705030763662686, "step": 6500}, {"loss": 1.0969, "grad_norm": 1.0387083292007446, "learning_rate": 0.0002, "epoch": 4.712269272529859, "step": 6510}, {"loss": 1.0493, "grad_norm": 0.7471951842308044, "learning_rate": 0.0002, "epoch": 4.7195077813970325, "step": 6520}, {"loss": 1.2632, "grad_norm": 0.8800424933433533, "learning_rate": 0.0002, "epoch": 4.726746290264206, "step": 6530}, {"loss": 1.2126, "grad_norm": 0.8136811852455139, "learning_rate": 0.0002, "epoch": 4.733984799131379, "step": 6540}, {"loss": 1.195, "grad_norm": 0.9910339713096619, "learning_rate": 0.0002, "epoch": 4.741223307998553, "step": 6550}, {"loss": 1.1201, "grad_norm": 1.0679163932800293, "learning_rate": 0.0002, "epoch": 4.748461816865726, "step": 6560}, {"loss": 1.0297, "grad_norm": 0.8468248248100281, "learning_rate": 0.0002, "epoch": 4.755700325732899, "step": 6570}, {"loss": 1.0858, "grad_norm": 0.8771235942840576, "learning_rate": 0.0002, "epoch": 4.762938834600073, "step": 6580}, {"loss": 1.077, "grad_norm": 0.7024846076965332, "learning_rate": 0.0002, "epoch": 4.770177343467246, "step": 6590}, {"loss": 1.0876, "grad_norm": 0.7836683392524719, "learning_rate": 0.0002, "epoch": 4.7774158523344195, "step": 6600}, {"loss": 1.1006, "grad_norm": 0.7717288136482239, "learning_rate": 0.0002, "epoch": 4.784654361201593, "step": 6610}, {"loss": 1.0376, "grad_norm": 0.884183943271637, "learning_rate": 0.0002, "epoch": 4.791892870068766, "step": 6620}, {"loss": 1.1757, "grad_norm": 1.383867621421814, "learning_rate": 0.0002, "epoch": 4.7991313789359396, "step": 6630}, {"loss": 1.0861, "grad_norm": 0.9741523861885071, "learning_rate": 0.0002, "epoch": 4.806369887803113, "step": 6640}, {"loss": 1.0884, "grad_norm": 0.9723693132400513, "learning_rate": 0.0002, "epoch": 4.813608396670286, "step": 6650}, {"loss": 1.2203, "grad_norm": 1.8324809074401855, "learning_rate": 0.0002, "epoch": 4.82084690553746, "step": 6660}, {"loss": 1.0292, "grad_norm": 0.904909074306488, "learning_rate": 0.0002, "epoch": 4.828085414404633, "step": 6670}, {"loss": 1.0349, "grad_norm": 0.7355411648750305, "learning_rate": 0.0002, "epoch": 4.835323923271806, "step": 6680}, {"loss": 1.0793, "grad_norm": 0.8934960961341858, "learning_rate": 0.0002, "epoch": 4.84256243213898, "step": 6690}, {"loss": 1.0375, "grad_norm": 1.4596954584121704, "learning_rate": 0.0002, "epoch": 4.849800941006153, "step": 6700}, {"loss": 1.1065, "grad_norm": 0.8310341238975525, "learning_rate": 0.0002, "epoch": 4.8570394498733265, "step": 6710}, {"loss": 1.1089, "grad_norm": 0.9709894061088562, "learning_rate": 0.0002, "epoch": 4.8642779587405, "step": 6720}, {"loss": 1.0069, "grad_norm": 0.852142333984375, "learning_rate": 0.0002, "epoch": 4.871516467607673, "step": 6730}, {"loss": 1.0507, "grad_norm": 1.0643625259399414, "learning_rate": 0.0002, "epoch": 4.878754976474847, "step": 6740}, {"loss": 1.056, "grad_norm": 0.9419508576393127, "learning_rate": 0.0002, "epoch": 4.88599348534202, "step": 6750}, {"loss": 1.1995, "grad_norm": 1.1818498373031616, "learning_rate": 0.0002, "epoch": 4.893231994209193, "step": 6760}, {"loss": 1.0925, "grad_norm": 0.9369569420814514, "learning_rate": 0.0002, "epoch": 4.900470503076367, "step": 6770}, {"loss": 1.1648, "grad_norm": 0.7012579441070557, "learning_rate": 0.0002, "epoch": 4.90770901194354, "step": 6780}, {"loss": 1.0926, "grad_norm": 0.9109319448471069, "learning_rate": 0.0002, "epoch": 4.914947520810713, "step": 6790}, {"loss": 1.0358, "grad_norm": 0.8077534437179565, "learning_rate": 0.0002, "epoch": 4.922186029677887, "step": 6800}, {"loss": 1.2549, "grad_norm": 0.7571148872375488, "learning_rate": 0.0002, "epoch": 4.92942453854506, "step": 6810}, {"loss": 0.9638, "grad_norm": 0.7325633764266968, "learning_rate": 0.0002, "epoch": 4.9366630474122335, "step": 6820}, {"loss": 1.0128, "grad_norm": 0.8465084433555603, "learning_rate": 0.0002, "epoch": 4.943901556279407, "step": 6830}, {"loss": 1.153, "grad_norm": 0.8753737807273865, "learning_rate": 0.0002, "epoch": 4.95114006514658, "step": 6840}, {"loss": 1.0247, "grad_norm": 0.9421748518943787, "learning_rate": 0.0002, "epoch": 4.958378574013754, "step": 6850}, {"loss": 1.1483, "grad_norm": 0.8245896697044373, "learning_rate": 0.0002, "epoch": 4.965617082880927, "step": 6860}, {"loss": 0.9905, "grad_norm": 0.8823089599609375, "learning_rate": 0.0002, "epoch": 4.9728555917481, "step": 6870}, {"loss": 1.1664, "grad_norm": 0.8406389355659485, "learning_rate": 0.0002, "epoch": 4.980094100615274, "step": 6880}, {"loss": 1.0944, "grad_norm": 0.9732868075370789, "learning_rate": 0.0002, "epoch": 4.987332609482447, "step": 6890}, {"loss": 1.1776, "grad_norm": 2.125141143798828, "learning_rate": 0.0002, "epoch": 4.99457111834962, "step": 6900}, {"eval_loss": 1.445176601409912, "eval_runtime": 27.2351, "eval_samples_per_second": 16.009, "eval_steps_per_second": 2.019, "epoch": 4.999638074556641, "step": 6907}, {"loss": 1.1362, "grad_norm": 0.9465792775154114, "learning_rate": 0.0002, "epoch": 5.001809627216793, "step": 6910}, {"loss": 0.982, "grad_norm": 1.2834891080856323, "learning_rate": 0.0002, "epoch": 5.009048136083966, "step": 6920}, {"loss": 0.9803, "grad_norm": 1.0297378301620483, "learning_rate": 0.0002, "epoch": 5.01628664495114, "step": 6930}, {"loss": 1.0447, "grad_norm": 1.1705161333084106, "learning_rate": 0.0002, "epoch": 5.023525153818313, "step": 6940}, {"loss": 1.0113, "grad_norm": 0.8293961882591248, "learning_rate": 0.0002, "epoch": 5.030763662685486, "step": 6950}, {"loss": 0.9203, "grad_norm": 1.0422210693359375, "learning_rate": 0.0002, "epoch": 5.03800217155266, "step": 6960}, {"loss": 1.0553, "grad_norm": 1.116104245185852, "learning_rate": 0.0002, "epoch": 5.045240680419833, "step": 6970}, {"loss": 0.9011, "grad_norm": 1.5118416547775269, "learning_rate": 0.0002, "epoch": 5.0524791892870065, "step": 6980}, {"loss": 0.9969, "grad_norm": 0.8383979797363281, "learning_rate": 0.0002, "epoch": 5.05971769815418, "step": 6990}, {"loss": 0.9659, "grad_norm": 1.3378649950027466, "learning_rate": 0.0002, "epoch": 5.066956207021353, "step": 7000}, {"loss": 1.0212, "grad_norm": 1.1840510368347168, "learning_rate": 0.0002, "epoch": 5.0741947158885266, "step": 7010}, {"loss": 0.9939, "grad_norm": 1.2354751825332642, "learning_rate": 0.0002, "epoch": 5.0814332247557, "step": 7020}, {"loss": 0.9831, "grad_norm": 1.3830451965332031, "learning_rate": 0.0002, "epoch": 5.088671733622873, "step": 7030}, {"loss": 1.1827, "grad_norm": 0.8101674318313599, "learning_rate": 0.0002, "epoch": 5.095910242490047, "step": 7040}, {"loss": 0.9255, "grad_norm": 0.897982656955719, "learning_rate": 0.0002, "epoch": 5.10314875135722, "step": 7050}, {"loss": 0.8784, "grad_norm": 1.2049678564071655, "learning_rate": 0.0002, "epoch": 