diff --git a/.gitattributes b/.gitattributes
index 18e44aa3f083b13e87a5a819318d3a5b89be97bc..099707ddd79adc0f11b4b66b516215ead5800837 100644
--- a/.gitattributes
+++ b/.gitattributes
@@ -3422,3 +3422,12 @@ gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0
 gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.8-num-5623-sd-1/checkpoint-8218/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.8-num-5623-sd-1/checkpoint-9392/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.8-num-5623-sd-1/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/tokenizer.json filter=lfs diff=lfs merge=lfs -text
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/README.md b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..830a14f7db2734beb59f320973504e45a3fe87f5
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/adapter_config.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e99bbcd43df1c19d98706c7e3be95c93844c5349
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/adapter_model.safetensors b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..109596fbe768580a1c4f560efa8a5a99f822a8aa
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d6e7766c9aae4d2fd34c0286b639660fa9abfa548c97dd1e795db9b07731b767
+size 29500848
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/README.md b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..830a14f7db2734beb59f320973504e45a3fe87f5
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/adapter_config.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e99bbcd43df1c19d98706c7e3be95c93844c5349
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/adapter_model.safetensors b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..297e33bb55e834deb7246e7a32009e5318049302
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:60cf6b1e50e1c52de15774e4d67a5155c40c04f7dbae8f5de9f3f70f1d4ef2b6
+size 29500848
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/optimizer.pt b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..607badf957d10930e5a9cc329b763d93dac11798
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b8888f98cbb338ac4a4d0abdcc1d2d1620eab5df580a348deafe766fc06acb72
+size 15064314
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/rng_state.pth b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..3f075498ff27fd5d1f0963673cf28eb97b2dc686
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:478113972e174c338abecc051a8c41b59427eb769e48c53edb8caee8216f7141
+size 14244
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/scheduler.pt b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..bfb6114c3284b46ade676f401deff9e55e7dc0f2
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:97aad0a099d88064f47969b65325e965443831cb49c6f63faa162e9b6f09345c
+size 1064
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/special_tokens_map.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/tokenizer.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..f58963a682665634ab180c28667e4faa8cf02ba2
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/tokenizer.model b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/tokenizer_config.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1adb4796c13b8d975555ecec45876ee75d1ae8b7
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/trainer_state.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..a7c14cf10afef52d1b5681477e45093e275e4a7c
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/trainer_state.json
@@ -0,0 +1,1030 @@
+{
+  "best_metric": 0.6257933378219604,
+  "best_model_checkpoint": "outputs-001/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395",
+  "epoch": 3.0,
+  "eval_steps": 10,
+  "global_step": 1395,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.021505376344086023,
+      "grad_norm": 0.9075053930282593,
+      "learning_rate": 0.0002,
+      "loss": 3.4172,
+      "step": 10
+    },
+    {
+      "epoch": 0.043010752688172046,
+      "grad_norm": 1.4321208000183105,
+      "learning_rate": 0.0002,
+      "loss": 2.5888,
+      "step": 20
+    },
+    {
+      "epoch": 0.06451612903225806,
+      "grad_norm": 1.7500602006912231,
+      "learning_rate": 0.0002,
+      "loss": 2.1195,
+      "step": 30
+    },
+    {
+      "epoch": 0.08602150537634409,
+      "grad_norm": 0.7606641054153442,
+      "learning_rate": 0.0002,
+      "loss": 1.9303,
+      "step": 40
+    },
+    {
+      "epoch": 0.10752688172043011,
+      "grad_norm": 1.2754929065704346,
+      "learning_rate": 0.0002,
+      "loss": 1.6112,
+      "step": 50
+    },
+    {
+      "epoch": 0.12903225806451613,
+      "grad_norm": 1.0936230421066284,
+      "learning_rate": 0.0002,
+      "loss": 1.4319,
+      "step": 60
+    },
+    {
+      "epoch": 0.15053763440860216,
+      "grad_norm": 1.144593596458435,
+      "learning_rate": 0.0002,
+      "loss": 1.3568,
+      "step": 70
+    },
+    {
+      "epoch": 0.17204301075268819,
+      "grad_norm": 1.2181956768035889,
+      "learning_rate": 0.0002,
+      "loss": 1.2028,
+      "step": 80
+    },
+    {
+      "epoch": 0.1935483870967742,
+      "grad_norm": 1.1260095834732056,
+      "learning_rate": 0.0002,
+      "loss": 1.1534,
+      "step": 90
+    },
+    {
+      "epoch": 0.21505376344086022,
+      "grad_norm": 1.1155284643173218,
+      "learning_rate": 0.0002,
+      "loss": 1.1089,
+      "step": 100
+    },
+    {
+      "epoch": 0.23655913978494625,
+      "grad_norm": 1.089565396308899,
+      "learning_rate": 0.0002,
+      "loss": 1.0883,
+      "step": 110
+    },
+    {
+      "epoch": 0.25806451612903225,
+      "grad_norm": 0.9833471775054932,
+      "learning_rate": 0.0002,
+      "loss": 1.0814,
+      "step": 120
+    },
+    {
+      "epoch": 0.27956989247311825,
+      "grad_norm": 1.0265629291534424,
+      "learning_rate": 0.0002,
+      "loss": 1.0239,
+      "step": 130
+    },
+    {
+      "epoch": 0.3010752688172043,
+      "grad_norm": 0.9344286322593689,
+      "learning_rate": 0.0002,
+      "loss": 0.9888,
+      "step": 140
+    },
+    {
+      "epoch": 0.3225806451612903,
+      "grad_norm": 0.9883386492729187,
+      "learning_rate": 0.0002,
+      "loss": 1.0043,
+      "step": 150
+    },
+    {
+      "epoch": 0.34408602150537637,
+      "grad_norm": 0.9299277067184448,
+      "learning_rate": 0.0002,
+      "loss": 0.9338,
+      "step": 160
+    },
+    {
+      "epoch": 0.3655913978494624,
+      "grad_norm": 1.390045404434204,
+      "learning_rate": 0.0002,
+      "loss": 0.9432,
+      "step": 170
+    },
+    {
+      "epoch": 0.3870967741935484,
+      "grad_norm": 1.0313078165054321,
+      "learning_rate": 0.0002,
+      "loss": 0.9008,
+      "step": 180
+    },
+    {
+      "epoch": 0.40860215053763443,
+      "grad_norm": 1.1792205572128296,
+      "learning_rate": 0.0002,
+      "loss": 0.9434,
+      "step": 190
+    },
+    {
+      "epoch": 0.43010752688172044,
+      "grad_norm": 1.049809217453003,
+      "learning_rate": 0.0002,
+      "loss": 0.8761,
+      "step": 200
+    },
+    {
+      "epoch": 0.45161290322580644,
+      "grad_norm": 0.990111768245697,
+      "learning_rate": 0.0002,
+      "loss": 0.8709,
+      "step": 210
+    },
+    {
+      "epoch": 0.4731182795698925,
+      "grad_norm": 0.9870412349700928,
+      "learning_rate": 0.0002,
+      "loss": 0.905,
+      "step": 220
+    },
+    {
+      "epoch": 0.4946236559139785,
+      "grad_norm": 0.8557345867156982,
+      "learning_rate": 0.0002,
+      "loss": 0.9129,
+      "step": 230
+    },
+    {
+      "epoch": 0.5161290322580645,
+      "grad_norm": 0.9746861457824707,
+      "learning_rate": 0.0002,
+      "loss": 0.8836,
+      "step": 240
+    },
+    {
+      "epoch": 0.5376344086021505,
+      "grad_norm": 0.9010438323020935,
+      "learning_rate": 0.0002,
+      "loss": 0.873,
+      "step": 250
+    },
+    {
+      "epoch": 0.5591397849462365,
+      "grad_norm": 0.9061082005500793,
+      "learning_rate": 0.0002,
+      "loss": 0.8241,
+      "step": 260
+    },
+    {
+      "epoch": 0.5806451612903226,
+      "grad_norm": 0.9311846494674683,
+      "learning_rate": 0.0002,
+      "loss": 0.8652,
+      "step": 270
+    },
+    {
+      "epoch": 0.6021505376344086,
+      "grad_norm": 0.9140254855155945,
+      "learning_rate": 0.0002,
+      "loss": 0.8256,
+      "step": 280
+    },
+    {
+      "epoch": 0.6236559139784946,
+      "grad_norm": 0.9722253084182739,
+      "learning_rate": 0.0002,
+      "loss": 0.8441,
+      "step": 290
+    },
+    {
+      "epoch": 0.6451612903225806,
+      "grad_norm": 0.8539168238639832,
+      "learning_rate": 0.0002,
+      "loss": 0.8314,
+      "step": 300
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.9053162932395935,
+      "learning_rate": 0.0002,
+      "loss": 0.8528,
+      "step": 310
+    },
+    {
+      "epoch": 0.6881720430107527,
+      "grad_norm": 0.8444252610206604,
+      "learning_rate": 0.0002,
+      "loss": 0.8209,
+      "step": 320
+    },
+    {
+      "epoch": 0.7096774193548387,
+      "grad_norm": 0.8127437829971313,
+      "learning_rate": 0.0002,
+      "loss": 0.8101,
+      "step": 330
+    },
+    {
+      "epoch": 0.7311827956989247,
+      "grad_norm": 0.886555016040802,
+      "learning_rate": 0.0002,
+      "loss": 0.8223,
+      "step": 340
+    },
+    {
+      "epoch": 0.7526881720430108,
+      "grad_norm": 0.8458548784255981,
+      "learning_rate": 0.0002,
+      "loss": 0.8368,
+      "step": 350
+    },
+    {
+      "epoch": 0.7741935483870968,
+      "grad_norm": 0.8683297634124756,
+      "learning_rate": 0.0002,
+      "loss": 0.8295,
+      "step": 360
+    },
+    {
+      "epoch": 0.7956989247311828,
+      "grad_norm": 0.8308405876159668,
+      "learning_rate": 0.0002,
+      "loss": 0.8232,
+      "step": 370
+    },
+    {
+      "epoch": 0.8172043010752689,
+      "grad_norm": 0.8305579423904419,
+      "learning_rate": 0.0002,
+      "loss": 0.7752,
+      "step": 380
+    },
+    {
+      "epoch": 0.8387096774193549,
+      "grad_norm": 0.8545567393302917,
+      "learning_rate": 0.0002,
+      "loss": 0.8267,
+      "step": 390
+    },
+    {
+      "epoch": 0.8602150537634409,
+      "grad_norm": 0.8486055731773376,
+      "learning_rate": 0.0002,
+      "loss": 0.8212,
+      "step": 400
+    },
+    {
+      "epoch": 0.8817204301075269,
+      "grad_norm": 0.8126763105392456,
+      "learning_rate": 0.0002,
+      "loss": 0.743,
+      "step": 410
+    },
+    {
+      "epoch": 0.9032258064516129,
+      "grad_norm": 0.8494045734405518,
+      "learning_rate": 0.0002,
+      "loss": 0.7993,
+      "step": 420
+    },
+    {
+      "epoch": 0.9247311827956989,
+      "grad_norm": 0.7639183402061462,
+      "learning_rate": 0.0002,
+      "loss": 0.8213,
+      "step": 430
+    },
+    {
+      "epoch": 0.946236559139785,
+      "grad_norm": 0.858101487159729,
+      "learning_rate": 0.0002,
+      "loss": 0.8015,
+      "step": 440
+    },
+    {
+      "epoch": 0.967741935483871,
+      "grad_norm": 0.8141381740570068,
+      "learning_rate": 0.0002,
+      "loss": 0.7629,
+      "step": 450
+    },
+    {
+      "epoch": 0.989247311827957,
+      "grad_norm": 0.8072513937950134,
+      "learning_rate": 0.0002,
+      "loss": 0.7357,
+      "step": 460
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.7740864157676697,
+      "eval_runtime": 21.383,
+      "eval_samples_per_second": 15.48,
+      "eval_steps_per_second": 1.964,
+      "step": 465
+    },
+    {
+      "epoch": 1.010752688172043,
+      "grad_norm": 0.8269494771957397,
+      "learning_rate": 0.0002,
+      "loss": 0.7701,
+      "step": 470
+    },
+    {
+      "epoch": 1.032258064516129,
+      "grad_norm": 0.7814009189605713,
+      "learning_rate": 0.0002,
+      "loss": 0.7532,
+      "step": 480
+    },
+    {
+      "epoch": 1.053763440860215,
+      "grad_norm": 0.8183923363685608,
+      "learning_rate": 0.0002,
+      "loss": 0.7689,
+      "step": 490
+    },
+    {
+      "epoch": 1.075268817204301,
+      "grad_norm": 0.8146600723266602,
+      "learning_rate": 0.0002,
+      "loss": 0.765,
+      "step": 500
+    },
+    {
+      "epoch": 1.096774193548387,
+      "grad_norm": 0.8635126352310181,
+      "learning_rate": 0.0002,
+      "loss": 0.7358,
+      "step": 510
+    },
+    {
+      "epoch": 1.118279569892473,
+      "grad_norm": 0.8520359396934509,
+      "learning_rate": 0.0002,
+      "loss": 0.7302,
+      "step": 520
+    },
+    {
+      "epoch": 1.139784946236559,
+      "grad_norm": 0.8026443123817444,
+      "learning_rate": 0.0002,
+      "loss": 0.7492,
+      "step": 530
+    },
+    {
+      "epoch": 1.1612903225806452,
+      "grad_norm": 0.8157258629798889,
+      "learning_rate": 0.0002,
+      "loss": 0.7518,
+      "step": 540
+    },
+    {
+      "epoch": 1.1827956989247312,
+      "grad_norm": 0.9450796246528625,
+      "learning_rate": 0.0002,
+      "loss": 0.7461,
+      "step": 550
+    },
+    {
+      "epoch": 1.2043010752688172,
+      "grad_norm": 0.8859835863113403,
+      "learning_rate": 0.0002,
+      "loss": 0.7128,
+      "step": 560
+    },
+    {
+      "epoch": 1.2258064516129032,
+      "grad_norm": 0.7819921970367432,
+      "learning_rate": 0.0002,
+      "loss": 0.7067,
+      "step": 570
+    },
+    {
+      "epoch": 1.2473118279569892,
+      "grad_norm": 0.7823445796966553,
+      "learning_rate": 0.0002,
+      "loss": 0.7577,
+      "step": 580
+    },
+    {
+      "epoch": 1.2688172043010753,
+      "grad_norm": 0.7931883931159973,
+      "learning_rate": 0.0002,
+      "loss": 0.7358,
+      "step": 590
+    },
+    {
+      "epoch": 1.2903225806451613,
+      "grad_norm": 0.7495734095573425,
+      "learning_rate": 0.0002,
+      "loss": 0.723,
+      "step": 600
+    },
+    {
+      "epoch": 1.3118279569892473,
+      "grad_norm": 0.9272717237472534,
+      "learning_rate": 0.0002,
+      "loss": 0.7386,
+      "step": 610
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 0.7968398332595825,
+      "learning_rate": 0.0002,
+      "loss": 0.7498,
+      "step": 620
+    },
+    {
+      "epoch": 1.3548387096774195,
+      "grad_norm": 0.7813659310340881,
+      "learning_rate": 0.0002,
+      "loss": 0.7635,
+      "step": 630
+    },
+    {
+      "epoch": 1.3763440860215055,
+      "grad_norm": 0.730925977230072,
+      "learning_rate": 0.0002,
+      "loss": 0.6665,
+      "step": 640
+    },
+    {
+      "epoch": 1.3978494623655915,
+      "grad_norm": 0.8011482954025269,
+      "learning_rate": 0.0002,
+      "loss": 0.7037,
+      "step": 650
+    },
+    {
+      "epoch": 1.4193548387096775,
+      "grad_norm": 0.7770085334777832,
+      "learning_rate": 0.0002,
+      "loss": 0.6931,
+      "step": 660
+    },
+    {
+      "epoch": 1.4408602150537635,
+      "grad_norm": 0.7432682514190674,
+      "learning_rate": 0.0002,
+      "loss": 0.6949,
+      "step": 670
+    },
+    {
+      "epoch": 1.4623655913978495,
+      "grad_norm": 0.8820092678070068,
+      "learning_rate": 0.0002,
+      "loss": 0.7444,
+      "step": 680
+    },
+    {
+      "epoch": 1.4838709677419355,
+      "grad_norm": 0.7786208987236023,
+      "learning_rate": 0.0002,
+      "loss": 0.6758,
+      "step": 690
+    },
+    {
+      "epoch": 1.5053763440860215,
+      "grad_norm": 0.7467480301856995,
+      "learning_rate": 0.0002,
+      "loss": 0.6702,
+      "step": 700
+    },
+    {
+      "epoch": 1.5268817204301075,
+      "grad_norm": 0.8147122263908386,
+      "learning_rate": 0.0002,
+      "loss": 0.7107,
+      "step": 710
+    },
+    {
+      "epoch": 1.5483870967741935,
+      "grad_norm": 0.796030580997467,
+      "learning_rate": 0.0002,
+      "loss": 0.7144,
+      "step": 720
+    },
+    {
+      "epoch": 1.5698924731182795,
+      "grad_norm": 0.8776171207427979,
+      "learning_rate": 0.0002,
+      "loss": 0.6936,
+      "step": 730
+    },
+    {
+      "epoch": 1.5913978494623655,
+      "grad_norm": 0.8056126236915588,
+      "learning_rate": 0.0002,
+      "loss": 0.7101,
+      "step": 740
+    },
+    {
+      "epoch": 1.6129032258064515,
+      "grad_norm": 0.8141863346099854,
+      "learning_rate": 0.0002,
+      "loss": 0.7162,
+      "step": 750
+    },
+    {
+      "epoch": 1.6344086021505375,
+      "grad_norm": 0.8100557327270508,
+      "learning_rate": 0.0002,
+      "loss": 0.7088,
+      "step": 760
+    },
+    {
+      "epoch": 1.6559139784946235,
+      "grad_norm": 0.8283200860023499,
+      "learning_rate": 0.0002,
+      "loss": 0.7212,
+      "step": 770
+    },
+    {
+      "epoch": 1.6774193548387095,
+      "grad_norm": 0.800865113735199,
+      "learning_rate": 0.0002,
+      "loss": 0.694,
+      "step": 780
+    },
+    {
+      "epoch": 1.6989247311827957,
+      "grad_norm": 0.8052287697792053,
+      "learning_rate": 0.0002,
+      "loss": 0.7076,
+      "step": 790
+    },
+    {
+      "epoch": 1.7204301075268817,
+      "grad_norm": 0.8619674444198608,
+      "learning_rate": 0.0002,
+      "loss": 0.7257,
+      "step": 800
+    },
+    {
+      "epoch": 1.7419354838709677,
+      "grad_norm": 0.8907215595245361,
+      "learning_rate": 0.0002,
+      "loss": 0.7141,
+      "step": 810
+    },
+    {
+      "epoch": 1.7634408602150538,
+      "grad_norm": 0.6976316571235657,
+      "learning_rate": 0.0002,
+      "loss": 0.7035,
+      "step": 820
+    },
+    {
+      "epoch": 1.7849462365591398,
+      "grad_norm": 0.7533746957778931,
+      "learning_rate": 0.0002,
+      "loss": 0.6916,
+      "step": 830
+    },
+    {
+      "epoch": 1.8064516129032258,
+      "grad_norm": 0.7326804399490356,
+      "learning_rate": 0.0002,
+      "loss": 0.7094,
+      "step": 840
+    },
+    {
+      "epoch": 1.827956989247312,
+      "grad_norm": 0.7782683372497559,
+      "learning_rate": 0.0002,
+      "loss": 0.6891,
+      "step": 850
+    },
+    {
+      "epoch": 1.849462365591398,
+      "grad_norm": 0.7424806356430054,
+      "learning_rate": 0.0002,
+      "loss": 0.6931,
+      "step": 860
+    },
+    {
+      "epoch": 1.870967741935484,
+      "grad_norm": 1.172325611114502,
+      "learning_rate": 0.0002,
+      "loss": 0.7354,
+      "step": 870
+    },
+    {
+      "epoch": 1.89247311827957,
+      "grad_norm": 0.771058201789856,
+      "learning_rate": 0.0002,
+      "loss": 0.6866,
+      "step": 880
+    },
+    {
+      "epoch": 1.913978494623656,
+      "grad_norm": 0.8624904155731201,
+      "learning_rate": 0.0002,
+      "loss": 0.7296,
+      "step": 890
+    },
+    {
+      "epoch": 1.935483870967742,
+      "grad_norm": 0.7062820792198181,
+      "learning_rate": 0.0002,
+      "loss": 0.7233,
+      "step": 900
+    },
+    {
+      "epoch": 1.956989247311828,
+      "grad_norm": 0.7560103535652161,
+      "learning_rate": 0.0002,
+      "loss": 0.6966,
+      "step": 910
+    },
+    {
+      "epoch": 1.978494623655914,
+      "grad_norm": 0.788899838924408,
+      "learning_rate": 0.0002,
+      "loss": 0.69,
+      "step": 920
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.6562113761901855,
+      "learning_rate": 0.0002,
+      "loss": 0.6505,
+      "step": 930
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.6885261535644531,
+      "eval_runtime": 21.4291,
+      "eval_samples_per_second": 15.446,
+      "eval_steps_per_second": 1.96,
+      "step": 930
+    },
+    {
+      "epoch": 2.021505376344086,
+      "grad_norm": 0.8216531872749329,
+      "learning_rate": 0.0002,
+      "loss": 0.6625,
+      "step": 940
+    },
+    {
+      "epoch": 2.043010752688172,
+      "grad_norm": 0.8317142724990845,
+      "learning_rate": 0.0002,
+      "loss": 0.6398,
+      "step": 950
+    },
+    {
+      "epoch": 2.064516129032258,
+      "grad_norm": 0.8446708917617798,
+      "learning_rate": 0.0002,
+      "loss": 0.649,
+      "step": 960
+    },
+    {
+      "epoch": 2.086021505376344,
+      "grad_norm": 0.735055148601532,
+      "learning_rate": 0.0002,
+      "loss": 0.657,
+      "step": 970
+    },
+    {
+      "epoch": 2.10752688172043,
+      "grad_norm": 0.7487243413925171,
+      "learning_rate": 0.0002,
+      "loss": 0.649,
+      "step": 980
+    },
+    {
+      "epoch": 2.129032258064516,
+      "grad_norm": 0.8573887944221497,
+      "learning_rate": 0.0002,
+      "loss": 0.6419,
+      "step": 990
+    },
+    {
+      "epoch": 2.150537634408602,
+      "grad_norm": 0.6284521818161011,
+      "learning_rate": 0.0002,
+      "loss": 0.6431,
+      "step": 1000
+    },
+    {
+      "epoch": 2.172043010752688,
+      "grad_norm": 0.754183292388916,
+      "learning_rate": 0.0002,
+      "loss": 0.6128,
+      "step": 1010
+    },
+    {
+      "epoch": 2.193548387096774,
+      "grad_norm": 0.9445359110832214,
+      "learning_rate": 0.0002,
+      "loss": 0.6253,
+      "step": 1020
+    },
+    {
+      "epoch": 2.21505376344086,
+      "grad_norm": 0.808508038520813,
+      "learning_rate": 0.0002,
+      "loss": 0.605,
+      "step": 1030
+    },
+    {
+      "epoch": 2.236559139784946,
+      "grad_norm": 0.9394679665565491,
+      "learning_rate": 0.0002,
+      "loss": 0.6786,
+      "step": 1040
+    },
+    {
+      "epoch": 2.258064516129032,
+      "grad_norm": 0.8151357769966125,
+      "learning_rate": 0.0002,
+      "loss": 0.6176,
+      "step": 1050
+    },
+    {
+      "epoch": 2.279569892473118,
+      "grad_norm": 0.7909848093986511,
+      "learning_rate": 0.0002,
+      "loss": 0.66,
+      "step": 1060
+    },
+    {
+      "epoch": 2.3010752688172045,
+      "grad_norm": 0.7506507039070129,
+      "learning_rate": 0.0002,
+      "loss": 0.6254,
+      "step": 1070
+    },
+    {
+      "epoch": 2.3225806451612905,
+      "grad_norm": 0.8240520358085632,
+      "learning_rate": 0.0002,
+      "loss": 0.6608,
+      "step": 1080
+    },
+    {
+      "epoch": 2.3440860215053765,
+      "grad_norm": 0.9342400431632996,
+      "learning_rate": 0.0002,
+      "loss": 0.6207,
+      "step": 1090
+    },
+    {
+      "epoch": 2.3655913978494625,
+      "grad_norm": 1.0598735809326172,
+      "learning_rate": 0.0002,
+      "loss": 0.6029,
+      "step": 1100
+    },
+    {
+      "epoch": 2.3870967741935485,
+      "grad_norm": 0.7907650470733643,
+      "learning_rate": 0.0002,
+      "loss": 0.6035,
+      "step": 1110
+    },
+    {
+      "epoch": 2.4086021505376345,
+      "grad_norm": 0.9388798475265503,
+      "learning_rate": 0.0002,
+      "loss": 0.6237,
+      "step": 1120
+    },
+    {
+      "epoch": 2.4301075268817205,
+      "grad_norm": 0.8985419869422913,
+      "learning_rate": 0.0002,
+      "loss": 0.6207,
+      "step": 1130
+    },
+    {
+      "epoch": 2.4516129032258065,
+      "grad_norm": 0.7471932768821716,
+      "learning_rate": 0.0002,
+      "loss": 0.5902,
+      "step": 1140
+    },
+    {
+      "epoch": 2.4731182795698925,
+      "grad_norm": 0.761131763458252,
+      "learning_rate": 0.0002,
+      "loss": 0.6446,
+      "step": 1150
+    },
+    {
+      "epoch": 2.4946236559139785,
+      "grad_norm": 0.7901819348335266,
+      "learning_rate": 0.0002,
+      "loss": 0.6088,
+      "step": 1160
+    },
+    {
+      "epoch": 2.5161290322580645,
+      "grad_norm": 0.9932922720909119,
+      "learning_rate": 0.0002,
+      "loss": 0.6142,
+      "step": 1170
+    },
+    {
+      "epoch": 2.5376344086021505,
+      "grad_norm": 0.7414287328720093,
+      "learning_rate": 0.0002,
+      "loss": 0.6407,
+      "step": 1180
+    },
+    {
+      "epoch": 2.5591397849462365,
+      "grad_norm": 0.8111771941184998,
+      "learning_rate": 0.0002,
+      "loss": 0.6161,
+      "step": 1190
+    },
+    {
+      "epoch": 2.5806451612903225,
+      "grad_norm": 0.7520156502723694,
+      "learning_rate": 0.0002,
+      "loss": 0.6006,
+      "step": 1200
+    },
+    {
+      "epoch": 2.6021505376344085,
+      "grad_norm": 0.9022907018661499,
+      "learning_rate": 0.0002,
+      "loss": 0.615,
+      "step": 1210
+    },
+    {
+      "epoch": 2.6236559139784945,
+      "grad_norm": 0.7746260166168213,
+      "learning_rate": 0.0002,
+      "loss": 0.6211,
+      "step": 1220
+    },
+    {
+      "epoch": 2.6451612903225805,
+      "grad_norm": 0.8482862114906311,
+      "learning_rate": 0.0002,
+      "loss": 0.616,
+      "step": 1230
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 0.7925458550453186,
+      "learning_rate": 0.0002,
+      "loss": 0.6417,
+      "step": 1240
+    },
+    {
+      "epoch": 2.688172043010753,
+      "grad_norm": 0.8369929194450378,
+      "learning_rate": 0.0002,
+      "loss": 0.6187,
+      "step": 1250
+    },
+    {
+      "epoch": 2.709677419354839,
+      "grad_norm": 0.8311542868614197,
+      "learning_rate": 0.0002,
+      "loss": 0.6138,
+      "step": 1260
+    },
+    {
+      "epoch": 2.731182795698925,
+      "grad_norm": 0.7204853296279907,
+      "learning_rate": 0.0002,
+      "loss": 0.5894,
+      "step": 1270
+    },
+    {
+      "epoch": 2.752688172043011,
+      "grad_norm": 0.8447284698486328,
+      "learning_rate": 0.0002,
+      "loss": 0.6325,
+      "step": 1280
+    },
+    {
+      "epoch": 2.774193548387097,
+      "grad_norm": 0.7738404273986816,
+      "learning_rate": 0.0002,
+      "loss": 0.5946,
+      "step": 1290
+    },
+    {
+      "epoch": 2.795698924731183,
+      "grad_norm": 0.8393287062644958,
+      "learning_rate": 0.0002,
+      "loss": 0.5678,
+      "step": 1300
+    },
+    {
+      "epoch": 2.817204301075269,
+      "grad_norm": 0.79121994972229,
+      "learning_rate": 0.0002,
+      "loss": 0.6092,
+      "step": 1310
+    },
+    {
+      "epoch": 2.838709677419355,
+      "grad_norm": 0.7331557869911194,
+      "learning_rate": 0.0002,
+      "loss": 0.5889,
+      "step": 1320
+    },
+    {
+      "epoch": 2.860215053763441,
+      "grad_norm": 0.9593998193740845,
+      "learning_rate": 0.0002,
+      "loss": 0.6048,
+      "step": 1330
+    },
+    {
+      "epoch": 2.881720430107527,
+      "grad_norm": 0.7215158343315125,
+      "learning_rate": 0.0002,
+      "loss": 0.6108,
+      "step": 1340
+    },
+    {
+      "epoch": 2.903225806451613,
+      "grad_norm": 0.840404212474823,
+      "learning_rate": 0.0002,
+      "loss": 0.5897,
+      "step": 1350
+    },
+    {
+      "epoch": 2.924731182795699,
+      "grad_norm": 0.870659351348877,
+      "learning_rate": 0.0002,
+      "loss": 0.6056,
+      "step": 1360
+    },
+    {
+      "epoch": 2.946236559139785,
+      "grad_norm": 0.8744975328445435,
+      "learning_rate": 0.0002,
+      "loss": 0.6205,
+      "step": 1370
+    },
+    {
+      "epoch": 2.967741935483871,
+      "grad_norm": 0.8030612468719482,
+      "learning_rate": 0.0002,
+      "loss": 0.5966,
+      "step": 1380
+    },
+    {
+      "epoch": 2.989247311827957,
+      "grad_norm": 0.825814962387085,
+      "learning_rate": 0.0002,
+      "loss": 0.6004,
+      "step": 1390
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 0.6257933378219604,
+      "eval_runtime": 21.3692,
+      "eval_samples_per_second": 15.49,
+      "eval_steps_per_second": 1.965,
+      "step": 1395
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 3720,
+  "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": 1.704968488747008e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/training_args.bin b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..483b53d4fc1c568a5fc890fa850e3450f390b208
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e4c4fe2be590ed03492316230adb3a1edca3e4066c55f3716c0352d7134c564
+size 5560
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/README.md b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..830a14f7db2734beb59f320973504e45a3fe87f5
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/adapter_config.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e99bbcd43df1c19d98706c7e3be95c93844c5349
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/adapter_model.safetensors b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..dfd77beeaba9bb06ee526b8e540726c081b0c6ea
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e541ab555a1d347ba51089b50ef294e08671de6d00f792adea734418e435599d
+size 29500848
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/optimizer.pt b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..794a25cf13ccb30dc82f107feaf3dddb9fbdb8d4
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:09e1da60b1d2d76754de5deaf49254cf1376561692f13d690f4adeb1394cec4f
+size 15064314
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/rng_state.pth b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..f7dd62630aaeb0615031e83bb54590332decdae8
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9acae9a74e6c3f8a91c318f8eb916624e759416f1baedf1f367b3c835fabf93c
+size 14244
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/scheduler.pt b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..cfa91775459dca484c64b3ca54e66a486cab7330
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab700ed56a8ffed38e7a4929a8f2f7616aeebff9b9062279a518b87700a13d34
+size 1064
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/special_tokens_map.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/tokenizer.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..f58963a682665634ab180c28667e4faa8cf02ba2
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/tokenizer.model b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/tokenizer_config.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1adb4796c13b8d975555ecec45876ee75d1ae8b7
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/trainer_state.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..1bfc74243b202020a163a9bb218285f4f984a6bf
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/trainer_state.json
@@ -0,0 +1,1367 @@
+{
+  "best_metric": 0.5737715363502502,
+  "best_model_checkpoint": "outputs-001/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860",
+  "epoch": 4.0,
+  "eval_steps": 10,
+  "global_step": 1860,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.021505376344086023,
+      "grad_norm": 0.9075053930282593,
+      "learning_rate": 0.0002,
+      "loss": 3.4172,
+      "step": 10
+    },
+    {
+      "epoch": 0.043010752688172046,
+      "grad_norm": 1.4321208000183105,
+      "learning_rate": 0.0002,
+      "loss": 2.5888,
+      "step": 20
+    },
+    {
+      "epoch": 0.06451612903225806,
+      "grad_norm": 1.7500602006912231,
+      "learning_rate": 0.0002,
+      "loss": 2.1195,
+      "step": 30
+    },
+    {
+      "epoch": 0.08602150537634409,
+      "grad_norm": 0.7606641054153442,
+      "learning_rate": 0.0002,
+      "loss": 1.9303,
+      "step": 40
+    },
+    {
+      "epoch": 0.10752688172043011,
+      "grad_norm": 1.2754929065704346,
+      "learning_rate": 0.0002,
+      "loss": 1.6112,
+      "step": 50
+    },
+    {
+      "epoch": 0.12903225806451613,
+      "grad_norm": 1.0936230421066284,
+      "learning_rate": 0.0002,
+      "loss": 1.4319,
+      "step": 60
+    },
+    {
+      "epoch": 0.15053763440860216,
+      "grad_norm": 1.144593596458435,
+      "learning_rate": 0.0002,
+      "loss": 1.3568,
+      "step": 70
+    },
+    {
+      "epoch": 0.17204301075268819,
+      "grad_norm": 1.2181956768035889,
+      "learning_rate": 0.0002,
+      "loss": 1.2028,
+      "step": 80
+    },
+    {
+      "epoch": 0.1935483870967742,
+      "grad_norm": 1.1260095834732056,
+      "learning_rate": 0.0002,
+      "loss": 1.1534,
+      "step": 90
+    },
+    {
+      "epoch": 0.21505376344086022,
+      "grad_norm": 1.1155284643173218,
+      "learning_rate": 0.0002,
+      "loss": 1.1089,
+      "step": 100
+    },
+    {
+      "epoch": 0.23655913978494625,
+      "grad_norm": 1.089565396308899,
+      "learning_rate": 0.0002,
+      "loss": 1.0883,
+      "step": 110
+    },
+    {
+      "epoch": 0.25806451612903225,
+      "grad_norm": 0.9833471775054932,
+      "learning_rate": 0.0002,
+      "loss": 1.0814,
+      "step": 120
+    },
+    {
+      "epoch": 0.27956989247311825,
+      "grad_norm": 1.0265629291534424,
+      "learning_rate": 0.0002,
+      "loss": 1.0239,
+      "step": 130
+    },
+    {
+      "epoch": 0.3010752688172043,
+      "grad_norm": 0.9344286322593689,
+      "learning_rate": 0.0002,
+      "loss": 0.9888,
+      "step": 140
+    },
+    {
+      "epoch": 0.3225806451612903,
+      "grad_norm": 0.9883386492729187,
+      "learning_rate": 0.0002,
+      "loss": 1.0043,
+      "step": 150
+    },
+    {
+      "epoch": 0.34408602150537637,
+      "grad_norm": 0.9299277067184448,
+      "learning_rate": 0.0002,
+      "loss": 0.9338,
+      "step": 160
+    },
+    {
+      "epoch": 0.3655913978494624,
+      "grad_norm": 1.390045404434204,
+      "learning_rate": 0.0002,
+      "loss": 0.9432,
+      "step": 170
+    },
+    {
+      "epoch": 0.3870967741935484,
+      "grad_norm": 1.0313078165054321,
+      "learning_rate": 0.0002,
+      "loss": 0.9008,
+      "step": 180
+    },
+    {
+      "epoch": 0.40860215053763443,
+      "grad_norm": 1.1792205572128296,
+      "learning_rate": 0.0002,
+      "loss": 0.9434,
+      "step": 190
+    },
+    {
+      "epoch": 0.43010752688172044,
+      "grad_norm": 1.049809217453003,
+      "learning_rate": 0.0002,
+      "loss": 0.8761,
+      "step": 200
+    },
+    {
+      "epoch": 0.45161290322580644,
+      "grad_norm": 0.990111768245697,
+      "learning_rate": 0.0002,
+      "loss": 0.8709,
+      "step": 210
+    },
+    {
+      "epoch": 0.4731182795698925,
+      "grad_norm": 0.9870412349700928,
+      "learning_rate": 0.0002,
+      "loss": 0.905,
+      "step": 220
+    },
+    {
+      "epoch": 0.4946236559139785,
+      "grad_norm": 0.8557345867156982,
+      "learning_rate": 0.0002,
+      "loss": 0.9129,
+      "step": 230
+    },
+    {
+      "epoch": 0.5161290322580645,
+      "grad_norm": 0.9746861457824707,
+      "learning_rate": 0.0002,
+      "loss": 0.8836,
+      "step": 240
+    },
+    {
+      "epoch": 0.5376344086021505,
+      "grad_norm": 0.9010438323020935,
+      "learning_rate": 0.0002,
+      "loss": 0.873,
+      "step": 250
+    },
+    {
+      "epoch": 0.5591397849462365,
+      "grad_norm": 0.9061082005500793,
+      "learning_rate": 0.0002,
+      "loss": 0.8241,
+      "step": 260
+    },
+    {
+      "epoch": 0.5806451612903226,
+      "grad_norm": 0.9311846494674683,
+      "learning_rate": 0.0002,
+      "loss": 0.8652,
+      "step": 270
+    },
+    {
+      "epoch": 0.6021505376344086,
+      "grad_norm": 0.9140254855155945,
+      "learning_rate": 0.0002,
+      "loss": 0.8256,
+      "step": 280
+    },
+    {
+      "epoch": 0.6236559139784946,
+      "grad_norm": 0.9722253084182739,
+      "learning_rate": 0.0002,
+      "loss": 0.8441,
+      "step": 290
+    },
+    {
+      "epoch": 0.6451612903225806,
+      "grad_norm": 0.8539168238639832,
+      "learning_rate": 0.0002,
+      "loss": 0.8314,
+      "step": 300
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.9053162932395935,
+      "learning_rate": 0.0002,
+      "loss": 0.8528,
+      "step": 310
+    },
+    {
+      "epoch": 0.6881720430107527,
+      "grad_norm": 0.8444252610206604,
+      "learning_rate": 0.0002,
+      "loss": 0.8209,
+      "step": 320
+    },
+    {
+      "epoch": 0.7096774193548387,
+      "grad_norm": 0.8127437829971313,
+      "learning_rate": 0.0002,
+      "loss": 0.8101,
+      "step": 330
+    },
+    {
+      "epoch": 0.7311827956989247,
+      "grad_norm": 0.886555016040802,
+      "learning_rate": 0.0002,
+      "loss": 0.8223,
+      "step": 340
+    },
+    {
+      "epoch": 0.7526881720430108,
+      "grad_norm": 0.8458548784255981,
+      "learning_rate": 0.0002,
+      "loss": 0.8368,
+      "step": 350
+    },
+    {
+      "epoch": 0.7741935483870968,
+      "grad_norm": 0.8683297634124756,
+      "learning_rate": 0.0002,
+      "loss": 0.8295,
+      "step": 360
+    },
+    {
+      "epoch": 0.7956989247311828,
+      "grad_norm": 0.8308405876159668,
+      "learning_rate": 0.0002,
+      "loss": 0.8232,
+      "step": 370
+    },
+    {
+      "epoch": 0.8172043010752689,
+      "grad_norm": 0.8305579423904419,
+      "learning_rate": 0.0002,
+      "loss": 0.7752,
+      "step": 380
+    },
+    {
+      "epoch": 0.8387096774193549,
+      "grad_norm": 0.8545567393302917,
+      "learning_rate": 0.0002,
+      "loss": 0.8267,
+      "step": 390
+    },
+    {
+      "epoch": 0.8602150537634409,
+      "grad_norm": 0.8486055731773376,
+      "learning_rate": 0.0002,
+      "loss": 0.8212,
+      "step": 400
+    },
+    {
+      "epoch": 0.8817204301075269,
+      "grad_norm": 0.8126763105392456,
+      "learning_rate": 0.0002,
+      "loss": 0.743,
+      "step": 410
+    },
+    {
+      "epoch": 0.9032258064516129,
+      "grad_norm": 0.8494045734405518,
+      "learning_rate": 0.0002,
+      "loss": 0.7993,
+      "step": 420
+    },
+    {
+      "epoch": 0.9247311827956989,
+      "grad_norm": 0.7639183402061462,
+      "learning_rate": 0.0002,
+      "loss": 0.8213,
+      "step": 430
+    },
+    {
+      "epoch": 0.946236559139785,
+      "grad_norm": 0.858101487159729,
+      "learning_rate": 0.0002,
+      "loss": 0.8015,
+      "step": 440
+    },
+    {
+      "epoch": 0.967741935483871,
+      "grad_norm": 0.8141381740570068,
+      "learning_rate": 0.0002,
+      "loss": 0.7629,
+      "step": 450
+    },
+    {
+      "epoch": 0.989247311827957,
+      "grad_norm": 0.8072513937950134,
+      "learning_rate": 0.0002,
+      "loss": 0.7357,
+      "step": 460
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.7740864157676697,
+      "eval_runtime": 21.383,
+      "eval_samples_per_second": 15.48,
+      "eval_steps_per_second": 1.964,
+      "step": 465
+    },
+    {
+      "epoch": 1.010752688172043,
+      "grad_norm": 0.8269494771957397,
+      "learning_rate": 0.0002,
+      "loss": 0.7701,
+      "step": 470
+    },
+    {
+      "epoch": 1.032258064516129,
+      "grad_norm": 0.7814009189605713,
+      "learning_rate": 0.0002,
+      "loss": 0.7532,
+      "step": 480
+    },
+    {
+      "epoch": 1.053763440860215,
+      "grad_norm": 0.8183923363685608,
+      "learning_rate": 0.0002,
+      "loss": 0.7689,
+      "step": 490
+    },
+    {
+      "epoch": 1.075268817204301,
+      "grad_norm": 0.8146600723266602,
+      "learning_rate": 0.0002,
+      "loss": 0.765,
+      "step": 500
+    },
+    {
+      "epoch": 1.096774193548387,
+      "grad_norm": 0.8635126352310181,
+      "learning_rate": 0.0002,
+      "loss": 0.7358,
+      "step": 510
+    },
+    {
+      "epoch": 1.118279569892473,
+      "grad_norm": 0.8520359396934509,
+      "learning_rate": 0.0002,
+      "loss": 0.7302,
+      "step": 520
+    },
+    {
+      "epoch": 1.139784946236559,
+      "grad_norm": 0.8026443123817444,
+      "learning_rate": 0.0002,
+      "loss": 0.7492,
+      "step": 530
+    },
+    {
+      "epoch": 1.1612903225806452,
+      "grad_norm": 0.8157258629798889,
+      "learning_rate": 0.0002,
+      "loss": 0.7518,
+      "step": 540
+    },
+    {
+      "epoch": 1.1827956989247312,
+      "grad_norm": 0.9450796246528625,
+      "learning_rate": 0.0002,
+      "loss": 0.7461,
+      "step": 550
+    },
+    {
+      "epoch": 1.2043010752688172,
+      "grad_norm": 0.8859835863113403,
+      "learning_rate": 0.0002,
+      "loss": 0.7128,
+      "step": 560
+    },
+    {
+      "epoch": 1.2258064516129032,
+      "grad_norm": 0.7819921970367432,
+      "learning_rate": 0.0002,
+      "loss": 0.7067,
+      "step": 570
+    },
+    {
+      "epoch": 1.2473118279569892,
+      "grad_norm": 0.7823445796966553,
+      "learning_rate": 0.0002,
+      "loss": 0.7577,
+      "step": 580
+    },
+    {
+      "epoch": 1.2688172043010753,
+      "grad_norm": 0.7931883931159973,
+      "learning_rate": 0.0002,
+      "loss": 0.7358,
+      "step": 590
+    },
+    {
+      "epoch": 1.2903225806451613,
+      "grad_norm": 0.7495734095573425,
+      "learning_rate": 0.0002,
+      "loss": 0.723,
+      "step": 600
+    },
+    {
+      "epoch": 1.3118279569892473,
+      "grad_norm": 0.9272717237472534,
+      "learning_rate": 0.0002,
+      "loss": 0.7386,
+      "step": 610
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 0.7968398332595825,
+      "learning_rate": 0.0002,
+      "loss": 0.7498,
+      "step": 620
+    },
+    {
+      "epoch": 1.3548387096774195,
+      "grad_norm": 0.7813659310340881,
+      "learning_rate": 0.0002,
+      "loss": 0.7635,
+      "step": 630
+    },
+    {
+      "epoch": 1.3763440860215055,
+      "grad_norm": 0.730925977230072,
+      "learning_rate": 0.0002,
+      "loss": 0.6665,
+      "step": 640
+    },
+    {
+      "epoch": 1.3978494623655915,
+      "grad_norm": 0.8011482954025269,
+      "learning_rate": 0.0002,
+      "loss": 0.7037,
+      "step": 650
+    },
+    {
+      "epoch": 1.4193548387096775,
+      "grad_norm": 0.7770085334777832,
+      "learning_rate": 0.0002,
+      "loss": 0.6931,
+      "step": 660
+    },
+    {
+      "epoch": 1.4408602150537635,
+      "grad_norm": 0.7432682514190674,
+      "learning_rate": 0.0002,
+      "loss": 0.6949,
+      "step": 670
+    },
+    {
+      "epoch": 1.4623655913978495,
+      "grad_norm": 0.8820092678070068,
+      "learning_rate": 0.0002,
+      "loss": 0.7444,
+      "step": 680
+    },
+    {
+      "epoch": 1.4838709677419355,
+      "grad_norm": 0.7786208987236023,
+      "learning_rate": 0.0002,
+      "loss": 0.6758,
+      "step": 690
+    },
+    {
+      "epoch": 1.5053763440860215,
+      "grad_norm": 0.7467480301856995,
+      "learning_rate": 0.0002,
+      "loss": 0.6702,
+      "step": 700
+    },
+    {
+      "epoch": 1.5268817204301075,
+      "grad_norm": 0.8147122263908386,
+      "learning_rate": 0.0002,
+      "loss": 0.7107,
+      "step": 710
+    },
+    {
+      "epoch": 1.5483870967741935,
+      "grad_norm": 0.796030580997467,
+      "learning_rate": 0.0002,
+      "loss": 0.7144,
+      "step": 720
+    },
+    {
+      "epoch": 1.5698924731182795,
+      "grad_norm": 0.8776171207427979,
+      "learning_rate": 0.0002,
+      "loss": 0.6936,
+      "step": 730
+    },
+    {
+      "epoch": 1.5913978494623655,
+      "grad_norm": 0.8056126236915588,
+      "learning_rate": 0.0002,
+      "loss": 0.7101,
+      "step": 740
+    },
+    {
+      "epoch": 1.6129032258064515,
+      "grad_norm": 0.8141863346099854,
+      "learning_rate": 0.0002,
+      "loss": 0.7162,
+      "step": 750
+    },
+    {
+      "epoch": 1.6344086021505375,
+      "grad_norm": 0.8100557327270508,
+      "learning_rate": 0.0002,
+      "loss": 0.7088,
+      "step": 760
+    },
+    {
+      "epoch": 1.6559139784946235,
+      "grad_norm": 0.8283200860023499,
+      "learning_rate": 0.0002,
+      "loss": 0.7212,
+      "step": 770
+    },
+    {
+      "epoch": 1.6774193548387095,
+      "grad_norm": 0.800865113735199,
+      "learning_rate": 0.0002,
+      "loss": 0.694,
+      "step": 780
+    },
+    {
+      "epoch": 1.6989247311827957,
+      "grad_norm": 0.8052287697792053,
+      "learning_rate": 0.0002,
+      "loss": 0.7076,
+      "step": 790
+    },
+    {
+      "epoch": 1.7204301075268817,
+      "grad_norm": 0.8619674444198608,
+      "learning_rate": 0.0002,
+      "loss": 0.7257,
+      "step": 800
+    },
+    {
+      "epoch": 1.7419354838709677,
+      "grad_norm": 0.8907215595245361,
+      "learning_rate": 0.0002,
+      "loss": 0.7141,
+      "step": 810
+    },
+    {
+      "epoch": 1.7634408602150538,
+      "grad_norm": 0.6976316571235657,
+      "learning_rate": 0.0002,
+      "loss": 0.7035,
+      "step": 820
+    },
+    {
+      "epoch": 1.7849462365591398,
+      "grad_norm": 0.7533746957778931,
+      "learning_rate": 0.0002,
+      "loss": 0.6916,
+      "step": 830
+    },
+    {
+      "epoch": 1.8064516129032258,
+      "grad_norm": 0.7326804399490356,
+      "learning_rate": 0.0002,
+      "loss": 0.7094,
+      "step": 840
+    },
+    {
+      "epoch": 1.827956989247312,
+      "grad_norm": 0.7782683372497559,
+      "learning_rate": 0.0002,
+      "loss": 0.6891,
+      "step": 850
+    },
+    {
+      "epoch": 1.849462365591398,
+      "grad_norm": 0.7424806356430054,
+      "learning_rate": 0.0002,
+      "loss": 0.6931,
+      "step": 860
+    },
+    {
+      "epoch": 1.870967741935484,
+      "grad_norm": 1.172325611114502,
+      "learning_rate": 0.0002,
+      "loss": 0.7354,
+      "step": 870
+    },
+    {
+      "epoch": 1.89247311827957,
+      "grad_norm": 0.771058201789856,
+      "learning_rate": 0.0002,
+      "loss": 0.6866,
+      "step": 880
+    },
+    {
+      "epoch": 1.913978494623656,
+      "grad_norm": 0.8624904155731201,
+      "learning_rate": 0.0002,
+      "loss": 0.7296,
+      "step": 890
+    },
+    {
+      "epoch": 1.935483870967742,
+      "grad_norm": 0.7062820792198181,
+      "learning_rate": 0.0002,
+      "loss": 0.7233,
+      "step": 900
+    },
+    {
+      "epoch": 1.956989247311828,
+      "grad_norm": 0.7560103535652161,
+      "learning_rate": 0.0002,
+      "loss": 0.6966,
+      "step": 910
+    },
+    {
+      "epoch": 1.978494623655914,
+      "grad_norm": 0.788899838924408,
+      "learning_rate": 0.0002,
+      "loss": 0.69,
+      "step": 920
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.6562113761901855,
+      "learning_rate": 0.0002,
+      "loss": 0.6505,
+      "step": 930
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.6885261535644531,
+      "eval_runtime": 21.4291,
+      "eval_samples_per_second": 15.446,
+      "eval_steps_per_second": 1.96,
+      "step": 930
+    },
+    {
+      "epoch": 2.021505376344086,
+      "grad_norm": 0.8216531872749329,
+      "learning_rate": 0.0002,
+      "loss": 0.6625,
+      "step": 940
+    },
+    {
+      "epoch": 2.043010752688172,
+      "grad_norm": 0.8317142724990845,
+      "learning_rate": 0.0002,
+      "loss": 0.6398,
+      "step": 950
+    },
+    {
+      "epoch": 2.064516129032258,
+      "grad_norm": 0.8446708917617798,
+      "learning_rate": 0.0002,
+      "loss": 0.649,
+      "step": 960
+    },
+    {
+      "epoch": 2.086021505376344,
+      "grad_norm": 0.735055148601532,
+      "learning_rate": 0.0002,
+      "loss": 0.657,
+      "step": 970
+    },
+    {
+      "epoch": 2.10752688172043,
+      "grad_norm": 0.7487243413925171,
+      "learning_rate": 0.0002,
+      "loss": 0.649,
+      "step": 980
+    },
+    {
+      "epoch": 2.129032258064516,
+      "grad_norm": 0.8573887944221497,
+      "learning_rate": 0.0002,
+      "loss": 0.6419,
+      "step": 990
+    },
+    {
+      "epoch": 2.150537634408602,
+      "grad_norm": 0.6284521818161011,
+      "learning_rate": 0.0002,
+      "loss": 0.6431,
+      "step": 1000
+    },
+    {
+      "epoch": 2.172043010752688,
+      "grad_norm": 0.754183292388916,
+      "learning_rate": 0.0002,
+      "loss": 0.6128,
+      "step": 1010
+    },
+    {
+      "epoch": 2.193548387096774,
+      "grad_norm": 0.9445359110832214,
+      "learning_rate": 0.0002,
+      "loss": 0.6253,
+      "step": 1020
+    },
+    {
+      "epoch": 2.21505376344086,
+      "grad_norm": 0.808508038520813,
+      "learning_rate": 0.0002,
+      "loss": 0.605,
+      "step": 1030
+    },
+    {
+      "epoch": 2.236559139784946,
+      "grad_norm": 0.9394679665565491,
+      "learning_rate": 0.0002,
+      "loss": 0.6786,
+      "step": 1040
+    },
+    {
+      "epoch": 2.258064516129032,
+      "grad_norm": 0.8151357769966125,
+      "learning_rate": 0.0002,
+      "loss": 0.6176,
+      "step": 1050
+    },
+    {
+      "epoch": 2.279569892473118,
+      "grad_norm": 0.7909848093986511,
+      "learning_rate": 0.0002,
+      "loss": 0.66,
+      "step": 1060
+    },
+    {
+      "epoch": 2.3010752688172045,
+      "grad_norm": 0.7506507039070129,
+      "learning_rate": 0.0002,
+      "loss": 0.6254,
+      "step": 1070
+    },
+    {
+      "epoch": 2.3225806451612905,
+      "grad_norm": 0.8240520358085632,
+      "learning_rate": 0.0002,
+      "loss": 0.6608,
+      "step": 1080
+    },
+    {
+      "epoch": 2.3440860215053765,
+      "grad_norm": 0.9342400431632996,
+      "learning_rate": 0.0002,
+      "loss": 0.6207,
+      "step": 1090
+    },
+    {
+      "epoch": 2.3655913978494625,
+      "grad_norm": 1.0598735809326172,
+      "learning_rate": 0.0002,
+      "loss": 0.6029,
+      "step": 1100
+    },
+    {
+      "epoch": 2.3870967741935485,
+      "grad_norm": 0.7907650470733643,
+      "learning_rate": 0.0002,
+      "loss": 0.6035,
+      "step": 1110
+    },
+    {
+      "epoch": 2.4086021505376345,
+      "grad_norm": 0.9388798475265503,
+      "learning_rate": 0.0002,
+      "loss": 0.6237,
+      "step": 1120
+    },
+    {
+      "epoch": 2.4301075268817205,
+      "grad_norm": 0.8985419869422913,
+      "learning_rate": 0.0002,
+      "loss": 0.6207,
+      "step": 1130
+    },
+    {
+      "epoch": 2.4516129032258065,
+      "grad_norm": 0.7471932768821716,
+      "learning_rate": 0.0002,
+      "loss": 0.5902,
+      "step": 1140
+    },
+    {
+      "epoch": 2.4731182795698925,
+      "grad_norm": 0.761131763458252,
+      "learning_rate": 0.0002,
+      "loss": 0.6446,
+      "step": 1150
+    },
+    {
+      "epoch": 2.4946236559139785,
+      "grad_norm": 0.7901819348335266,
+      "learning_rate": 0.0002,
+      "loss": 0.6088,
+      "step": 1160
+    },
+    {
+      "epoch": 2.5161290322580645,
+      "grad_norm": 0.9932922720909119,
+      "learning_rate": 0.0002,
+      "loss": 0.6142,
+      "step": 1170
+    },
+    {
+      "epoch": 2.5376344086021505,
+      "grad_norm": 0.7414287328720093,
+      "learning_rate": 0.0002,
+      "loss": 0.6407,
+      "step": 1180
+    },
+    {
+      "epoch": 2.5591397849462365,
+      "grad_norm": 0.8111771941184998,
+      "learning_rate": 0.0002,
+      "loss": 0.6161,
+      "step": 1190
+    },
+    {
+      "epoch": 2.5806451612903225,
+      "grad_norm": 0.7520156502723694,
+      "learning_rate": 0.0002,
+      "loss": 0.6006,
+      "step": 1200
+    },
+    {
+      "epoch": 2.6021505376344085,
+      "grad_norm": 0.9022907018661499,
+      "learning_rate": 0.0002,
+      "loss": 0.615,
+      "step": 1210
+    },
+    {
+      "epoch": 2.6236559139784945,
+      "grad_norm": 0.7746260166168213,
+      "learning_rate": 0.0002,
+      "loss": 0.6211,
+      "step": 1220
+    },
+    {
+      "epoch": 2.6451612903225805,
+      "grad_norm": 0.8482862114906311,
+      "learning_rate": 0.0002,
+      "loss": 0.616,
+      "step": 1230
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 0.7925458550453186,
+      "learning_rate": 0.0002,
+      "loss": 0.6417,
+      "step": 1240
+    },
+    {
+      "epoch": 2.688172043010753,
+      "grad_norm": 0.8369929194450378,
+      "learning_rate": 0.0002,
+      "loss": 0.6187,
+      "step": 1250
+    },
+    {
+      "epoch": 2.709677419354839,
+      "grad_norm": 0.8311542868614197,
+      "learning_rate": 0.0002,
+      "loss": 0.6138,
+      "step": 1260
+    },
+    {
+      "epoch": 2.731182795698925,
+      "grad_norm": 0.7204853296279907,
+      "learning_rate": 0.0002,
+      "loss": 0.5894,
+      "step": 1270
+    },
+    {
+      "epoch": 2.752688172043011,
+      "grad_norm": 0.8447284698486328,
+      "learning_rate": 0.0002,
+      "loss": 0.6325,
+      "step": 1280
+    },
+    {
+      "epoch": 2.774193548387097,
+      "grad_norm": 0.7738404273986816,
+      "learning_rate": 0.0002,
+      "loss": 0.5946,
+      "step": 1290
+    },
+    {
+      "epoch": 2.795698924731183,
+      "grad_norm": 0.8393287062644958,
+      "learning_rate": 0.0002,
+      "loss": 0.5678,
+      "step": 1300
+    },
+    {
+      "epoch": 2.817204301075269,
+      "grad_norm": 0.79121994972229,
+      "learning_rate": 0.0002,
+      "loss": 0.6092,
+      "step": 1310
+    },
+    {
+      "epoch": 2.838709677419355,
+      "grad_norm": 0.7331557869911194,
+      "learning_rate": 0.0002,
+      "loss": 0.5889,
+      "step": 1320
+    },
+    {
+      "epoch": 2.860215053763441,
+      "grad_norm": 0.9593998193740845,
+      "learning_rate": 0.0002,
+      "loss": 0.6048,
+      "step": 1330
+    },
+    {
+      "epoch": 2.881720430107527,
+      "grad_norm": 0.7215158343315125,
+      "learning_rate": 0.0002,
+      "loss": 0.6108,
+      "step": 1340
+    },
+    {
+      "epoch": 2.903225806451613,
+      "grad_norm": 0.840404212474823,
+      "learning_rate": 0.0002,
+      "loss": 0.5897,
+      "step": 1350
+    },
+    {
+      "epoch": 2.924731182795699,
+      "grad_norm": 0.870659351348877,
+      "learning_rate": 0.0002,
+      "loss": 0.6056,
+      "step": 1360
+    },
+    {
+      "epoch": 2.946236559139785,
+      "grad_norm": 0.8744975328445435,
+      "learning_rate": 0.0002,
+      "loss": 0.6205,
+      "step": 1370
+    },
+    {
+      "epoch": 2.967741935483871,
+      "grad_norm": 0.8030612468719482,
+      "learning_rate": 0.0002,
+      "loss": 0.5966,
+      "step": 1380
+    },
+    {
+      "epoch": 2.989247311827957,
+      "grad_norm": 0.825814962387085,
+      "learning_rate": 0.0002,
+      "loss": 0.6004,
+      "step": 1390
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 0.6257933378219604,
+      "eval_runtime": 21.3692,
+      "eval_samples_per_second": 15.49,
+      "eval_steps_per_second": 1.965,
+      "step": 1395
+    },
+    {
+      "epoch": 3.010752688172043,
+      "grad_norm": 0.8650677800178528,
+      "learning_rate": 0.0002,
+      "loss": 0.5696,
+      "step": 1400
+    },
+    {
+      "epoch": 3.032258064516129,
+      "grad_norm": 0.8364197015762329,
+      "learning_rate": 0.0002,
+      "loss": 0.5483,
+      "step": 1410
+    },
+    {
+      "epoch": 3.053763440860215,
+      "grad_norm": 0.8278448581695557,
+      "learning_rate": 0.0002,
+      "loss": 0.5606,
+      "step": 1420
+    },
+    {
+      "epoch": 3.075268817204301,
+      "grad_norm": 0.8806642889976501,
+      "learning_rate": 0.0002,
+      "loss": 0.5572,
+      "step": 1430
+    },
+    {
+      "epoch": 3.096774193548387,
+      "grad_norm": 0.8180029988288879,
+      "learning_rate": 0.0002,
+      "loss": 0.585,
+      "step": 1440
+    },
+    {
+      "epoch": 3.118279569892473,
+      "grad_norm": 0.8561782836914062,
+      "learning_rate": 0.0002,
+      "loss": 0.5667,
+      "step": 1450
+    },
+    {
+      "epoch": 3.139784946236559,
+      "grad_norm": 0.8377029299736023,
+      "learning_rate": 0.0002,
+      "loss": 0.5246,
+      "step": 1460
+    },
+    {
+      "epoch": 3.161290322580645,
+      "grad_norm": 0.885779082775116,
+      "learning_rate": 0.0002,
+      "loss": 0.5464,
+      "step": 1470
+    },
+    {
+      "epoch": 3.182795698924731,
+      "grad_norm": 0.9388518333435059,
+      "learning_rate": 0.0002,
+      "loss": 0.541,
+      "step": 1480
+    },
+    {
+      "epoch": 3.204301075268817,
+      "grad_norm": 0.8816235661506653,
+      "learning_rate": 0.0002,
+      "loss": 0.5447,
+      "step": 1490
+    },
+    {
+      "epoch": 3.225806451612903,
+      "grad_norm": 0.9885783791542053,
+      "learning_rate": 0.0002,
+      "loss": 0.5466,
+      "step": 1500
+    },
+    {
+      "epoch": 3.247311827956989,
+      "grad_norm": 0.8635850548744202,
+      "learning_rate": 0.0002,
+      "loss": 0.5455,
+      "step": 1510
+    },
+    {
+      "epoch": 3.268817204301075,
+      "grad_norm": 0.829853355884552,
+      "learning_rate": 0.0002,
+      "loss": 0.5419,
+      "step": 1520
+    },
+    {
+      "epoch": 3.2903225806451615,
+      "grad_norm": 0.9037486910820007,
+      "learning_rate": 0.0002,
+      "loss": 0.54,
+      "step": 1530
+    },
+    {
+      "epoch": 3.3118279569892475,
+      "grad_norm": 0.8173713684082031,
+      "learning_rate": 0.0002,
+      "loss": 0.5375,
+      "step": 1540
+    },
+    {
+      "epoch": 3.3333333333333335,
+      "grad_norm": 0.796953022480011,
+      "learning_rate": 0.0002,
+      "loss": 0.5405,
+      "step": 1550
+    },
+    {
+      "epoch": 3.3548387096774195,
+      "grad_norm": 0.7894400358200073,
+      "learning_rate": 0.0002,
+      "loss": 0.5505,
+      "step": 1560
+    },
+    {
+      "epoch": 3.3763440860215055,
+      "grad_norm": 0.9434949159622192,
+      "learning_rate": 0.0002,
+      "loss": 0.5395,
+      "step": 1570
+    },
+    {
+      "epoch": 3.3978494623655915,
+      "grad_norm": 0.8666760325431824,
+      "learning_rate": 0.0002,
+      "loss": 0.5271,
+      "step": 1580
+    },
+    {
+      "epoch": 3.4193548387096775,
+      "grad_norm": 0.7782467007637024,
+      "learning_rate": 0.0002,
+      "loss": 0.5439,
+      "step": 1590
+    },
+    {
+      "epoch": 3.4408602150537635,
+      "grad_norm": 0.8849126696586609,
+      "learning_rate": 0.0002,
+      "loss": 0.5161,
+      "step": 1600
+    },
+    {
+      "epoch": 3.4623655913978495,
+      "grad_norm": 0.7863831520080566,
+      "learning_rate": 0.0002,
+      "loss": 0.5353,
+      "step": 1610
+    },
+    {
+      "epoch": 3.4838709677419355,
+      "grad_norm": 1.0403116941452026,
+      "learning_rate": 0.0002,
+      "loss": 0.5308,
+      "step": 1620
+    },
+    {
+      "epoch": 3.5053763440860215,
+      "grad_norm": 0.8307499289512634,
+      "learning_rate": 0.0002,
+      "loss": 0.5339,
+      "step": 1630
+    },
+    {
+      "epoch": 3.5268817204301075,
+      "grad_norm": 0.9132118821144104,
+      "learning_rate": 0.0002,
+      "loss": 0.5361,
+      "step": 1640
+    },
+    {
+      "epoch": 3.5483870967741935,
+      "grad_norm": 0.9322578310966492,
+      "learning_rate": 0.0002,
+      "loss": 0.5828,
+      "step": 1650
+    },
+    {
+      "epoch": 3.5698924731182795,
+      "grad_norm": 0.9782460331916809,
+      "learning_rate": 0.0002,
+      "loss": 0.546,
+      "step": 1660
+    },
+    {
+      "epoch": 3.5913978494623655,
+      "grad_norm": 0.7189919352531433,
+      "learning_rate": 0.0002,
+      "loss": 0.5424,
+      "step": 1670
+    },
+    {
+      "epoch": 3.6129032258064515,
+      "grad_norm": 0.9689221382141113,
+      "learning_rate": 0.0002,
+      "loss": 0.5514,
+      "step": 1680
+    },
+    {
+      "epoch": 3.6344086021505375,
+      "grad_norm": 0.9684675335884094,
+      "learning_rate": 0.0002,
+      "loss": 0.5379,
+      "step": 1690
+    },
+    {
+      "epoch": 3.6559139784946235,
+      "grad_norm": 0.8851472735404968,
+      "learning_rate": 0.0002,
+      "loss": 0.5748,
+      "step": 1700
+    },
+    {
+      "epoch": 3.6774193548387095,
+      "grad_norm": 0.7709833383560181,
+      "learning_rate": 0.0002,
+      "loss": 0.5412,
+      "step": 1710
+    },
+    {
+      "epoch": 3.698924731182796,
+      "grad_norm": 0.818236231803894,
+      "learning_rate": 0.0002,
+      "loss": 0.521,
+      "step": 1720
+    },
+    {
+      "epoch": 3.720430107526882,
+      "grad_norm": 0.870642364025116,
+      "learning_rate": 0.0002,
+      "loss": 0.5445,
+      "step": 1730
+    },
+    {
+      "epoch": 3.741935483870968,
+      "grad_norm": 1.0245511531829834,
+      "learning_rate": 0.0002,
+      "loss": 0.5307,
+      "step": 1740
+    },
+    {
+      "epoch": 3.763440860215054,
+      "grad_norm": 0.8607558608055115,
+      "learning_rate": 0.0002,
+      "loss": 0.5593,
+      "step": 1750
+    },
+    {
+      "epoch": 3.78494623655914,
+      "grad_norm": 0.8511829972267151,
+      "learning_rate": 0.0002,
+      "loss": 0.536,
+      "step": 1760
+    },
+    {
+      "epoch": 3.806451612903226,
+      "grad_norm": 0.7969087362289429,
+      "learning_rate": 0.0002,
+      "loss": 0.5193,
+      "step": 1770
+    },
+    {
+      "epoch": 3.827956989247312,
+      "grad_norm": 0.8457245826721191,
+      "learning_rate": 0.0002,
+      "loss": 0.5578,
+      "step": 1780
+    },
+    {
+      "epoch": 3.849462365591398,
+      "grad_norm": 0.8893467783927917,
+      "learning_rate": 0.0002,
+      "loss": 0.5337,
+      "step": 1790
+    },
+    {
+      "epoch": 3.870967741935484,
+      "grad_norm": 0.8593819737434387,
+      "learning_rate": 0.0002,
+      "loss": 0.5024,
+      "step": 1800
+    },
+    {
+      "epoch": 3.89247311827957,
+      "grad_norm": 0.7574560642242432,
+      "learning_rate": 0.0002,
+      "loss": 0.5134,
+      "step": 1810
+    },
+    {