5.110387260224393, "step": 7060}, {"loss": 1.0182, "grad_norm": 1.5912116765975952, "learning_rate": 0.0002, "epoch": 5.117625769091567, "step": 7070}, {"loss": 1.0909, "grad_norm": 0.9261530041694641, "learning_rate": 0.0002, "epoch": 5.12486427795874, "step": 7080}, {"loss": 0.9603, "grad_norm": 1.1454812288284302, "learning_rate": 0.0002, "epoch": 5.1321027868259135, "step": 7090}, {"loss": 0.9149, "grad_norm": 1.0049978494644165, "learning_rate": 0.0002, "epoch": 5.139341295693087, "step": 7100}, {"loss": 0.9463, "grad_norm": 1.4513251781463623, "learning_rate": 0.0002, "epoch": 5.14657980456026, "step": 7110}, {"loss": 0.8995, "grad_norm": 0.9800849556922913, "learning_rate": 0.0002, "epoch": 5.153818313427434, "step": 7120}, {"loss": 0.9835, "grad_norm": 0.9698708653450012, "learning_rate": 0.0002, "epoch": 5.161056822294607, "step": 7130}, {"loss": 0.9672, "grad_norm": 1.1126646995544434, "learning_rate": 0.0002, "epoch": 5.16829533116178, "step": 7140}, {"loss": 0.9384, "grad_norm": 0.9248330593109131, "learning_rate": 0.0002, "epoch": 5.175533840028954, "step": 7150}, {"loss": 0.826, "grad_norm": 0.7967255711555481, "learning_rate": 0.0002, "epoch": 5.182772348896127, "step": 7160}, {"loss": 1.0078, "grad_norm": 0.9933333992958069, "learning_rate": 0.0002, "epoch": 5.1900108577633, "step": 7170}, {"loss": 1.0276, "grad_norm": 1.0080649852752686, "learning_rate": 0.0002, "epoch": 5.197249366630474, "step": 7180}, {"loss": 1.0201, "grad_norm": 1.3954921960830688, "learning_rate": 0.0002, "epoch": 5.204487875497647, "step": 7190}, {"loss": 1.0863, "grad_norm": 1.2386271953582764, "learning_rate": 0.0002, "epoch": 5.2117263843648205, "step": 7200}, {"loss": 0.8863, "grad_norm": 1.2379488945007324, "learning_rate": 0.0002, "epoch": 5.218964893231994, "step": 7210}, {"loss": 1.0518, "grad_norm": 0.9882503747940063, "learning_rate": 0.0002, "epoch": 5.226203402099167, "step": 7220}, {"loss": 0.9834, "grad_norm": 1.1728729009628296, "learning_rate": 0.0002, "epoch": 5.233441910966341, "step": 7230}, {"loss": 0.9269, "grad_norm": 0.9849673509597778, "learning_rate": 0.0002, "epoch": 5.240680419833514, "step": 7240}, {"loss": 0.9935, "grad_norm": 1.177639365196228, "learning_rate": 0.0002, "epoch": 5.247918928700687, "step": 7250}, {"loss": 1.0639, "grad_norm": 1.2395055294036865, "learning_rate": 0.0002, "epoch": 5.255157437567861, "step": 7260}, {"loss": 1.0138, "grad_norm": 1.3999171257019043, "learning_rate": 0.0002, "epoch": 5.262395946435034, "step": 7270}, {"loss": 0.9745, "grad_norm": 0.7698732018470764, "learning_rate": 0.0002, "epoch": 5.269634455302207, "step": 7280}, {"loss": 1.0389, "grad_norm": 0.9167453646659851, "learning_rate": 0.0002, "epoch": 5.276872964169381, "step": 7290}, {"loss": 0.9858, "grad_norm": 1.113830804824829, "learning_rate": 0.0002, "epoch": 5.284111473036554, "step": 7300}, {"loss": 0.9577, "grad_norm": 0.9644396901130676, "learning_rate": 0.0002, "epoch": 5.2913499819037275, "step": 7310}, {"loss": 1.0556, "grad_norm": 1.462435007095337, "learning_rate": 0.0002, "epoch": 5.298588490770901, "step": 7320}, {"loss": 0.871, "grad_norm": 0.9406287670135498, "learning_rate": 0.0002, "epoch": 5.305826999638074, "step": 7330}, {"loss": 1.0022, "grad_norm": 0.9698247909545898, "learning_rate": 0.0002, "epoch": 5.313065508505248, "step": 7340}, {"loss": 0.915, "grad_norm": 1.12003755569458, "learning_rate": 0.0002, "epoch": 5.320304017372421, "step": 7350}, {"loss": 0.9838, "grad_norm": 1.598681926727295, "learning_rate": 0.0002, "epoch": 5.327542526239594, "step": 7360}, {"loss": 1.0, "grad_norm": 1.0450010299682617, "learning_rate": 0.0002, "epoch": 5.334781035106768, "step": 7370}, {"loss": 0.9983, "grad_norm": 0.8680008053779602, "learning_rate": 0.0002, "epoch": 5.342019543973941, "step": 7380}, {"loss": 0.9851, "grad_norm": 1.0115476846694946, "learning_rate": 0.0002, "epoch": 5.349258052841114, "step": 7390}, {"loss": 1.0702, "grad_norm": 0.9589748382568359, "learning_rate": 0.0002, "epoch": 5.356496561708288, "step": 7400}, {"loss": 0.9366, "grad_norm": 0.6729998588562012, "learning_rate": 0.0002, "epoch": 5.363735070575461, "step": 7410}, {"loss": 1.0126, "grad_norm": 0.9246699213981628, "learning_rate": 0.0002, "epoch": 5.3709735794426345, "step": 7420}, {"loss": 0.9815, "grad_norm": 1.1266791820526123, "learning_rate": 0.0002, "epoch": 5.378212088309808, "step": 7430}, {"loss": 1.1166, "grad_norm": 1.8056942224502563, "learning_rate": 0.0002, "epoch": 5.385450597176981, "step": 7440}, {"loss": 0.9604, "grad_norm": 0.9802932739257812, "learning_rate": 0.0002, "epoch": 5.392689106044155, "step": 7450}, {"loss": 0.9656, "grad_norm": 1.0504707098007202, "learning_rate": 0.0002, "epoch": 5.399927614911328, "step": 7460}, {"loss": 1.0132, "grad_norm": 1.1915022134780884, "learning_rate": 0.0002, "epoch": 5.407166123778501, "step": 7470}, {"loss": 1.0041, "grad_norm": 1.1856611967086792, "learning_rate": 0.0002, "epoch": 5.414404632645675, "step": 7480}, {"loss": 0.9747, "grad_norm": 1.292152762413025, "learning_rate": 0.0002, "epoch": 5.421643141512848, "step": 7490}, {"loss": 0.9659, "grad_norm": 1.2675740718841553, "learning_rate": 0.0002, "epoch": 5.4288816503800215, "step": 7500}, {"loss": 1.0271, "grad_norm": 1.4034695625305176, "learning_rate": 0.0002, "epoch": 5.436120159247195, "step": 7510}, {"loss": 1.0318, "grad_norm": 0.984588623046875, "learning_rate": 0.0002, "epoch": 5.443358668114368, "step": 7520}, {"loss": 1.0726, "grad_norm": 0.8419108390808105, "learning_rate": 0.0002, "epoch": 5.450597176981542, "step": 7530}, {"loss": 1.0499, "grad_norm": 1.0270143747329712, "learning_rate": 0.0002, "epoch": 5.457835685848715, "step": 7540}, {"loss": 0.9804, "grad_norm": 2.2158689498901367, "learning_rate": 0.0002, "epoch": 5.465074194715888, "step": 7550}, {"loss": 0.9856, "grad_norm": 1.0740524530410767, "learning_rate": 0.0002, "epoch": 5.472312703583062, "step": 7560}, {"loss": 1.0522, "grad_norm": 1.3804482221603394, "learning_rate": 0.0002, "epoch": 5.479551212450235, "step": 7570}, {"loss": 1.0297, "grad_norm": 0.9428979754447937, "learning_rate": 0.0002, "epoch": 5.486789721317408, "step": 7580}, {"loss": 1.0906, "grad_norm": 0.9548295736312866, "learning_rate": 0.0002, "epoch": 5.494028230184582, "step": 7590}, {"loss": 0.8853, "grad_norm": 1.0691065788269043, "learning_rate": 0.0002, "epoch": 5.501266739051755, "step": 7600}, {"loss": 1.0375, "grad_norm": 1.0987380743026733, "learning_rate": 0.0002, "epoch": 5.5085052479189285, "step": 7610}, {"loss": 1.0162, "grad_norm": 0.9483979344367981, "learning_rate": 0.0002, "epoch": 5.515743756786102, "step": 7620}, {"loss": 1.105, "grad_norm": 1.16624915599823, "learning_rate": 0.0002, "epoch": 5.522982265653275, "step": 7630}, {"loss": 