+      "epoch": 3.913978494623656,
+      "grad_norm": 0.8681567311286926,
+      "learning_rate": 0.0002,
+      "loss": 0.5263,
+      "step": 1820
+    },
+    {
+      "epoch": 3.935483870967742,
+      "grad_norm": 0.9068132042884827,
+      "learning_rate": 0.0002,
+      "loss": 0.532,
+      "step": 1830
+    },
+    {
+      "epoch": 3.956989247311828,
+      "grad_norm": 0.8668948411941528,
+      "learning_rate": 0.0002,
+      "loss": 0.5427,
+      "step": 1840
+    },
+    {
+      "epoch": 3.978494623655914,
+      "grad_norm": 1.046032428741455,
+      "learning_rate": 0.0002,
+      "loss": 0.5349,
+      "step": 1850
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.904780387878418,
+      "learning_rate": 0.0002,
+      "loss": 0.5087,
+      "step": 1860
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 0.5737715363502502,
+      "eval_runtime": 21.4915,
+      "eval_samples_per_second": 15.401,
+      "eval_steps_per_second": 1.954,
+      "step": 1860
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 3720,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.273291318329344e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/training_args.bin b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..483b53d4fc1c568a5fc890fa850e3450f390b208
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e4c4fe2be590ed03492316230adb3a1edca3e4066c55f3716c0352d7134c564
+size 5560
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/README.md b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..830a14f7db2734beb59f320973504e45a3fe87f5
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/adapter_config.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e99bbcd43df1c19d98706c7e3be95c93844c5349
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/adapter_model.safetensors b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..aa056367b45557b778367b6675f412237c70cea3
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1a3a8816014f6b56e239eee03e09b0769511aa83826ed046f5efd719d015e965
+size 29500848
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/optimizer.pt b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..6abe8edcb5aee11be12a9f19776830275ab66a8c
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a65edbec23b2e333180b69103be1e3e81617ef94611db51c279fd4ecd891f153
+size 15064314
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/rng_state.pth b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..3edef16d50aeba6872f99d589bb61a56ffa1b359
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:76d94a0b8ba8ae587f26bf24336817481dcc65d73576ae598d1102cde342b949
+size 14244
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/scheduler.pt b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..9ce9ee917955ebc932f8a9f74e3434c7f701155b
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2afcae7bbfb81927fb154285fa5a98cc3acb8e8fa78aa6b1ed36a42f6b92d2ba
+size 1064
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/special_tokens_map.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/tokenizer.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..f58963a682665634ab180c28667e4faa8cf02ba2
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/tokenizer.model b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/tokenizer_config.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1adb4796c13b8d975555ecec45876ee75d1ae8b7
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/trainer_state.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..d920ecc7e29eba2359fd6943f13e5ace96af55a0
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/trainer_state.json
@@ -0,0 +1,1697 @@
+{
+  "best_metric": 0.5363914370536804,
+  "best_model_checkpoint": "outputs-001/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325",
+  "epoch": 5.0,
+  "eval_steps": 10,
+  "global_step": 2325,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.021505376344086023,
+      "grad_norm": 0.9075053930282593,
+      "learning_rate": 0.0002,
+      "loss": 3.4172,
+      "step": 10
+    },
+    {
+      "epoch": 0.043010752688172046,
+      "grad_norm": 1.4321208000183105,
+      "learning_rate": 0.0002,
+      "loss": 2.5888,
+      "step": 20
+    },
+    {
+      "epoch": 0.06451612903225806,
+      "grad_norm": 1.7500602006912231,
+      "learning_rate": 0.0002,
+      "loss": 2.1195,
+      "step": 30
+    },
+    {
+      "epoch": 0.08602150537634409,
+      "grad_norm": 0.7606641054153442,
+      "learning_rate": 0.0002,
+      "loss": 1.9303,
+      "step": 40
+    },
+    {
+      "epoch": 0.10752688172043011,
+      "grad_norm": 1.2754929065704346,
+      "learning_rate": 0.0002,
+      "loss": 1.6112,
+      "step": 50
+    },
+    {
+      "epoch": 0.12903225806451613,
+      "grad_norm": 1.0936230421066284,
+      "learning_rate": 0.0002,
+      "loss": 1.4319,
+      "step": 60
+    },
+    {
+      "epoch": 0.15053763440860216,
+      "grad_norm": 1.144593596458435,
+      "learning_rate": 0.0002,
+      "loss": 1.3568,
+      "step": 70
+    },
+    {
+      "epoch": 0.17204301075268819,
+      "grad_norm": 1.2181956768035889,
+      "learning_rate": 0.0002,
+      "loss": 1.2028,
+      "step": 80
+    },
+    {
+      "epoch": 0.1935483870967742,
+      "grad_norm": 1.1260095834732056,
+      "learning_rate": 0.0002,
+      "loss": 1.1534,
+      "step": 90
+    },
+    {
+      "epoch": 0.21505376344086022,
+      "grad_norm": 1.1155284643173218,
+      "learning_rate": 0.0002,
+      "loss": 1.1089,
+      "step": 100
+    },
+    {
+      "epoch": 0.23655913978494625,
+      "grad_norm": 1.089565396308899,
+      "learning_rate": 0.0002,
+      "loss": 1.0883,
+      "step": 110
+    },
+    {
+      "epoch": 0.25806451612903225,
+      "grad_norm": 0.9833471775054932,
+      "learning_rate": 0.0002,
+      "loss": 1.0814,
+      "step": 120
+    },
+    {
+      "epoch": 0.27956989247311825,
+      "grad_norm": 1.0265629291534424,
+      "learning_rate": 0.0002,
+      "loss": 1.0239,
+      "step": 130
+    },
+    {
+      "epoch": 0.3010752688172043,
+      "grad_norm": 0.9344286322593689,
+      "learning_rate": 0.0002,
+      "loss": 0.9888,
+      "step": 140
+    },
+    {
+      "epoch": 0.3225806451612903,
+      "grad_norm": 0.9883386492729187,
+      "learning_rate": 0.0002,
+      "loss": 1.0043,
+      "step": 150
+    },
+    {
+      "epoch": 0.34408602150537637,
+      "grad_norm": 0.9299277067184448,
+      "learning_rate": 0.0002,
+      "loss": 0.9338,
+      "step": 160
+    },
+    {
+      "epoch": 0.3655913978494624,
+      "grad_norm": 1.390045404434204,
+      "learning_rate": 0.0002,
+      "loss": 0.9432,
+      "step": 170
+    },
+    {
+      "epoch": 0.3870967741935484,
+      "grad_norm": 1.0313078165054321,
+      "learning_rate": 0.0002,
+      "loss": 0.9008,
+      "step": 180
+    },
+    {
+      "epoch": 0.40860215053763443,
+      "grad_norm": 1.1792205572128296,
+      "learning_rate": 0.0002,
+      "loss": 0.9434,
+      "step": 190
+    },
+    {
+      "epoch": 0.43010752688172044,
+      "grad_norm": 1.049809217453003,
+      "learning_rate": 0.0002,
+      "loss": 0.8761,
+      "step": 200
+    },
+    {
+      "epoch": 0.45161290322580644,
+      "grad_norm": 0.990111768245697,
+      "learning_rate": 0.0002,
+      "loss": 0.8709,
+      "step": 210
+    },
+    {
+      "epoch": 0.4731182795698925,
+      "grad_norm": 0.9870412349700928,
+      "learning_rate": 0.0002,
+      "loss": 0.905,
+      "step": 220
+    },
+    {
+      "epoch": 0.4946236559139785,
+      "grad_norm": 0.8557345867156982,
+      "learning_rate": 0.0002,
+      "loss": 0.9129,
+      "step": 230
+    },
+    {
+      "epoch": 0.5161290322580645,
+      "grad_norm": 0.9746861457824707,
+      "learning_rate": 0.0002,
+      "loss": 0.8836,
+      "step": 240
+    },
+    {
+      "epoch": 0.5376344086021505,
+      "grad_norm": 0.9010438323020935,
+      "learning_rate": 0.0002,
+      "loss": 0.873,
+      "step": 250
+    },
+    {
+      "epoch": 0.5591397849462365,
+      "grad_norm": 0.9061082005500793,
+      "learning_rate": 0.0002,
+      "loss": 0.8241,
+      "step": 260
+    },
+    {
+      "epoch": 0.5806451612903226,
+      "grad_norm": 0.9311846494674683,
+      "learning_rate": 0.0002,
+      "loss": 0.8652,
+      "step": 270
+    },
+    {
+      "epoch": 0.6021505376344086,
+      "grad_norm": 0.9140254855155945,
+      "learning_rate": 0.0002,
+      "loss": 0.8256,
+      "step": 280
+    },
+    {
+      "epoch": 0.6236559139784946,
+      "grad_norm": 0.9722253084182739,
+      "learning_rate": 0.0002,
+      "loss": 0.8441,
+      "step": 290
+    },
+    {
+      "epoch": 0.6451612903225806,
+      "grad_norm": 0.8539168238639832,
+      "learning_rate": 0.0002,
+      "loss": 0.8314,
+      "step": 300
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.9053162932395935,
+      "learning_rate": 0.0002,
+      "loss": 0.8528,
+      "step": 310
+    },
+    {
+      "epoch": 0.6881720430107527,
+      "grad_norm": 0.8444252610206604,
+      "learning_rate": 0.0002,
+      "loss": 0.8209,
+      "step": 320
+    },
+    {
+      "epoch": 0.7096774193548387,
+      "grad_norm": 0.8127437829971313,
+      "learning_rate": 0.0002,
+      "loss": 0.8101,
+      "step": 330
+    },
+    {
+      "epoch": 0.7311827956989247,
+      "grad_norm": 0.886555016040802,
+      "learning_rate": 0.0002,
+      "loss": 0.8223,
+      "step": 340
+    },
+    {
+      "epoch": 0.7526881720430108,
+      "grad_norm": 0.8458548784255981,
+      "learning_rate": 0.0002,
+      "loss": 0.8368,
+      "step": 350
+    },
+    {
+      "epoch": 0.7741935483870968,
+      "grad_norm": 0.8683297634124756,
+      "learning_rate": 0.0002,
+      "loss": 0.8295,
+      "step": 360
+    },
+    {
+      "epoch": 0.7956989247311828,
+      "grad_norm": 0.8308405876159668,
+      "learning_rate": 0.0002,
+      "loss": 0.8232,
+      "step": 370
+    },
+    {
+      "epoch": 0.8172043010752689,
+      "grad_norm": 0.8305579423904419,
+      "learning_rate": 0.0002,
+      "loss": 0.7752,
+      "step": 380
+    },
+    {
+      "epoch": 0.8387096774193549,
+      "grad_norm": 0.8545567393302917,
+      "learning_rate": 0.0002,
+      "loss": 0.8267,
+      "step": 390
+    },
+    {
+      "epoch": 0.8602150537634409,
+      "grad_norm": 0.8486055731773376,
+      "learning_rate": 0.0002,
+      "loss": 0.8212,
+      "step": 400
+    },
+    {
+      "epoch": 0.8817204301075269,
+      "grad_norm": 0.8126763105392456,
+      "learning_rate": 0.0002,
+      "loss": 0.743,
+      "step": 410
+    },
+    {
+      "epoch": 0.9032258064516129,
+      "grad_norm": 0.8494045734405518,
+      "learning_rate": 0.0002,
+      "loss": 0.7993,
+      "step": 420
+    },
+    {
+      "epoch": 0.9247311827956989,
+      "grad_norm": 0.7639183402061462,
+      "learning_rate": 0.0002,
+      "loss": 0.8213,
+      "step": 430
+    },
+    {
+      "epoch": 0.946236559139785,
+      "grad_norm": 0.858101487159729,
+      "learning_rate": 0.0002,
+      "loss": 0.8015,
+      "step": 440
+    },
+    {
+      "epoch": 0.967741935483871,
+      "grad_norm": 0.8141381740570068,
+      "learning_rate": 0.0002,
+      "loss": 0.7629,
+      "step": 450
+    },
+    {
+      "epoch": 0.989247311827957,
+      "grad_norm": 0.8072513937950134,
+      "learning_rate": 0.0002,
+      "loss": 0.7357,
+      "step": 460
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.7740864157676697,
+      "eval_runtime": 21.383,
+      "eval_samples_per_second": 15.48,
+      "eval_steps_per_second": 1.964,
+      "step": 465
+    },
+    {
+      "epoch": 1.010752688172043,
+      "grad_norm": 0.8269494771957397,
+      "learning_rate": 0.0002,
+      "loss": 0.7701,
+      "step": 470
+    },
+    {
+      "epoch": 1.032258064516129,
+      "grad_norm": 0.7814009189605713,
+      "learning_rate": 0.0002,
+      "loss": 0.7532,
+      "step": 480
+    },
+    {
+      "epoch": 1.053763440860215,
+      "grad_norm": 0.8183923363685608,
+      "learning_rate": 0.0002,
+      "loss": 0.7689,
+      "step": 490
+    },
+    {
+      "epoch": 1.075268817204301,
+      "grad_norm": 0.8146600723266602,
+      "learning_rate": 0.0002,
+      "loss": 0.765,
+      "step": 500
+    },
+    {
+      "epoch": 1.096774193548387,
+      "grad_norm": 0.8635126352310181,
+      "learning_rate": 0.0002,
+      "loss": 0.7358,
+      "step": 510
+    },
+    {
+      "epoch": 1.118279569892473,
+      "grad_norm": 0.8520359396934509,
+      "learning_rate": 0.0002,
+      "loss": 0.7302,
+      "step": 520
+    },
+    {
+      "epoch": 1.139784946236559,
+      "grad_norm": 0.8026443123817444,
+      "learning_rate": 0.0002,
+      "loss": 0.7492,
+      "step": 530
+    },
+    {
+      "epoch": 1.1612903225806452,
+      "grad_norm": 0.8157258629798889,
+      "learning_rate": 0.0002,
+      "loss": 0.7518,
+      "step": 540
+    },
+    {
+      "epoch": 1.1827956989247312,
+      "grad_norm": 0.9450796246528625,
+      "learning_rate": 0.0002,
+      "loss": 0.7461,
+      "step": 550
+    },
+    {
+      "epoch": 1.2043010752688172,
+      "grad_norm": 0.8859835863113403,
+      "learning_rate": 0.0002,
+      "loss": 0.7128,
+      "step": 560
+    },
+    {
+      "epoch": 1.2258064516129032,
+      "grad_norm": 0.7819921970367432,
+      "learning_rate": 0.0002,
+      "loss": 0.7067,
+      "step": 570
+    },
+    {
+      "epoch": 1.2473118279569892,
+      "grad_norm": 0.7823445796966553,
+      "learning_rate": 0.0002,
+      "loss": 0.7577,
+      "step": 580
+    },
+    {
+      "epoch": 1.2688172043010753,
+      "grad_norm": 0.7931883931159973,
+      "learning_rate": 0.0002,
+      "loss": 0.7358,
+      "step": 590
+    },
+    {
+      "epoch": 1.2903225806451613,
+      "grad_norm": 0.7495734095573425,
+      "learning_rate": 0.0002,
+      "loss": 0.723,
+      "step": 600
+    },
+    {
+      "epoch": 1.3118279569892473,
+      "grad_norm": 0.9272717237472534,
+      "learning_rate": 0.0002,
+      "loss": 0.7386,
+      "step": 610
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 0.7968398332595825,
+      "learning_rate": 0.0002,
+      "loss": 0.7498,
+      "step": 620
+    },
+    {
+      "epoch": 1.3548387096774195,
+      "grad_norm": 0.7813659310340881,
+      "learning_rate": 0.0002,
+      "loss": 0.7635,
+      "step": 630
+    },
+    {
+      "epoch": 1.3763440860215055,
+      "grad_norm": 0.730925977230072,
+      "learning_rate": 0.0002,
+      "loss": 0.6665,
+      "step": 640
+    },
+    {
+      "epoch": 1.3978494623655915,
+      "grad_norm": 0.8011482954025269,
+      "learning_rate": 0.0002,
+      "loss": 0.7037,
+      "step": 650
+    },
+    {
+      "epoch": 1.4193548387096775,
+      "grad_norm": 0.7770085334777832,
+      "learning_rate": 0.0002,
+      "loss": 0.6931,
+      "step": 660
+    },
+    {
+      "epoch": 1.4408602150537635,
+      "grad_norm": 0.7432682514190674,
+      "learning_rate": 0.0002,
+      "loss": 0.6949,
+      "step": 670
+    },
+    {
+      "epoch": 1.4623655913978495,
+      "grad_norm": 0.8820092678070068,
+      "learning_rate": 0.0002,
+      "loss": 0.7444,
+      "step": 680
+    },
+    {
+      "epoch": 1.4838709677419355,
+      "grad_norm": 0.7786208987236023,
+      "learning_rate": 0.0002,
+      "loss": 0.6758,
+      "step": 690
+    },
+    {
+      "epoch": 1.5053763440860215,
+      "grad_norm": 0.7467480301856995,
+      "learning_rate": 0.0002,
+      "loss": 0.6702,
+      "step": 700
+    },
+    {
+      "epoch": 1.5268817204301075,
+      "grad_norm": 0.8147122263908386,
+      "learning_rate": 0.0002,
+      "loss": 0.7107,
+      "step": 710
+    },
+    {
+      "epoch": 1.5483870967741935,
+      "grad_norm": 0.796030580997467,
+      "learning_rate": 0.0002,
+      "loss": 0.7144,
+      "step": 720
+    },
+    {
+      "epoch": 1.5698924731182795,
+      "grad_norm": 0.8776171207427979,
+      "learning_rate": 0.0002,
+      "loss": 0.6936,
+      "step": 730
+    },
+    {
+      "epoch": 1.5913978494623655,
+      "grad_norm": 0.8056126236915588,
+      "learning_rate": 0.0002,
+      "loss": 0.7101,
+      "step": 740
+    },
+    {
+      "epoch": 1.6129032258064515,
+      "grad_norm": 0.8141863346099854,
+      "learning_rate": 0.0002,
+      "loss": 0.7162,
+      "step": 750
+    },
+    {
+      "epoch": 1.6344086021505375,
+      "grad_norm": 0.8100557327270508,
+      "learning_rate": 0.0002,
+      "loss": 0.7088,
+      "step": 760
+    },
+    {
+      "epoch": 1.6559139784946235,
+      "grad_norm": 0.8283200860023499,
+      "learning_rate": 0.0002,
+      "loss": 0.7212,
+      "step": 770
+    },
+    {
+      "epoch": 1.6774193548387095,
+      "grad_norm": 0.800865113735199,
+      "learning_rate": 0.0002,
+      "loss": 0.694,
+      "step": 780
+    },
+    {
+      "epoch": 1.6989247311827957,
+      "grad_norm": 0.8052287697792053,
+      "learning_rate": 0.0002,
+      "loss": 0.7076,
+      "step": 790
+    },
+    {
+      "epoch": 1.7204301075268817,
+      "grad_norm": 0.8619674444198608,
+      "learning_rate": 0.0002,
+      "loss": 0.7257,
+      "step": 800
+    },
+    {
+      "epoch": 1.7419354838709677,
+      "grad_norm": 0.8907215595245361,
+      "learning_rate": 0.0002,
+      "loss": 0.7141,
+      "step": 810
+    },
+    {
+      "epoch": 1.7634408602150538,
+      "grad_norm": 0.6976316571235657,
+      "learning_rate": 0.0002,
+      "loss": 0.7035,
+      "step": 820
+    },
+    {
+      "epoch": 1.7849462365591398,
+      "grad_norm": 0.7533746957778931,
+      "learning_rate": 0.0002,
+      "loss": 0.6916,
+      "step": 830
+    },
+    {
+      "epoch": 1.8064516129032258,
+      "grad_norm": 0.7326804399490356,
+      "learning_rate": 0.0002,
+      "loss": 0.7094,
+      "step": 840
+    },
+    {
+      "epoch": 1.827956989247312,
+      "grad_norm": 0.7782683372497559,
+      "learning_rate": 0.0002,
+      "loss": 0.6891,
+      "step": 850
+    },
+    {
+      "epoch": 1.849462365591398,
+      "grad_norm": 0.7424806356430054,
+      "learning_rate": 0.0002,
+      "loss": 0.6931,
+      "step": 860
+    },
+    {
+      "epoch": 1.870967741935484,
+      "grad_norm": 1.172325611114502,
+      "learning_rate": 0.0002,
+      "loss": 0.7354,
+      "step": 870
+    },
+    {
+      "epoch": 1.89247311827957,
+      "grad_norm": 0.771058201789856,
+      "learning_rate": 0.0002,
+      "loss": 0.6866,
+      "step": 880
+    },
+    {
+      "epoch": 1.913978494623656,
+      "grad_norm": 0.8624904155731201,
+      "learning_rate": 0.0002,
+      "loss": 0.7296,
+      "step": 890
+    },
+    {
+      "epoch": 1.935483870967742,
+      "grad_norm": 0.7062820792198181,
+      "learning_rate": 0.0002,
+      "loss": 0.7233,
+      "step": 900
+    },
+    {
+      "epoch": 1.956989247311828,
+      "grad_norm": 0.7560103535652161,
+      "learning_rate": 0.0002,
+      "loss": 0.6966,
+      "step": 910
+    },
+    {
+      "epoch": 1.978494623655914,
+      "grad_norm": 0.788899838924408,
+      "learning_rate": 0.0002,
+      "loss": 0.69,
+      "step": 920
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.6562113761901855,
+      "learning_rate": 0.0002,
+      "loss": 0.6505,
+      "step": 930
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.6885261535644531,
+      "eval_runtime": 21.4291,
+      "eval_samples_per_second": 15.446,
+      "eval_steps_per_second": 1.96,
+      "step": 930
+    },
+    {
+      "epoch": 2.021505376344086,
+      "grad_norm": 0.8216531872749329,
+      "learning_rate": 0.0002,
+      "loss": 0.6625,
+      "step": 940
+    },
+    {
+      "epoch": 2.043010752688172,
+      "grad_norm": 0.8317142724990845,
+      "learning_rate": 0.0002,
+      "loss": 0.6398,
+      "step": 950
+    },
+    {
+      "epoch": 2.064516129032258,
+      "grad_norm": 0.8446708917617798,
+      "learning_rate": 0.0002,
+      "loss": 0.649,
+      "step": 960
+    },
+    {
+      "epoch": 2.086021505376344,
+      "grad_norm": 0.735055148601532,
+      "learning_rate": 0.0002,
+      "loss": 0.657,
+      "step": 970
+    },
+    {
+      "epoch": 2.10752688172043,
+      "grad_norm": 0.7487243413925171,
+      "learning_rate": 0.0002,
+      "loss": 0.649,
+      "step": 980
+    },
+    {
+      "epoch": 2.129032258064516,
+      "grad_norm": 0.8573887944221497,
+      "learning_rate": 0.0002,
+      "loss": 0.6419,
+      "step": 990
+    },
+    {
+      "epoch": 2.150537634408602,
+      "grad_norm": 0.6284521818161011,
+      "learning_rate": 0.0002,
+      "loss": 0.6431,
+      "step": 1000
+    },
+    {
+      "epoch": 2.172043010752688,
+      "grad_norm": 0.754183292388916,
+      "learning_rate": 0.0002,
+      "loss": 0.6128,
+      "step": 1010
+    },
+    {
+      "epoch": 2.193548387096774,
+      "grad_norm": 0.9445359110832214,
+      "learning_rate": 0.0002,
+      "loss": 0.6253,
+      "step": 1020
+    },
+    {
+      "epoch": 2.21505376344086,
+      "grad_norm": 0.808508038520813,
+      "learning_rate": 0.0002,
+      "loss": 0.605,
+      "step": 1030
+    },
+    {
+      "epoch": 2.236559139784946,
+      "grad_norm": 0.9394679665565491,
+      "learning_rate": 0.0002,
+      "loss": 0.6786,
+      "step": 1040
+    },
+    {
+      "epoch": 2.258064516129032,
+      "grad_norm": 0.8151357769966125,
+      "learning_rate": 0.0002,
+      "loss": 0.6176,
+      "step": 1050
+    },
+    {
+      "epoch": 2.279569892473118,
+      "grad_norm": 0.7909848093986511,
+      "learning_rate": 0.0002,
+      "loss": 0.66,
+      "step": 1060
+    },
+    {
+      "epoch": 2.3010752688172045,
+      "grad_norm": 0.7506507039070129,
+      "learning_rate": 0.0002,
+      "loss": 0.6254,
+      "step": 1070
+    },
+    {
+      "epoch": 2.3225806451612905,
+      "grad_norm": 0.8240520358085632,
+      "learning_rate": 0.0002,
+      "loss": 0.6608,
+      "step": 1080
+    },
+    {
+      "epoch": 2.3440860215053765,
+      "grad_norm": 0.9342400431632996,
+      "learning_rate": 0.0002,
+      "loss": 0.6207,
+      "step": 1090
+    },
+    {
+      "epoch": 2.3655913978494625,
+      "grad_norm": 1.0598735809326172,
+      "learning_rate": 0.0002,
+      "loss": 0.6029,
+      "step": 1100
+    },
+    {
+      "epoch": 2.3870967741935485,
+      "grad_norm": 0.7907650470733643,
+      "learning_rate": 0.0002,
+      "loss": 0.6035,
+      "step": 1110
+    },
+    {
+      "epoch": 2.4086021505376345,
+      "grad_norm": 0.9388798475265503,
+      "learning_rate": 0.0002,
+      "loss": 0.6237,
+      "step": 1120
+    },
+    {
+      "epoch": 2.4301075268817205,
+      "grad_norm": 0.8985419869422913,
+      "learning_rate": 0.0002,
+      "loss": 0.6207,
+      "step": 1130
+    },
+    {
+      "epoch": 2.4516129032258065,
+      "grad_norm": 0.7471932768821716,
+      "learning_rate": 0.0002,
+      "loss": 0.5902,
+      "step": 1140
+    },
+    {
+      "epoch": 2.4731182795698925,
+      "grad_norm": 0.761131763458252,
+      "learning_rate": 0.0002,
+      "loss": 0.6446,
+      "step": 1150
+    },
+    {
+      "epoch": 2.4946236559139785,
+      "grad_norm": 0.7901819348335266,
+      "learning_rate": 0.0002,
+      "loss": 0.6088,
+      "step": 1160
+    },
+    {
+      "epoch": 2.5161290322580645,
+      "grad_norm": 0.9932922720909119,
+      "learning_rate": 0.0002,
+      "loss": 0.6142,
+      "step": 1170
+    },
+    {
+      "epoch": 2.5376344086021505,
+      "grad_norm": 0.7414287328720093,
+      "learning_rate": 0.0002,
+      "loss": 0.6407,
+      "step": 1180
+    },
+    {
+      "epoch": 2.5591397849462365,
+      "grad_norm": 0.8111771941184998,
+      "learning_rate": 0.0002,
+      "loss": 0.6161,
+      "step": 1190
+    },
+    {
+      "epoch": 2.5806451612903225,
+      "grad_norm": 0.7520156502723694,
+      "learning_rate": 0.0002,
+      "loss": 0.6006,
+      "step": 1200
+    },
+    {
+      "epoch": 2.6021505376344085,
+      "grad_norm": 0.9022907018661499,
+      "learning_rate": 0.0002,
+      "loss": 0.615,
+      "step": 1210
+    },
+    {
+      "epoch": 2.6236559139784945,
+      "grad_norm": 0.7746260166168213,
+      "learning_rate": 0.0002,
+      "loss": 0.6211,
+      "step": 1220
+    },
+    {
+      "epoch": 2.6451612903225805,
+      "grad_norm": 0.8482862114906311,
+      "learning_rate": 0.0002,
+      "loss": 0.616,
+      "step": 1230
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 0.7925458550453186,
+      "learning_rate": 0.0002,
+      "loss": 0.6417,
+      "step": 1240
+    },
+    {
+      "epoch": 2.688172043010753,
+      "grad_norm": 0.8369929194450378,
+      "learning_rate": 0.0002,
+      "loss": 0.6187,
+      "step": 1250
+    },
+    {
+      "epoch": 2.709677419354839,
+      "grad_norm": 0.8311542868614197,
+      "learning_rate": 0.0002,
+      "loss": 0.6138,
+      "step": 1260
+    },
+    {
+      "epoch": 2.731182795698925,
+      "grad_norm": 0.7204853296279907,
+      "learning_rate": 0.0002,
+      "loss": 0.5894,
+      "step": 1270
+    },
+    {
+      "epoch": 2.752688172043011,
+      "grad_norm": 0.8447284698486328,
+      "learning_rate": 0.0002,
+      "loss": 0.6325,
+      "step": 1280
+    },
+    {
+      "epoch": 2.774193548387097,
+      "grad_norm": 0.7738404273986816,
+      "learning_rate": 0.0002,
+      "loss": 0.5946,
+      "step": 1290
+    },
+    {
+      "epoch": 2.795698924731183,
+      "grad_norm": 0.8393287062644958,
+      "learning_rate": 0.0002,
+      "loss": 0.5678,
+      "step": 1300
+    },
+    {
+      "epoch": 2.817204301075269,
+      "grad_norm": 0.79121994972229,
+      "learning_rate": 0.0002,
+      "loss": 0.6092,
+      "step": 1310
+    },
+    {
+      "epoch": 2.838709677419355,
+      "grad_norm": 0.7331557869911194,
+      "learning_rate": 0.0002,
+      "loss": 0.5889,
+      "step": 1320
+    },
+    {
+      "epoch": 2.860215053763441,
+      "grad_norm": 0.9593998193740845,
+      "learning_rate": 0.0002,
+      "loss": 0.6048,
+      "step": 1330
+    },
+    {
+      "epoch": 2.881720430107527,
+      "grad_norm": 0.7215158343315125,
+      "learning_rate": 0.0002,
+      "loss": 0.6108,
+      "step": 1340
+    },
+    {
+      "epoch": 2.903225806451613,
+      "grad_norm": 0.840404212474823,
+      "learning_rate": 0.0002,
+      "loss": 0.5897,
+      "step": 1350
+    },
+    {
+      "epoch": 2.924731182795699,
+      "grad_norm": 0.870659351348877,
+      "learning_rate": 0.0002,
+      "loss": 0.6056,
+      "step": 1360
+    },
+    {
+      "epoch": 2.946236559139785,
+      "grad_norm": 0.8744975328445435,
+      "learning_rate": 0.0002,
+      "loss": 0.6205,
+      "step": 1370
+    },
+    {
+      "epoch": 2.967741935483871,
+      "grad_norm": 0.8030612468719482,
+      "learning_rate": 0.0002,
+      "loss": 0.5966,
+      "step": 1380
+    },
+    {
+      "epoch": 2.989247311827957,
+      "grad_norm": 0.825814962387085,
+      "learning_rate": 0.0002,
+      "loss": 0.6004,
+      "step": 1390
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 0.6257933378219604,
+      "eval_runtime": 21.3692,
+      "eval_samples_per_second": 15.49,
+      "eval_steps_per_second": 1.965,
+      "step": 1395
+    },
+    {
+      "epoch": 3.010752688172043,
+      "grad_norm": 0.8650677800178528,
+      "learning_rate": 0.0002,
+      "loss": 0.5696,
+      "step": 1400
+    },
+    {
+      "epoch": 3.032258064516129,
+      "grad_norm": 0.8364197015762329,
+      "learning_rate": 0.0002,
+      "loss": 0.5483,
+      "step": 1410
+    },
+    {
+      "epoch": 3.053763440860215,
+      "grad_norm": 0.8278448581695557,
+      "learning_rate": 0.0002,
+      "loss": 0.5606,
+      "step": 1420
+    },
+    {
+      "epoch": 3.075268817204301,
+      "grad_norm": 0.8806642889976501,
+      "learning_rate": 0.0002,
+      "loss": 0.5572,
+      "step": 1430
+    },
+    {
+      "epoch": 3.096774193548387,
+      "grad_norm": 0.8180029988288879,
+      "learning_rate": 0.0002,
+      "loss": 0.585,
+      "step": 1440
+    },
+    {
+      "epoch": 3.118279569892473,
+      "grad_norm": 0.8561782836914062,
+      "learning_rate": 0.0002,
+      "loss": 0.5667,
+      "step": 1450
+    },
+    {
+      "epoch": 3.139784946236559,
+      "grad_norm": 0.8377029299736023,
+      "learning_rate": 0.0002,
+      "loss": 0.5246,
+      "step": 1460
+    },
+    {
+      "epoch": 3.161290322580645,
+      "grad_norm": 0.885779082775116,
+      "learning_rate": 0.0002,
+      "loss": 0.5464,
+      "step": 1470
+    },
+    {
+      "epoch": 3.182795698924731,
+      "grad_norm": 0.9388518333435059,
+      "learning_rate": 0.0002,
+      "loss": 0.541,
+      "step": 1480
+    },
+    {
+      "epoch": 3.204301075268817,
+      "grad_norm": 0.8816235661506653,
+      "learning_rate": 0.0002,
+      "loss": 0.5447,
+      "step": 1490
+    },
+    {
+      "epoch": 3.225806451612903,
+      "grad_norm": 0.9885783791542053,
+      "learning_rate": 0.0002,
+      "loss": 0.5466,
+      "step": 1500
+    },
+    {
+      "epoch": 3.247311827956989,
+      "grad_norm": 0.8635850548744202,
+      "learning_rate": 0.0002,
+      "loss": 0.5455,
+      "step": 1510
+    },
+    {
+      "epoch": 3.268817204301075,
+      "grad_norm": 0.829853355884552,
+      "learning_rate": 0.0002,
+      "loss": 0.5419,
+      "step": 1520
+    },
+    {
+      "epoch": 3.2903225806451615,
+      "grad_norm": 0.9037486910820007,
+      "learning_rate": 0.0002,
+      "loss": 0.54,
+      "step": 1530
+    },
+    {
+      "epoch": 3.3118279569892475,
+      "grad_norm": 0.8173713684082031,
+      "learning_rate": 0.0002,
+      "loss": 0.5375,
+      "step": 1540
+    },
+    {
+      "epoch": 3.3333333333333335,
+      "grad_norm": 0.796953022480011,
+      "learning_rate": 0.0002,
+      "loss": 0.5405,
+      "step": 1550
+    },
+    {
+      "epoch": 3.3548387096774195,
+      "grad_norm": 0.7894400358200073,
+      "learning_rate": 0.0002,
+      "loss": 0.5505,
+      "step": 1560
+    },
+    {
+      "epoch": 3.3763440860215055,
+      "grad_norm": 0.9434949159622192,
+      "learning_rate": 0.0002,
+      "loss": 0.5395,
+      "step": 1570
+    },
+    {
+      "epoch": 3.3978494623655915,
+      "grad_norm": 0.8666760325431824,
+      "learning_rate": 0.0002,
+      "loss": 0.5271,
+      "step": 1580
+    },
+    {
+      "epoch": 3.4193548387096775,
+      "grad_norm": 0.7782467007637024,
+      "learning_rate": 0.0002,
+      "loss": 0.5439,
+      "step": 1590
+    },
+    {
+      "epoch": 3.4408602150537635,
+      "grad_norm": 0.8849126696586609,
+      "learning_rate": 0.0002,
+      "loss": 0.5161,
+      "step": 1600
+    },
+    {
+      "epoch": 3.4623655913978495,
+      "grad_norm": 0.7863831520080566,
+      "learning_rate": 0.0002,
+      "loss": 0.5353,
+      "step": 1610
+    },
+    {
+      "epoch": 3.4838709677419355,
+      "grad_norm": 1.0403116941452026,
+      "learning_rate": 0.0002,
+      "loss": 0.5308,
+      "step": 1620
+    },
+    {
+      "epoch": 3.5053763440860215,
+      "grad_norm": 0.8307499289512634,
+      "learning_rate": 0.0002,
+      "loss": 0.5339,
+      "step": 1630
+    },
+    {
+      "epoch": 3.5268817204301075,
+      "grad_norm": 0.9132118821144104,
+      "learning_rate": 0.0002,
+      "loss": 0.5361,
+      "step": 1640
+    },
+    {
+      "epoch": 3.5483870967741935,
+      "grad_norm": 0.9322578310966492,
+      "learning_rate": 0.0002,
+      "loss": 0.5828,
+      "step": 1650
+    },
+    {
+      "epoch": 3.5698924731182795,
+      "grad_norm": 0.9782460331916809,
+      "learning_rate": 0.0002,
+      "loss": 0.546,
+      "step": 1660
+    },
+    {
+      "epoch": 3.5913978494623655,
+      "grad_norm": 0.7189919352531433,
+      "learning_rate": 0.0002,
+      "loss": 0.5424,
+      "step": 1670
+    },
+    {
+      "epoch": 3.6129032258064515,
+      "grad_norm": 0.9689221382141113,
+      "learning_rate": 0.0002,
+      "loss": 0.5514,
+      "step": 1680
+    },
+    {
+      "epoch": 3.6344086021505375,
+      "grad_norm": 0.9684675335884094,
+      "learning_rate": 0.0002,
+      "loss": 0.5379,
+      "step": 1690
+    },
+    {
+      "epoch": 3.6559139784946235,
+      "grad_norm": 0.8851472735404968,
+      "learning_rate": 0.0002,
+      "loss": 0.5748,
+      "step": 1700
+    },
+    {
+      "epoch": 3.6774193548387095,
+      "grad_norm": 0.7709833383560181,
+      "learning_rate": 0.0002,
+      "loss": 0.5412,
+      "step": 1710
+    },
+    {
+      "epoch": 3.698924731182796,
+      "grad_norm": 0.818236231803894,
+      "learning_rate": 0.0002,
+      "loss": 0.521,
+      "step": 1720
+    },
+    {
+      "epoch": 3.720430107526882,
+      "grad_norm": 0.870642364025116,
+      "learning_rate": 0.0002,
+      "loss": 0.5445,
+      "step": 1730
+    },
+    {
+      "epoch": 3.741935483870968,
+      "grad_norm": 1.0245511531829834,
+      "learning_rate": 0.0002,
+      "loss": 0.5307,
+      "step": 1740
+    },
+    {
+      "epoch": 3.763440860215054,
+      "grad_norm": 0.8607558608055115,
+      "learning_rate": 0.0002,
+      "loss": 0.5593,
+      "step": 1750
+    },
+    {
+      "epoch": 3.78494623655914,
+      "grad_norm": 0.8511829972267151,
+      "learning_rate": 0.0002,
+      "loss": 0.536,
+      "step": 1760
+    },
+    {
+      "epoch": 3.806451612903226,
+      "grad_norm": 0.7969087362289429,
+      "learning_rate": 0.0002,
+      "loss": 0.5193,
+      "step": 1770
+    },
+    {
+      "epoch": 3.827956989247312,
+      "grad_norm": 0.8457245826721191,
+      "learning_rate": 0.0002,
+      "loss": 0.5578,
+      "step": 1780
+    },
+    {
+      "epoch": 3.849462365591398,
+      "grad_norm": 0.8893467783927917,
+      "learning_rate": 0.0002,
+      "loss": 0.5337,
+      "step": 1790
+    },
+    {
+      "epoch": 3.870967741935484,
+      "grad_norm": 0.8593819737434387,
+      "learning_rate": 0.0002,
+      "loss": 0.5024,
+      "step": 1800
+    },
+    {
+      "epoch": 3.89247311827957,
+      "grad_norm": 0.7574560642242432,
+      "learning_rate": 0.0002,
+      "loss": 0.5134,
+      "step": 1810
+    },
+    {
+      "epoch": 3.913978494623656,
+      "grad_norm": 0.8681567311286926,
+      "learning_rate": 0.0002,
+      "loss": 0.5263,
+      "step": 1820
+    },
+    {
+      "epoch": 3.935483870967742,
+      "grad_norm": 0.9068132042884827,
+      "learning_rate": 0.0002,
+      "loss": 0.532,
+      "step": 1830
+    },
+    {
+      "epoch": 3.956989247311828,
+      "grad_norm": 0.8668948411941528,
+      "learning_rate": 0.0002,
+      "loss": 0.5427,
+      "step": 1840
+    },
+    {
+      "epoch": 3.978494623655914,
+      "grad_norm": 1.046032428741455,
+      "learning_rate": 0.0002,
+      "loss": 0.5349,
+      "step": 1850
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.904780387878418,
+      "learning_rate": 0.0002,
+      "loss": 0.5087,
+      "step": 1860
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 0.5737715363502502,
+      "eval_runtime": 21.4915,
+      "eval_samples_per_second": 15.401,
+      "eval_steps_per_second": 1.954,
+      "step": 1860
+    },
+    {
+      "epoch": 4.021505376344086,
+      "grad_norm": 0.8611752986907959,
+      "learning_rate": 0.0002,
+      "loss": 0.4843,
+      "step": 1870
+    },
+    {
+      "epoch": 4.043010752688172,
+      "grad_norm": 0.838782548904419,
+      "learning_rate": 0.0002,
+      "loss": 0.4814,
+      "step": 1880
+    },
+    {
+      "epoch": 4.064516129032258,
+      "grad_norm": 0.9119709134101868,
+      "learning_rate": 0.0002,
+      "loss": 0.474,
+      "step": 1890
+    },
+    {
+      "epoch": 4.086021505376344,
+      "grad_norm": 0.8026251196861267,
+      "learning_rate": 0.0002,
+      "loss": 0.4951,
+      "step": 1900
+    },
+    {
+      "epoch": 4.10752688172043,
+      "grad_norm": 0.8773705363273621,
+      "learning_rate": 0.0002,
+      "loss": 0.491,
+      "step": 1910
+    },
+    {
+      "epoch": 4.129032258064516,
+      "grad_norm": 0.8762255907058716,
+      "learning_rate": 0.0002,
+      "loss": 0.474,
+      "step": 1920
+    },
+    {
+      "epoch": 4.150537634408602,
+      "grad_norm": 0.8371861577033997,
+      "learning_rate": 0.0002,
+      "loss": 0.4816,
+      "step": 1930
+    },
+    {
+      "epoch": 4.172043010752688,
+      "grad_norm": 0.9703728556632996,
+      "learning_rate": 0.0002,
+      "loss": 0.472,
+      "step": 1940
+    },
+    {
+      "epoch": 4.193548387096774,
+      "grad_norm": 0.8802874684333801,
+      "learning_rate": 0.0002,
+      "loss": 0.4772,
+      "step": 1950
+    },
+    {
+      "epoch": 4.21505376344086,
+      "grad_norm": 1.0103057622909546,
+      "learning_rate": 0.0002,
+      "loss": 0.5032,
+      "step": 1960
+    },
+    {
+      "epoch": 4.236559139784946,
+      "grad_norm": 0.9212995171546936,
+      "learning_rate": 0.0002,
+      "loss": 0.4945,
+      "step": 1970
+    },
+    {
+      "epoch": 4.258064516129032,
+      "grad_norm": 1.009544849395752,
+      "learning_rate": 0.0002,
+      "loss": 0.4753,
+      "step": 1980
+    },
+    {
+      "epoch": 4.279569892473118,
+      "grad_norm": 0.8535077571868896,
+      "learning_rate": 0.0002,
+      "loss": 0.4789,
+      "step": 1990
+    },
+    {
+      "epoch": 4.301075268817204,
+      "grad_norm": 0.8363022804260254,
+      "learning_rate": 0.0002,
+      "loss": 0.4782,
+      "step": 2000
+    },
+    {
+      "epoch": 4.32258064516129,
+      "grad_norm": 0.9041762948036194,
+      "learning_rate": 0.0002,
+      "loss": 0.4875,
+      "step": 2010
+    },
+    {
+      "epoch": 4.344086021505376,
+      "grad_norm": 0.960790753364563,
+      "learning_rate": 0.0002,
+      "loss": 0.4779,
+      "step": 2020
+    },
+    {
+      "epoch": 4.365591397849462,
+      "grad_norm": 0.8823095560073853,
+      "learning_rate": 0.0002,
+      "loss": 0.4626,
+      "step": 2030
+    },
+    {
+      "epoch": 4.387096774193548,
+      "grad_norm": 0.952100396156311,
+      "learning_rate": 0.0002,
+      "loss": 0.4883,
+      "step": 2040
+    },
+    {
+      "epoch": 4.408602150537634,
+      "grad_norm": 1.0793498754501343,
+      "learning_rate": 0.0002,
+      "loss": 0.4789,
+      "step": 2050
+    },
+    {
+      "epoch": 4.43010752688172,
+      "grad_norm": 0.8987208008766174,
+      "learning_rate": 0.0002,
+      "loss": 0.4827,
+      "step": 2060
+    },
+    {
+      "epoch": 4.451612903225806,
+      "grad_norm": 0.8539772033691406,
+      "learning_rate": 0.0002,
+      "loss": 0.4594,
+      "step": 2070
+    },
+    {
+      "epoch": 4.473118279569892,
+      "grad_norm": 0.9160863757133484,
+      "learning_rate": 0.0002,
+      "loss": 0.4752,
+      "step": 2080
+    },
+    {
+      "epoch": 4.494623655913978,
+      "grad_norm": 0.9946850538253784,
+      "learning_rate": 0.0002,
+      "loss": 0.5033,
+      "step": 2090
+    },
+    {
+      "epoch": 4.516129032258064,
+      "grad_norm": 0.908039391040802,
+      "learning_rate": 0.0002,
+      "loss": 0.4842,
+      "step": 2100
+    },
+    {
+      "epoch": 4.53763440860215,
+      "grad_norm": 1.1462254524230957,
+      "learning_rate": 0.0002,
+      "loss": 0.4861,
+      "step": 2110
+    },
+    {
+      "epoch": 4.559139784946236,
+      "grad_norm": 0.8392056226730347,
+      "learning_rate": 0.0002,
+      "loss": 0.4892,
+      "step": 2120
+    },
+    {
+      "epoch": 4.580645161290323,
+      "grad_norm": 0.9673896431922913,
+      "learning_rate": 0.0002,
+      "loss": 0.4824,
+      "step": 2130
+    },
+    {
+      "epoch": 4.602150537634409,
+      "grad_norm": 0.9047091603279114,
+      "learning_rate": 0.0002,
+      "loss": 0.4665,
+      "step": 2140
+    },
+    {
+      "epoch": 4.623655913978495,
+      "grad_norm": 0.9013425707817078,
+      "learning_rate": 0.0002,
+      "loss": 0.4714,
+      "step": 2150
+    },
+    {
+      "epoch": 4.645161290322581,
+      "grad_norm": 0.8899165391921997,
+      "learning_rate": 0.0002,
+      "loss": 0.472,
+      "step": 2160
+    },
+    {
+      "epoch": 4.666666666666667,
+      "grad_norm": 0.748602569103241,
+      "learning_rate": 0.0002,
+      "loss": 0.4635,
+      "step": 2170
+    },
+    {
+      "epoch": 4.688172043010753,
+      "grad_norm": 0.8694155216217041,
+      "learning_rate": 0.0002,
+      "loss": 0.4695,
+      "step": 2180
+    },
+    {
+      "epoch": 4.709677419354839,
+      "grad_norm": 0.9134316444396973,
+      "learning_rate": 0.0002,
+      "loss": 0.4929,
+      "step": 2190
+    },
+    {
+      "epoch": 4.731182795698925,
+      "grad_norm": 0.8504763245582581,
+      "learning_rate": 0.0002,
+      "loss": 0.4855,
+      "step": 2200
+    },
+    {
+      "epoch": 4.752688172043011,
+      "grad_norm": 1.0321544408798218,
+      "learning_rate": 0.0002,
+      "loss": 0.4517,
+      "step": 2210
+    },
+    {
+      "epoch": 4.774193548387097,
+      "grad_norm": 0.9368237257003784,
+      "learning_rate": 0.0002,
+      "loss": 0.4796,
+      "step": 2220
+    },
+    {
+      "epoch": 4.795698924731183,
+      "grad_norm": 0.9319947361946106,
+      "learning_rate": 0.0002,
+      "loss": 0.4837,
+      "step": 2230
+    },
+    {
+      "epoch": 4.817204301075269,
+      "grad_norm": 0.904333770275116,
+      "learning_rate": 0.0002,
+      "loss": 0.4696,
+      "step": 2240
+    },
+    {
+      "epoch": 4.838709677419355,
+      "grad_norm": 0.8097078204154968,
+      "learning_rate": 0.0002,
+      "loss": 0.4746,
+      "step": 2250
+    },
+    {
+      "epoch": 4.860215053763441,
+      "grad_norm": 0.9128859043121338,
+      "learning_rate": 0.0002,
+      "loss": 0.4438,
+      "step": 2260
+    },
+    {
+      "epoch": 4.881720430107527,
+      "grad_norm": 0.883129894733429,
+      "learning_rate": 0.0002,
+      "loss": 0.4693,
+      "step": 2270
+    },
+    {
+      "epoch": 4.903225806451613,
+      "grad_norm": 0.85712730884552,
+      "learning_rate": 0.0002,
+      "loss": 0.4494,
+      "step": 2280
+    },
+    {
+      "epoch": 4.924731182795699,
+      "grad_norm": 1.2101863622665405,
+      "learning_rate": 0.0002,
+      "loss": 0.4593,
+      "step": 2290
+    },
+    {
+      "epoch": 4.946236559139785,
+      "grad_norm": 0.917966902256012,
+      "learning_rate": 0.0002,
+      "loss": 0.4779,
+      "step": 2300
+    },
+    {
+      "epoch": 4.967741935483871,
+      "grad_norm": 0.7740724086761475,
+      "learning_rate": 0.0002,
+      "loss": 0.4666,
+      "step": 2310
+    },
+    {
+      "epoch": 4.989247311827957,
+      "grad_norm": 1.0199906826019287,
+      "learning_rate": 0.0002,
+      "loss": 0.4629,
+      "step": 2320
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 0.5363914370536804,
+      "eval_runtime": 21.3941,
+      "eval_samples_per_second": 15.472,
+      "eval_steps_per_second": 1.963,
+      "step": 2325
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 3720,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.84161414791168e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/training_args.bin b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..483b53d4fc1c568a5fc890fa850e3450f390b208
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e4c4fe2be590ed03492316230adb3a1edca3e4066c55f3716c0352d7134c564
+size 5560
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/README.md b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..830a14f7db2734beb59f320973504e45a3fe87f5
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/adapter_config.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e99bbcd43df1c19d98706c7e3be95c93844c5349
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/adapter_model.safetensors b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..a3470024eb7a2c3549695e46e24bf3e15184388a
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:000663c19f6be785e72fa959fb587169066699d674d8bfee57cbf973cb87e984
+size 29500848
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/optimizer.pt b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..28d6e106f46ddda3b3229e66e2d0a1f4bace4e38
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a3c8a8ef88f82223655c75a2ed9e3c85789689fd8bed182ccbb066f87bfcc34
+size 15064314
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/rng_state.pth b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..b067621f71756ec961f604616d7abedc3836d6a5
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8237fb590cc4741b8cdb25f103d4ff4351eda1d6bd6372159499b51d44b3fbfb
+size 14244
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/scheduler.pt b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..4ac258a9c1113698d9863b2652a5ee759ceaa463
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b3b25b7e8485ad3adf27ba5c1b150b8d23655686491e68d0ee87d9c1bea784d9
+size 1064
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/special_tokens_map.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/tokenizer.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..f58963a682665634ab180c28667e4faa8cf02ba2
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/tokenizer.model b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/tokenizer_config.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1adb4796c13b8d975555ecec45876ee75d1ae8b7
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/trainer_state.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..620fc3de22756106f08cb63afdc47ec190d59ccb
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/trainer_state.json
@@ -0,0 +1,2034 @@
+{
+  "best_metric": 0.5057176947593689,
+  "best_model_checkpoint": "outputs-001/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790",
+  "epoch": 6.0,
+  "eval_steps": 10,
+  "global_step": 2790,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.021505376344086023,
+      "grad_norm": 0.9075053930282593,
+      "learning_rate": 0.0002,
+      "loss": 3.4172,
+      "step": 10
+    },
+    {
+      "epoch": 0.043010752688172046,
+      "grad_norm": 1.4321208000183105,
+      "learning_rate": 0.0002,
+      "loss": 2.5888,
+      "step": 20
+    },
+    {
+      "epoch": 0.06451612903225806,
+      "grad_norm": 1.7500602006912231,
+      "learning_rate": 0.0002,
+      "loss": 2.1195,
+      "step": 30
+    },
+    {
+      "epoch": 0.08602150537634409,
+      "grad_norm": 0.7606641054153442,
+      "learning_rate": 0.0002,
+      "loss": 1.9303,
+      "step": 40
+    },
+    {
+      "epoch": 0.10752688172043011,
+      "grad_norm": 1.2754929065704346,
+      "learning_rate": 0.0002,
+      "loss": 1.6112,
+      "step": 50
+    },
+    {
+      "epoch": 0.12903225806451613,
+      "grad_norm": 1.0936230421066284,
+      "learning_rate": 0.0002,
+      "loss": 1.4319,
+      "step": 60
+    },
+    {
+      "epoch": 0.15053763440860216,
+      "grad_norm": 1.144593596458435,
+      "learning_rate": 0.0002,
+      "loss": 1.3568,
+      "step": 70
+    },
+    {
+      "epoch": 0.17204301075268819,
+      "grad_norm": 1.2181956768035889,
+      "learning_rate": 0.0002,
+      "loss": 1.2028,
+      "step": 80
+    },
+    {
+      "epoch": 0.1935483870967742,
+      "grad_norm": 1.1260095834732056,
+      "learning_rate": 0.0002,
+      "loss": 1.1534,
+      "step": 90
+    },
+    {
+      "epoch": 0.21505376344086022,
+      "grad_norm": 1.1155284643173218,
+      "learning_rate": 0.0002,
+      "loss": 1.1089,
+      "step": 100
+    },
+    {
+      "epoch": 0.23655913978494625,
+      "grad_norm": 1.089565396308899,
+      "learning_rate": 0.0002,
+      "loss": 1.0883,
+      "step": 110
+    },
+    {
+      "epoch": 0.25806451612903225,
+      "grad_norm": 0.9833471775054932,
+      "learning_rate": 0.0002,
+      "loss": 1.0814,
+      "step": 120
+    },
+    {
+      "epoch": 0.27956989247311825,
+      "grad_norm": 1.0265629291534424,
+      "learning_rate": 0.0002,
+      "loss": 1.0239,
+      "step": 130
+    },
+    {
+      "epoch": 0.3010752688172043,
+      "grad_norm": 0.9344286322593689,
+      "learning_rate": 0.0002,
+      "loss": 0.9888,
+      "step": 140
+    },
+    {
+      "epoch": 0.3225806451612903,
+      "grad_norm": 0.9883386492729187,
+      "learning_rate": 0.0002,
+      "loss": 1.0043,
+      "step": 150
+    },
+    {
+      "epoch": 0.34408602150537637,
+      "grad_norm": 0.9299277067184448,
+      "learning_rate": 0.0002,
+      "loss": 0.9338,
+      "step": 160
+    },
+    {
+      "epoch": 0.3655913978494624,
+      "grad_norm": 1.390045404434204,
+      "learning_rate": 0.0002,
+      "loss": 0.9432,
+      "step": 170
+    },
+    {
+      "epoch": 0.3870967741935484,
+      "grad_norm": 1.0313078165054321,
+      "learning_rate": 0.0002,
+      "loss": 0.9008,
+      "step": 180
+    },
+    {
+      "epoch": 0.40860215053763443,
+      "grad_norm": 1.1792205572128296,
+      "learning_rate": 0.0002,
+      "loss": 0.9434,
+      "step": 190
+    },
+    {
+      "epoch": 0.43010752688172044,
+      "grad_norm": 1.049809217453003,
+      "learning_rate": 0.0002,
+      "loss": 0.8761,
+      "step": 200
+    },
+    {
+      "epoch": 0.45161290322580644,
+      "grad_norm": 0.990111768245697,
+      "learning_rate": 0.0002,
+      "loss": 0.8709,
+      "step": 210
+    },
+    {
+      "epoch": 0.4731182795698925,
+      "grad_norm": 0.9870412349700928,
+      "learning_rate": 0.0002,
+      "loss": 0.905,
+      "step": 220
+    },
+    {
+      "epoch": 0.4946236559139785,
+      "grad_norm": 0.8557345867156982,
+      "learning_rate": 0.0002,
+      "loss": 0.9129,
+      "step": 230
+    },
+    {
+      "epoch": 0.5161290322580645,
+      "grad_norm": 0.9746861457824707,
+      "learning_rate": 0.0002,
+      "loss": 0.8836,
+      "step": 240
+    },
+    {
+      "epoch": 0.5376344086021505,
+      "grad_norm": 0.9010438323020935,
+      "learning_rate": 0.0002,
+      "loss": 0.873,
+      "step": 250
+    },
+    {
+      "epoch": 0.5591397849462365,
+      "grad_norm": 0.9061082005500793,
+      "learning_rate": 0.0002,
+      "loss": 0.8241,
+      "step": 260
+    },
+    {
+      "epoch": 0.5806451612903226,
+      "grad_norm": 0.9311846494674683,
+      "learning_rate": 0.0002,
+      "loss": 0.8652,
+      "step": 270
+    },
+    {
+      "epoch": 0.6021505376344086,
+      "grad_norm": 0.9140254855155945,
+      "learning_rate": 0.0002,
+      "loss": 0.8256,
+      "step": 280
+    },
+    {
+      "epoch": 0.6236559139784946,
+      "grad_norm": 0.9722253084182739,
+      "learning_rate": 0.0002,
+      "loss": 0.8441,
+      "step": 290
+    },
+    {
+      "epoch": 0.6451612903225806,
+      "grad_norm": 0.8539168238639832,
+      "learning_rate": 0.0002,
+      "loss": 0.8314,
+      "step": 300
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.9053162932395935,
+      "learning_rate": 0.0002,
+      "loss": 0.8528,
+      "step": 310
+    },
+    {
+      "epoch": 0.6881720430107527,
+      "grad_norm": 0.8444252610206604,
+      "learning_rate": 0.0002,
+      "loss": 0.8209,
+      "step": 320
+    },
+    {
+      "epoch": 0.7096774193548387,
+      "grad_norm": 0.8127437829971313,
+      "learning_rate": 0.0002,
+      "loss": 0.8101,
+      "step": 330
+    },
+    {
+      "epoch": 0.7311827956989247,
+      "grad_norm": 0.886555016040802,
+      "learning_rate": 0.0002,
+      "loss": 0.8223,
+      "step": 340
+    },
+    {
+      "epoch": 0.7526881720430108,
+      "grad_norm": 0.8458548784255981,
+      "learning_rate": 0.0002,
+      "loss": 0.8368,
+      "step": 350
+    },
+    {
+      "epoch": 0.7741935483870968,
+      "grad_norm": 0.8683297634124756,
+      "learning_rate": 0.0002,
+      "loss": 0.8295,
+      "step": 360
+    },
+    {
+      "epoch": 0.7956989247311828,
+      "grad_norm": 0.8308405876159668,
+      "learning_rate": 0.0002,
+      "loss": 0.8232,
+      "step": 370
+    },
+    {
+      "epoch": 0.8172043010752689,
+      "grad_norm": 0.8305579423904419,
+      "learning_rate": 0.0002,
+      "loss": 0.7752,
+      "step": 380
+    },
+    {
+      "epoch": 0.8387096774193549,
+      "grad_norm": 0.8545567393302917,
+      "learning_rate": 0.0002,
+      "loss": 0.8267,
+      "step": 390
+    },
+    {
+      "epoch": 0.8602150537634409,
+      "grad_norm": 0.8486055731773376,
+      "learning_rate": 0.0002,
+      "loss": 0.8212,
+      "step": 400
+    },
+    {
+      "epoch": 0.8817204301075269,
+      "grad_norm": 0.8126763105392456,
+      "learning_rate": 0.0002,
+      "loss": 0.743,
+      "step": 410
+    },
+    {
+      "epoch": 0.9032258064516129,
+      "grad_norm": 0.8494045734405518,
+      "learning_rate": 0.0002,
+      "loss": 0.7993,
+      "step": 420
+    },
+    {
+      "epoch": 0.9247311827956989,
+      "grad_norm": 0.7639183402061462,
+      "learning_rate": 0.0002,
+      "loss": 0.8213,
+      "step": 430
+    },
+    {
+      "epoch": 0.946236559139785,
+      "grad_norm": 0.858101487159729,
+      "learning_rate": 0.0002,
+      "loss": 0.8015,
+      "step": 440
+    },
+    {
+      "epoch": 0.967741935483871,
+      "grad_norm": 0.8141381740570068,
+      "learning_rate": 0.0002,
+      "loss": 0.7629,
+      "step": 450
+    },
+    {
+      "epoch": 0.989247311827957,
+      "grad_norm": 0.8072513937950134,
+      "learning_rate": 0.0002,
+      "loss": 0.7357,
+      "step": 460
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.7740864157676697,
+      "eval_runtime": 21.383,
+      "eval_samples_per_second": 15.48,
+      "eval_steps_per_second": 1.964,
+      "step": 465
+    },
+    {
+      "epoch": 1.010752688172043,
+      "grad_norm": 0.8269494771957397,
+      "learning_rate": 0.0002,
+      "loss": 0.7701,
+      "step": 470
+    },
+    {
+      "epoch": 1.032258064516129,
+      "grad_norm": 0.7814009189605713,
+      "learning_rate": 0.0002,
+      "loss": 0.7532,
+      "step": 480
+    },
+    {
+      "epoch": 1.053763440860215,
+      "grad_norm": 0.8183923363685608,
+      "learning_rate": 0.0002,
+      "loss": 0.7689,
+      "step": 490
+    },
+    {
+      "epoch": 1.075268817204301,
+      "grad_norm": 0.8146600723266602,
+      "learning_rate": 0.0002,
+      "loss": 0.765,
+      "step": 500
+    },
+    {
+      "epoch": 1.096774193548387,
+      "grad_norm": 0.8635126352310181,
+      "learning_rate": 0.0002,
+      "loss": 0.7358,
+      "step": 510
+    },
+    {
+      "epoch": 1.118279569892473,
+      "grad_norm": 0.8520359396934509,
+      "learning_rate": 0.0002,
+      "loss": 0.7302,
+      "step": 520
+    },
+    {
+      "epoch": 1.139784946236559,
+      "grad_norm": 0.8026443123817444,
+      "learning_rate": 0.0002,
+      "loss": 0.7492,
+      "step": 530
+    },
+    {
+      "epoch": 1.1612903225806452,
+      "grad_norm": 0.8157258629798889,
+      "learning_rate": 0.0002,
+      "loss": 0.7518,
+      "step": 540
+    },
+    {
+      "epoch": 1.1827956989247312,
+      "grad_norm": 0.9450796246528625,
+      "learning_rate": 0.0002,
+      "loss": 0.7461,
+      "step": 550
+    },
+    {
+      "epoch": 1.2043010752688172,
+      "grad_norm": 0.8859835863113403,
+      "learning_rate": 0.0002,
+      "loss": 0.7128,
+      "step": 560
+    },
+    {
+      "epoch": 1.2258064516129032,
+      "grad_norm": 0.7819921970367432,
+      "learning_rate": 0.0002,
+      "loss": 0.7067,
+      "step": 570
+    },
+    {
+      "epoch": 1.2473118279569892,
+      "grad_norm": 0.7823445796966553,
+      "learning_rate": 0.0002,
+      "loss": 0.7577,
+      "step": 580
+    },
+    {
+      "epoch": 1.2688172043010753,
+      "grad_norm": 0.7931883931159973,
+      "learning_rate": 0.0002,
+      "loss": 0.7358,
+      "step": 590
+    },
+    {
+      "epoch": 1.2903225806451613,
+      "grad_norm": 0.7495734095573425,
+      "learning_rate": 0.0002,
+      "loss": 0.723,
+      "step": 600
+    },
+    {
+      "epoch": 1.3118279569892473,
+      "grad_norm": 0.9272717237472534,
+      "learning_rate": 0.0002,
+      "loss": 0.7386,
+      "step": 610
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 0.7968398332595825,
+      "learning_rate": 0.0002,
+      "loss": 0.7498,
+      "step": 620
+    },
+    {
+      "epoch": 1.3548387096774195,
+      "grad_norm": 0.7813659310340881,
+      "learning_rate": 0.0002,
+      "loss": 0.7635,
+      "step": 630
+    },
+    {
+      "epoch": 1.3763440860215055,
+      "grad_norm": 0.730925977230072,
+      "learning_rate": 0.0002,
+      "loss": 0.6665,
+      "step": 640
+    },
+    {
+      "epoch": 1.3978494623655915,
+      "grad_norm": 0.8011482954025269,
+      "learning_rate": 0.0002,
+      "loss": 0.7037,
+      "step": 650
+    },
+    {
+      "epoch": 1.4193548387096775,
+      "grad_norm": 0.7770085334777832,
+      "learning_rate": 0.0002,
+      "loss": 0.6931,
+      "step": 660
+    },
+    {
+      "epoch": 1.4408602150537635,
+      "grad_norm": 0.7432682514190674,
+      "learning_rate": 0.0002,
+      "loss": 0.6949,
+      "step": 670
+    },
+    {
+      "epoch": 1.4623655913978495,
+      "grad_norm": 0.8820092678070068,
+      "learning_rate": 0.0002,
+      "loss": 0.7444,
+      "step": 680
+    },
+    {
+      "epoch": 1.4838709677419355,
+      "grad_norm": 0.7786208987236023,
+      "learning_rate": 0.0002,
+      "loss": 0.6758,
+      "step": 690
+    },
+    {
+      "epoch": 1.5053763440860215,
+      "grad_norm": 0.7467480301856995,
+      "learning_rate": 0.0002,
+      "loss": 0.6702,
+      "step": 700
+    },
+    {
+      "epoch": 1.5268817204301075,
+      "grad_norm": 0.8147122263908386,
+      "learning_rate": 0.0002,
+      "loss": 0.7107,
+      "step": 710
+    },
+    {
+      "epoch": 1.5483870967741935,
+      "grad_norm": 0.796030580997467,
+      "learning_rate": 0.0002,
+      "loss": 0.7144,
+      "step": 720
+    },
+    {
+      "epoch": 1.5698924731182795,
+      "grad_norm": 0.8776171207427979,
+      "learning_rate": 0.0002,
+      "loss": 0.6936,
+      "step": 730
+    },
+    {
+      "epoch": 1.5913978494623655,
+      "grad_norm": 0.8056126236915588,
+      "learning_rate": 0.0002,
+      "loss": 0.7101,
+      "step": 740
+    },
+    {
+      "epoch": 1.6129032258064515,
+      "grad_norm": 0.8141863346099854,
+      "learning_rate": 0.0002,
+      "loss": 0.7162,
+      "step": 750
+    },
+    {
+      "epoch": 1.6344086021505375,
+      "grad_norm": 0.8100557327270508,
+      "learning_rate": 0.0002,
+      "loss": 0.7088,
+      "step": 760
+    },
+    {
+      "epoch": 1.6559139784946235,
+      "grad_norm": 0.8283200860023499,
+      "learning_rate": 0.0002,
+      "loss": 0.7212,
+      "step": 770
+    },
+    {
+      "epoch": 1.6774193548387095,
+      "grad_norm": 0.800865113735199,
+      "learning_rate": 0.0002,
+      "loss": 0.694,
+      "step": 780
+    },
+    {
+      "epoch": 1.6989247311827957,
+      "grad_norm": 0.8052287697792053,
+      "learning_rate": 0.0002,
+      "loss": 0.7076,
+      "step": 790
+    },
+    {
+      "epoch": 1.7204301075268817,
+      "grad_norm": 0.8619674444198608,
+      "learning_rate": 0.0002,
+      "loss": 0.7257,
+      "step": 800
+    },
+    {
+      "epoch": 1.7419354838709677,
+      "grad_norm": 0.8907215595245361,
+      "learning_rate": 0.0002,
+      "loss": 0.7141,
+      "step": 810
+    },
+    {
+      "epoch": 1.7634408602150538,
+      "grad_norm": 0.6976316571235657,
+      "learning_rate": 0.0002,
+      "loss": 0.7035,
+      "step": 820
+    },
+    {
+      "epoch": 1.7849462365591398,
+      "grad_norm": 0.7533746957778931,
+      "learning_rate": 0.0002,
+      "loss": 0.6916,
+      "step": 830
+    },
+    {
+      "epoch": 1.8064516129032258,
+      "grad_norm": 0.7326804399490356,
+      "learning_rate": 0.0002,
+      "loss": 0.7094,
+      "step": 840
+    },
+    {
+      "epoch": 1.827956989247312,
+      "grad_norm": 0.7782683372497559,
+      "learning_rate": 0.0002,
+      "loss": 0.6891,
+      "step": 850
+    },
+    {
+      "epoch": 1.849462365591398,
+      "grad_norm": 0.7424806356430054,
+      "learning_rate": 0.0002,
+      "loss": 0.6931,
+      "step": 860
+    },
+    {
+      "epoch": 1.870967741935484,
+      "grad_norm": 1.172325611114502,
+      "learning_rate": 0.0002,
+      "loss": 0.7354,
+      "step": 870
+    },
+    {