0.8695, "grad_norm": 0.8563777208328247, "learning_rate": 0.0002, "epoch": 5.530220774520449, "step": 7640}, {"loss": 0.9297, "grad_norm": 1.268186092376709, "learning_rate": 0.0002, "epoch": 5.537459283387622, "step": 7650}, {"loss": 1.1152, "grad_norm": 1.0752092599868774, "learning_rate": 0.0002, "epoch": 5.544697792254795, "step": 7660}, {"loss": 0.9344, "grad_norm": 1.210389256477356, "learning_rate": 0.0002, "epoch": 5.551936301121969, "step": 7670}, {"loss": 1.0349, "grad_norm": 1.669063925743103, "learning_rate": 0.0002, "epoch": 5.559174809989142, "step": 7680}, {"loss": 0.9833, "grad_norm": 1.038020133972168, "learning_rate": 0.0002, "epoch": 5.566413318856315, "step": 7690}, {"loss": 0.8907, "grad_norm": 1.316673994064331, "learning_rate": 0.0002, "epoch": 5.573651827723489, "step": 7700}, {"loss": 0.9614, "grad_norm": 1.029935359954834, "learning_rate": 0.0002, "epoch": 5.580890336590662, "step": 7710}, {"loss": 1.0409, "grad_norm": 0.9401940703392029, "learning_rate": 0.0002, "epoch": 5.5881288454578355, "step": 7720}, {"loss": 0.9272, "grad_norm": 2.4811816215515137, "learning_rate": 0.0002, "epoch": 5.595367354325009, "step": 7730}, {"loss": 0.992, "grad_norm": 1.0329105854034424, "learning_rate": 0.0002, "epoch": 5.602605863192182, "step": 7740}, {"loss": 0.9493, "grad_norm": 1.479629635810852, "learning_rate": 0.0002, "epoch": 5.609844372059356, "step": 7750}, {"loss": 1.0727, "grad_norm": 1.9232319593429565, "learning_rate": 0.0002, "epoch": 5.617082880926529, "step": 7760}, {"loss": 1.0741, "grad_norm": 1.0055509805679321, "learning_rate": 0.0002, "epoch": 5.624321389793702, "step": 7770}, {"loss": 1.0731, "grad_norm": 1.0037437677383423, "learning_rate": 0.0002, "epoch": 5.631559898660876, "step": 7780}, {"loss": 1.0913, "grad_norm": 1.4245030879974365, "learning_rate": 0.0002, "epoch": 5.638798407528049, "step": 7790}, {"loss": 0.9711, "grad_norm": 1.080687403678894, "learning_rate": 0.0002, "epoch": 5.646036916395222, "step": 7800}, {"loss": 1.0276, "grad_norm": 1.354953408241272, "learning_rate": 0.0002, "epoch": 5.653275425262396, "step": 7810}, {"loss": 1.0534, "grad_norm": 0.8966761231422424, "learning_rate": 0.0002, "epoch": 5.660513934129569, "step": 7820}, {"loss": 1.0662, "grad_norm": 1.0675480365753174, "learning_rate": 0.0002, "epoch": 5.6677524429967425, "step": 7830}, {"loss": 1.1077, "grad_norm": 1.2104216814041138, "learning_rate": 0.0002, "epoch": 5.674990951863916, "step": 7840}, {"loss": 0.9627, "grad_norm": 1.105790376663208, "learning_rate": 0.0002, "epoch": 5.682229460731089, "step": 7850}, {"loss": 1.0483, "grad_norm": 1.0915391445159912, "learning_rate": 0.0002, "epoch": 5.689467969598263, "step": 7860}, {"loss": 1.0291, "grad_norm": 0.8957812786102295, "learning_rate": 0.0002, "epoch": 5.696706478465436, "step": 7870}, {"loss": 0.9785, "grad_norm": 1.9189311265945435, "learning_rate": 0.0002, "epoch": 5.703944987332609, "step": 7880}, {"loss": 1.0076, "grad_norm": 1.0867321491241455, "learning_rate": 0.0002, "epoch": 5.711183496199783, "step": 7890}, {"loss": 1.0236, "grad_norm": 1.0233147144317627, "learning_rate": 0.0002, "epoch": 5.718422005066956, "step": 7900}, {"loss": 0.9872, "grad_norm": 1.16460382938385, "learning_rate": 0.0002, "epoch": 5.7256605139341294, "step": 7910}, {"loss": 1.0762, "grad_norm": 1.1098358631134033, "learning_rate": 0.0002, "epoch": 5.732899022801303, "step": 7920}, {"loss": 0.9937, "grad_norm": 0.8555701375007629, "learning_rate": 0.0002, "epoch": 5.740137531668476, "step": 7930}, {"loss": 1.0081, "grad_norm": 0.9885705709457397, "learning_rate": 0.0002, "epoch": 5.7473760405356495, "step": 7940}, {"loss": 0.9909, "grad_norm": 0.9184203147888184, "learning_rate": 0.0002, "epoch": 5.754614549402823, "step": 7950}, {"loss": 1.0767, "grad_norm": 0.9653698205947876, "learning_rate": 0.0002, "epoch": 5.761853058269996, "step": 7960}, {"loss": 0.9317, "grad_norm": 1.0014251470565796, "learning_rate": 0.0002, "epoch": 5.76909156713717, "step": 7970}, {"loss": 1.0271, "grad_norm": 1.004701018333435, "learning_rate": 0.0002, "epoch": 5.776330076004343, "step": 7980}, {"loss": 1.0397, "grad_norm": 0.950577974319458, "learning_rate": 0.0002, "epoch": 5.783568584871516, "step": 7990}, {"loss": 0.9725, "grad_norm": 1.2986834049224854, "learning_rate": 0.0002, "epoch": 5.79080709373869, "step": 8000}, {"loss": 1.039, "grad_norm": 1.3353424072265625, "learning_rate": 0.0002, "epoch": 5.798045602605863, "step": 8010}, {"loss": 1.0626, "grad_norm": 0.7650562524795532, "learning_rate": 0.0002, "epoch": 5.8052841114730365, "step": 8020}, {"loss": 1.0802, "grad_norm": 1.0156235694885254, "learning_rate": 0.0002, "epoch": 5.81252262034021, "step": 8030}, {"loss": 1.0185, "grad_norm": 1.3092900514602661, "learning_rate": 0.0002, "epoch": 5.819761129207383, "step": 8040}, {"loss": 0.9905, "grad_norm": 1.184428095817566, "learning_rate": 0.0002, "epoch": 5.826999638074557, "step": 8050}, {"loss": 1.0548, "grad_norm": 0.979401707649231, "learning_rate": 0.0002, "epoch": 5.83423814694173, "step": 8060}, {"loss": 0.9721, "grad_norm": 1.3557400703430176, "learning_rate": 0.0002, "epoch": 5.841476655808903, "step": 8070}, {"loss": 1.0235, "grad_norm": 0.8429333567619324, "learning_rate": 0.0002, "epoch": 5.848715164676077, "step": 8080}, {"loss": 0.952, "grad_norm": 1.3167692422866821, "learning_rate": 0.0002, "epoch": 5.85595367354325, "step": 8090}, {"loss": 0.9609, "grad_norm": 0.9750998020172119, "learning_rate": 0.0002, "epoch": 5.863192182410423, "step": 8100}, {"loss": 1.0789, "grad_norm": 1.1869813203811646, "learning_rate": 0.0002, "epoch": 5.870430691277597, "step": 8110}, {"loss": 1.0331, "grad_norm": 1.508615255355835, "learning_rate": 0.0002, "epoch": 5.87766920014477, "step": 8120}, {"loss": 1.0171, "grad_norm": 0.9439908266067505, "learning_rate": 0.0002, "epoch": 5.8849077090119435, "step": 8130}, {"loss": 0.9682, "grad_norm": 0.910508930683136, "learning_rate": 0.0002, "epoch": 5.892146217879117, "step": 8140}, {"loss": 1.0032, "grad_norm": 1.111501932144165, "learning_rate": 0.0002, "epoch": 5.89938472674629, "step": 8150}, {"loss": 1.0266, "grad_norm": 0.726554274559021, "learning_rate": 0.0002, "epoch": 5.906623235613464, "step": 8160}, {"loss": 1.0681, "grad_norm": 1.1084556579589844, "learning_rate": 0.0002, "epoch": 5.913861744480637, "step": 8170}, {"loss": 0.969, "grad_norm": 0.9695167541503906, "learning_rate": 0.0002, "epoch": 5.92110025334781, "step": 8180}, {"loss": 0.9858, "grad_norm": 1.1169592142105103, "learning_rate": 0.0002, "epoch": 5.928338762214984, "step": 8190}, {"loss": 1.0924, "grad_norm": 1.5116780996322632, "learning_rate": 0.0002, "epoch": 5.935577271082157, "step": 8200}, {"loss": 0.878, "grad_norm": 1.0073388814926147, "learning_rate": 