+      "epoch": 1.89247311827957,
+      "grad_norm": 0.771058201789856,
+      "learning_rate": 0.0002,
+      "loss": 0.6866,
+      "step": 880
+    },
+    {
+      "epoch": 1.913978494623656,
+      "grad_norm": 0.8624904155731201,
+      "learning_rate": 0.0002,
+      "loss": 0.7296,
+      "step": 890
+    },
+    {
+      "epoch": 1.935483870967742,
+      "grad_norm": 0.7062820792198181,
+      "learning_rate": 0.0002,
+      "loss": 0.7233,
+      "step": 900
+    },
+    {
+      "epoch": 1.956989247311828,
+      "grad_norm": 0.7560103535652161,
+      "learning_rate": 0.0002,
+      "loss": 0.6966,
+      "step": 910
+    },
+    {
+      "epoch": 1.978494623655914,
+      "grad_norm": 0.788899838924408,
+      "learning_rate": 0.0002,
+      "loss": 0.69,
+      "step": 920
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.6562113761901855,
+      "learning_rate": 0.0002,
+      "loss": 0.6505,
+      "step": 930
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.6885261535644531,
+      "eval_runtime": 21.4291,
+      "eval_samples_per_second": 15.446,
+      "eval_steps_per_second": 1.96,
+      "step": 930
+    },
+    {
+      "epoch": 2.021505376344086,
+      "grad_norm": 0.8216531872749329,
+      "learning_rate": 0.0002,
+      "loss": 0.6625,
+      "step": 940
+    },
+    {
+      "epoch": 2.043010752688172,
+      "grad_norm": 0.8317142724990845,
+      "learning_rate": 0.0002,
+      "loss": 0.6398,
+      "step": 950
+    },
+    {
+      "epoch": 2.064516129032258,
+      "grad_norm": 0.8446708917617798,
+      "learning_rate": 0.0002,
+      "loss": 0.649,
+      "step": 960
+    },
+    {
+      "epoch": 2.086021505376344,
+      "grad_norm": 0.735055148601532,
+      "learning_rate": 0.0002,
+      "loss": 0.657,
+      "step": 970
+    },
+    {
+      "epoch": 2.10752688172043,
+      "grad_norm": 0.7487243413925171,
+      "learning_rate": 0.0002,
+      "loss": 0.649,
+      "step": 980
+    },
+    {
+      "epoch": 2.129032258064516,
+      "grad_norm": 0.8573887944221497,
+      "learning_rate": 0.0002,
+      "loss": 0.6419,
+      "step": 990
+    },
+    {
+      "epoch": 2.150537634408602,
+      "grad_norm": 0.6284521818161011,
+      "learning_rate": 0.0002,
+      "loss": 0.6431,
+      "step": 1000
+    },
+    {
+      "epoch": 2.172043010752688,
+      "grad_norm": 0.754183292388916,
+      "learning_rate": 0.0002,
+      "loss": 0.6128,
+      "step": 1010
+    },
+    {
+      "epoch": 2.193548387096774,
+      "grad_norm": 0.9445359110832214,
+      "learning_rate": 0.0002,
+      "loss": 0.6253,
+      "step": 1020
+    },
+    {
+      "epoch": 2.21505376344086,
+      "grad_norm": 0.808508038520813,
+      "learning_rate": 0.0002,
+      "loss": 0.605,
+      "step": 1030
+    },
+    {
+      "epoch": 2.236559139784946,
+      "grad_norm": 0.9394679665565491,
+      "learning_rate": 0.0002,
+      "loss": 0.6786,
+      "step": 1040
+    },
+    {
+      "epoch": 2.258064516129032,
+      "grad_norm": 0.8151357769966125,
+      "learning_rate": 0.0002,
+      "loss": 0.6176,
+      "step": 1050
+    },
+    {
+      "epoch": 2.279569892473118,
+      "grad_norm": 0.7909848093986511,
+      "learning_rate": 0.0002,
+      "loss": 0.66,
+      "step": 1060
+    },
+    {
+      "epoch": 2.3010752688172045,
+      "grad_norm": 0.7506507039070129,
+      "learning_rate": 0.0002,
+      "loss": 0.6254,
+      "step": 1070
+    },
+    {
+      "epoch": 2.3225806451612905,
+      "grad_norm": 0.8240520358085632,
+      "learning_rate": 0.0002,
+      "loss": 0.6608,
+      "step": 1080
+    },
+    {
+      "epoch": 2.3440860215053765,
+      "grad_norm": 0.9342400431632996,
+      "learning_rate": 0.0002,
+      "loss": 0.6207,
+      "step": 1090
+    },
+    {
+      "epoch": 2.3655913978494625,
+      "grad_norm": 1.0598735809326172,
+      "learning_rate": 0.0002,
+      "loss": 0.6029,
+      "step": 1100
+    },
+    {
+      "epoch": 2.3870967741935485,
+      "grad_norm": 0.7907650470733643,
+      "learning_rate": 0.0002,
+      "loss": 0.6035,
+      "step": 1110
+    },
+    {
+      "epoch": 2.4086021505376345,
+      "grad_norm": 0.9388798475265503,
+      "learning_rate": 0.0002,
+      "loss": 0.6237,
+      "step": 1120
+    },
+    {
+      "epoch": 2.4301075268817205,
+      "grad_norm": 0.8985419869422913,
+      "learning_rate": 0.0002,
+      "loss": 0.6207,
+      "step": 1130
+    },
+    {
+      "epoch": 2.4516129032258065,
+      "grad_norm": 0.7471932768821716,
+      "learning_rate": 0.0002,
+      "loss": 0.5902,
+      "step": 1140
+    },
+    {
+      "epoch": 2.4731182795698925,
+      "grad_norm": 0.761131763458252,
+      "learning_rate": 0.0002,
+      "loss": 0.6446,
+      "step": 1150
+    },
+    {
+      "epoch": 2.4946236559139785,
+      "grad_norm": 0.7901819348335266,
+      "learning_rate": 0.0002,
+      "loss": 0.6088,
+      "step": 1160
+    },
+    {
+      "epoch": 2.5161290322580645,
+      "grad_norm": 0.9932922720909119,
+      "learning_rate": 0.0002,
+      "loss": 0.6142,
+      "step": 1170
+    },
+    {
+      "epoch": 2.5376344086021505,
+      "grad_norm": 0.7414287328720093,
+      "learning_rate": 0.0002,
+      "loss": 0.6407,
+      "step": 1180
+    },
+    {
+      "epoch": 2.5591397849462365,
+      "grad_norm": 0.8111771941184998,
+      "learning_rate": 0.0002,
+      "loss": 0.6161,
+      "step": 1190
+    },
+    {
+      "epoch": 2.5806451612903225,
+      "grad_norm": 0.7520156502723694,
+      "learning_rate": 0.0002,
+      "loss": 0.6006,
+      "step": 1200
+    },
+    {
+      "epoch": 2.6021505376344085,
+      "grad_norm": 0.9022907018661499,
+      "learning_rate": 0.0002,
+      "loss": 0.615,
+      "step": 1210
+    },
+    {
+      "epoch": 2.6236559139784945,
+      "grad_norm": 0.7746260166168213,
+      "learning_rate": 0.0002,
+      "loss": 0.6211,
+      "step": 1220
+    },
+    {
+      "epoch": 2.6451612903225805,
+      "grad_norm": 0.8482862114906311,
+      "learning_rate": 0.0002,
+      "loss": 0.616,
+      "step": 1230
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 0.7925458550453186,
+      "learning_rate": 0.0002,
+      "loss": 0.6417,
+      "step": 1240
+    },
+    {
+      "epoch": 2.688172043010753,
+      "grad_norm": 0.8369929194450378,
+      "learning_rate": 0.0002,
+      "loss": 0.6187,
+      "step": 1250
+    },
+    {
+      "epoch": 2.709677419354839,
+      "grad_norm": 0.8311542868614197,
+      "learning_rate": 0.0002,
+      "loss": 0.6138,
+      "step": 1260
+    },
+    {
+      "epoch": 2.731182795698925,
+      "grad_norm": 0.7204853296279907,
+      "learning_rate": 0.0002,
+      "loss": 0.5894,
+      "step": 1270
+    },
+    {
+      "epoch": 2.752688172043011,
+      "grad_norm": 0.8447284698486328,
+      "learning_rate": 0.0002,
+      "loss": 0.6325,
+      "step": 1280
+    },
+    {
+      "epoch": 2.774193548387097,
+      "grad_norm": 0.7738404273986816,
+      "learning_rate": 0.0002,
+      "loss": 0.5946,
+      "step": 1290
+    },
+    {
+      "epoch": 2.795698924731183,
+      "grad_norm": 0.8393287062644958,
+      "learning_rate": 0.0002,
+      "loss": 0.5678,
+      "step": 1300
+    },
+    {
+      "epoch": 2.817204301075269,
+      "grad_norm": 0.79121994972229,
+      "learning_rate": 0.0002,
+      "loss": 0.6092,
+      "step": 1310
+    },
+    {
+      "epoch": 2.838709677419355,
+      "grad_norm": 0.7331557869911194,
+      "learning_rate": 0.0002,
+      "loss": 0.5889,
+      "step": 1320
+    },
+    {
+      "epoch": 2.860215053763441,
+      "grad_norm": 0.9593998193740845,
+      "learning_rate": 0.0002,
+      "loss": 0.6048,
+      "step": 1330
+    },
+    {
+      "epoch": 2.881720430107527,
+      "grad_norm": 0.7215158343315125,
+      "learning_rate": 0.0002,
+      "loss": 0.6108,
+      "step": 1340
+    },
+    {
+      "epoch": 2.903225806451613,
+      "grad_norm": 0.840404212474823,
+      "learning_rate": 0.0002,
+      "loss": 0.5897,
+      "step": 1350
+    },
+    {
+      "epoch": 2.924731182795699,
+      "grad_norm": 0.870659351348877,
+      "learning_rate": 0.0002,
+      "loss": 0.6056,
+      "step": 1360
+    },
+    {
+      "epoch": 2.946236559139785,
+      "grad_norm": 0.8744975328445435,
+      "learning_rate": 0.0002,
+      "loss": 0.6205,
+      "step": 1370
+    },
+    {
+      "epoch": 2.967741935483871,
+      "grad_norm": 0.8030612468719482,
+      "learning_rate": 0.0002,
+      "loss": 0.5966,
+      "step": 1380
+    },
+    {
+      "epoch": 2.989247311827957,
+      "grad_norm": 0.825814962387085,
+      "learning_rate": 0.0002,
+      "loss": 0.6004,
+      "step": 1390
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 0.6257933378219604,
+      "eval_runtime": 21.3692,
+      "eval_samples_per_second": 15.49,
+      "eval_steps_per_second": 1.965,
+      "step": 1395
+    },
+    {
+      "epoch": 3.010752688172043,
+      "grad_norm": 0.8650677800178528,
+      "learning_rate": 0.0002,
+      "loss": 0.5696,
+      "step": 1400
+    },
+    {
+      "epoch": 3.032258064516129,
+      "grad_norm": 0.8364197015762329,
+      "learning_rate": 0.0002,
+      "loss": 0.5483,
+      "step": 1410
+    },
+    {
+      "epoch": 3.053763440860215,
+      "grad_norm": 0.8278448581695557,
+      "learning_rate": 0.0002,
+      "loss": 0.5606,
+      "step": 1420
+    },
+    {
+      "epoch": 3.075268817204301,
+      "grad_norm": 0.8806642889976501,
+      "learning_rate": 0.0002,
+      "loss": 0.5572,
+      "step": 1430
+    },
+    {
+      "epoch": 3.096774193548387,
+      "grad_norm": 0.8180029988288879,
+      "learning_rate": 0.0002,
+      "loss": 0.585,
+      "step": 1440
+    },
+    {
+      "epoch": 3.118279569892473,
+      "grad_norm": 0.8561782836914062,
+      "learning_rate": 0.0002,
+      "loss": 0.5667,
+      "step": 1450
+    },
+    {
+      "epoch": 3.139784946236559,
+      "grad_norm": 0.8377029299736023,
+      "learning_rate": 0.0002,
+      "loss": 0.5246,
+      "step": 1460
+    },
+    {
+      "epoch": 3.161290322580645,
+      "grad_norm": 0.885779082775116,
+      "learning_rate": 0.0002,
+      "loss": 0.5464,
+      "step": 1470
+    },
+    {
+      "epoch": 3.182795698924731,
+      "grad_norm": 0.9388518333435059,
+      "learning_rate": 0.0002,
+      "loss": 0.541,
+      "step": 1480
+    },
+    {
+      "epoch": 3.204301075268817,
+      "grad_norm": 0.8816235661506653,
+      "learning_rate": 0.0002,
+      "loss": 0.5447,
+      "step": 1490
+    },
+    {
+      "epoch": 3.225806451612903,
+      "grad_norm": 0.9885783791542053,
+      "learning_rate": 0.0002,
+      "loss": 0.5466,
+      "step": 1500
+    },
+    {
+      "epoch": 3.247311827956989,
+      "grad_norm": 0.8635850548744202,
+      "learning_rate": 0.0002,
+      "loss": 0.5455,
+      "step": 1510
+    },
+    {
+      "epoch": 3.268817204301075,
+      "grad_norm": 0.829853355884552,
+      "learning_rate": 0.0002,
+      "loss": 0.5419,
+      "step": 1520
+    },
+    {
+      "epoch": 3.2903225806451615,
+      "grad_norm": 0.9037486910820007,
+      "learning_rate": 0.0002,
+      "loss": 0.54,
+      "step": 1530
+    },
+    {
+      "epoch": 3.3118279569892475,
+      "grad_norm": 0.8173713684082031,
+      "learning_rate": 0.0002,
+      "loss": 0.5375,
+      "step": 1540
+    },
+    {
+      "epoch": 3.3333333333333335,
+      "grad_norm": 0.796953022480011,
+      "learning_rate": 0.0002,
+      "loss": 0.5405,
+      "step": 1550
+    },
+    {
+      "epoch": 3.3548387096774195,
+      "grad_norm": 0.7894400358200073,
+      "learning_rate": 0.0002,
+      "loss": 0.5505,
+      "step": 1560
+    },
+    {
+      "epoch": 3.3763440860215055,
+      "grad_norm": 0.9434949159622192,
+      "learning_rate": 0.0002,
+      "loss": 0.5395,
+      "step": 1570
+    },
+    {
+      "epoch": 3.3978494623655915,
+      "grad_norm": 0.8666760325431824,
+      "learning_rate": 0.0002,
+      "loss": 0.5271,
+      "step": 1580
+    },
+    {
+      "epoch": 3.4193548387096775,
+      "grad_norm": 0.7782467007637024,
+      "learning_rate": 0.0002,
+      "loss": 0.5439,
+      "step": 1590
+    },
+    {
+      "epoch": 3.4408602150537635,
+      "grad_norm": 0.8849126696586609,
+      "learning_rate": 0.0002,
+      "loss": 0.5161,
+      "step": 1600
+    },
+    {
+      "epoch": 3.4623655913978495,
+      "grad_norm": 0.7863831520080566,
+      "learning_rate": 0.0002,
+      "loss": 0.5353,
+      "step": 1610
+    },
+    {
+      "epoch": 3.4838709677419355,
+      "grad_norm": 1.0403116941452026,
+      "learning_rate": 0.0002,
+      "loss": 0.5308,
+      "step": 1620
+    },
+    {
+      "epoch": 3.5053763440860215,
+      "grad_norm": 0.8307499289512634,
+      "learning_rate": 0.0002,
+      "loss": 0.5339,
+      "step": 1630
+    },
+    {
+      "epoch": 3.5268817204301075,
+      "grad_norm": 0.9132118821144104,
+      "learning_rate": 0.0002,
+      "loss": 0.5361,
+      "step": 1640
+    },
+    {
+      "epoch": 3.5483870967741935,
+      "grad_norm": 0.9322578310966492,
+      "learning_rate": 0.0002,
+      "loss": 0.5828,
+      "step": 1650
+    },
+    {
+      "epoch": 3.5698924731182795,
+      "grad_norm": 0.9782460331916809,
+      "learning_rate": 0.0002,
+      "loss": 0.546,
+      "step": 1660
+    },
+    {
+      "epoch": 3.5913978494623655,
+      "grad_norm": 0.7189919352531433,
+      "learning_rate": 0.0002,
+      "loss": 0.5424,
+      "step": 1670
+    },
+    {
+      "epoch": 3.6129032258064515,
+      "grad_norm": 0.9689221382141113,
+      "learning_rate": 0.0002,
+      "loss": 0.5514,
+      "step": 1680
+    },
+    {
+      "epoch": 3.6344086021505375,
+      "grad_norm": 0.9684675335884094,
+      "learning_rate": 0.0002,
+      "loss": 0.5379,
+      "step": 1690
+    },
+    {
+      "epoch": 3.6559139784946235,
+      "grad_norm": 0.8851472735404968,
+      "learning_rate": 0.0002,
+      "loss": 0.5748,
+      "step": 1700
+    },
+    {
+      "epoch": 3.6774193548387095,
+      "grad_norm": 0.7709833383560181,
+      "learning_rate": 0.0002,
+      "loss": 0.5412,
+      "step": 1710
+    },
+    {
+      "epoch": 3.698924731182796,
+      "grad_norm": 0.818236231803894,
+      "learning_rate": 0.0002,
+      "loss": 0.521,
+      "step": 1720
+    },
+    {
+      "epoch": 3.720430107526882,
+      "grad_norm": 0.870642364025116,
+      "learning_rate": 0.0002,
+      "loss": 0.5445,
+      "step": 1730
+    },
+    {
+      "epoch": 3.741935483870968,
+      "grad_norm": 1.0245511531829834,
+      "learning_rate": 0.0002,
+      "loss": 0.5307,
+      "step": 1740
+    },
+    {
+      "epoch": 3.763440860215054,
+      "grad_norm": 0.8607558608055115,
+      "learning_rate": 0.0002,
+      "loss": 0.5593,
+      "step": 1750
+    },
+    {
+      "epoch": 3.78494623655914,
+      "grad_norm": 0.8511829972267151,
+      "learning_rate": 0.0002,
+      "loss": 0.536,
+      "step": 1760
+    },
+    {
+      "epoch": 3.806451612903226,
+      "grad_norm": 0.7969087362289429,
+      "learning_rate": 0.0002,
+      "loss": 0.5193,
+      "step": 1770
+    },
+    {
+      "epoch": 3.827956989247312,
+      "grad_norm": 0.8457245826721191,
+      "learning_rate": 0.0002,
+      "loss": 0.5578,
+      "step": 1780
+    },
+    {
+      "epoch": 3.849462365591398,
+      "grad_norm": 0.8893467783927917,
+      "learning_rate": 0.0002,
+      "loss": 0.5337,
+      "step": 1790
+    },
+    {
+      "epoch": 3.870967741935484,
+      "grad_norm": 0.8593819737434387,
+      "learning_rate": 0.0002,
+      "loss": 0.5024,
+      "step": 1800
+    },
+    {
+      "epoch": 3.89247311827957,
+      "grad_norm": 0.7574560642242432,
+      "learning_rate": 0.0002,
+      "loss": 0.5134,
+      "step": 1810
+    },
+    {
+      "epoch": 3.913978494623656,
+      "grad_norm": 0.8681567311286926,
+      "learning_rate": 0.0002,
+      "loss": 0.5263,
+      "step": 1820
+    },
+    {
+      "epoch": 3.935483870967742,
+      "grad_norm": 0.9068132042884827,
+      "learning_rate": 0.0002,
+      "loss": 0.532,
+      "step": 1830
+    },
+    {
+      "epoch": 3.956989247311828,
+      "grad_norm": 0.8668948411941528,
+      "learning_rate": 0.0002,
+      "loss": 0.5427,
+      "step": 1840
+    },
+    {
+      "epoch": 3.978494623655914,
+      "grad_norm": 1.046032428741455,
+      "learning_rate": 0.0002,
+      "loss": 0.5349,
+      "step": 1850
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.904780387878418,
+      "learning_rate": 0.0002,
+      "loss": 0.5087,
+      "step": 1860
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 0.5737715363502502,
+      "eval_runtime": 21.4915,
+      "eval_samples_per_second": 15.401,
+      "eval_steps_per_second": 1.954,
+      "step": 1860
+    },
+    {
+      "epoch": 4.021505376344086,
+      "grad_norm": 0.8611752986907959,
+      "learning_rate": 0.0002,
+      "loss": 0.4843,
+      "step": 1870
+    },
+    {
+      "epoch": 4.043010752688172,
+      "grad_norm": 0.838782548904419,
+      "learning_rate": 0.0002,
+      "loss": 0.4814,
+      "step": 1880
+    },
+    {
+      "epoch": 4.064516129032258,
+      "grad_norm": 0.9119709134101868,
+      "learning_rate": 0.0002,
+      "loss": 0.474,
+      "step": 1890
+    },
+    {
+      "epoch": 4.086021505376344,
+      "grad_norm": 0.8026251196861267,
+      "learning_rate": 0.0002,
+      "loss": 0.4951,
+      "step": 1900
+    },
+    {
+      "epoch": 4.10752688172043,
+      "grad_norm": 0.8773705363273621,
+      "learning_rate": 0.0002,
+      "loss": 0.491,
+      "step": 1910
+    },
+    {
+      "epoch": 4.129032258064516,
+      "grad_norm": 0.8762255907058716,
+      "learning_rate": 0.0002,
+      "loss": 0.474,
+      "step": 1920
+    },
+    {
+      "epoch": 4.150537634408602,
+      "grad_norm": 0.8371861577033997,
+      "learning_rate": 0.0002,
+      "loss": 0.4816,
+      "step": 1930
+    },
+    {
+      "epoch": 4.172043010752688,
+      "grad_norm": 0.9703728556632996,
+      "learning_rate": 0.0002,
+      "loss": 0.472,
+      "step": 1940
+    },
+    {
+      "epoch": 4.193548387096774,
+      "grad_norm": 0.8802874684333801,
+      "learning_rate": 0.0002,
+      "loss": 0.4772,
+      "step": 1950
+    },
+    {
+      "epoch": 4.21505376344086,
+      "grad_norm": 1.0103057622909546,
+      "learning_rate": 0.0002,
+      "loss": 0.5032,
+      "step": 1960
+    },
+    {
+      "epoch": 4.236559139784946,
+      "grad_norm": 0.9212995171546936,
+      "learning_rate": 0.0002,
+      "loss": 0.4945,
+      "step": 1970
+    },
+    {
+      "epoch": 4.258064516129032,
+      "grad_norm": 1.009544849395752,
+      "learning_rate": 0.0002,
+      "loss": 0.4753,
+      "step": 1980
+    },
+    {
+      "epoch": 4.279569892473118,
+      "grad_norm": 0.8535077571868896,
+      "learning_rate": 0.0002,
+      "loss": 0.4789,
+      "step": 1990
+    },
+    {
+      "epoch": 4.301075268817204,
+      "grad_norm": 0.8363022804260254,
+      "learning_rate": 0.0002,
+      "loss": 0.4782,
+      "step": 2000
+    },
+    {
+      "epoch": 4.32258064516129,
+      "grad_norm": 0.9041762948036194,
+      "learning_rate": 0.0002,
+      "loss": 0.4875,
+      "step": 2010
+    },
+    {
+      "epoch": 4.344086021505376,
+      "grad_norm": 0.960790753364563,
+      "learning_rate": 0.0002,
+      "loss": 0.4779,
+      "step": 2020
+    },
+    {
+      "epoch": 4.365591397849462,
+      "grad_norm": 0.8823095560073853,
+      "learning_rate": 0.0002,
+      "loss": 0.4626,
+      "step": 2030
+    },
+    {
+      "epoch": 4.387096774193548,
+      "grad_norm": 0.952100396156311,
+      "learning_rate": 0.0002,
+      "loss": 0.4883,
+      "step": 2040
+    },
+    {
+      "epoch": 4.408602150537634,
+      "grad_norm": 1.0793498754501343,
+      "learning_rate": 0.0002,
+      "loss": 0.4789,
+      "step": 2050
+    },
+    {
+      "epoch": 4.43010752688172,
+      "grad_norm": 0.8987208008766174,
+      "learning_rate": 0.0002,
+      "loss": 0.4827,
+      "step": 2060
+    },
+    {
+      "epoch": 4.451612903225806,
+      "grad_norm": 0.8539772033691406,
+      "learning_rate": 0.0002,
+      "loss": 0.4594,
+      "step": 2070
+    },
+    {
+      "epoch": 4.473118279569892,
+      "grad_norm": 0.9160863757133484,
+      "learning_rate": 0.0002,
+      "loss": 0.4752,
+      "step": 2080
+    },
+    {
+      "epoch": 4.494623655913978,
+      "grad_norm": 0.9946850538253784,
+      "learning_rate": 0.0002,
+      "loss": 0.5033,
+      "step": 2090
+    },
+    {
+      "epoch": 4.516129032258064,
+      "grad_norm": 0.908039391040802,
+      "learning_rate": 0.0002,
+      "loss": 0.4842,
+      "step": 2100
+    },
+    {
+      "epoch": 4.53763440860215,
+      "grad_norm": 1.1462254524230957,
+      "learning_rate": 0.0002,
+      "loss": 0.4861,
+      "step": 2110
+    },
+    {
+      "epoch": 4.559139784946236,
+      "grad_norm": 0.8392056226730347,
+      "learning_rate": 0.0002,
+      "loss": 0.4892,
+      "step": 2120
+    },
+    {
+      "epoch": 4.580645161290323,
+      "grad_norm": 0.9673896431922913,
+      "learning_rate": 0.0002,
+      "loss": 0.4824,
+      "step": 2130
+    },
+    {
+      "epoch": 4.602150537634409,
+      "grad_norm": 0.9047091603279114,
+      "learning_rate": 0.0002,
+      "loss": 0.4665,
+      "step": 2140
+    },
+    {
+      "epoch": 4.623655913978495,
+      "grad_norm": 0.9013425707817078,
+      "learning_rate": 0.0002,
+      "loss": 0.4714,
+      "step": 2150
+    },
+    {
+      "epoch": 4.645161290322581,
+      "grad_norm": 0.8899165391921997,
+      "learning_rate": 0.0002,
+      "loss": 0.472,
+      "step": 2160
+    },
+    {
+      "epoch": 4.666666666666667,
+      "grad_norm": 0.748602569103241,
+      "learning_rate": 0.0002,
+      "loss": 0.4635,
+      "step": 2170
+    },
+    {
+      "epoch": 4.688172043010753,
+      "grad_norm": 0.8694155216217041,
+      "learning_rate": 0.0002,
+      "loss": 0.4695,
+      "step": 2180
+    },
+    {
+      "epoch": 4.709677419354839,
+      "grad_norm": 0.9134316444396973,
+      "learning_rate": 0.0002,
+      "loss": 0.4929,
+      "step": 2190
+    },
+    {
+      "epoch": 4.731182795698925,
+      "grad_norm": 0.8504763245582581,
+      "learning_rate": 0.0002,
+      "loss": 0.4855,
+      "step": 2200
+    },
+    {
+      "epoch": 4.752688172043011,
+      "grad_norm": 1.0321544408798218,
+      "learning_rate": 0.0002,
+      "loss": 0.4517,
+      "step": 2210
+    },
+    {
+      "epoch": 4.774193548387097,
+      "grad_norm": 0.9368237257003784,
+      "learning_rate": 0.0002,
+      "loss": 0.4796,
+      "step": 2220
+    },
+    {
+      "epoch": 4.795698924731183,
+      "grad_norm": 0.9319947361946106,
+      "learning_rate": 0.0002,
+      "loss": 0.4837,
+      "step": 2230
+    },
+    {
+      "epoch": 4.817204301075269,
+      "grad_norm": 0.904333770275116,
+      "learning_rate": 0.0002,
+      "loss": 0.4696,
+      "step": 2240
+    },
+    {
+      "epoch": 4.838709677419355,
+      "grad_norm": 0.8097078204154968,
+      "learning_rate": 0.0002,
+      "loss": 0.4746,
+      "step": 2250
+    },
+    {
+      "epoch": 4.860215053763441,
+      "grad_norm": 0.9128859043121338,
+      "learning_rate": 0.0002,
+      "loss": 0.4438,
+      "step": 2260
+    },
+    {
+      "epoch": 4.881720430107527,
+      "grad_norm": 0.883129894733429,
+      "learning_rate": 0.0002,
+      "loss": 0.4693,
+      "step": 2270
+    },
+    {
+      "epoch": 4.903225806451613,
+      "grad_norm": 0.85712730884552,
+      "learning_rate": 0.0002,
+      "loss": 0.4494,
+      "step": 2280
+    },
+    {
+      "epoch": 4.924731182795699,
+      "grad_norm": 1.2101863622665405,
+      "learning_rate": 0.0002,
+      "loss": 0.4593,
+      "step": 2290
+    },
+    {
+      "epoch": 4.946236559139785,
+      "grad_norm": 0.917966902256012,
+      "learning_rate": 0.0002,
+      "loss": 0.4779,
+      "step": 2300
+    },
+    {
+      "epoch": 4.967741935483871,
+      "grad_norm": 0.7740724086761475,
+      "learning_rate": 0.0002,
+      "loss": 0.4666,
+      "step": 2310
+    },
+    {
+      "epoch": 4.989247311827957,
+      "grad_norm": 1.0199906826019287,
+      "learning_rate": 0.0002,
+      "loss": 0.4629,
+      "step": 2320
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 0.5363914370536804,
+      "eval_runtime": 21.3941,
+      "eval_samples_per_second": 15.472,
+      "eval_steps_per_second": 1.963,
+      "step": 2325
+    },
+    {
+      "epoch": 5.010752688172043,
+      "grad_norm": 0.8580502271652222,
+      "learning_rate": 0.0002,
+      "loss": 0.4543,
+      "step": 2330
+    },
+    {
+      "epoch": 5.032258064516129,
+      "grad_norm": 0.7702704668045044,
+      "learning_rate": 0.0002,
+      "loss": 0.404,
+      "step": 2340
+    },
+    {
+      "epoch": 5.053763440860215,
+      "grad_norm": 0.9417401552200317,
+      "learning_rate": 0.0002,
+      "loss": 0.4408,
+      "step": 2350
+    },
+    {
+      "epoch": 5.075268817204301,
+      "grad_norm": 0.9461463689804077,
+      "learning_rate": 0.0002,
+      "loss": 0.4306,
+      "step": 2360
+    },
+    {
+      "epoch": 5.096774193548387,
+      "grad_norm": 0.8931282162666321,
+      "learning_rate": 0.0002,
+      "loss": 0.4251,
+      "step": 2370
+    },
+    {
+      "epoch": 5.118279569892473,
+      "grad_norm": 1.000909447669983,
+      "learning_rate": 0.0002,
+      "loss": 0.4249,
+      "step": 2380
+    },
+    {
+      "epoch": 5.139784946236559,
+      "grad_norm": 0.8640249967575073,
+      "learning_rate": 0.0002,
+      "loss": 0.4231,
+      "step": 2390
+    },
+    {
+      "epoch": 5.161290322580645,
+      "grad_norm": 1.0451020002365112,
+      "learning_rate": 0.0002,
+      "loss": 0.4272,
+      "step": 2400
+    },
+    {
+      "epoch": 5.182795698924731,
+      "grad_norm": 0.7896912097930908,
+      "learning_rate": 0.0002,
+      "loss": 0.4177,
+      "step": 2410
+    },
+    {
+      "epoch": 5.204301075268817,
+      "grad_norm": 0.8424463272094727,
+      "learning_rate": 0.0002,
+      "loss": 0.4116,
+      "step": 2420
+    },
+    {
+      "epoch": 5.225806451612903,
+      "grad_norm": 1.0852105617523193,
+      "learning_rate": 0.0002,
+      "loss": 0.4225,
+      "step": 2430
+    },
+    {
+      "epoch": 5.247311827956989,
+      "grad_norm": 0.9285983443260193,
+      "learning_rate": 0.0002,
+      "loss": 0.4352,
+      "step": 2440
+    },
+    {
+      "epoch": 5.268817204301075,
+      "grad_norm": 0.9119299054145813,
+      "learning_rate": 0.0002,
+      "loss": 0.4262,
+      "step": 2450
+    },
+    {
+      "epoch": 5.290322580645161,
+      "grad_norm": 0.8790456056594849,
+      "learning_rate": 0.0002,
+      "loss": 0.4494,
+      "step": 2460
+    },
+    {
+      "epoch": 5.311827956989247,
+      "grad_norm": 0.8726504445075989,
+      "learning_rate": 0.0002,
+      "loss": 0.4421,
+      "step": 2470
+    },
+    {
+      "epoch": 5.333333333333333,
+      "grad_norm": 0.9415227770805359,
+      "learning_rate": 0.0002,
+      "loss": 0.4372,
+      "step": 2480
+    },
+    {
+      "epoch": 5.354838709677419,
+      "grad_norm": 0.9133324027061462,
+      "learning_rate": 0.0002,
+      "loss": 0.4223,
+      "step": 2490
+    },
+    {
+      "epoch": 5.376344086021505,
+      "grad_norm": 0.9567879438400269,
+      "learning_rate": 0.0002,
+      "loss": 0.4401,
+      "step": 2500
+    },
+    {
+      "epoch": 5.397849462365591,
+      "grad_norm": 0.9239469766616821,
+      "learning_rate": 0.0002,
+      "loss": 0.4094,
+      "step": 2510
+    },
+    {
+      "epoch": 5.419354838709677,
+      "grad_norm": 1.0293527841567993,
+      "learning_rate": 0.0002,
+      "loss": 0.4416,
+      "step": 2520
+    },
+    {
+      "epoch": 5.440860215053763,
+      "grad_norm": 0.8618718981742859,
+      "learning_rate": 0.0002,
+      "loss": 0.4311,
+      "step": 2530
+    },
+    {
+      "epoch": 5.462365591397849,
+      "grad_norm": 0.740166187286377,
+      "learning_rate": 0.0002,
+      "loss": 0.462,
+      "step": 2540
+    },
+    {
+      "epoch": 5.483870967741936,
+      "grad_norm": 0.901566743850708,
+      "learning_rate": 0.0002,
+      "loss": 0.4172,
+      "step": 2550
+    },
+    {
+      "epoch": 5.505376344086022,
+      "grad_norm": 0.7957597970962524,
+      "learning_rate": 0.0002,
+      "loss": 0.4315,
+      "step": 2560
+    },
+    {
+      "epoch": 5.526881720430108,
+      "grad_norm": 1.1139343976974487,
+      "learning_rate": 0.0002,
+      "loss": 0.4263,
+      "step": 2570
+    },
+    {
+      "epoch": 5.548387096774194,
+      "grad_norm": 0.989765465259552,
+      "learning_rate": 0.0002,
+      "loss": 0.4056,
+      "step": 2580
+    },
+    {
+      "epoch": 5.56989247311828,
+      "grad_norm": 0.9416969418525696,
+      "learning_rate": 0.0002,
+      "loss": 0.4311,
+      "step": 2590
+    },
+    {
+      "epoch": 5.591397849462366,
+      "grad_norm": 0.9184830784797668,
+      "learning_rate": 0.0002,
+      "loss": 0.4363,
+      "step": 2600
+    },
+    {
+      "epoch": 5.612903225806452,
+      "grad_norm": 1.0512700080871582,
+      "learning_rate": 0.0002,
+      "loss": 0.432,
+      "step": 2610
+    },
+    {
+      "epoch": 5.634408602150538,
+      "grad_norm": 0.901462197303772,
+      "learning_rate": 0.0002,
+      "loss": 0.4227,
+      "step": 2620
+    },
+    {
+      "epoch": 5.655913978494624,
+      "grad_norm": 0.9732566475868225,
+      "learning_rate": 0.0002,
+      "loss": 0.4332,
+      "step": 2630
+    },
+    {
+      "epoch": 5.67741935483871,
+      "grad_norm": 0.8180275559425354,
+      "learning_rate": 0.0002,
+      "loss": 0.4223,
+      "step": 2640
+    },
+    {
+      "epoch": 5.698924731182796,
+      "grad_norm": 1.1354765892028809,
+      "learning_rate": 0.0002,
+      "loss": 0.4311,
+      "step": 2650
+    },
+    {
+      "epoch": 5.720430107526882,
+      "grad_norm": 0.9161503314971924,
+      "learning_rate": 0.0002,
+      "loss": 0.4409,
+      "step": 2660
+    },
+    {
+      "epoch": 5.741935483870968,
+      "grad_norm": 1.0561772584915161,
+      "learning_rate": 0.0002,
+      "loss": 0.4394,
+      "step": 2670
+    },
+    {
+      "epoch": 5.763440860215054,
+      "grad_norm": 0.7712787389755249,
+      "learning_rate": 0.0002,
+      "loss": 0.424,
+      "step": 2680
+    },
+    {
+      "epoch": 5.78494623655914,
+      "grad_norm": 0.9674550294876099,
+      "learning_rate": 0.0002,
+      "loss": 0.4326,
+      "step": 2690
+    },
+    {
+      "epoch": 5.806451612903226,
+      "grad_norm": 0.7531843781471252,
+      "learning_rate": 0.0002,
+      "loss": 0.4459,
+      "step": 2700
+    },
+    {
+      "epoch": 5.827956989247312,
+      "grad_norm": 1.1332131624221802,
+      "learning_rate": 0.0002,
+      "loss": 0.4276,
+      "step": 2710
+    },
+    {
+      "epoch": 5.849462365591398,
+      "grad_norm": 0.9367414116859436,
+      "learning_rate": 0.0002,
+      "loss": 0.4113,
+      "step": 2720
+    },
+    {
+      "epoch": 5.870967741935484,
+      "grad_norm": 0.8267706632614136,
+      "learning_rate": 0.0002,
+      "loss": 0.4227,
+      "step": 2730
+    },
+    {
+      "epoch": 5.89247311827957,
+      "grad_norm": 1.1040657758712769,
+      "learning_rate": 0.0002,
+      "loss": 0.4218,
+      "step": 2740
+    },
+    {
+      "epoch": 5.913978494623656,
+      "grad_norm": 0.8879582285881042,
+      "learning_rate": 0.0002,
+      "loss": 0.4129,
+      "step": 2750
+    },
+    {
+      "epoch": 5.935483870967742,
+      "grad_norm": 0.9264667630195618,
+      "learning_rate": 0.0002,
+      "loss": 0.4241,
+      "step": 2760
+    },
+    {
+      "epoch": 5.956989247311828,
+      "grad_norm": 0.9373905658721924,
+      "learning_rate": 0.0002,
+      "loss": 0.4318,
+      "step": 2770
+    },
+    {
+      "epoch": 5.978494623655914,
+      "grad_norm": 1.0063740015029907,
+      "learning_rate": 0.0002,
+      "loss": 0.423,
+      "step": 2780
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 0.8291367292404175,
+      "learning_rate": 0.0002,
+      "loss": 0.4382,
+      "step": 2790
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 0.5057176947593689,
+      "eval_runtime": 21.3206,
+      "eval_samples_per_second": 15.525,
+      "eval_steps_per_second": 1.97,
+      "step": 2790
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 3720,
+  "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": 3.409936977494016e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/training_args.bin b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..483b53d4fc1c568a5fc890fa850e3450f390b208
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e4c4fe2be590ed03492316230adb3a1edca3e4066c55f3716c0352d7134c564
+size 5560
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/README.md b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..830a14f7db2734beb59f320973504e45a3fe87f5
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/adapter_config.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e99bbcd43df1c19d98706c7e3be95c93844c5349
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/adapter_model.safetensors b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..0b8e87ae7c16b8643438e3c9cbd31e06aa096121
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:54f4a168fbd2c583f9daa69a9737f492d382725904351c5a22fd1b960c408e37
+size 29500848
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/optimizer.pt b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..bf0e0220021fce5faa860af5f9bdfae428031807
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6bbbfcdecc3011c7fc856dbded7d96c1b54e87d6f765b458fa6f3b8e08fc6a1e
+size 15064314
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/rng_state.pth b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..25ffe2b662b93a483d3ef1bf10d6b4d9ce9f8786
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f73bb6248668e8e530e914affcab3aeae1a987ed024a2c75b1b13c7d20e1a3ca
+size 14244
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/scheduler.pt b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..637f69c1cfc0c0dafdad68cbe5ec085272752500
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5dec82ffffb79214bda8d32893fbea6069d6fb1f5366d6289391a63b2e486940
+size 1064
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/special_tokens_map.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/tokenizer.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..f58963a682665634ab180c28667e4faa8cf02ba2
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/tokenizer.model b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/tokenizer_config.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1adb4796c13b8d975555ecec45876ee75d1ae8b7
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/trainer_state.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..5a0027a9dea35dcdba6086c15e3c2a407e286c27
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/trainer_state.json
@@ -0,0 +1,2364 @@
+{
+  "best_metric": 0.48264747858047485,
+  "best_model_checkpoint": "outputs-001/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255",
+  "epoch": 7.0,
+  "eval_steps": 10,
+  "global_step": 3255,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.021505376344086023,
+      "grad_norm": 0.9075053930282593,
+      "learning_rate": 0.0002,
+      "loss": 3.4172,
+      "step": 10
+    },
+    {
+      "epoch": 0.043010752688172046,
+      "grad_norm": 1.4321208000183105,
+      "learning_rate": 0.0002,
+      "loss": 2.5888,
+      "step": 20
+    },
+    {
+      "epoch": 0.06451612903225806,
+      "grad_norm": 1.7500602006912231,
+      "learning_rate": 0.0002,
+      "loss": 2.1195,
+      "step": 30
+    },
+    {
+      "epoch": 0.08602150537634409,
+      "grad_norm": 0.7606641054153442,
+      "learning_rate": 0.0002,
+      "loss": 1.9303,
+      "step": 40
+    },
+    {
+      "epoch": 0.10752688172043011,
+      "grad_norm": 1.2754929065704346,
+      "learning_rate": 0.0002,
+      "loss": 1.6112,
+      "step": 50
+    },
+    {
+      "epoch": 0.12903225806451613,
+      "grad_norm": 1.0936230421066284,
+      "learning_rate": 0.0002,
+      "loss": 1.4319,
+      "step": 60
+    },
+    {
+      "epoch": 0.15053763440860216,
+      "grad_norm": 1.144593596458435,
+      "learning_rate": 0.0002,
+      "loss": 1.3568,
+      "step": 70
+    },
+    {
+      "epoch": 0.17204301075268819,
+      "grad_norm": 1.2181956768035889,
+      "learning_rate": 0.0002,
+      "loss": 1.2028,
+      "step": 80
+    },
+    {
+      "epoch": 0.1935483870967742,
+      "grad_norm": 1.1260095834732056,
+      "learning_rate": 0.0002,
+      "loss": 1.1534,
+      "step": 90
+    },
+    {
+      "epoch": 0.21505376344086022,
+      "grad_norm": 1.1155284643173218,
+      "learning_rate": 0.0002,
+      "loss": 1.1089,
+      "step": 100
+    },
+    {
+      "epoch": 0.23655913978494625,
+      "grad_norm": 1.089565396308899,
+      "learning_rate": 0.0002,
+      "loss": 1.0883,
+      "step": 110
+    },
+    {
+      "epoch": 0.25806451612903225,
+      "grad_norm": 0.9833471775054932,
+      "learning_rate": 0.0002,
+      "loss": 1.0814,
+      "step": 120
+    },
+    {
+      "epoch": 0.27956989247311825,
+      "grad_norm": 1.0265629291534424,
+      "learning_rate": 0.0002,
+      "loss": 1.0239,
+      "step": 130
+    },
+    {
+      "epoch": 0.3010752688172043,
+      "grad_norm": 0.9344286322593689,
+      "learning_rate": 0.0002,
+      "loss": 0.9888,
+      "step": 140
+    },
+    {
+      "epoch": 0.3225806451612903,
+      "grad_norm": 0.9883386492729187,
+      "learning_rate": 0.0002,
+      "loss": 1.0043,
+      "step": 150
+    },
+    {
+      "epoch": 0.34408602150537637,
+      "grad_norm": 0.9299277067184448,
+      "learning_rate": 0.0002,
+      "loss": 0.9338,
+      "step": 160
+    },
+    {
+      "epoch": 0.3655913978494624,
+      "grad_norm": 1.390045404434204,
+      "learning_rate": 0.0002,
+      "loss": 0.9432,
+      "step": 170
+    },
+    {
+      "epoch": 0.3870967741935484,
+      "grad_norm": 1.0313078165054321,
+      "learning_rate": 0.0002,
+      "loss": 0.9008,
+      "step": 180
+    },
+    {
+      "epoch": 0.40860215053763443,
+      "grad_norm": 1.1792205572128296,
+      "learning_rate": 0.0002,
+      "loss": 0.9434,
+      "step": 190
+    },
+    {
+      "epoch": 0.43010752688172044,
+      "grad_norm": 1.049809217453003,
+      "learning_rate": 0.0002,
+      "loss": 0.8761,
+      "step": 200
+    },
+    {
+      "epoch": 0.45161290322580644,
+      "grad_norm": 0.990111768245697,
+      "learning_rate": 0.0002,
+      "loss": 0.8709,
+      "step": 210
+    },
+    {
+      "epoch": 0.4731182795698925,
+      "grad_norm": 0.9870412349700928,
+      "learning_rate": 0.0002,
+      "loss": 0.905,
+      "step": 220
+    },
+    {
+      "epoch": 0.4946236559139785,
+      "grad_norm": 0.8557345867156982,
+      "learning_rate": 0.0002,
+      "loss": 0.9129,
+      "step": 230
+    },
+    {
+      "epoch": 0.5161290322580645,
+      "grad_norm": 0.9746861457824707,
+      "learning_rate": 0.0002,
+      "loss": 0.8836,
+      "step": 240
+    },
+    {
+      "epoch": 0.5376344086021505,
+      "grad_norm": 0.9010438323020935,
+      "learning_rate": 0.0002,
+      "loss": 0.873,
+      "step": 250
+    },
+    {
+      "epoch": 0.5591397849462365,
+      "grad_norm": 0.9061082005500793,
+      "learning_rate": 0.0002,
+      "loss": 0.8241,
+      "step": 260
+    },
+    {
+      "epoch": 0.5806451612903226,
+      "grad_norm": 0.9311846494674683,
+      "learning_rate": 0.0002,
+      "loss": 0.8652,
+      "step": 270
+    },
+    {
+      "epoch": 0.6021505376344086,
+      "grad_norm": 0.9140254855155945,
+      "learning_rate": 0.0002,
+      "loss": 0.8256,
+      "step": 280
+    },
+    {
+      "epoch": 0.6236559139784946,
+      "grad_norm": 0.9722253084182739,
+      "learning_rate": 0.0002,
+      "loss": 0.8441,
+      "step": 290
+    },
+    {
+      "epoch": 0.6451612903225806,
+      "grad_norm": 0.8539168238639832,
+      "learning_rate": 0.0002,
+      "loss": 0.8314,
+      "step": 300
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.9053162932395935,
+      "learning_rate": 0.0002,
+      "loss": 0.8528,
+      "step": 310
+    },
+    {
+      "epoch": 0.6881720430107527,
+      "grad_norm": 0.8444252610206604,
+      "learning_rate": 0.0002,
+      "loss": 0.8209,
+      "step": 320
+    },
+    {
+      "epoch": 0.7096774193548387,
+      "grad_norm": 0.8127437829971313,
+      "learning_rate": 0.0002,
+      "loss": 0.8101,
+      "step": 330
+    },
+    {
+      "epoch": 0.7311827956989247,
+      "grad_norm": 0.886555016040802,
+      "learning_rate": 0.0002,
+      "loss": 0.8223,
+      "step": 340
+    },
+    {
+      "epoch": 0.7526881720430108,
+      "grad_norm": 0.8458548784255981,
+      "learning_rate": 0.0002,
+      "loss": 0.8368,
+      "step": 350
+    },
+    {
+      "epoch": 0.7741935483870968,
+      "grad_norm": 0.8683297634124756,
+      "learning_rate": 0.0002,
+      "loss": 0.8295,
+      "step": 360
+    },
+    {
+      "epoch": 0.7956989247311828,
+      "grad_norm": 0.8308405876159668,
+      "learning_rate": 0.0002,
+      "loss": 0.8232,
+      "step": 370
+    },
+    {
+      "epoch": 0.8172043010752689,
+      "grad_norm": 0.8305579423904419,
+      "learning_rate": 0.0002,
+      "loss": 0.7752,
+      "step": 380
+    },
+    {
+      "epoch": 0.8387096774193549,
+      "grad_norm": 0.8545567393302917,
+      "learning_rate": 0.0002,
+      "loss": 0.8267,
+      "step": 390
+    },
+    {
+      "epoch": 0.8602150537634409,
+      "grad_norm": 0.8486055731773376,
+      "learning_rate": 0.0002,
+      "loss": 0.8212,
+      "step": 400
+    },
+    {
+      "epoch": 0.8817204301075269,
+      "grad_norm": 0.8126763105392456,
+      "learning_rate": 0.0002,
+      "loss": 0.743,
+      "step": 410
+    },
+    {
+      "epoch": 0.9032258064516129,
+      "grad_norm": 0.8494045734405518,
+      "learning_rate": 0.0002,
+      "loss": 0.7993,
+      "step": 420
+    },
+    {
+      "epoch": 0.9247311827956989,
+      "grad_norm": 0.7639183402061462,
+      "learning_rate": 0.0002,
+      "loss": 0.8213,
+      "step": 430
+    },
+    {
+      "epoch": 0.946236559139785,
+      "grad_norm": 0.858101487159729,
+      "learning_rate": 0.0002,
+      "loss": 0.8015,
+      "step": 440
+    },
+    {
+      "epoch": 0.967741935483871,
+      "grad_norm": 0.8141381740570068,
+      "learning_rate": 0.0002,
+      "loss": 0.7629,
+      "step": 450
+    },
+    {
+      "epoch": 0.989247311827957,
+      "grad_norm": 0.8072513937950134,
+      "learning_rate": 0.0002,
+      "loss": 0.7357,
+      "step": 460
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.7740864157676697,
+      "eval_runtime": 21.383,
+      "eval_samples_per_second": 15.48,
+      "eval_steps_per_second": 1.964,
+      "step": 465
+    },
+    {
+      "epoch": 1.010752688172043,
+      "grad_norm": 0.8269494771957397,
+      "learning_rate": 0.0002,
+      "loss": 0.7701,
+      "step": 470
+    },
+    {
+      "epoch": 1.032258064516129,
+      "grad_norm": 0.7814009189605713,
+      "learning_rate": 0.0002,
+      "loss": 0.7532,
+      "step": 480
+    },
+    {
+      "epoch": 1.053763440860215,
+      "grad_norm": 0.8183923363685608,
+      "learning_rate": 0.0002,
+      "loss": 0.7689,
+      "step": 490
+    },
+    {
+      "epoch": 1.075268817204301,
+      "grad_norm": 0.8146600723266602,
+      "learning_rate": 0.0002,
+      "loss": 0.765,
+      "step": 500
+    },
+    {
+      "epoch": 1.096774193548387,
+      "grad_norm": 0.8635126352310181,
+      "learning_rate": 0.0002,
+      "loss": 0.7358,
+      "step": 510
+    },
+    {
+      "epoch": 1.118279569892473,
+      "grad_norm": 0.8520359396934509,
+      "learning_rate": 0.0002,
+      "loss": 0.7302,
+      "step": 520
+    },
+    {
+      "epoch": 1.139784946236559,
+      "grad_norm": 0.8026443123817444,
+      "learning_rate": 0.0002,
+      "loss": 0.7492,
+      "step": 530
+    },
+    {
+      "epoch": 1.1612903225806452,
+      "grad_norm": 0.8157258629798889,
+      "learning_rate": 0.0002,
+      "loss": 0.7518,
+      "step": 540
+    },
+    {
+      "epoch": 1.1827956989247312,
+      "grad_norm": 0.9450796246528625,
+      "learning_rate": 0.0002,
+      "loss": 0.7461,
+      "step": 550
+    },
+    {
+      "epoch": 1.2043010752688172,
+      "grad_norm": 0.8859835863113403,
+      "learning_rate": 0.0002,
+      "loss": 0.7128,
+      "step": 560
+    },
+    {
+      "epoch": 1.2258064516129032,
+      "grad_norm": 0.7819921970367432,
+      "learning_rate": 0.0002,
+      "loss": 0.7067,
+      "step": 570
+    },
+    {
+      "epoch": 1.2473118279569892,
+      "grad_norm": 0.7823445796966553,
+      "learning_rate": 0.0002,
+      "loss": 0.7577,
+      "step": 580
+    },
+    {
+      "epoch": 1.2688172043010753,
+      "grad_norm": 0.7931883931159973,
+      "learning_rate": 0.0002,
+      "loss": 0.7358,
+      "step": 590
+    },
+    {
+      "epoch": 1.2903225806451613,
+      "grad_norm": 0.7495734095573425,
+      "learning_rate": 0.0002,
+      "loss": 0.723,
+      "step": 600
+    },
+    {
+      "epoch": 1.3118279569892473,
+      "grad_norm": 0.9272717237472534,
+      "learning_rate": 0.0002,
+      "loss": 0.7386,
+      "step": 610
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 0.7968398332595825,
+      "learning_rate": 0.0002,
+      "loss": 0.7498,
+      "step": 620
+    },
+    {
+      "epoch": 1.3548387096774195,
+      "grad_norm": 0.7813659310340881,
+      "learning_rate": 0.0002,
+      "loss": 0.7635,
+      "step": 630
+    },
+    {
+      "epoch": 1.3763440860215055,
+      "grad_norm": 0.730925977230072,
+      "learning_rate": 0.0002,
+      "loss": 0.6665,
+      "step": 640
+    },
+    {
+      "epoch": 1.3978494623655915,
+      "grad_norm": 0.8011482954025269,
+      "learning_rate": 0.0002,
+      "loss": 0.7037,
+      "step": 650
+    },
+    {
+      "epoch": 1.4193548387096775,
+      "grad_norm": 0.7770085334777832,
+      "learning_rate": 0.0002,
+      "loss": 0.6931,
+      "step": 660
+    },
+    {
+      "epoch": 1.4408602150537635,
+      "grad_norm": 0.7432682514190674,
+      "learning_rate": 0.0002,
+      "loss": 0.6949,
+      "step": 670
+    },
+    {
+      "epoch": 1.4623655913978495,
+      "grad_norm": 0.8820092678070068,
+      "learning_rate": 0.0002,
+      "loss": 0.7444,
+      "step": 680
+    },
+    {
+      "epoch": 1.4838709677419355,
+      "grad_norm": 0.7786208987236023,
+      "learning_rate": 0.0002,
+      "loss": 0.6758,
+      "step": 690
+    },
+    {
+      "epoch": 1.5053763440860215,
+      "grad_norm": 0.7467480301856995,
+      "learning_rate": 0.0002,
+      "loss": 0.6702,
+      "step": 700
+    },
+    {
+      "epoch": 1.5268817204301075,
+      "grad_norm": 0.8147122263908386,
+      "learning_rate": 0.0002,
+      "loss": 0.7107,
+      "step": 710
+    },
+    {
+      "epoch": 1.5483870967741935,
+      "grad_norm": 0.796030580997467,
+      "learning_rate": 0.0002,
+      "loss": 0.7144,
+      "step": 720
+    },
+    {
+      "epoch": 1.5698924731182795,
+      "grad_norm": 0.8776171207427979,
+      "learning_rate": 0.0002,
+      "loss": 0.6936,
+      "step": 730
+    },
+    {
+      "epoch": 1.5913978494623655,
+      "grad_norm": 0.8056126236915588,
+      "learning_rate": 0.0002,
+      "loss": 0.7101,
+      "step": 740
+    },
+    {
+      "epoch": 1.6129032258064515,
+      "grad_norm": 0.8141863346099854,
+      "learning_rate": 0.0002,
+      "loss": 0.7162,
+      "step": 750
+    },
+    {
+      "epoch": 1.6344086021505375,
+      "grad_norm": 0.8100557327270508,
+      "learning_rate": 0.0002,
+      "loss": 0.7088,
+      "step": 760
+    },
+    {
+      "epoch": 1.6559139784946235,
+      "grad_norm": 0.8283200860023499,
+      "learning_rate": 0.0002,
+      "loss": 0.7212,
+      "step": 770
+    },
+    {
+      "epoch": 1.6774193548387095,
+      "grad_norm": 0.800865113735199,
+      "learning_rate": 0.0002,
+      "loss": 0.694,
+      "step": 780
+    },
+    {
+      "epoch": 1.6989247311827957,
+      "grad_norm": 0.8052287697792053,
+      "learning_rate": 0.0002,
+      "loss": 0.7076,
+      "step": 790
+    },
+    {
+      "epoch": 1.7204301075268817,
+      "grad_norm": 0.8619674444198608,
+      "learning_rate": 0.0002,
+      "loss": 0.7257,
+      "step": 800
+    },
+    {
+      "epoch": 1.7419354838709677,
+      "grad_norm": 0.8907215595245361,
+      "learning_rate": 0.0002,
+      "loss": 0.7141,
+      "step": 810
+    },
+    {
+      "epoch": 1.7634408602150538,
+      "grad_norm": 0.6976316571235657,
+      "learning_rate": 0.0002,
+      "loss": 0.7035,
+      "step": 820
+    },
+    {
+      "epoch": 1.7849462365591398,
+      "grad_norm": 0.7533746957778931,
+      "learning_rate": 0.0002,
+      "loss": 0.6916,
+      "step": 830
+    },
+    {
+      "epoch": 1.8064516129032258,
+      "grad_norm": 0.7326804399490356,
+      "learning_rate": 0.0002,
+      "loss": 0.7094,
+      "step": 840
+    },
+    {
+      "epoch": 1.827956989247312,
+      "grad_norm": 0.7782683372497559,
+      "learning_rate": 0.0002,
+      "loss": 0.6891,
+      "step": 850
+    },
+    {
+      "epoch": 1.849462365591398,
+      "grad_norm": 0.7424806356430054,
+      "learning_rate": 0.0002,
+      "loss": 0.6931,
+      "step": 860
+    },
+    {
+      "epoch": 1.870967741935484,
+      "grad_norm": 1.172325611114502,
+      "learning_rate": 0.0002,
+      "loss": 0.7354,
+      "step": 870
+    },
+    {
+      "epoch": 1.89247311827957,
+      "grad_norm": 0.771058201789856,
+      "learning_rate": 0.0002,
+      "loss": 0.6866,
+      "step": 880
+    },
+    {
+      "epoch": 1.913978494623656,
+      "grad_norm": 0.8624904155731201,
+      "learning_rate": 0.0002,
+      "loss": 0.7296,
+      "step": 890
+    },
+    {
+      "epoch": 1.935483870967742,
+      "grad_norm": 0.7062820792198181,
+      "learning_rate": 0.0002,
+      "loss": 0.7233,
+      "step": 900
+    },
+    {
+      "epoch": 1.956989247311828,
+      "grad_norm": 0.7560103535652161,
+      "learning_rate": 0.0002,
+      "loss": 0.6966,
+      "step": 910
+    },
+    {
+      "epoch": 1.978494623655914,
+      "grad_norm": 0.788899838924408,
+      "learning_rate": 0.0002,
+      "loss": 0.69,
+      "step": 920
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.6562113761901855,
+      "learning_rate": 0.0002,
+      "loss": 0.6505,
+      "step": 930
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.6885261535644531,
+      "eval_runtime": 21.4291,
+      "eval_samples_per_second": 15.446,
+      "eval_steps_per_second": 1.96,
+      "step": 930
+    },
+    {
+      "epoch": 2.021505376344086,
+      "grad_norm": 0.8216531872749329,
+      "learning_rate": 0.0002,
+      "loss": 0.6625,
+      "step": 940
+    },
+    {
+      "epoch": 2.043010752688172,
+      "grad_norm": 0.8317142724990845,
+      "learning_rate": 0.0002,
+      "loss": 0.6398,
+      "step": 950
+    },
+    {
+      "epoch": 2.064516129032258,
+      "grad_norm": 0.8446708917617798,
+      "learning_rate": 0.0002,
+      "loss": 0.649,
+      "step": 960
+    },
+    {
+      "epoch": 2.086021505376344,
+      "grad_norm": 0.735055148601532,
+      "learning_rate": 0.0002,
+      "loss": 0.657,
+      "step": 970
+    },
+    {
+      "epoch": 2.10752688172043,
+      "grad_norm": 0.7487243413925171,
+      "learning_rate": 0.0002,
+      "loss": 0.649,
+      "step": 980
+    },
+    {
+      "epoch": 2.129032258064516,
+      "grad_norm": 0.8573887944221497,
+      "learning_rate": 0.0002,
+      "loss": 0.6419,
+      "step": 990
+    },
+    {
+      "epoch": 2.150537634408602,
+      "grad_norm": 0.6284521818161011,
+      "learning_rate": 0.0002,
+      "loss": 0.6431,
+      "step": 1000
+    },
+    {
+      "epoch": 2.172043010752688,
+      "grad_norm": 0.754183292388916,
+      "learning_rate": 0.0002,
+      "loss": 0.6128,
+      "step": 1010
+    },
+    {
+      "epoch": 2.193548387096774,
+      "grad_norm": 0.9445359110832214,
+      "learning_rate": 0.0002,
+      "loss": 0.6253,
+      "step": 1020
+    },
+    {
+      "epoch": 2.21505376344086,
+      "grad_norm": 0.808508038520813,
+      "learning_rate": 0.0002,
+      "loss": 0.605,
+      "step": 1030
+    },
+    {
+      "epoch": 2.236559139784946,
+      "grad_norm": 0.9394679665565491,
+      "learning_rate": 0.0002,
+      "loss": 0.6786,
+      "step": 1040
+    },
+    {
+      "epoch": 2.258064516129032,
+      "grad_norm": 0.8151357769966125,
+      "learning_rate": 0.0002,
+      "loss": 0.6176,
+      "step": 1050
+    },
+    {
+      "epoch": 2.279569892473118,
+      "grad_norm": 0.7909848093986511,
+      "learning_rate": 0.0002,
+      "loss": 0.66,
+      "step": 1060
+    },
+    {
+      "epoch": 2.3010752688172045,
+      "grad_norm": 0.7506507039070129,
+      "learning_rate": 0.0002,
+      "loss": 0.6254,
+      "step": 1070
+    },
+    {
+      "epoch": 2.3225806451612905,
+      "grad_norm": 0.8240520358085632,
+      "learning_rate": 0.0002,
+      "loss": 0.6608,
+      "step": 1080
+    },
+    {
+      "epoch": 2.3440860215053765,
+      "grad_norm": 0.9342400431632996,
+      "learning_rate": 0.0002,
+      "loss": 0.6207,
+      "step": 1090
+    },
+    {
+      "epoch": 2.3655913978494625,
+      "grad_norm": 1.0598735809326172,
+      "learning_rate": 0.0002,
+      "loss": 0.6029,
+      "step": 1100
+    },
+    {
+      "epoch": 2.3870967741935485,
+      "grad_norm": 0.7907650470733643,
+      "learning_rate": 0.0002,
+      "loss": 0.6035,
+      "step": 1110
+    },
+    {
+      "epoch": 2.4086021505376345,
+      "grad_norm": 0.9388798475265503,
+      "learning_rate": 0.0002,
+      "loss": 0.6237,
+      "step": 1120
+    },
+    {
+      "epoch": 2.4301075268817205,
+      "grad_norm": 0.8985419869422913,
+      "learning_rate": 0.0002,
+      "loss": 0.6207,
+      "step": 1130
+    },
+    {
+      "epoch": 2.4516129032258065,
+      "grad_norm": 0.7471932768821716,
+      "learning_rate": 0.0002,
+      "loss": 0.5902,
+      "step": 1140
+    },
+    {
+      "epoch": 2.4731182795698925,
+      "grad_norm": 0.761131763458252,
+      "learning_rate": 0.0002,
+      "loss": 0.6446,
+      "step": 1150
+    },
+    {
+      "epoch": 2.4946236559139785,
+      "grad_norm": 0.7901819348335266,
+      "learning_rate": 0.0002,
+      "loss": 0.6088,
+      "step": 1160
+    },
+    {
+      "epoch": 2.5161290322580645,
+      "grad_norm": 0.9932922720909119,
+      "learning_rate": 0.0002,
+      "loss": 0.6142,
+      "step": 1170
+    },
+    {
+      "epoch": 2.5376344086021505,
+      "grad_norm": 0.7414287328720093,
+      "learning_rate": 0.0002,
+      "loss": 0.6407,
+      "step": 1180
+    },
+    {
+      "epoch": 2.5591397849462365,
+      "grad_norm": 0.8111771941184998,
+      "learning_rate": 0.0002,
+      "loss": 0.6161,
+      "step": 1190
+    },
+    {
+      "epoch": 2.5806451612903225,
+      "grad_norm": 0.7520156502723694,
+      "learning_rate": 0.0002,
+      "loss": 0.6006,
+      "step": 1200
+    },
+    {
+      "epoch": 2.6021505376344085,
+      "grad_norm": 0.9022907018661499,
+      "learning_rate": 0.0002,
+      "loss": 0.615,
+      "step": 1210
+    },
+    {
+      "epoch": 2.6236559139784945,
+      "grad_norm": 0.7746260166168213,
+      "learning_rate": 0.0002,
+      "loss": 0.6211,
+      "step": 1220
+    },
+    {
+      "epoch": 2.6451612903225805,
+      "grad_norm": 0.8482862114906311,
+      "learning_rate": 0.0002,
+      "loss": 0.616,
+      "step": 1230
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 0.7925458550453186,
+      "learning_rate": 0.0002,
+      "loss": 0.6417,
+      "step": 1240
+    },
+    {
+      "epoch": 2.688172043010753,
+      "grad_norm": 0.8369929194450378,
+      "learning_rate": 0.0002,
+      "loss": 0.6187,
+      "step": 1250
+    },
+    {
+      "epoch": 2.709677419354839,
+      "grad_norm": 0.8311542868614197,
+      "learning_rate": 0.0002,
+      "loss": 0.6138,
+      "step": 1260
+    },
+    {
+      "epoch": 2.731182795698925,
+      "grad_norm": 0.7204853296279907,
+      "learning_rate": 0.0002,
+      "loss": 0.5894,
+      "step": 1270
+    },
+    {
+      "epoch": 2.752688172043011,
+      "grad_norm": 0.8447284698486328,
+      "learning_rate": 0.0002,
+      "loss": 0.6325,
+      "step": 1280
+    },
+    {
+      "epoch": 2.774193548387097,
+      "grad_norm": 0.7738404273986816,
+      "learning_rate": 0.0002,
+      "loss": 0.5946,
+      "step": 1290
+    },
+    {
+      "epoch": 2.795698924731183,
+      "grad_norm": 0.8393287062644958,
+      "learning_rate": 0.0002,
+      "loss": 0.5678,
+      "step": 1300
+    },
+    {
+      "epoch": 2.817204301075269,
+      "grad_norm": 0.79121994972229,
+      "learning_rate": 0.0002,
+      "loss": 0.6092,
+      "step": 1310
+    },
+    {
+      "epoch": 2.838709677419355,
+      "grad_norm": 0.7331557869911194,
+      "learning_rate": 0.0002,
+      "loss": 0.5889,
+      "step": 1320
+    },
+    {
+      "epoch": 2.860215053763441,
+      "grad_norm": 0.9593998193740845,
+      "learning_rate": 0.0002,
+      "loss": 0.6048,
+      "step": 1330
+    },
+    {
+      "epoch": 2.881720430107527,
+      "grad_norm": 0.7215158343315125,
+      "learning_rate": 0.0002,
+      "loss": 0.6108,
+      "step": 1340
+    },
+    {
+      "epoch": 2.903225806451613,
+      "grad_norm": 0.840404212474823,
+      "learning_rate": 0.0002,
+      "loss": 0.5897,
+      "step": 1350
+    },
+    {
+      "epoch": 2.924731182795699,
+      "grad_norm": 0.870659351348877,
+      "learning_rate": 0.0002,
+      "loss": 0.6056,
+      "step": 1360
+    },
+    {
+      "epoch": 2.946236559139785,
+      "grad_norm": 0.8744975328445435,
+      "learning_rate": 0.0002,
+      "loss": 0.6205,
+      "step": 1370
+    },
+    {
+      "epoch": 2.967741935483871,
+      "grad_norm": 0.8030612468719482,
+      "learning_rate": 0.0002,
+      "loss": 0.5966,
+      "step": 1380
+    },
+    {
+      "epoch": 2.989247311827957,
+      "grad_norm": 0.825814962387085,
+      "learning_rate": 0.0002,
+      "loss": 0.6004,
+      "step": 1390
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 0.6257933378219604,
+      "eval_runtime": 21.3692,
+      "eval_samples_per_second": 15.49,
+      "eval_steps_per_second": 1.965,
+      "step": 1395
+    },
+    {
+      "epoch": 3.010752688172043,
+      "grad_norm": 0.8650677800178528,
+      "learning_rate": 0.0002,
+      "loss": 0.5696,
+      "step": 1400
+    },
+    {
+      "epoch": 3.032258064516129,
+      "grad_norm": 0.8364197015762329,
+      "learning_rate": 0.0002,
+      "loss": 0.5483,
+      "step": 1410
+    },
+    {
+      "epoch": 3.053763440860215,
+      "grad_norm": 0.8278448581695557,
+      "learning_rate": 0.0002,
+      "loss": 0.5606,
+      "step": 1420
+    },
+    {
+      "epoch": 3.075268817204301,
+      "grad_norm": 0.8806642889976501,
+      "learning_rate": 0.0002,
+      "loss": 0.5572,
+      "step": 1430
+    },
+    {
+      "epoch": 3.096774193548387,
+      "grad_norm": 0.8180029988288879,
+      "learning_rate": 0.0002,
+      "loss": 0.585,
+      "step": 1440
+    },
+    {
+      "epoch": 3.118279569892473,
+      "grad_norm": 0.8561782836914062,
+      "learning_rate": 0.0002,
+      "loss": 0.5667,
+      "step": 1450
+    },
+    {
+      "epoch": 3.139784946236559,
+      "grad_norm": 0.8377029299736023,
+      "learning_rate": 0.0002,
+      "loss": 0.5246,
+      "step": 1460
+    },
+    {
+      "epoch": 3.161290322580645,
+      "grad_norm": 0.885779082775116,
+      "learning_rate": 0.0002,
+      "loss": 0.5464,
+      "step": 1470
+    },
+    {
+      "epoch": 3.182795698924731,
+      "grad_norm": 0.9388518333435059,
+      "learning_rate": 0.0002,
+      "loss": 0.541,
+      "step": 1480
+    },
+    {
+      "epoch": 3.204301075268817,
+      "grad_norm": 0.8816235661506653,