0.0002, "epoch": 5.94281577994933, "step": 8210}, {"loss": 1.0462, "grad_norm": 0.9323263168334961, "learning_rate": 0.0002, "epoch": 5.950054288816504, "step": 8220}, {"loss": 1.0291, "grad_norm": 0.9422887563705444, "learning_rate": 0.0002, "epoch": 5.957292797683677, "step": 8230}, {"loss": 0.953, "grad_norm": 0.9691047668457031, "learning_rate": 0.0002, "epoch": 5.9645313065508505, "step": 8240}, {"loss": 0.9842, "grad_norm": 0.9650622606277466, "learning_rate": 0.0002, "epoch": 5.971769815418024, "step": 8250}, {"loss": 0.907, "grad_norm": 1.077958345413208, "learning_rate": 0.0002, "epoch": 5.979008324285197, "step": 8260}, {"loss": 0.9162, "grad_norm": 0.8946306109428406, "learning_rate": 0.0002, "epoch": 5.986246833152371, "step": 8270}, {"loss": 1.0439, "grad_norm": 1.34098219871521, "learning_rate": 0.0002, "epoch": 5.993485342019544, "step": 8280}, {"eval_loss": 1.4714229106903076, "eval_runtime": 26.301, "eval_samples_per_second": 16.577, "eval_steps_per_second": 2.091, "epoch": 6.0, "step": 8289}, {"loss": 1.1403, "grad_norm": 0.9737564325332642, "learning_rate": 0.0002, "epoch": 6.000723850886717, "step": 8290}, {"loss": 0.8875, "grad_norm": 1.2205945253372192, "learning_rate": 0.0002, "epoch": 6.007962359753891, "step": 8300}, {"loss": 0.8623, "grad_norm": 1.3529434204101562, "learning_rate": 0.0002, "epoch": 6.015200868621064, "step": 8310}, {"loss": 0.9427, "grad_norm": 1.2300174236297607, "learning_rate": 0.0002, "epoch": 6.022439377488237, "step": 8320}, {"loss": 0.9322, "grad_norm": 0.9248194098472595, "learning_rate": 0.0002, "epoch": 6.029677886355411, "step": 8330}, {"loss": 0.9302, "grad_norm": 1.1140035390853882, "learning_rate": 0.0002, "epoch": 6.036916395222584, "step": 8340}, {"loss": 0.8255, "grad_norm": 1.2097352743148804, "learning_rate": 0.0002, "epoch": 6.0441549040897575, "step": 8350}, {"loss": 0.8792, "grad_norm": 0.9472483396530151, "learning_rate": 0.0002, "epoch": 6.051393412956931, "step": 8360}, {"loss": 0.8865, "grad_norm": 1.0195368528366089, "learning_rate": 0.0002, "epoch": 6.058631921824104, "step": 8370}, {"loss": 0.8858, "grad_norm": 1.182735562324524, "learning_rate": 0.0002, "epoch": 6.065870430691278, "step": 8380}, {"loss": 0.9455, "grad_norm": 1.1042858362197876, "learning_rate": 0.0002, "epoch": 6.073108939558451, "step": 8390}, {"loss": 0.9723, "grad_norm": 0.8606401085853577, "learning_rate": 0.0002, "epoch": 6.080347448425624, "step": 8400}, {"loss": 0.8436, "grad_norm": 1.1015676259994507, "learning_rate": 0.0002, "epoch": 6.087585957292798, "step": 8410}, {"loss": 0.8845, "grad_norm": 1.690224289894104, "learning_rate": 0.0002, "epoch": 6.094824466159971, "step": 8420}, {"loss": 0.8484, "grad_norm": 1.1928749084472656, "learning_rate": 0.0002, "epoch": 6.1020629750271445, "step": 8430}, {"loss": 0.9546, "grad_norm": 1.0816864967346191, "learning_rate": 0.0002, "epoch": 6.109301483894318, "step": 8440}, {"loss": 0.8286, "grad_norm": 1.1638226509094238, "learning_rate": 0.0002, "epoch": 6.116539992761491, "step": 8450}, {"loss": 0.8749, "grad_norm": 1.3782968521118164, "learning_rate": 0.0002, "epoch": 6.1237785016286646, "step": 8460}, {"loss": 0.7956, "grad_norm": 1.2030094861984253, "learning_rate": 0.0002, "epoch": 6.131017010495838, "step": 8470}, {"loss": 0.8393, "grad_norm": 1.3227659463882446, "learning_rate": 0.0002, "epoch": 6.138255519363011, "step": 8480}, {"loss": 0.9175, "grad_norm": 1.104384422302246, "learning_rate": 0.0002, "epoch": 6.145494028230185, "step": 8490}, {"loss": 0.861, "grad_norm": 1.518805980682373, "learning_rate": 0.0002, "epoch": 6.152732537097358, "step": 8500}, {"loss": 0.9169, "grad_norm": 1.2029093503952026, "learning_rate": 0.0002, "epoch": 6.159971045964531, "step": 8510}, {"loss": 0.8701, "grad_norm": 1.2991217374801636, "learning_rate": 0.0002, "epoch": 6.167209554831705, "step": 8520}, {"loss": 0.9748, "grad_norm": 1.7002956867218018, "learning_rate": 0.0002, "epoch": 6.174448063698878, "step": 8530}, {"loss": 0.8881, "grad_norm": 1.6653581857681274, "learning_rate": 0.0002, "epoch": 6.1816865725660515, "step": 8540}, {"loss": 0.817, "grad_norm": 1.0493303537368774, "learning_rate": 0.0002, "epoch": 6.188925081433225, "step": 8550}, {"loss": 0.8726, "grad_norm": 1.539345622062683, "learning_rate": 0.0002, "epoch": 6.196163590300398, "step": 8560}, {"loss": 0.9452, "grad_norm": 1.2757070064544678, "learning_rate": 0.0002, "epoch": 6.203402099167572, "step": 8570}, {"loss": 0.8773, "grad_norm": 1.2416890859603882, "learning_rate": 0.0002, "epoch": 6.210640608034745, "step": 8580}, {"loss": 0.815, "grad_norm": 1.617621898651123, "learning_rate": 0.0002, "epoch": 6.217879116901918, "step": 8590}, {"loss": 0.9137, "grad_norm": 1.058962106704712, "learning_rate": 0.0002, "epoch": 6.225117625769092, "step": 8600}, {"loss": 0.8164, "grad_norm": 1.1489088535308838, "learning_rate": 0.0002, "epoch": 6.232356134636265, "step": 8610}, {"loss": 0.9476, "grad_norm": 0.9391577243804932, "learning_rate": 0.0002, "epoch": 6.239594643503438, "step": 8620}, {"loss": 0.932, "grad_norm": 1.363706111907959, "learning_rate": 0.0002, "epoch": 6.246833152370612, "step": 8630}, {"loss": 0.8917, "grad_norm": 0.779502809047699, "learning_rate": 0.0002, "epoch": 6.254071661237785, "step": 8640}, {"loss": 0.9196, "grad_norm": 2.000821590423584, "learning_rate": 0.0002, "epoch": 6.2613101701049585, "step": 8650}, {"loss": 0.9794, "grad_norm": 1.1521023511886597, "learning_rate": 0.0002, "epoch": 6.268548678972132, "step": 8660}, {"loss": 0.9147, "grad_norm": 1.3734570741653442, "learning_rate": 0.0002, "epoch": 6.275787187839305, "step": 8670}, {"loss": 0.795, "grad_norm": 0.9550670385360718, "learning_rate": 0.0002, "epoch": 6.283025696706479, "step": 8680}, {"loss": 0.9049, "grad_norm": 0.8937032222747803, "learning_rate": 0.0002, "epoch": 6.290264205573652, "step": 8690}, {"loss": 0.8526, "grad_norm": 1.3352779150009155, "learning_rate": 0.0002, "epoch": 6.297502714440825, "step": 8700}, {"loss": 0.8572, "grad_norm": 1.3057222366333008, "learning_rate": 0.0002, "epoch": 6.304741223307999, "step": 8710}, {"loss": 0.8825, "grad_norm": 0.9078314304351807, "learning_rate": 0.0002, "epoch": 6.311979732175172, "step": 8720}, {"loss": 0.8666, "grad_norm": 1.6663457155227661, "learning_rate": 0.0002, "epoch": 6.319218241042345, "step": 8730}, {"loss": 0.927, "grad_norm": 1.2043739557266235, "learning_rate": 0.0002, "epoch": 6.326456749909519, "step": 8740}, {"loss": 0.8014, "grad_norm": 0.9165967702865601, "learning_rate": 0.0002, "epoch": 6.333695258776692, "step": 8750}, {"loss": 0.9761, "grad_norm": 1.016452670097351, "learning_rate": 0.0002, "epoch": 6.3409337676438655, "step": 8760}, {"loss": 1.022, "grad_norm": 1.2209261655807495, "learning_rate": 0.0002, "epoch": 