+      "learning_rate": 0.0002,
+      "loss": 0.5447,
+      "step": 1490
+    },
+    {
+      "epoch": 3.225806451612903,
+      "grad_norm": 0.9885783791542053,
+      "learning_rate": 0.0002,
+      "loss": 0.5466,
+      "step": 1500
+    },
+    {
+      "epoch": 3.247311827956989,
+      "grad_norm": 0.8635850548744202,
+      "learning_rate": 0.0002,
+      "loss": 0.5455,
+      "step": 1510
+    },
+    {
+      "epoch": 3.268817204301075,
+      "grad_norm": 0.829853355884552,
+      "learning_rate": 0.0002,
+      "loss": 0.5419,
+      "step": 1520
+    },
+    {
+      "epoch": 3.2903225806451615,
+      "grad_norm": 0.9037486910820007,
+      "learning_rate": 0.0002,
+      "loss": 0.54,
+      "step": 1530
+    },
+    {
+      "epoch": 3.3118279569892475,
+      "grad_norm": 0.8173713684082031,
+      "learning_rate": 0.0002,
+      "loss": 0.5375,
+      "step": 1540
+    },
+    {
+      "epoch": 3.3333333333333335,
+      "grad_norm": 0.796953022480011,
+      "learning_rate": 0.0002,
+      "loss": 0.5405,
+      "step": 1550
+    },
+    {
+      "epoch": 3.3548387096774195,
+      "grad_norm": 0.7894400358200073,
+      "learning_rate": 0.0002,
+      "loss": 0.5505,
+      "step": 1560
+    },
+    {
+      "epoch": 3.3763440860215055,
+      "grad_norm": 0.9434949159622192,
+      "learning_rate": 0.0002,
+      "loss": 0.5395,
+      "step": 1570
+    },
+    {
+      "epoch": 3.3978494623655915,
+      "grad_norm": 0.8666760325431824,
+      "learning_rate": 0.0002,
+      "loss": 0.5271,
+      "step": 1580
+    },
+    {
+      "epoch": 3.4193548387096775,
+      "grad_norm": 0.7782467007637024,
+      "learning_rate": 0.0002,
+      "loss": 0.5439,
+      "step": 1590
+    },
+    {
+      "epoch": 3.4408602150537635,
+      "grad_norm": 0.8849126696586609,
+      "learning_rate": 0.0002,
+      "loss": 0.5161,
+      "step": 1600
+    },
+    {
+      "epoch": 3.4623655913978495,
+      "grad_norm": 0.7863831520080566,
+      "learning_rate": 0.0002,
+      "loss": 0.5353,
+      "step": 1610
+    },
+    {
+      "epoch": 3.4838709677419355,
+      "grad_norm": 1.0403116941452026,
+      "learning_rate": 0.0002,
+      "loss": 0.5308,
+      "step": 1620
+    },
+    {
+      "epoch": 3.5053763440860215,
+      "grad_norm": 0.8307499289512634,
+      "learning_rate": 0.0002,
+      "loss": 0.5339,
+      "step": 1630
+    },
+    {
+      "epoch": 3.5268817204301075,
+      "grad_norm": 0.9132118821144104,
+      "learning_rate": 0.0002,
+      "loss": 0.5361,
+      "step": 1640
+    },
+    {
+      "epoch": 3.5483870967741935,
+      "grad_norm": 0.9322578310966492,
+      "learning_rate": 0.0002,
+      "loss": 0.5828,
+      "step": 1650
+    },
+    {
+      "epoch": 3.5698924731182795,
+      "grad_norm": 0.9782460331916809,
+      "learning_rate": 0.0002,
+      "loss": 0.546,
+      "step": 1660
+    },
+    {
+      "epoch": 3.5913978494623655,
+      "grad_norm": 0.7189919352531433,
+      "learning_rate": 0.0002,
+      "loss": 0.5424,
+      "step": 1670
+    },
+    {
+      "epoch": 3.6129032258064515,
+      "grad_norm": 0.9689221382141113,
+      "learning_rate": 0.0002,
+      "loss": 0.5514,
+      "step": 1680
+    },
+    {
+      "epoch": 3.6344086021505375,
+      "grad_norm": 0.9684675335884094,
+      "learning_rate": 0.0002,
+      "loss": 0.5379,
+      "step": 1690
+    },
+    {
+      "epoch": 3.6559139784946235,
+      "grad_norm": 0.8851472735404968,
+      "learning_rate": 0.0002,
+      "loss": 0.5748,
+      "step": 1700
+    },
+    {
+      "epoch": 3.6774193548387095,
+      "grad_norm": 0.7709833383560181,
+      "learning_rate": 0.0002,
+      "loss": 0.5412,
+      "step": 1710
+    },
+    {
+      "epoch": 3.698924731182796,
+      "grad_norm": 0.818236231803894,
+      "learning_rate": 0.0002,
+      "loss": 0.521,
+      "step": 1720
+    },
+    {
+      "epoch": 3.720430107526882,
+      "grad_norm": 0.870642364025116,
+      "learning_rate": 0.0002,
+      "loss": 0.5445,
+      "step": 1730
+    },
+    {
+      "epoch": 3.741935483870968,
+      "grad_norm": 1.0245511531829834,
+      "learning_rate": 0.0002,
+      "loss": 0.5307,
+      "step": 1740
+    },
+    {
+      "epoch": 3.763440860215054,
+      "grad_norm": 0.8607558608055115,
+      "learning_rate": 0.0002,
+      "loss": 0.5593,
+      "step": 1750
+    },
+    {
+      "epoch": 3.78494623655914,
+      "grad_norm": 0.8511829972267151,
+      "learning_rate": 0.0002,
+      "loss": 0.536,
+      "step": 1760
+    },
+    {
+      "epoch": 3.806451612903226,
+      "grad_norm": 0.7969087362289429,
+      "learning_rate": 0.0002,
+      "loss": 0.5193,
+      "step": 1770
+    },
+    {
+      "epoch": 3.827956989247312,
+      "grad_norm": 0.8457245826721191,
+      "learning_rate": 0.0002,
+      "loss": 0.5578,
+      "step": 1780
+    },
+    {
+      "epoch": 3.849462365591398,
+      "grad_norm": 0.8893467783927917,
+      "learning_rate": 0.0002,
+      "loss": 0.5337,
+      "step": 1790
+    },
+    {
+      "epoch": 3.870967741935484,
+      "grad_norm": 0.8593819737434387,
+      "learning_rate": 0.0002,
+      "loss": 0.5024,
+      "step": 1800
+    },
+    {
+      "epoch": 3.89247311827957,
+      "grad_norm": 0.7574560642242432,
+      "learning_rate": 0.0002,
+      "loss": 0.5134,
+      "step": 1810
+    },
+    {
+      "epoch": 3.913978494623656,
+      "grad_norm": 0.8681567311286926,
+      "learning_rate": 0.0002,
+      "loss": 0.5263,
+      "step": 1820
+    },
+    {
+      "epoch": 3.935483870967742,
+      "grad_norm": 0.9068132042884827,
+      "learning_rate": 0.0002,
+      "loss": 0.532,
+      "step": 1830
+    },
+    {
+      "epoch": 3.956989247311828,
+      "grad_norm": 0.8668948411941528,
+      "learning_rate": 0.0002,
+      "loss": 0.5427,
+      "step": 1840
+    },
+    {
+      "epoch": 3.978494623655914,
+      "grad_norm": 1.046032428741455,
+      "learning_rate": 0.0002,
+      "loss": 0.5349,
+      "step": 1850
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.904780387878418,
+      "learning_rate": 0.0002,
+      "loss": 0.5087,
+      "step": 1860
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 0.5737715363502502,
+      "eval_runtime": 21.4915,
+      "eval_samples_per_second": 15.401,
+      "eval_steps_per_second": 1.954,
+      "step": 1860
+    },
+    {
+      "epoch": 4.021505376344086,
+      "grad_norm": 0.8611752986907959,
+      "learning_rate": 0.0002,
+      "loss": 0.4843,
+      "step": 1870
+    },
+    {
+      "epoch": 4.043010752688172,
+      "grad_norm": 0.838782548904419,
+      "learning_rate": 0.0002,
+      "loss": 0.4814,
+      "step": 1880
+    },
+    {
+      "epoch": 4.064516129032258,
+      "grad_norm": 0.9119709134101868,
+      "learning_rate": 0.0002,
+      "loss": 0.474,
+      "step": 1890
+    },
+    {
+      "epoch": 4.086021505376344,
+      "grad_norm": 0.8026251196861267,
+      "learning_rate": 0.0002,
+      "loss": 0.4951,
+      "step": 1900
+    },
+    {
+      "epoch": 4.10752688172043,
+      "grad_norm": 0.8773705363273621,
+      "learning_rate": 0.0002,
+      "loss": 0.491,
+      "step": 1910
+    },
+    {
+      "epoch": 4.129032258064516,
+      "grad_norm": 0.8762255907058716,
+      "learning_rate": 0.0002,
+      "loss": 0.474,
+      "step": 1920
+    },
+    {
+      "epoch": 4.150537634408602,
+      "grad_norm": 0.8371861577033997,
+      "learning_rate": 0.0002,
+      "loss": 0.4816,
+      "step": 1930
+    },
+    {
+      "epoch": 4.172043010752688,
+      "grad_norm": 0.9703728556632996,
+      "learning_rate": 0.0002,
+      "loss": 0.472,
+      "step": 1940
+    },
+    {
+      "epoch": 4.193548387096774,
+      "grad_norm": 0.8802874684333801,
+      "learning_rate": 0.0002,
+      "loss": 0.4772,
+      "step": 1950
+    },
+    {
+      "epoch": 4.21505376344086,
+      "grad_norm": 1.0103057622909546,
+      "learning_rate": 0.0002,
+      "loss": 0.5032,
+      "step": 1960
+    },
+    {
+      "epoch": 4.236559139784946,
+      "grad_norm": 0.9212995171546936,
+      "learning_rate": 0.0002,
+      "loss": 0.4945,
+      "step": 1970
+    },
+    {
+      "epoch": 4.258064516129032,
+      "grad_norm": 1.009544849395752,
+      "learning_rate": 0.0002,
+      "loss": 0.4753,
+      "step": 1980
+    },
+    {
+      "epoch": 4.279569892473118,
+      "grad_norm": 0.8535077571868896,
+      "learning_rate": 0.0002,
+      "loss": 0.4789,
+      "step": 1990
+    },
+    {
+      "epoch": 4.301075268817204,
+      "grad_norm": 0.8363022804260254,
+      "learning_rate": 0.0002,
+      "loss": 0.4782,
+      "step": 2000
+    },
+    {
+      "epoch": 4.32258064516129,
+      "grad_norm": 0.9041762948036194,
+      "learning_rate": 0.0002,
+      "loss": 0.4875,
+      "step": 2010
+    },
+    {
+      "epoch": 4.344086021505376,
+      "grad_norm": 0.960790753364563,
+      "learning_rate": 0.0002,
+      "loss": 0.4779,
+      "step": 2020
+    },
+    {
+      "epoch": 4.365591397849462,
+      "grad_norm": 0.8823095560073853,
+      "learning_rate": 0.0002,
+      "loss": 0.4626,
+      "step": 2030
+    },
+    {
+      "epoch": 4.387096774193548,
+      "grad_norm": 0.952100396156311,
+      "learning_rate": 0.0002,
+      "loss": 0.4883,
+      "step": 2040
+    },
+    {
+      "epoch": 4.408602150537634,
+      "grad_norm": 1.0793498754501343,
+      "learning_rate": 0.0002,
+      "loss": 0.4789,
+      "step": 2050
+    },
+    {
+      "epoch": 4.43010752688172,
+      "grad_norm": 0.8987208008766174,
+      "learning_rate": 0.0002,
+      "loss": 0.4827,
+      "step": 2060
+    },
+    {
+      "epoch": 4.451612903225806,
+      "grad_norm": 0.8539772033691406,
+      "learning_rate": 0.0002,
+      "loss": 0.4594,
+      "step": 2070
+    },
+    {
+      "epoch": 4.473118279569892,
+      "grad_norm": 0.9160863757133484,
+      "learning_rate": 0.0002,
+      "loss": 0.4752,
+      "step": 2080
+    },
+    {
+      "epoch": 4.494623655913978,
+      "grad_norm": 0.9946850538253784,
+      "learning_rate": 0.0002,
+      "loss": 0.5033,
+      "step": 2090
+    },
+    {
+      "epoch": 4.516129032258064,
+      "grad_norm": 0.908039391040802,
+      "learning_rate": 0.0002,
+      "loss": 0.4842,
+      "step": 2100
+    },
+    {
+      "epoch": 4.53763440860215,
+      "grad_norm": 1.1462254524230957,
+      "learning_rate": 0.0002,
+      "loss": 0.4861,
+      "step": 2110
+    },
+    {
+      "epoch": 4.559139784946236,
+      "grad_norm": 0.8392056226730347,
+      "learning_rate": 0.0002,
+      "loss": 0.4892,
+      "step": 2120
+    },
+    {
+      "epoch": 4.580645161290323,
+      "grad_norm": 0.9673896431922913,
+      "learning_rate": 0.0002,
+      "loss": 0.4824,
+      "step": 2130
+    },
+    {
+      "epoch": 4.602150537634409,
+      "grad_norm": 0.9047091603279114,
+      "learning_rate": 0.0002,
+      "loss": 0.4665,
+      "step": 2140
+    },
+    {
+      "epoch": 4.623655913978495,
+      "grad_norm": 0.9013425707817078,
+      "learning_rate": 0.0002,
+      "loss": 0.4714,
+      "step": 2150
+    },
+    {
+      "epoch": 4.645161290322581,
+      "grad_norm": 0.8899165391921997,
+      "learning_rate": 0.0002,
+      "loss": 0.472,
+      "step": 2160
+    },
+    {
+      "epoch": 4.666666666666667,
+      "grad_norm": 0.748602569103241,
+      "learning_rate": 0.0002,
+      "loss": 0.4635,
+      "step": 2170
+    },
+    {
+      "epoch": 4.688172043010753,
+      "grad_norm": 0.8694155216217041,
+      "learning_rate": 0.0002,
+      "loss": 0.4695,
+      "step": 2180
+    },
+    {
+      "epoch": 4.709677419354839,
+      "grad_norm": 0.9134316444396973,
+      "learning_rate": 0.0002,
+      "loss": 0.4929,
+      "step": 2190
+    },
+    {
+      "epoch": 4.731182795698925,
+      "grad_norm": 0.8504763245582581,
+      "learning_rate": 0.0002,
+      "loss": 0.4855,
+      "step": 2200
+    },
+    {
+      "epoch": 4.752688172043011,
+      "grad_norm": 1.0321544408798218,
+      "learning_rate": 0.0002,
+      "loss": 0.4517,
+      "step": 2210
+    },
+    {
+      "epoch": 4.774193548387097,
+      "grad_norm": 0.9368237257003784,
+      "learning_rate": 0.0002,
+      "loss": 0.4796,
+      "step": 2220
+    },
+    {
+      "epoch": 4.795698924731183,
+      "grad_norm": 0.9319947361946106,
+      "learning_rate": 0.0002,
+      "loss": 0.4837,
+      "step": 2230
+    },
+    {
+      "epoch": 4.817204301075269,
+      "grad_norm": 0.904333770275116,
+      "learning_rate": 0.0002,
+      "loss": 0.4696,
+      "step": 2240
+    },
+    {
+      "epoch": 4.838709677419355,
+      "grad_norm": 0.8097078204154968,
+      "learning_rate": 0.0002,
+      "loss": 0.4746,
+      "step": 2250
+    },
+    {
+      "epoch": 4.860215053763441,
+      "grad_norm": 0.9128859043121338,
+      "learning_rate": 0.0002,
+      "loss": 0.4438,
+      "step": 2260
+    },
+    {
+      "epoch": 4.881720430107527,
+      "grad_norm": 0.883129894733429,
+      "learning_rate": 0.0002,
+      "loss": 0.4693,
+      "step": 2270
+    },
+    {
+      "epoch": 4.903225806451613,
+      "grad_norm": 0.85712730884552,
+      "learning_rate": 0.0002,
+      "loss": 0.4494,
+      "step": 2280
+    },
+    {
+      "epoch": 4.924731182795699,
+      "grad_norm": 1.2101863622665405,
+      "learning_rate": 0.0002,
+      "loss": 0.4593,
+      "step": 2290
+    },
+    {
+      "epoch": 4.946236559139785,
+      "grad_norm": 0.917966902256012,
+      "learning_rate": 0.0002,
+      "loss": 0.4779,
+      "step": 2300
+    },
+    {
+      "epoch": 4.967741935483871,
+      "grad_norm": 0.7740724086761475,
+      "learning_rate": 0.0002,
+      "loss": 0.4666,
+      "step": 2310
+    },
+    {
+      "epoch": 4.989247311827957,
+      "grad_norm": 1.0199906826019287,
+      "learning_rate": 0.0002,
+      "loss": 0.4629,
+      "step": 2320
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 0.5363914370536804,
+      "eval_runtime": 21.3941,
+      "eval_samples_per_second": 15.472,
+      "eval_steps_per_second": 1.963,
+      "step": 2325
+    },
+    {
+      "epoch": 5.010752688172043,
+      "grad_norm": 0.8580502271652222,
+      "learning_rate": 0.0002,
+      "loss": 0.4543,
+      "step": 2330
+    },
+    {
+      "epoch": 5.032258064516129,
+      "grad_norm": 0.7702704668045044,
+      "learning_rate": 0.0002,
+      "loss": 0.404,
+      "step": 2340
+    },
+    {
+      "epoch": 5.053763440860215,
+      "grad_norm": 0.9417401552200317,
+      "learning_rate": 0.0002,
+      "loss": 0.4408,
+      "step": 2350
+    },
+    {
+      "epoch": 5.075268817204301,
+      "grad_norm": 0.9461463689804077,
+      "learning_rate": 0.0002,
+      "loss": 0.4306,
+      "step": 2360
+    },
+    {
+      "epoch": 5.096774193548387,
+      "grad_norm": 0.8931282162666321,
+      "learning_rate": 0.0002,
+      "loss": 0.4251,
+      "step": 2370
+    },
+    {
+      "epoch": 5.118279569892473,
+      "grad_norm": 1.000909447669983,
+      "learning_rate": 0.0002,
+      "loss": 0.4249,
+      "step": 2380
+    },
+    {
+      "epoch": 5.139784946236559,
+      "grad_norm": 0.8640249967575073,
+      "learning_rate": 0.0002,
+      "loss": 0.4231,
+      "step": 2390
+    },
+    {
+      "epoch": 5.161290322580645,
+      "grad_norm": 1.0451020002365112,
+      "learning_rate": 0.0002,
+      "loss": 0.4272,
+      "step": 2400
+    },
+    {
+      "epoch": 5.182795698924731,
+      "grad_norm": 0.7896912097930908,
+      "learning_rate": 0.0002,
+      "loss": 0.4177,
+      "step": 2410
+    },
+    {
+      "epoch": 5.204301075268817,
+      "grad_norm": 0.8424463272094727,
+      "learning_rate": 0.0002,
+      "loss": 0.4116,
+      "step": 2420
+    },
+    {
+      "epoch": 5.225806451612903,
+      "grad_norm": 1.0852105617523193,
+      "learning_rate": 0.0002,
+      "loss": 0.4225,
+      "step": 2430
+    },
+    {
+      "epoch": 5.247311827956989,
+      "grad_norm": 0.9285983443260193,
+      "learning_rate": 0.0002,
+      "loss": 0.4352,
+      "step": 2440
+    },
+    {
+      "epoch": 5.268817204301075,
+      "grad_norm": 0.9119299054145813,
+      "learning_rate": 0.0002,
+      "loss": 0.4262,
+      "step": 2450
+    },
+    {
+      "epoch": 5.290322580645161,
+      "grad_norm": 0.8790456056594849,
+      "learning_rate": 0.0002,
+      "loss": 0.4494,
+      "step": 2460
+    },
+    {
+      "epoch": 5.311827956989247,
+      "grad_norm": 0.8726504445075989,
+      "learning_rate": 0.0002,
+      "loss": 0.4421,
+      "step": 2470
+    },
+    {
+      "epoch": 5.333333333333333,
+      "grad_norm": 0.9415227770805359,
+      "learning_rate": 0.0002,
+      "loss": 0.4372,
+      "step": 2480
+    },
+    {
+      "epoch": 5.354838709677419,
+      "grad_norm": 0.9133324027061462,
+      "learning_rate": 0.0002,
+      "loss": 0.4223,
+      "step": 2490
+    },
+    {
+      "epoch": 5.376344086021505,
+      "grad_norm": 0.9567879438400269,
+      "learning_rate": 0.0002,
+      "loss": 0.4401,
+      "step": 2500
+    },
+    {
+      "epoch": 5.397849462365591,
+      "grad_norm": 0.9239469766616821,
+      "learning_rate": 0.0002,
+      "loss": 0.4094,
+      "step": 2510
+    },
+    {
+      "epoch": 5.419354838709677,
+      "grad_norm": 1.0293527841567993,
+      "learning_rate": 0.0002,
+      "loss": 0.4416,
+      "step": 2520
+    },
+    {
+      "epoch": 5.440860215053763,
+      "grad_norm": 0.8618718981742859,
+      "learning_rate": 0.0002,
+      "loss": 0.4311,
+      "step": 2530
+    },
+    {
+      "epoch": 5.462365591397849,
+      "grad_norm": 0.740166187286377,
+      "learning_rate": 0.0002,
+      "loss": 0.462,
+      "step": 2540
+    },
+    {
+      "epoch": 5.483870967741936,
+      "grad_norm": 0.901566743850708,
+      "learning_rate": 0.0002,
+      "loss": 0.4172,
+      "step": 2550
+    },
+    {
+      "epoch": 5.505376344086022,
+      "grad_norm": 0.7957597970962524,
+      "learning_rate": 0.0002,
+      "loss": 0.4315,
+      "step": 2560
+    },
+    {
+      "epoch": 5.526881720430108,
+      "grad_norm": 1.1139343976974487,
+      "learning_rate": 0.0002,
+      "loss": 0.4263,
+      "step": 2570
+    },
+    {
+      "epoch": 5.548387096774194,
+      "grad_norm": 0.989765465259552,
+      "learning_rate": 0.0002,
+      "loss": 0.4056,
+      "step": 2580
+    },
+    {
+      "epoch": 5.56989247311828,
+      "grad_norm": 0.9416969418525696,
+      "learning_rate": 0.0002,
+      "loss": 0.4311,
+      "step": 2590
+    },
+    {
+      "epoch": 5.591397849462366,
+      "grad_norm": 0.9184830784797668,
+      "learning_rate": 0.0002,
+      "loss": 0.4363,
+      "step": 2600
+    },
+    {
+      "epoch": 5.612903225806452,
+      "grad_norm": 1.0512700080871582,
+      "learning_rate": 0.0002,
+      "loss": 0.432,
+      "step": 2610
+    },
+    {
+      "epoch": 5.634408602150538,
+      "grad_norm": 0.901462197303772,
+      "learning_rate": 0.0002,
+      "loss": 0.4227,
+      "step": 2620
+    },
+    {
+      "epoch": 5.655913978494624,
+      "grad_norm": 0.9732566475868225,
+      "learning_rate": 0.0002,
+      "loss": 0.4332,
+      "step": 2630
+    },
+    {
+      "epoch": 5.67741935483871,
+      "grad_norm": 0.8180275559425354,
+      "learning_rate": 0.0002,
+      "loss": 0.4223,
+      "step": 2640
+    },
+    {
+      "epoch": 5.698924731182796,
+      "grad_norm": 1.1354765892028809,
+      "learning_rate": 0.0002,
+      "loss": 0.4311,
+      "step": 2650
+    },
+    {
+      "epoch": 5.720430107526882,
+      "grad_norm": 0.9161503314971924,
+      "learning_rate": 0.0002,
+      "loss": 0.4409,
+      "step": 2660
+    },
+    {
+      "epoch": 5.741935483870968,
+      "grad_norm": 1.0561772584915161,
+      "learning_rate": 0.0002,
+      "loss": 0.4394,
+      "step": 2670
+    },
+    {
+      "epoch": 5.763440860215054,
+      "grad_norm": 0.7712787389755249,
+      "learning_rate": 0.0002,
+      "loss": 0.424,
+      "step": 2680
+    },
+    {
+      "epoch": 5.78494623655914,
+      "grad_norm": 0.9674550294876099,
+      "learning_rate": 0.0002,
+      "loss": 0.4326,
+      "step": 2690
+    },
+    {
+      "epoch": 5.806451612903226,
+      "grad_norm": 0.7531843781471252,
+      "learning_rate": 0.0002,
+      "loss": 0.4459,
+      "step": 2700
+    },
+    {
+      "epoch": 5.827956989247312,
+      "grad_norm": 1.1332131624221802,
+      "learning_rate": 0.0002,
+      "loss": 0.4276,
+      "step": 2710
+    },
+    {
+      "epoch": 5.849462365591398,
+      "grad_norm": 0.9367414116859436,
+      "learning_rate": 0.0002,
+      "loss": 0.4113,
+      "step": 2720
+    },
+    {
+      "epoch": 5.870967741935484,
+      "grad_norm": 0.8267706632614136,
+      "learning_rate": 0.0002,
+      "loss": 0.4227,
+      "step": 2730
+    },
+    {
+      "epoch": 5.89247311827957,
+      "grad_norm": 1.1040657758712769,
+      "learning_rate": 0.0002,
+      "loss": 0.4218,
+      "step": 2740
+    },
+    {
+      "epoch": 5.913978494623656,
+      "grad_norm": 0.8879582285881042,
+      "learning_rate": 0.0002,
+      "loss": 0.4129,
+      "step": 2750
+    },
+    {
+      "epoch": 5.935483870967742,
+      "grad_norm": 0.9264667630195618,
+      "learning_rate": 0.0002,
+      "loss": 0.4241,
+      "step": 2760
+    },
+    {
+      "epoch": 5.956989247311828,
+      "grad_norm": 0.9373905658721924,
+      "learning_rate": 0.0002,
+      "loss": 0.4318,
+      "step": 2770
+    },
+    {
+      "epoch": 5.978494623655914,
+      "grad_norm": 1.0063740015029907,
+      "learning_rate": 0.0002,
+      "loss": 0.423,
+      "step": 2780
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 0.8291367292404175,
+      "learning_rate": 0.0002,
+      "loss": 0.4382,
+      "step": 2790
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 0.5057176947593689,
+      "eval_runtime": 21.3206,
+      "eval_samples_per_second": 15.525,
+      "eval_steps_per_second": 1.97,
+      "step": 2790
+    },
+    {
+      "epoch": 6.021505376344086,
+      "grad_norm": 1.0137434005737305,
+      "learning_rate": 0.0002,
+      "loss": 0.3907,
+      "step": 2800
+    },
+    {
+      "epoch": 6.043010752688172,
+      "grad_norm": 0.7550579905509949,
+      "learning_rate": 0.0002,
+      "loss": 0.3793,
+      "step": 2810
+    },
+    {
+      "epoch": 6.064516129032258,
+      "grad_norm": 1.0664116144180298,
+      "learning_rate": 0.0002,
+      "loss": 0.4003,
+      "step": 2820
+    },
+    {
+      "epoch": 6.086021505376344,
+      "grad_norm": 0.7908814549446106,
+      "learning_rate": 0.0002,
+      "loss": 0.3876,
+      "step": 2830
+    },
+    {
+      "epoch": 6.10752688172043,
+      "grad_norm": 0.8101639747619629,
+      "learning_rate": 0.0002,
+      "loss": 0.3884,
+      "step": 2840
+    },
+    {
+      "epoch": 6.129032258064516,
+      "grad_norm": 0.7882567048072815,
+      "learning_rate": 0.0002,
+      "loss": 0.3835,
+      "step": 2850
+    },
+    {
+      "epoch": 6.150537634408602,
+      "grad_norm": 1.0134103298187256,
+      "learning_rate": 0.0002,
+      "loss": 0.3827,
+      "step": 2860
+    },
+    {
+      "epoch": 6.172043010752688,
+      "grad_norm": 0.9240215420722961,
+      "learning_rate": 0.0002,
+      "loss": 0.3963,
+      "step": 2870
+    },
+    {
+      "epoch": 6.193548387096774,
+      "grad_norm": 0.8322992920875549,
+      "learning_rate": 0.0002,
+      "loss": 0.4049,
+      "step": 2880
+    },
+    {
+      "epoch": 6.21505376344086,
+      "grad_norm": 0.9238720536231995,
+      "learning_rate": 0.0002,
+      "loss": 0.381,
+      "step": 2890
+    },
+    {
+      "epoch": 6.236559139784946,
+      "grad_norm": 0.9361863732337952,
+      "learning_rate": 0.0002,
+      "loss": 0.3852,
+      "step": 2900
+    },
+    {
+      "epoch": 6.258064516129032,
+      "grad_norm": 0.9670863747596741,
+      "learning_rate": 0.0002,
+      "loss": 0.3917,
+      "step": 2910
+    },
+    {
+      "epoch": 6.279569892473118,
+      "grad_norm": 0.7724685668945312,
+      "learning_rate": 0.0002,
+      "loss": 0.3826,
+      "step": 2920
+    },
+    {
+      "epoch": 6.301075268817204,
+      "grad_norm": 0.8125540614128113,
+      "learning_rate": 0.0002,
+      "loss": 0.3988,
+      "step": 2930
+    },
+    {
+      "epoch": 6.32258064516129,
+      "grad_norm": 0.9483002424240112,
+      "learning_rate": 0.0002,
+      "loss": 0.3778,
+      "step": 2940
+    },
+    {
+      "epoch": 6.344086021505376,
+      "grad_norm": 1.098374843597412,
+      "learning_rate": 0.0002,
+      "loss": 0.3823,
+      "step": 2950
+    },
+    {
+      "epoch": 6.365591397849462,
+      "grad_norm": 1.0169378519058228,
+      "learning_rate": 0.0002,
+      "loss": 0.3886,
+      "step": 2960
+    },
+    {
+      "epoch": 6.387096774193548,
+      "grad_norm": 0.8594151139259338,
+      "learning_rate": 0.0002,
+      "loss": 0.3936,
+      "step": 2970
+    },
+    {
+      "epoch": 6.408602150537634,
+      "grad_norm": 0.9507288336753845,
+      "learning_rate": 0.0002,
+      "loss": 0.3871,
+      "step": 2980
+    },
+    {
+      "epoch": 6.43010752688172,
+      "grad_norm": 0.9212459325790405,
+      "learning_rate": 0.0002,
+      "loss": 0.3852,
+      "step": 2990
+    },
+    {
+      "epoch": 6.451612903225806,
+      "grad_norm": 0.9696952104568481,
+      "learning_rate": 0.0002,
+      "loss": 0.3929,
+      "step": 3000
+    },
+    {
+      "epoch": 6.473118279569892,
+      "grad_norm": 0.8872610330581665,
+      "learning_rate": 0.0002,
+      "loss": 0.3933,
+      "step": 3010
+    },
+    {
+      "epoch": 6.494623655913978,
+      "grad_norm": 0.9207532405853271,
+      "learning_rate": 0.0002,
+      "loss": 0.393,
+      "step": 3020
+    },
+    {
+      "epoch": 6.516129032258064,
+      "grad_norm": 0.9116262793540955,
+      "learning_rate": 0.0002,
+      "loss": 0.3848,
+      "step": 3030
+    },
+    {
+      "epoch": 6.53763440860215,
+      "grad_norm": 0.83391934633255,
+      "learning_rate": 0.0002,
+      "loss": 0.3964,
+      "step": 3040
+    },
+    {
+      "epoch": 6.559139784946236,
+      "grad_norm": 0.890931248664856,
+      "learning_rate": 0.0002,
+      "loss": 0.3758,
+      "step": 3050
+    },
+    {
+      "epoch": 6.580645161290323,
+      "grad_norm": 1.0100581645965576,
+      "learning_rate": 0.0002,
+      "loss": 0.3944,
+      "step": 3060
+    },
+    {
+      "epoch": 6.602150537634409,
+      "grad_norm": 0.783526599407196,
+      "learning_rate": 0.0002,
+      "loss": 0.3992,
+      "step": 3070
+    },
+    {
+      "epoch": 6.623655913978495,
+      "grad_norm": 1.324326515197754,
+      "learning_rate": 0.0002,
+      "loss": 0.4144,
+      "step": 3080
+    },
+    {
+      "epoch": 6.645161290322581,
+      "grad_norm": 0.9102319478988647,
+      "learning_rate": 0.0002,
+      "loss": 0.3986,
+      "step": 3090
+    },
+    {
+      "epoch": 6.666666666666667,
+      "grad_norm": 0.96951824426651,
+      "learning_rate": 0.0002,
+      "loss": 0.3873,
+      "step": 3100
+    },
+    {
+      "epoch": 6.688172043010753,
+      "grad_norm": 0.9786809086799622,
+      "learning_rate": 0.0002,
+      "loss": 0.3931,
+      "step": 3110
+    },
+    {
+      "epoch": 6.709677419354839,
+      "grad_norm": 1.0301238298416138,
+      "learning_rate": 0.0002,
+      "loss": 0.3714,
+      "step": 3120
+    },
+    {
+      "epoch": 6.731182795698925,
+      "grad_norm": 1.1690906286239624,
+      "learning_rate": 0.0002,
+      "loss": 0.3823,
+      "step": 3130
+    },
+    {
+      "epoch": 6.752688172043011,
+      "grad_norm": 0.963306725025177,
+      "learning_rate": 0.0002,
+      "loss": 0.3936,
+      "step": 3140
+    },
+    {
+      "epoch": 6.774193548387097,
+      "grad_norm": 0.8565770983695984,
+      "learning_rate": 0.0002,
+      "loss": 0.3975,
+      "step": 3150
+    },
+    {
+      "epoch": 6.795698924731183,
+      "grad_norm": 0.8887158632278442,
+      "learning_rate": 0.0002,
+      "loss": 0.3903,
+      "step": 3160
+    },
+    {
+      "epoch": 6.817204301075269,
+      "grad_norm": 0.8234561085700989,
+      "learning_rate": 0.0002,
+      "loss": 0.4098,
+      "step": 3170
+    },
+    {
+      "epoch": 6.838709677419355,
+      "grad_norm": 0.9000219702720642,
+      "learning_rate": 0.0002,
+      "loss": 0.4041,
+      "step": 3180
+    },
+    {
+      "epoch": 6.860215053763441,
+      "grad_norm": 1.1366009712219238,
+      "learning_rate": 0.0002,
+      "loss": 0.3933,
+      "step": 3190
+    },
+    {
+      "epoch": 6.881720430107527,
+      "grad_norm": 0.8747097849845886,
+      "learning_rate": 0.0002,
+      "loss": 0.3972,
+      "step": 3200
+    },
+    {
+      "epoch": 6.903225806451613,
+      "grad_norm": 0.8533893823623657,
+      "learning_rate": 0.0002,
+      "loss": 0.404,
+      "step": 3210
+    },
+    {
+      "epoch": 6.924731182795699,
+      "grad_norm": 0.8127949237823486,
+      "learning_rate": 0.0002,
+      "loss": 0.3906,
+      "step": 3220
+    },
+    {
+      "epoch": 6.946236559139785,
+      "grad_norm": 0.8872477412223816,
+      "learning_rate": 0.0002,
+      "loss": 0.3747,
+      "step": 3230
+    },
+    {
+      "epoch": 6.967741935483871,
+      "grad_norm": 0.8541608452796936,
+      "learning_rate": 0.0002,
+      "loss": 0.3817,
+      "step": 3240
+    },
+    {
+      "epoch": 6.989247311827957,
+      "grad_norm": 0.8390752673149109,
+      "learning_rate": 0.0002,
+      "loss": 0.3863,
+      "step": 3250
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 0.48264747858047485,
+      "eval_runtime": 21.3942,
+      "eval_samples_per_second": 15.471,
+      "eval_steps_per_second": 1.963,
+      "step": 3255
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 3720,
+  "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": 3.978259807076352e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/training_args.bin b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..483b53d4fc1c568a5fc890fa850e3450f390b208