6.348172276511039, "step": 8770}, {"loss": 0.8012, "grad_norm": 1.3380663394927979, "learning_rate": 0.0002, "epoch": 6.355410785378212, "step": 8780}, {"loss": 0.9553, "grad_norm": 2.3311562538146973, "learning_rate": 0.0002, "epoch": 6.362649294245386, "step": 8790}, {"loss": 0.8676, "grad_norm": 1.0330604314804077, "learning_rate": 0.0002, "epoch": 6.369887803112559, "step": 8800}, {"loss": 0.98, "grad_norm": 0.9655511975288391, "learning_rate": 0.0002, "epoch": 6.377126311979732, "step": 8810}, {"loss": 1.0324, "grad_norm": 1.1065765619277954, "learning_rate": 0.0002, "epoch": 6.384364820846906, "step": 8820}, {"loss": 1.0078, "grad_norm": 1.2631285190582275, "learning_rate": 0.0002, "epoch": 6.391603329714079, "step": 8830}, {"loss": 0.8989, "grad_norm": 0.92459636926651, "learning_rate": 0.0002, "epoch": 6.398841838581252, "step": 8840}, {"loss": 0.8536, "grad_norm": 0.9982633590698242, "learning_rate": 0.0002, "epoch": 6.406080347448426, "step": 8850}, {"loss": 0.8949, "grad_norm": 1.0746768712997437, "learning_rate": 0.0002, "epoch": 6.413318856315599, "step": 8860}, {"loss": 0.8547, "grad_norm": 1.3024073839187622, "learning_rate": 0.0002, "epoch": 6.4205573651827725, "step": 8870}, {"loss": 0.9618, "grad_norm": 1.2764527797698975, "learning_rate": 0.0002, "epoch": 6.427795874049946, "step": 8880}, {"loss": 0.8905, "grad_norm": 0.8318809270858765, "learning_rate": 0.0002, "epoch": 6.435034382917119, "step": 8890}, {"loss": 0.917, "grad_norm": 1.7350783348083496, "learning_rate": 0.0002, "epoch": 6.442272891784293, "step": 8900}, {"loss": 1.0229, "grad_norm": 1.3430488109588623, "learning_rate": 0.0002, "epoch": 6.449511400651466, "step": 8910}, {"loss": 0.9678, "grad_norm": 1.5907495021820068, "learning_rate": 0.0002, "epoch": 6.456749909518639, "step": 8920}, {"loss": 0.9639, "grad_norm": 1.8579202890396118, "learning_rate": 0.0002, "epoch": 6.463988418385813, "step": 8930}, {"loss": 0.9302, "grad_norm": 1.2233413457870483, "learning_rate": 0.0002, "epoch": 6.471226927252986, "step": 8940}, {"loss": 0.9169, "grad_norm": 1.009103775024414, "learning_rate": 0.0002, "epoch": 6.4784654361201595, "step": 8950}, {"loss": 0.8969, "grad_norm": 1.1265181303024292, "learning_rate": 0.0002, "epoch": 6.485703944987333, "step": 8960}, {"loss": 0.8374, "grad_norm": 1.1733338832855225, "learning_rate": 0.0002, "epoch": 6.492942453854506, "step": 8970}, {"loss": 0.8764, "grad_norm": 1.0444518327713013, "learning_rate": 0.0002, "epoch": 6.50018096272168, "step": 8980}, {"loss": 0.9582, "grad_norm": 1.2296479940414429, "learning_rate": 0.0002, "epoch": 6.507419471588853, "step": 8990}, {"loss": 0.8557, "grad_norm": 1.370417833328247, "learning_rate": 0.0002, "epoch": 6.514657980456026, "step": 9000}, {"loss": 0.9787, "grad_norm": 1.4787620306015015, "learning_rate": 0.0002, "epoch": 6.5218964893232, "step": 9010}, {"loss": 0.967, "grad_norm": 0.8550514578819275, "learning_rate": 0.0002, "epoch": 6.529134998190373, "step": 9020}, {"loss": 0.9755, "grad_norm": 1.2327991724014282, "learning_rate": 0.0002, "epoch": 6.536373507057546, "step": 9030}, {"loss": 0.9248, "grad_norm": 1.0915621519088745, "learning_rate": 0.0002, "epoch": 6.54361201592472, "step": 9040}, {"loss": 1.0024, "grad_norm": 1.7243309020996094, "learning_rate": 0.0002, "epoch": 6.550850524791893, "step": 9050}, {"loss": 1.0123, "grad_norm": 0.954359769821167, "learning_rate": 0.0002, "epoch": 6.5580890336590665, "step": 9060}, {"loss": 0.8261, "grad_norm": 1.066051959991455, "learning_rate": 0.0002, "epoch": 6.56532754252624, "step": 9070}, {"loss": 0.944, "grad_norm": 1.200271487236023, "learning_rate": 0.0002, "epoch": 6.572566051393413, "step": 9080}, {"loss": 0.9788, "grad_norm": 1.4331457614898682, "learning_rate": 0.0002, "epoch": 6.579804560260587, "step": 9090}, {"loss": 1.0216, "grad_norm": 1.0892444849014282, "learning_rate": 0.0002, "epoch": 6.58704306912776, "step": 9100}, {"loss": 0.8557, "grad_norm": 1.849726915359497, "learning_rate": 0.0002, "epoch": 6.594281577994933, "step": 9110}, {"loss": 0.9495, "grad_norm": 1.1228708028793335, "learning_rate": 0.0002, "epoch": 6.601520086862107, "step": 9120}, {"loss": 1.0169, "grad_norm": 1.0928595066070557, "learning_rate": 0.0002, "epoch": 6.60875859572928, "step": 9130}, {"loss": 0.9342, "grad_norm": 1.2138155698776245, "learning_rate": 0.0002, "epoch": 6.615997104596453, "step": 9140}, {"loss": 0.8715, "grad_norm": 1.5155235528945923, "learning_rate": 0.0002, "epoch": 6.623235613463627, "step": 9150}, {"loss": 0.9806, "grad_norm": 1.3194212913513184, "learning_rate": 0.0002, "epoch": 6.6304741223308, "step": 9160}, {"loss": 0.8958, "grad_norm": 1.045623779296875, "learning_rate": 0.0002, "epoch": 6.6377126311979735, "step": 9170}, {"loss": 0.8698, "grad_norm": 0.9647570252418518, "learning_rate": 0.0002, "epoch": 6.644951140065147, "step": 9180}, {"loss": 0.8829, "grad_norm": 1.0818220376968384, "learning_rate": 0.0002, "epoch": 6.65218964893232, "step": 9190}, {"loss": 0.9745, "grad_norm": 1.2792822122573853, "learning_rate": 0.0002, "epoch": 6.659428157799494, "step": 9200}, {"loss": 0.8854, "grad_norm": 1.2764191627502441, "learning_rate": 0.0002, "epoch": 6.666666666666667, "step": 9210}, {"loss": 0.9709, "grad_norm": 1.0552066564559937, "learning_rate": 0.0002, "epoch": 6.67390517553384, "step": 9220}, {"loss": 0.8855, "grad_norm": 1.082476019859314, "learning_rate": 0.0002, "epoch": 6.681143684401014, "step": 9230}, {"loss": 0.9779, "grad_norm": 1.3313323259353638, "learning_rate": 0.0002, "epoch": 6.688382193268187, "step": 9240}, {"loss": 1.005, "grad_norm": 1.130048394203186, "learning_rate": 0.0002, "epoch": 6.69562070213536, "step": 9250}, {"loss": 0.9969, "grad_norm": 1.1997296810150146, "learning_rate": 0.0002, "epoch": 6.702859211002534, "step": 9260}, {"loss": 0.8691, "grad_norm": 1.0591834783554077, "learning_rate": 0.0002, "epoch": 6.710097719869707, "step": 9270}, {"loss": 0.9603, "grad_norm": 1.2722901105880737, "learning_rate": 0.0002, "epoch": 6.7173362287368805, "step": 9280}, {"loss": 0.9227, "grad_norm": 1.1150950193405151, "learning_rate": 0.0002, "epoch": 6.724574737604054, "step": 9290}, {"loss": 0.95, "grad_norm": 1.1575992107391357, "learning_rate": 0.0002, "epoch": 6.731813246471227, "step": 9300}, {"loss": 0.9822, "grad_norm": 0.9371691346168518, "learning_rate": 0.0002, "epoch": 6.739051755338401, "step": 9310}, {"loss": 0.9773, "grad_norm": 1.4924226999282837, "learning_rate": 0.0002, "epoch": 6.746290264205574, "step": 9320}, {"loss": 0.969, "grad_norm": 1.1524218320846558, "learning_rate": 0.0002, "epoch": 6.753528773072747, "step": 9330}, {"loss": 0.9271, "grad_norm": 0.9500471949577332, "learning_rate": 0.0002, "epoch": 6.760767281939921, "step": 9340}, {"loss": 0.9029, "grad_norm": 1.2062290906906128, "learning_rate": 0.0002, "epoch": 6.768005790807094, "step": 9350}, {"loss": 0.9121, "grad_norm": 1.212631106376648, "learning_rate": 0.0002, "epoch": 6.7752442996742674, "step": 9360}, {"loss": 0.8486, "grad_norm": 1.9135472774505615, "learning_rate": 0.0002, "epoch": 6.782482808541441, "step": 9370}, {"loss": 0.9332, "grad_norm": 0.9682775139808655, "learning_rate": 0.0002, "epoch": 6.789721317408614, "step": 9380}, {"loss": 0.8548, "grad_norm": 1.1405237913131714, "learning_rate": 0.0002, "epoch": 6.7969598262757875, "step": 9390}, {"loss": 0.8922, "grad_norm": 1.6855751276016235, "learning_rate": 0.0002, "epoch": 6.804198335142961, "step": 9400}, {"loss": 0.9417, "grad_norm": 1.6590169668197632, "learning_rate": 0.0002, "epoch": 6.811436844010134, "step": 9410}, {"loss": 0.868, "grad_norm": 1.8795170783996582, "learning_rate": 0.0002, "epoch": 6.818675352877308, "step": 9420}, {"loss": 0.9142, "grad_norm": 1.1087183952331543, "learning_rate": 0.0002, "epoch": 6.825913861744481, "step": 9430}, {"loss": 1.1427, "grad_norm": 1.4178446531295776, "learning_rate": 0.0002, "epoch": 6.833152370611654, "step": 9440}, {"loss": 0.8325, "grad_norm": 1.0792350769042969, "learning_rate": 0.0002, "epoch": 6.840390879478828, "step": 9450}, {"loss": 1.0078, "grad_norm": 1.2159196138381958, "learning_rate": 0.0002, "epoch": 6.847629388346001, "step": 9460}, {"loss": 0.9536, "grad_norm": 0.9998821020126343, "learning_rate": 0.0002, "epoch": 6.8548678972131745, "step": 9470}, {"loss": 0.9277, "grad_norm": 0.7940687537193298, "learning_rate": 0.0002, "epoch": 6.862106406080348, "step": 9480}, {"loss": 0.8612, "grad_norm": 0.9572826027870178, "learning_rate": 0.0002, "epoch": 6.869344914947521, "step": 9490}, {"loss": 0.9611, "grad_norm": 1.1086537837982178, "learning_rate": 0.0002, "epoch": 6.876583423814694, "step": 9500}, {"loss": 0.9276, "grad_norm": 1.1934887170791626, "learning_rate": 0.0002, "epoch": 6.883821932681867, "step": 9510}, {"loss": 0.8416, "grad_norm": 1.207324504852295, "learning_rate": 0.0002, "epoch": 6.89106044154904, "step": 9520}, {"loss": 0.9378, "grad_norm": 1.1303677558898926, "learning_rate": 0.0002, "epoch": 6.898298950416214, "step": 9530}, {"loss": 0.9599, "grad_norm": 1.4958926439285278, "learning_rate": 0.0002, "epoch": 6.905537459283387, "step": 9540}, {"loss": 0.9365, "grad_norm": 1.2141553163528442, "learning_rate": 0.0002, "epoch": 6.9127759681505605, "step": 9550}, {"loss": 1.0291, "grad_norm": 1.6544346809387207, "learning_rate": 0.0002, "epoch": 6.920014477017734, "step": 9560}, {"loss": 0.8439, "grad_norm": 1.0540320873260498, "learning_rate": 0.0002, "epoch": 6.927252985884907, "step": 9570}, {"loss": 0.9831, "grad_norm": 1.3095581531524658, "learning_rate": 0.0002, "epoch": 6.934491494752081, "step": 9580}, {"loss": 0.8694, "grad_norm": 1.4509341716766357, "learning_rate": 0.0002, "epoch": 6.941730003619254, "step": 9590}, {"loss": 0.983, "grad_norm": 1.1091740131378174, "learning_rate": 0.0002, "epoch": 6.948968512486427, "step": 9600}, {"loss": 0.9126, "grad_norm": 1.102929949760437, "learning_rate": 0.0002, "epoch": 6.956207021353601, "step": 9610}, {"loss": 0.9622, "grad_norm": 1.1377743482589722, "learning_rate": 0.0002, "epoch": 6.963445530220774, "step": 9620}, {"loss": 0.9045, "grad_norm": 1.2070361375808716, "learning_rate": 0.0002, "epoch": 6.970684039087947, "step": 9630}, {"loss": 0.9714, "grad_norm": 1.30153489112854, "learning_rate": 0.0002, "epoch": 6.977922547955121, "step": 9640}, {"loss": 0.9555, "grad_norm": 1.4641543626785278, "learning_rate": 0.0002, "epoch": 6.985161056822294, "step": 9650}, {"loss": 0.9177, "grad_norm": 1.0497819185256958, "learning_rate": 0.0002, "epoch": 6.9923995656894675, "step": 9660}, {"loss": 0.8369, "grad_norm": 1.2500354051589966, "learning_rate": 0.0002, "epoch": 6.999638074556641, "step": 9670}, {"eval_loss": 1.518465518951416, "eval_runtime": 26.4525, "eval_samples_per_second": 16.482, "eval_steps_per_second": 2.079, "epoch": 6.999638074556641, "step": 9670}, {"loss": 0.7792, "grad_norm": 1.0240943431854248, "learning_rate": 0.0002, "epoch": 7.006876583423814, "step": 9680}, {"loss": 0.7812, "grad_norm": 1.2250308990478516, "learning_rate": 0.0002, "epoch": 7.014115092290988, "step": 9690}, {"loss": 0.8463, "grad_norm": 1.397510290145874, "learning_rate": 0.0002, "epoch": 7.021353601158161, "step": 9700}, {"loss": 0.738, "grad_norm": 1.9754822254180908, "learning_rate": 0.0002, "epoch": 7.028592110025334, "step": 9710}, {"loss": 0.8017, "grad_norm": 1.7932360172271729, "learning_rate": 0.0002, "epoch": 7.035830618892508, "step": 9720}, {"loss": 0.7427, "grad_norm": 0.8552590608596802, "learning_rate": 0.0002, "epoch": 7.043069127759681, "step": 9730}, {"loss": 0.8204, "grad_norm": 1.758886694908142, "learning_rate": 0.0002, "epoch": 7.0503076366268544, "step": 9740}, {"loss": 0.8444, "grad_norm": 1.5239284038543701, "learning_rate": 0.0002, "epoch": 7.057546145494028, "step": 9750}, {"loss": 0.9078, "grad_norm": 1.2506821155548096, "learning_rate": 0.0002, "epoch": 7.064784654361201, "step": 9760}, {"loss": 0.7512, "grad_norm": 1.274202823638916, "learning_rate": 0.0002, "epoch": 7.0720231632283745, "step": 9770}, {"loss": 0.7693, "grad_norm": 1.296419620513916, "learning_rate": 0.0002, "epoch": 7.079261672095548, "step": 9780}, {"loss": 0.8311, "grad_norm": 1.3418468236923218, "learning_rate": 0.0002, "epoch": 7.086500180962721, "step": 9790}, {"loss": 0.8302, "grad_norm": 1.1935685873031616, "learning_rate": 0.0002, "epoch": 7.093738689829895, "step": 9800}, {"loss": 0.8037, "grad_norm": 1.2341830730438232, "learning_rate": 0.0002, "epoch": 7.100977198697068, "step": 9810}, {"loss": 0.7887, "grad_norm": 1.3398581743240356, "learning_rate": 0.0002, "epoch": 7.108215707564241, "step": 9820}, {"loss": 0.824, "grad_norm": 1.1919665336608887, "learning_rate": 0.0002, "epoch": 7.115454216431415, "step": 9830}, {"loss": 0.7296, "grad_norm": 0.9331274032592773, "learning_rate": 0.0002, "epoch": 7.122692725298588, "step": 9840}, {"loss": 0.8208, "grad_norm": 1.0933221578598022, "learning_rate": 0.0002, "epoch": 7.1299312341657615, "step": 9850}, {"loss": 0.7717, "grad_norm": 1.0350896120071411, "learning_rate": 0.0002, "epoch": 7.137169743032935, "step": 9860}, {"loss": 0.903, "grad_norm": 1.334342360496521, "learning_rate": 0.0002, "epoch": 7.144408251900108, "step": 9870}, {"loss": 0.7759, "grad_norm": 1.5271754264831543, "learning_rate": 0.0002, "epoch": 7.151646760767282, "step": 9880}, {"loss": 0.8223, "grad_norm": 1.0453001260757446, "learning_rate": 0.0002, "epoch": 7.158885269634455, "step": 9890}, {"loss": 0.8334, "grad_norm": 1.204174518585205, "learning_rate": 0.0002, "epoch": 7.166123778501628, "step": 9900}, {"loss": 0.82, "grad_norm": 1.0774344205856323, "learning_rate": 0.0002, "epoch": 7.173362287368802, "step": 9910}, {"loss": 0.7772, "grad_norm": 1.282188892364502, "learning_rate": 0.0002, "epoch": 7.180600796235975, "step": 9920}, {"loss": 0.7619, "grad_norm": 1.1413695812225342, "learning_rate": 0.0002, "epoch": 7.187839305103148, "step": 9930}, {"loss": 0.7194, "grad_norm": 1.2970763444900513, "learning_rate": 0.0002, "epoch": 7.195077813970322, "step": 9940}, {"loss": 0.857, "grad_norm": 1.2535417079925537, "learning_rate": 0.0002, "epoch": 7.202316322837495, "step": 9950}, {"loss": 0.839, "grad_norm": 1.3520581722259521, "learning_rate": 0.0002, "epoch": 7.2095548317046685, "step": 9960}, {"loss": 0.8308, "grad_norm": 1.288572072982788, "learning_rate": 0.0002, "epoch": 7.216793340571842, "step": 9970}, {"loss": 0.8102, "grad_norm": 1.4298021793365479, "learning_rate": 0.0002, "epoch": 7.224031849439015, "step": 9980}, {"loss": 0.9437, "grad_norm": 1.4797194004058838, "learning_rate": 0.0002, "epoch": 7.231270358306189, "step": 9990}, {"loss": 0.8486, "grad_norm": 1.5020978450775146, "learning_rate": 0.0002, "epoch": 7.238508867173362, "step": 10000}, {"loss": 0.8584, "grad_norm": 1.1417840719223022, "learning_rate": 0.0002, "epoch": 7.245747376040535, "step": 10010}, {"loss": 0.8547, "grad_norm": 1.746782898902893, "learning_rate": 0.0002, "epoch": 7.252985884907709, "step": 10020}, {"loss": 0.8721, "grad_norm": 1.019222617149353, "learning_rate": 0.0002, "epoch": 7.260224393774882, "step": 10030}, {"loss": 0.836, "grad_norm": 1.3712849617004395, "learning_rate": 0.0002, "epoch": 7.267462902642055, "step": 10040}, {"loss": 0.9228, "grad_norm": 1.5264288187026978, "learning_rate": 0.0002, "epoch": 7.274701411509229, "step": 10050}, {"loss": 0.8803, "grad_norm": 1.2784953117370605, "learning_rate": 0.0002, "epoch": 7.281939920376402, "step": 10060}, {"loss": 0.9144, "grad_norm": 1.0246731042861938, "learning_rate": 0.0002, "epoch": 7.2891784292435755, "step": 10070}, {"loss": 0.7846, "grad_norm": 1.2060108184814453, "learning_rate": 0.0002, "epoch": 7.296416938110749, "step": 10080}, {"loss": 0.8715, "grad_norm": 1.0908410549163818, "learning_rate": 0.0002, "epoch": 7.303655446977922, "step": 10090}, {"loss": 0.7516, "grad_norm": 1.2308661937713623, "learning_rate": 0.0002, "epoch": 7.310893955845096, "step": 10100}, {"loss": 0.7477, "grad_norm": 1.185610055923462, "learning_rate": 0.0002, "epoch": 7.318132464712269, "step": 10110}, {"loss": 0.8495, "grad_norm": 1.0026527643203735, "learning_rate": 0.0002, "epoch": 7.325370973579442, "step": 10120}, {"loss": 0.9155, "grad_norm": 1.3346470594406128, "learning_rate": 0.0002, "epoch": 7.332609482446616, "step": 10130}, {"loss": 0.975, "grad_norm": 1.5946321487426758, "learning_rate": 0.0002, "epoch": 7.339847991313789, "step": 10140}, {"loss": 0.7774, "grad_norm": 1.3622175455093384, "learning_rate": 0.0002, "epoch": 7.347086500180962, "step": 10150}, {"loss": 0.8455, "grad_norm": 1.0937085151672363, "learning_rate": 0.0002, "epoch": 7.354325009048136, "step": 10160}, {"loss": 0.8548, "grad_norm": 1.6057474613189697, "learning_rate": 0.0002, "epoch": 7.361563517915309, "step": 10170}, {"loss": 0.8345, "grad_norm": 1.234887719154358, "learning_rate": 0.0002, "epoch": 7.3688020267824825, "step": 10180}, {"loss": 0.9041, "grad_norm": 1.2238616943359375, "learning_rate": 0.0002, "epoch": 7.376040535649656, "step": 10190}, {"loss": 0.8568, "grad_norm": 1.2640055418014526, "learning_rate": 0.0002, "epoch": 7.383279044516829, "step": 10200}, {"loss": 0.8233, "grad_norm": 1.2917805910110474, "learning_rate": 0.0002, "epoch": 7.390517553384003, "step": 10210}, {"loss": 0.8748, "grad_norm": 1.096583366394043, "learning_rate": 0.0002, "epoch": 7.397756062251176, "step": 10220}, {"loss": 0.7882, "grad_norm": 1.1854201555252075, "learning_rate": 0.0002, "epoch": 7.404994571118349, "step": 10230}, {"loss": 0.7746, "grad_norm": 1.2318766117095947, "learning_rate": 0.0002, "epoch": 7.412233079985523, "step": 10240}, {"loss": 0.9144, "grad_norm": 1.395302414894104, "learning_rate": 0.0002, "epoch": 7.419471588852696, "step": 10250}, {"loss": 0.7673, "grad_norm": 1.118148922920227, "learning_rate": 0.0002, "epoch": 7.4267100977198695, "step": 10260}, {"loss": 0.8743, "grad_norm": 1.1969468593597412, "learning_rate": 0.0002, "epoch": 7.433948606587043, "step": 10270}, {"loss": 0.8955, "grad_norm": 1.434050440788269, "learning_rate": 0.0002, "epoch": 7.441187115454216, "step": 10280}, {"loss": 0.7716, "grad_norm": 1.2344770431518555, "learning_rate": 0.0002, "epoch": 7.4484256243213895, "step": 10290}, {"loss": 0.801, "grad_norm": 1.2186434268951416, "learning_rate": 0.0002, "epoch": 7.455664133188563, "step": 10300}, {"loss": 0.8937, "grad_norm": 1.482475757598877, "learning_rate": 0.0002, "epoch": 7.462902642055736, "step": 10310}, {"loss": 0.7954, "grad_norm": 1.8391777276992798, "learning_rate": 0.0002, "epoch": 7.47014115092291, "step": 10320}, {"loss": 0.8711, "grad_norm": 1.9489128589630127, "learning_rate": 0.0002, "epoch": 7.477379659790083, "step": 10330}, {"loss": 0.8455, "grad_norm": 1.369743824005127, "learning_rate": 0.0002, "epoch": 7.484618168657256, "step": 10340}, {"loss": 0.9501, "grad_norm": 1.3188602924346924, "learning_rate": 0.0002, "epoch": 7.49185667752443, "step": 10350}, {"loss": 0.8293, "grad_norm": 1.1885292530059814, "learning_rate": 0.0002, "epoch": 7.499095186391603, "step": 10360}, {"loss": 0.8832, "grad_norm": 1.4873403310775757, "learning_rate": 0.0002, "epoch": 7.5063336952587765, "step": 10370}, {"loss": 0.8531, "grad_norm": 1.8681598901748657, "learning_rate": 0.0002, "epoch": 7.51357220412595, "step": 10380}, {"loss": 0.8172, "grad_norm": 1.398186445236206, "learning_rate": 0.0002, "epoch": 7.520810712993123, "step": 10390}, {"loss": 0.9239, "grad_norm": 1.272192358970642, "learning_rate": 0.0002, "epoch": 7.528049221860297, "step": 10400}, {"loss": 0.818, "grad_norm": 0.9671797752380371, "learning_rate": 0.0002, "epoch": 7.53528773072747, "step": 10410}, {"loss": 0.8393, "grad_norm": 0.9752382040023804, "learning_rate": 0.0002, "epoch": 7.542526239594643, "step": 10420}, {"loss": 0.8011, "grad_norm": 1.2241966724395752, "learning_rate": 0.0002, "epoch": 7.549764748461817, "step": 10430}, {"loss": 0.8497, "grad_norm": 1.4615166187286377, "learning_rate": 0.0002, "epoch": 7.55700325732899, "step": 10440}, {"loss": 0.8475, "grad_norm": 1.123205542564392, "learning_rate": 0.0002, "epoch": 7.564241766196163, "step": 10450}, {"loss": 0.893, "grad_norm": 1.3798918724060059, "learning_rate": 0.0002, "epoch": 7.571480275063337, "step": 10460}, {"loss": 0.8109, "grad_norm": 1.3772553205490112, "learning_rate": 0.0002, "epoch": 7.57871878393051, "step": 10470}, {"loss": 0.7907, "grad_norm": 1.4591912031173706, "learning_rate": 0.0002, "epoch": 7.5859572927976835, "step": 10480}, {"loss": 0.878, "grad_norm": 1.4248491525650024, "learning_rate": 0.0002, "epoch": 7.593195801664857, "step": 10490}, {"loss": 0.7818, "grad_norm": 1.2663065195083618, "learning_rate": 0.0002, "epoch": 7.60043431053203, "step": 10500}, {"loss": 0.8349, "grad_norm": 1.1095938682556152, "learning_rate": 0.0002, "epoch": 7.607672819399204, "step": 10510}, {"loss": 0.7793, "grad_norm": 1.8462796211242676, "learning_rate": 0.0002, "epoch": 7.614911328266377, "step": 10520}, {"loss": 0.7695, "grad_norm": 1.1936118602752686, "learning_rate": 0.0002, "epoch": 7.62214983713355, "step": 10530}, {"loss": 0.8911, "grad_norm": 1.3520885705947876, "learning_rate": 0.0002, "epoch": 7.629388346000724, "step": 10540}, {"loss": 0.8589, "grad_norm": 1.2915338277816772, "learning_rate": 0.0002, "epoch": 7.636626854867897, "step": 10550}, {"loss": 0.8932, "grad_norm": 1.125656008720398, "learning_rate": 0.0002, "epoch": 7.64386536373507, "step": 10560}, {"loss": 0.8689, "grad_norm": 1.419791579246521, "learning_rate": 0.0002, "epoch": 7.651103872602244, "step": 10570}, {"loss": 0.8113, "grad_norm": 1.2106866836547852, "learning_rate": 0.0002, "epoch": 7.658342381469417, "step": 10580}, {"loss": 0.7383, "grad_norm": 1.359818458557129, "learning_rate": 0.0002, "epoch": 7.6655808903365905, "step": 10590}, {"loss": 0.8986, "grad_norm": 1.3971713781356812, "learning_rate": 0.0002, "epoch": 7.672819399203764, "step": 10600}, {"loss": 0.942, "grad_norm": 1.287888765335083, "learning_rate": 0.0002, "epoch": 7.680057908070937, "step": 10610}, {"loss": 0.8835, "grad_norm": 0.9856569766998291, "learning_rate": 0.0002, "epoch": 7.687296416938111, "step": 10620}, {"loss": 0.8463, "grad_norm": 1.5403797626495361, "learning_rate": 0.0002, "epoch": 7.694534925805284, "step": 10630}, {"loss": 0.7888, "grad_norm": 1.204551339149475, "learning_rate": 0.0002, "epoch": 7.701773434672457, "step": 10640}, {"loss": 0.7582, "grad_norm": 1.3801014423370361, "learning_rate": 0.0002, "epoch": 7.709011943539631, "step": 10650}, {"loss": 0.8715, "grad_norm": 1.3335949182510376, "learning_rate": 0.0002, "epoch": 7.716250452406804, "step": 10660}, {"loss": 0.9412, "grad_norm": 1.1740102767944336, "learning_rate": 0.0002, "epoch": 7.723488961273977, "step": 10670}, {"loss": 0.832, "grad_norm": 1.1663082838058472, "learning_rate": 0.0002, "epoch": 7.730727470141151, "step": 10680}, {"loss": 0.9191, "grad_norm": 1.3149393796920776, "learning_rate": 0.0002, "epoch": 7.737965979008324, "step": 10690}, {"loss": 0.8697, "grad_norm": 1.3169108629226685, "learning_rate": 0.0002, "epoch": 7.7452044878754975, "step": 10700}, {"loss": 0.8623, "grad_norm": 1.4583594799041748, "learning_rate": 0.0002, "epoch": 7.752442996742671, "step": 10710}, {"loss": 0.75, "grad_norm": 1.1077126264572144, "learning_rate": 0.0002, "epoch": 7.759681505609844, "step": 10720}, {"loss": 0.9109, "grad_norm": 1.5475820302963257, "learning_rate": 0.0002, "epoch": 7.766920014477018, "step": 10730}, {"loss": 0.8153, "grad_norm": 1.2319282293319702, "learning_rate": 0.0002, "epoch": 7.774158523344191, "step": 10740}, {"loss": 0.8324, "grad_norm": 0.9938047528266907, "learning_rate": 0.0002, "epoch": 7.781397032211364, "step": 10750}, {"loss": 0.9356, "grad_norm": 1.2498962879180908, "learning_rate": 0.0002, "epoch": 7.788635541078538, "step": 10760}, {"loss": 0.8744, "grad_norm": 2.192695379257202, "learning_rate": 0.0002, "epoch": 7.795874049945711, "step": 10770}, {"loss": 0.9072, "grad_norm": 1.1851826906204224, "learning_rate": 0.0002, "epoch": 7.8031125588128845, "step": 10780}, {"loss": 0.8933, "grad_norm": 1.0591034889221191, "learning_rate": 0.0002, "epoch": 7.810351067680058, "step": 10790}, {"loss": 0.8298, "grad_norm": 0.9350354671478271, "learning_rate": 0.0002, "epoch": 7.817589576547231, "step": 10800}, {"loss": 0.8701, "grad_norm": 1.5080015659332275, "learning_rate": 0.0002, "epoch": 7.8248280854144046, "step": 10810}, {"loss": 0.9315, "grad_norm": 2.136425495147705, "learning_rate": 0.0002, "epoch": 7.832066594281578, "step": 10820}, {"loss": 0.8204, "grad_norm": 1.5646673440933228, "learning_rate": 0.0002, "epoch": 7.839305103148751, "step": 10830}, {"loss": 0.7794, "grad_norm": 1.381301999092102, "learning_rate": 0.0002, "epoch": 7.846543612015925, "step": 10840}, {"loss": 0.9088, "grad_norm": 1.9323210716247559, "learning_rate": 0.0002, "epoch": 7.853782120883098, "step": 10850}, {"loss": 0.8854, "grad_norm": 1.020809531211853, "learning_rate": 0.0002, "epoch": 7.861020629750271, "step": 10860}, {"loss": 0.8144, "grad_norm": 1.1488909721374512, "learning_rate": 0.0002, "epoch": 7.868259138617445, "step": 10870}, {"loss": 0.883, "grad_norm": 1.4068763256072998, "learning_rate": 0.0002, "epoch": 7.875497647484618, "step": 10880}, {"loss": 0.7431, "grad_norm": 0.9201020002365112, "learning_rate": 0.0002, "epoch": 7.8827361563517915, "step": 10890}, {"loss": 0.8358, "grad_norm": 1.3163132667541504, "learning_rate": 0.0002, "epoch": 7.889974665218965, "step": 10900}, {"loss": 0.9908, "grad_norm": 1.65055513381958, "learning_rate": 0.0002, "epoch": 7.897213174086138, "step": 10910}, {"loss": 0.9883, "grad_norm": 1.1068748235702515, "learning_rate": 0.0002, "epoch": 7.904451682953312, "step": 10920}, {"loss": 0.8553, "grad_norm": 1.8744254112243652, "learning_rate": 0.0002, "epoch": 7.911690191820485, "step": 10930}, {"loss": 0.8969, "grad_norm": 1.3279157876968384, "learning_rate": 0.0002, "epoch": 7.918928700687658, "step": 10940}, {"loss": 0.8642, "grad_norm": 1.0890769958496094, "learning_rate": 0.0002, "epoch": 7.926167209554832, "step": 10950}, {"loss": 0.8742, "grad_norm": 1.3951836824417114, "learning_rate": 0.0002, "epoch": 7.933405718422005, "step": 10960}, {"loss": 0.895, "grad_norm": 1.2761356830596924, "learning_rate": 0.0002, "epoch": 7.940644227289178, "step": 10970}, {"loss": 0.8303, "grad_norm": 1.2073882818222046, "learning_rate": 0.0002, "epoch": 7.947882736156352, "step": 10980}, {"loss": 0.8675, "grad_norm": 1.1899374723434448, "learning_rate": 0.0002, "epoch": 7.955121245023525, "step": 10990}, {"loss": 0.8373, "grad_norm": 1.3041194677352905, "learning_rate": 0.0002, "epoch": 7.9623597538906985, "step": 11000}, {"loss": 0.8207, "grad_norm": 1.3564491271972656, "learning_rate": 0.0002, "epoch": 7.969598262757872, "step": 11010}, {"loss": 0.895, "grad_norm": 1.1411082744598389, "learning_rate": 0.0002, "epoch": 7.976836771625045, "step": 11020}, {"loss": 0.811, "grad_norm": 1.1378493309020996, "learning_rate": 0.0002, "epoch": 7.984075280492219, "step": 11030}, {"loss": 0.8257, "grad_norm": 1.5072855949401855, "learning_rate": 0.0002, "epoch": 7.991313789359392, "step": 11040}]}