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e4c4fe2be590ed03492316230adb3a1edca3e4066c55f3716c0352d7134c564
+size 5560
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/README.md b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..830a14f7db2734beb59f320973504e45a3fe87f5
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/adapter_config.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e99bbcd43df1c19d98706c7e3be95c93844c5349
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/adapter_model.safetensors b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..109596fbe768580a1c4f560efa8a5a99f822a8aa
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d6e7766c9aae4d2fd34c0286b639660fa9abfa548c97dd1e795db9b07731b767
+size 29500848
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/optimizer.pt b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..5bdebbc054ca2b0200df38941952fedea3c6a295
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:60ca519ee855c6817801ec024584c4698868e71abdffc173ae7f81078aeaa2e2
+size 15064314
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/rng_state.pth b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..23026a8b2b73844837d8930972411617be8de756
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:824426ac36d9031404d4740da6b956e37fa68cac1f3a35686b97495a4c2b41ce
+size 14244
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/scheduler.pt b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..f4e307b1dec6f5df3468f48f3e16d5c992503374
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:02613a6afb86ae474ea068b51099dfd9dc4abfb57724ef7ae1f766d5528370f2
+size 1064
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/special_tokens_map.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/tokenizer.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..f58963a682665634ab180c28667e4faa8cf02ba2
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/tokenizer.model b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/tokenizer_config.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1adb4796c13b8d975555ecec45876ee75d1ae8b7
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/trainer_state.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..7c70b1322223ff3381e8cd51fb03d267481762b3
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/trainer_state.json
@@ -0,0 +1,2701 @@
+{
+  "best_metric": 0.46552130579948425,
+  "best_model_checkpoint": "outputs-001/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720",
+  "epoch": 8.0,
+  "eval_steps": 10,
+  "global_step": 3720,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.021505376344086023,
+      "grad_norm": 0.9075053930282593,
+      "learning_rate": 0.0002,
+      "loss": 3.4172,
+      "step": 10
+    },
+    {
+      "epoch": 0.043010752688172046,
+      "grad_norm": 1.4321208000183105,
+      "learning_rate": 0.0002,
+      "loss": 2.5888,
+      "step": 20
+    },
+    {
+      "epoch": 0.06451612903225806,
+      "grad_norm": 1.7500602006912231,
+      "learning_rate": 0.0002,
+      "loss": 2.1195,
+      "step": 30
+    },
+    {
+      "epoch": 0.08602150537634409,
+      "grad_norm": 0.7606641054153442,
+      "learning_rate": 0.0002,
+      "loss": 1.9303,
+      "step": 40
+    },
+    {
+      "epoch": 0.10752688172043011,
+      "grad_norm": 1.2754929065704346,
+      "learning_rate": 0.0002,
+      "loss": 1.6112,
+      "step": 50
+    },
+    {
+      "epoch": 0.12903225806451613,
+      "grad_norm": 1.0936230421066284,
+      "learning_rate": 0.0002,
+      "loss": 1.4319,
+      "step": 60
+    },
+    {
+      "epoch": 0.15053763440860216,
+      "grad_norm": 1.144593596458435,
+      "learning_rate": 0.0002,
+      "loss": 1.3568,
+      "step": 70
+    },
+    {
+      "epoch": 0.17204301075268819,
+      "grad_norm": 1.2181956768035889,
+      "learning_rate": 0.0002,
+      "loss": 1.2028,
+      "step": 80
+    },
+    {
+      "epoch": 0.1935483870967742,
+      "grad_norm": 1.1260095834732056,
+      "learning_rate": 0.0002,
+      "loss": 1.1534,
+      "step": 90
+    },
+    {
+      "epoch": 0.21505376344086022,
+      "grad_norm": 1.1155284643173218,
+      "learning_rate": 0.0002,
+      "loss": 1.1089,
+      "step": 100
+    },
+    {
+      "epoch": 0.23655913978494625,
+      "grad_norm": 1.089565396308899,
+      "learning_rate": 0.0002,
+      "loss": 1.0883,
+      "step": 110
+    },
+    {
+      "epoch": 0.25806451612903225,
+      "grad_norm": 0.9833471775054932,
+      "learning_rate": 0.0002,
+      "loss": 1.0814,
+      "step": 120
+    },
+    {
+      "epoch": 0.27956989247311825,
+      "grad_norm": 1.0265629291534424,
+      "learning_rate": 0.0002,
+      "loss": 1.0239,
+      "step": 130
+    },
+    {
+      "epoch": 0.3010752688172043,
+      "grad_norm": 0.9344286322593689,
+      "learning_rate": 0.0002,
+      "loss": 0.9888,
+      "step": 140
+    },
+    {
+      "epoch": 0.3225806451612903,
+      "grad_norm": 0.9883386492729187,
+      "learning_rate": 0.0002,
+      "loss": 1.0043,
+      "step": 150
+    },
+    {
+      "epoch": 0.34408602150537637,
+      "grad_norm": 0.9299277067184448,
+      "learning_rate": 0.0002,
+      "loss": 0.9338,
+      "step": 160
+    },
+    {
+      "epoch": 0.3655913978494624,
+      "grad_norm": 1.390045404434204,
+      "learning_rate": 0.0002,
+      "loss": 0.9432,
+      "step": 170
+    },
+    {
+      "epoch": 0.3870967741935484,
+      "grad_norm": 1.0313078165054321,
+      "learning_rate": 0.0002,
+      "loss": 0.9008,
+      "step": 180
+    },
+    {
+      "epoch": 0.40860215053763443,
+      "grad_norm": 1.1792205572128296,
+      "learning_rate": 0.0002,
+      "loss": 0.9434,
+      "step": 190
+    },
+    {
+      "epoch": 0.43010752688172044,
+      "grad_norm": 1.049809217453003,
+      "learning_rate": 0.0002,
+      "loss": 0.8761,
+      "step": 200
+    },
+    {
+      "epoch": 0.45161290322580644,
+      "grad_norm": 0.990111768245697,
+      "learning_rate": 0.0002,
+      "loss": 0.8709,
+      "step": 210
+    },
+    {
+      "epoch": 0.4731182795698925,
+      "grad_norm": 0.9870412349700928,
+      "learning_rate": 0.0002,
+      "loss": 0.905,
+      "step": 220
+    },
+    {
+      "epoch": 0.4946236559139785,
+      "grad_norm": 0.8557345867156982,
+      "learning_rate": 0.0002,
+      "loss": 0.9129,
+      "step": 230
+    },
+    {
+      "epoch": 0.5161290322580645,
+      "grad_norm": 0.9746861457824707,
+      "learning_rate": 0.0002,
+      "loss": 0.8836,
+      "step": 240
+    },
+    {
+      "epoch": 0.5376344086021505,
+      "grad_norm": 0.9010438323020935,
+      "learning_rate": 0.0002,
+      "loss": 0.873,
+      "step": 250
+    },
+    {
+      "epoch": 0.5591397849462365,
+      "grad_norm": 0.9061082005500793,
+      "learning_rate": 0.0002,
+      "loss": 0.8241,
+      "step": 260
+    },
+    {
+      "epoch": 0.5806451612903226,
+      "grad_norm": 0.9311846494674683,
+      "learning_rate": 0.0002,
+      "loss": 0.8652,
+      "step": 270
+    },
+    {
+      "epoch": 0.6021505376344086,
+      "grad_norm": 0.9140254855155945,
+      "learning_rate": 0.0002,
+      "loss": 0.8256,
+      "step": 280
+    },
+    {
+      "epoch": 0.6236559139784946,
+      "grad_norm": 0.9722253084182739,
+      "learning_rate": 0.0002,
+      "loss": 0.8441,
+      "step": 290
+    },
+    {
+      "epoch": 0.6451612903225806,
+      "grad_norm": 0.8539168238639832,
+      "learning_rate": 0.0002,
+      "loss": 0.8314,
+      "step": 300
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.9053162932395935,
+      "learning_rate": 0.0002,
+      "loss": 0.8528,
+      "step": 310
+    },
+    {
+      "epoch": 0.6881720430107527,
+      "grad_norm": 0.8444252610206604,
+      "learning_rate": 0.0002,
+      "loss": 0.8209,
+      "step": 320
+    },
+    {
+      "epoch": 0.7096774193548387,
+      "grad_norm": 0.8127437829971313,
+      "learning_rate": 0.0002,
+      "loss": 0.8101,
+      "step": 330
+    },
+    {
+      "epoch": 0.7311827956989247,
+      "grad_norm": 0.886555016040802,
+      "learning_rate": 0.0002,
+      "loss": 0.8223,
+      "step": 340
+    },
+    {
+      "epoch": 0.7526881720430108,
+      "grad_norm": 0.8458548784255981,
+      "learning_rate": 0.0002,
+      "loss": 0.8368,
+      "step": 350
+    },
+    {
+      "epoch": 0.7741935483870968,
+      "grad_norm": 0.8683297634124756,
+      "learning_rate": 0.0002,
+      "loss": 0.8295,
+      "step": 360
+    },
+    {
+      "epoch": 0.7956989247311828,
+      "grad_norm": 0.8308405876159668,
+      "learning_rate": 0.0002,
+      "loss": 0.8232,
+      "step": 370
+    },
+    {
+      "epoch": 0.8172043010752689,
+      "grad_norm": 0.8305579423904419,
+      "learning_rate": 0.0002,
+      "loss": 0.7752,
+      "step": 380
+    },
+    {
+      "epoch": 0.8387096774193549,
+      "grad_norm": 0.8545567393302917,
+      "learning_rate": 0.0002,
+      "loss": 0.8267,
+      "step": 390
+    },
+    {
+      "epoch": 0.8602150537634409,
+      "grad_norm": 0.8486055731773376,
+      "learning_rate": 0.0002,
+      "loss": 0.8212,
+      "step": 400
+    },
+    {
+      "epoch": 0.8817204301075269,
+      "grad_norm": 0.8126763105392456,
+      "learning_rate": 0.0002,
+      "loss": 0.743,
+      "step": 410
+    },
+    {
+      "epoch": 0.9032258064516129,
+      "grad_norm": 0.8494045734405518,
+      "learning_rate": 0.0002,
+      "loss": 0.7993,
+      "step": 420
+    },
+    {
+      "epoch": 0.9247311827956989,
+      "grad_norm": 0.7639183402061462,
+      "learning_rate": 0.0002,
+      "loss": 0.8213,
+      "step": 430
+    },
+    {
+      "epoch": 0.946236559139785,
+      "grad_norm": 0.858101487159729,
+      "learning_rate": 0.0002,
+      "loss": 0.8015,
+      "step": 440
+    },
+    {
+      "epoch": 0.967741935483871,
+      "grad_norm": 0.8141381740570068,
+      "learning_rate": 0.0002,
+      "loss": 0.7629,
+      "step": 450
+    },
+    {
+      "epoch": 0.989247311827957,
+      "grad_norm": 0.8072513937950134,
+      "learning_rate": 0.0002,
+      "loss": 0.7357,
+      "step": 460
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.7740864157676697,
+      "eval_runtime": 21.383,
+      "eval_samples_per_second": 15.48,
+      "eval_steps_per_second": 1.964,
+      "step": 465
+    },
+    {
+      "epoch": 1.010752688172043,
+      "grad_norm": 0.8269494771957397,
+      "learning_rate": 0.0002,
+      "loss": 0.7701,
+      "step": 470
+    },
+    {
+      "epoch": 1.032258064516129,
+      "grad_norm": 0.7814009189605713,
+      "learning_rate": 0.0002,
+      "loss": 0.7532,
+      "step": 480
+    },
+    {
+      "epoch": 1.053763440860215,
+      "grad_norm": 0.8183923363685608,
+      "learning_rate": 0.0002,
+      "loss": 0.7689,
+      "step": 490
+    },
+    {
+      "epoch": 1.075268817204301,
+      "grad_norm": 0.8146600723266602,
+      "learning_rate": 0.0002,
+      "loss": 0.765,
+      "step": 500
+    },
+    {
+      "epoch": 1.096774193548387,
+      "grad_norm": 0.8635126352310181,
+      "learning_rate": 0.0002,
+      "loss": 0.7358,
+      "step": 510
+    },
+    {
+      "epoch": 1.118279569892473,
+      "grad_norm": 0.8520359396934509,
+      "learning_rate": 0.0002,
+      "loss": 0.7302,
+      "step": 520
+    },
+    {
+      "epoch": 1.139784946236559,
+      "grad_norm": 0.8026443123817444,
+      "learning_rate": 0.0002,
+      "loss": 0.7492,
+      "step": 530
+    },
+    {
+      "epoch": 1.1612903225806452,
+      "grad_norm": 0.8157258629798889,
+      "learning_rate": 0.0002,
+      "loss": 0.7518,
+      "step": 540
+    },
+    {
+      "epoch": 1.1827956989247312,
+      "grad_norm": 0.9450796246528625,
+      "learning_rate": 0.0002,
+      "loss": 0.7461,
+      "step": 550
+    },
+    {
+      "epoch": 1.2043010752688172,
+      "grad_norm": 0.8859835863113403,
+      "learning_rate": 0.0002,
+      "loss": 0.7128,
+      "step": 560
+    },
+    {
+      "epoch": 1.2258064516129032,
+      "grad_norm": 0.7819921970367432,
+      "learning_rate": 0.0002,
+      "loss": 0.7067,
+      "step": 570
+    },
+    {
+      "epoch": 1.2473118279569892,
+      "grad_norm": 0.7823445796966553,
+      "learning_rate": 0.0002,
+      "loss": 0.7577,
+      "step": 580
+    },
+    {
+      "epoch": 1.2688172043010753,
+      "grad_norm": 0.7931883931159973,
+      "learning_rate": 0.0002,
+      "loss": 0.7358,
+      "step": 590
+    },
+    {
+      "epoch": 1.2903225806451613,
+      "grad_norm": 0.7495734095573425,
+      "learning_rate": 0.0002,
+      "loss": 0.723,
+      "step": 600
+    },
+    {
+      "epoch": 1.3118279569892473,
+      "grad_norm": 0.9272717237472534,
+      "learning_rate": 0.0002,
+      "loss": 0.7386,
+      "step": 610
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 0.7968398332595825,
+      "learning_rate": 0.0002,
+      "loss": 0.7498,
+      "step": 620
+    },
+    {
+      "epoch": 1.3548387096774195,
+      "grad_norm": 0.7813659310340881,
+      "learning_rate": 0.0002,
+      "loss": 0.7635,
+      "step": 630
+    },
+    {
+      "epoch": 1.3763440860215055,
+      "grad_norm": 0.730925977230072,
+      "learning_rate": 0.0002,
+      "loss": 0.6665,
+      "step": 640
+    },
+    {
+      "epoch": 1.3978494623655915,
+      "grad_norm": 0.8011482954025269,
+      "learning_rate": 0.0002,
+      "loss": 0.7037,
+      "step": 650
+    },
+    {
+      "epoch": 1.4193548387096775,
+      "grad_norm": 0.7770085334777832,
+      "learning_rate": 0.0002,
+      "loss": 0.6931,
+      "step": 660
+    },
+    {
+      "epoch": 1.4408602150537635,
+      "grad_norm": 0.7432682514190674,
+      "learning_rate": 0.0002,
+      "loss": 0.6949,
+      "step": 670
+    },
+    {
+      "epoch": 1.4623655913978495,
+      "grad_norm": 0.8820092678070068,
+      "learning_rate": 0.0002,
+      "loss": 0.7444,
+      "step": 680
+    },
+    {
+      "epoch": 1.4838709677419355,
+      "grad_norm": 0.7786208987236023,
+      "learning_rate": 0.0002,
+      "loss": 0.6758,
+      "step": 690
+    },
+    {
+      "epoch": 1.5053763440860215,
+      "grad_norm": 0.7467480301856995,
+      "learning_rate": 0.0002,
+      "loss": 0.6702,
+      "step": 700
+    },
+    {
+      "epoch": 1.5268817204301075,
+      "grad_norm": 0.8147122263908386,
+      "learning_rate": 0.0002,
+      "loss": 0.7107,
+      "step": 710
+    },
+    {
+      "epoch": 1.5483870967741935,
+      "grad_norm": 0.796030580997467,
+      "learning_rate": 0.0002,
+      "loss": 0.7144,
+      "step": 720
+    },
+    {
+      "epoch": 1.5698924731182795,
+      "grad_norm": 0.8776171207427979,
+      "learning_rate": 0.0002,
+      "loss": 0.6936,
+      "step": 730
+    },
+    {
+      "epoch": 1.5913978494623655,
+      "grad_norm": 0.8056126236915588,
+      "learning_rate": 0.0002,
+      "loss": 0.7101,
+      "step": 740
+    },
+    {
+      "epoch": 1.6129032258064515,
+      "grad_norm": 0.8141863346099854,
+      "learning_rate": 0.0002,
+      "loss": 0.7162,
+      "step": 750
+    },
+    {
+      "epoch": 1.6344086021505375,
+      "grad_norm": 0.8100557327270508,
+      "learning_rate": 0.0002,
+      "loss": 0.7088,
+      "step": 760
+    },
+    {
+      "epoch": 1.6559139784946235,
+      "grad_norm": 0.8283200860023499,
+      "learning_rate": 0.0002,
+      "loss": 0.7212,
+      "step": 770
+    },
+    {
+      "epoch": 1.6774193548387095,
+      "grad_norm": 0.800865113735199,
+      "learning_rate": 0.0002,
+      "loss": 0.694,
+      "step": 780
+    },
+    {
+      "epoch": 1.6989247311827957,
+      "grad_norm": 0.8052287697792053,
+      "learning_rate": 0.0002,
+      "loss": 0.7076,
+      "step": 790
+    },
+    {
+      "epoch": 1.7204301075268817,
+      "grad_norm": 0.8619674444198608,
+      "learning_rate": 0.0002,
+      "loss": 0.7257,
+      "step": 800
+    },
+    {
+      "epoch": 1.7419354838709677,
+      "grad_norm": 0.8907215595245361,
+      "learning_rate": 0.0002,
+      "loss": 0.7141,
+      "step": 810
+    },
+    {
+      "epoch": 1.7634408602150538,
+      "grad_norm": 0.6976316571235657,
+      "learning_rate": 0.0002,
+      "loss": 0.7035,
+      "step": 820
+    },
+    {
+      "epoch": 1.7849462365591398,
+      "grad_norm": 0.7533746957778931,
+      "learning_rate": 0.0002,
+      "loss": 0.6916,
+      "step": 830
+    },
+    {
+      "epoch": 1.8064516129032258,
+      "grad_norm": 0.7326804399490356,
+      "learning_rate": 0.0002,
+      "loss": 0.7094,
+      "step": 840
+    },
+    {
+      "epoch": 1.827956989247312,
+      "grad_norm": 0.7782683372497559,
+      "learning_rate": 0.0002,
+      "loss": 0.6891,
+      "step": 850
+    },
+    {
+      "epoch": 1.849462365591398,
+      "grad_norm": 0.7424806356430054,
+      "learning_rate": 0.0002,
+      "loss": 0.6931,
+      "step": 860
+    },
+    {
+      "epoch": 1.870967741935484,
+      "grad_norm": 1.172325611114502,
+      "learning_rate": 0.0002,
+      "loss": 0.7354,
+      "step": 870
+    },
+    {
+      "epoch": 1.89247311827957,
+      "grad_norm": 0.771058201789856,
+      "learning_rate": 0.0002,
+      "loss": 0.6866,
+      "step": 880
+    },
+    {
+      "epoch": 1.913978494623656,
+      "grad_norm": 0.8624904155731201,
+      "learning_rate": 0.0002,
+      "loss": 0.7296,
+      "step": 890
+    },
+    {
+      "epoch": 1.935483870967742,
+      "grad_norm": 0.7062820792198181,
+      "learning_rate": 0.0002,
+      "loss": 0.7233,
+      "step": 900
+    },
+    {
+      "epoch": 1.956989247311828,
+      "grad_norm": 0.7560103535652161,
+      "learning_rate": 0.0002,
+      "loss": 0.6966,
+      "step": 910
+    },
+    {
+      "epoch": 1.978494623655914,
+      "grad_norm": 0.788899838924408,
+      "learning_rate": 0.0002,
+      "loss": 0.69,
+      "step": 920
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.6562113761901855,
+      "learning_rate": 0.0002,
+      "loss": 0.6505,
+      "step": 930
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.6885261535644531,
+      "eval_runtime": 21.4291,
+      "eval_samples_per_second": 15.446,
+      "eval_steps_per_second": 1.96,
+      "step": 930
+    },
+    {
+      "epoch": 2.021505376344086,
+      "grad_norm": 0.8216531872749329,
+      "learning_rate": 0.0002,
+      "loss": 0.6625,
+      "step": 940
+    },
+    {
+      "epoch": 2.043010752688172,
+      "grad_norm": 0.8317142724990845,
+      "learning_rate": 0.0002,
+      "loss": 0.6398,
+      "step": 950
+    },
+    {
+      "epoch": 2.064516129032258,
+      "grad_norm": 0.8446708917617798,
+      "learning_rate": 0.0002,
+      "loss": 0.649,
+      "step": 960
+    },
+    {
+      "epoch": 2.086021505376344,
+      "grad_norm": 0.735055148601532,
+      "learning_rate": 0.0002,
+      "loss": 0.657,
+      "step": 970
+    },
+    {
+      "epoch": 2.10752688172043,
+      "grad_norm": 0.7487243413925171,
+      "learning_rate": 0.0002,
+      "loss": 0.649,
+      "step": 980
+    },
+    {
+      "epoch": 2.129032258064516,
+      "grad_norm": 0.8573887944221497,
+      "learning_rate": 0.0002,
+      "loss": 0.6419,
+      "step": 990
+    },
+    {
+      "epoch": 2.150537634408602,
+      "grad_norm": 0.6284521818161011,
+      "learning_rate": 0.0002,
+      "loss": 0.6431,
+      "step": 1000
+    },
+    {
+      "epoch": 2.172043010752688,
+      "grad_norm": 0.754183292388916,
+      "learning_rate": 0.0002,
+      "loss": 0.6128,
+      "step": 1010
+    },
+    {
+      "epoch": 2.193548387096774,
+      "grad_norm": 0.9445359110832214,
+      "learning_rate": 0.0002,
+      "loss": 0.6253,
+      "step": 1020
+    },
+    {
+      "epoch": 2.21505376344086,
+      "grad_norm": 0.808508038520813,
+      "learning_rate": 0.0002,
+      "loss": 0.605,
+      "step": 1030
+    },
+    {
+      "epoch": 2.236559139784946,
+      "grad_norm": 0.9394679665565491,
+      "learning_rate": 0.0002,
+      "loss": 0.6786,
+      "step": 1040
+    },
+    {
+      "epoch": 2.258064516129032,
+      "grad_norm": 0.8151357769966125,
+      "learning_rate": 0.0002,
+      "loss": 0.6176,
+      "step": 1050
+    },
+    {
+      "epoch": 2.279569892473118,
+      "grad_norm": 0.7909848093986511,
+      "learning_rate": 0.0002,
+      "loss": 0.66,
+      "step": 1060
+    },
+    {
+      "epoch": 2.3010752688172045,
+      "grad_norm": 0.7506507039070129,
+      "learning_rate": 0.0002,
+      "loss": 0.6254,
+      "step": 1070
+    },
+    {
+      "epoch": 2.3225806451612905,
+      "grad_norm": 0.8240520358085632,
+      "learning_rate": 0.0002,
+      "loss": 0.6608,
+      "step": 1080
+    },
+    {
+      "epoch": 2.3440860215053765,
+      "grad_norm": 0.9342400431632996,
+      "learning_rate": 0.0002,
+      "loss": 0.6207,
+      "step": 1090
+    },
+    {
+      "epoch": 2.3655913978494625,
+      "grad_norm": 1.0598735809326172,
+      "learning_rate": 0.0002,
+      "loss": 0.6029,
+      "step": 1100
+    },
+    {
+      "epoch": 2.3870967741935485,
+      "grad_norm": 0.7907650470733643,
+      "learning_rate": 0.0002,
+      "loss": 0.6035,
+      "step": 1110
+    },
+    {
+      "epoch": 2.4086021505376345,
+      "grad_norm": 0.9388798475265503,
+      "learning_rate": 0.0002,
+      "loss": 0.6237,
+      "step": 1120
+    },
+    {
+      "epoch": 2.4301075268817205,
+      "grad_norm": 0.8985419869422913,
+      "learning_rate": 0.0002,
+      "loss": 0.6207,
+      "step": 1130
+    },
+    {
+      "epoch": 2.4516129032258065,
+      "grad_norm": 0.7471932768821716,
+      "learning_rate": 0.0002,
+      "loss": 0.5902,
+      "step": 1140
+    },
+    {
+      "epoch": 2.4731182795698925,
+      "grad_norm": 0.761131763458252,
+      "learning_rate": 0.0002,
+      "loss": 0.6446,
+      "step": 1150
+    },
+    {
+      "epoch": 2.4946236559139785,
+      "grad_norm": 0.7901819348335266,
+      "learning_rate": 0.0002,
+      "loss": 0.6088,
+      "step": 1160
+    },
+    {
+      "epoch": 2.5161290322580645,
+      "grad_norm": 0.9932922720909119,
+      "learning_rate": 0.0002,
+      "loss": 0.6142,
+      "step": 1170
+    },
+    {
+      "epoch": 2.5376344086021505,
+      "grad_norm": 0.7414287328720093,
+      "learning_rate": 0.0002,
+      "loss": 0.6407,
+      "step": 1180
+    },
+    {
+      "epoch": 2.5591397849462365,
+      "grad_norm": 0.8111771941184998,
+      "learning_rate": 0.0002,
+      "loss": 0.6161,
+      "step": 1190
+    },
+    {
+      "epoch": 2.5806451612903225,
+      "grad_norm": 0.7520156502723694,
+      "learning_rate": 0.0002,
+      "loss": 0.6006,
+      "step": 1200
+    },
+    {
+      "epoch": 2.6021505376344085,
+      "grad_norm": 0.9022907018661499,
+      "learning_rate": 0.0002,
+      "loss": 0.615,
+      "step": 1210
+    },
+    {
+      "epoch": 2.6236559139784945,
+      "grad_norm": 0.7746260166168213,
+      "learning_rate": 0.0002,
+      "loss": 0.6211,
+      "step": 1220
+    },
+    {
+      "epoch": 2.6451612903225805,
+      "grad_norm": 0.8482862114906311,
+      "learning_rate": 0.0002,
+      "loss": 0.616,
+      "step": 1230
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 0.7925458550453186,
+      "learning_rate": 0.0002,
+      "loss": 0.6417,
+      "step": 1240
+    },
+    {
+      "epoch": 2.688172043010753,
+      "grad_norm": 0.8369929194450378,
+      "learning_rate": 0.0002,
+      "loss": 0.6187,
+      "step": 1250
+    },
+    {
+      "epoch": 2.709677419354839,
+      "grad_norm": 0.8311542868614197,
+      "learning_rate": 0.0002,
+      "loss": 0.6138,
+      "step": 1260
+    },
+    {
+      "epoch": 2.731182795698925,
+      "grad_norm": 0.7204853296279907,
+      "learning_rate": 0.0002,
+      "loss": 0.5894,
+      "step": 1270
+    },
+    {
+      "epoch": 2.752688172043011,
+      "grad_norm": 0.8447284698486328,
+      "learning_rate": 0.0002,
+      "loss": 0.6325,
+      "step": 1280
+    },
+    {
+      "epoch": 2.774193548387097,
+      "grad_norm": 0.7738404273986816,
+      "learning_rate": 0.0002,
+      "loss": 0.5946,
+      "step": 1290
+    },
+    {
+      "epoch": 2.795698924731183,
+      "grad_norm": 0.8393287062644958,
+      "learning_rate": 0.0002,
+      "loss": 0.5678,
+      "step": 1300
+    },
+    {
+      "epoch": 2.817204301075269,
+      "grad_norm": 0.79121994972229,
+      "learning_rate": 0.0002,
+      "loss": 0.6092,
+      "step": 1310
+    },
+    {
+      "epoch": 2.838709677419355,
+      "grad_norm": 0.7331557869911194,
+      "learning_rate": 0.0002,
+      "loss": 0.5889,
+      "step": 1320
+    },
+    {
+      "epoch": 2.860215053763441,
+      "grad_norm": 0.9593998193740845,
+      "learning_rate": 0.0002,
+      "loss": 0.6048,
+      "step": 1330
+    },
+    {
+      "epoch": 2.881720430107527,
+      "grad_norm": 0.7215158343315125,
+      "learning_rate": 0.0002,
+      "loss": 0.6108,
+      "step": 1340
+    },
+    {
+      "epoch": 2.903225806451613,
+      "grad_norm": 0.840404212474823,
+      "learning_rate": 0.0002,
+      "loss": 0.5897,
+      "step": 1350
+    },
+    {
+      "epoch": 2.924731182795699,
+      "grad_norm": 0.870659351348877,
+      "learning_rate": 0.0002,
+      "loss": 0.6056,
+      "step": 1360
+    },
+    {
+      "epoch": 2.946236559139785,
+      "grad_norm": 0.8744975328445435,
+      "learning_rate": 0.0002,
+      "loss": 0.6205,
+      "step": 1370
+    },
+    {
+      "epoch": 2.967741935483871,
+      "grad_norm": 0.8030612468719482,
+      "learning_rate": 0.0002,
+      "loss": 0.5966,
+      "step": 1380
+    },
+    {
+      "epoch": 2.989247311827957,
+      "grad_norm": 0.825814962387085,
+      "learning_rate": 0.0002,
+      "loss": 0.6004,
+      "step": 1390
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 0.6257933378219604,
+      "eval_runtime": 21.3692,
+      "eval_samples_per_second": 15.49,
+      "eval_steps_per_second": 1.965,
+      "step": 1395
+    },
+    {
+      "epoch": 3.010752688172043,
+      "grad_norm": 0.8650677800178528,
+      "learning_rate": 0.0002,
+      "loss": 0.5696,
+      "step": 1400
+    },
+    {
+      "epoch": 3.032258064516129,
+      "grad_norm": 0.8364197015762329,
+      "learning_rate": 0.0002,
+      "loss": 0.5483,
+      "step": 1410
+    },
+    {
+      "epoch": 3.053763440860215,
+      "grad_norm": 0.8278448581695557,
+      "learning_rate": 0.0002,
+      "loss": 0.5606,
+      "step": 1420
+    },
+    {
+      "epoch": 3.075268817204301,
+      "grad_norm": 0.8806642889976501,
+      "learning_rate": 0.0002,
+      "loss": 0.5572,
+      "step": 1430
+    },
+    {
+      "epoch": 3.096774193548387,
+      "grad_norm": 0.8180029988288879,
+      "learning_rate": 0.0002,
+      "loss": 0.585,
+      "step": 1440
+    },
+    {
+      "epoch": 3.118279569892473,
+      "grad_norm": 0.8561782836914062,
+      "learning_rate": 0.0002,
+      "loss": 0.5667,
+      "step": 1450
+    },
+    {
+      "epoch": 3.139784946236559,
+      "grad_norm": 0.8377029299736023,
+      "learning_rate": 0.0002,
+      "loss": 0.5246,
+      "step": 1460
+    },
+    {
+      "epoch": 3.161290322580645,
+      "grad_norm": 0.885779082775116,
+      "learning_rate": 0.0002,
+      "loss": 0.5464,
+      "step": 1470
+    },
+    {
+      "epoch": 3.182795698924731,
+      "grad_norm": 0.9388518333435059,
+      "learning_rate": 0.0002,
+      "loss": 0.541,
+      "step": 1480
+    },
+    {
+      "epoch": 3.204301075268817,
+      "grad_norm": 0.8816235661506653,
+      "learning_rate": 0.0002,
+      "loss": 0.5447,
+      "step": 1490
+    },
+    {
+      "epoch": 3.225806451612903,
+      "grad_norm": 0.9885783791542053,
+      "learning_rate": 0.0002,
+      "loss": 0.5466,
+      "step": 1500
+    },
+    {
+      "epoch": 3.247311827956989,
+      "grad_norm": 0.8635850548744202,
+      "learning_rate": 0.0002,
+      "loss": 0.5455,
+      "step": 1510
+    },
+    {
+      "epoch": 3.268817204301075,
+      "grad_norm": 0.829853355884552,
+      "learning_rate": 0.0002,
+      "loss": 0.5419,
+      "step": 1520
+    },
+    {
+      "epoch": 3.2903225806451615,
+      "grad_norm": 0.9037486910820007,
+      "learning_rate": 0.0002,
+      "loss": 0.54,
+      "step": 1530
+    },
+    {
+      "epoch": 3.3118279569892475,
+      "grad_norm": 0.8173713684082031,
+      "learning_rate": 0.0002,
+      "loss": 0.5375,
+      "step": 1540
+    },
+    {
+      "epoch": 3.3333333333333335,
+      "grad_norm": 0.796953022480011,
+      "learning_rate": 0.0002,
+      "loss": 0.5405,
+      "step": 1550
+    },
+    {
+      "epoch": 3.3548387096774195,
+      "grad_norm": 0.7894400358200073,
+      "learning_rate": 0.0002,
+      "loss": 0.5505,
+      "step": 1560
+    },
+    {
+      "epoch": 3.3763440860215055,
+      "grad_norm": 0.9434949159622192,
+      "learning_rate": 0.0002,
+      "loss": 0.5395,
+      "step": 1570
+    },
+    {
+      "epoch": 3.3978494623655915,
+      "grad_norm": 0.8666760325431824,
+      "learning_rate": 0.0002,
+      "loss": 0.5271,
+      "step": 1580
+    },
+    {
+      "epoch": 3.4193548387096775,
+      "grad_norm": 0.7782467007637024,
+      "learning_rate": 0.0002,
+      "loss": 0.5439,
+      "step": 1590
+    },
+    {
+      "epoch": 3.4408602150537635,
+      "grad_norm": 0.8849126696586609,
+      "learning_rate": 0.0002,
+      "loss": 0.5161,
+      "step": 1600
+    },
+    {
+      "epoch": 3.4623655913978495,
+      "grad_norm": 0.7863831520080566,
+      "learning_rate": 0.0002,
+      "loss": 0.5353,
+      "step": 1610
+    },
+    {
+      "epoch": 3.4838709677419355,
+      "grad_norm": 1.0403116941452026,
+      "learning_rate": 0.0002,
+      "loss": 0.5308,
+      "step": 1620
+    },
+    {
+      "epoch": 3.5053763440860215,
+      "grad_norm": 0.8307499289512634,
+      "learning_rate": 0.0002,
+      "loss": 0.5339,
+      "step": 1630
+    },
+    {
+      "epoch": 3.5268817204301075,
+      "grad_norm": 0.9132118821144104,
+      "learning_rate": 0.0002,
+      "loss": 0.5361,
+      "step": 1640
+    },
+    {
+      "epoch": 3.5483870967741935,
+      "grad_norm": 0.9322578310966492,
+      "learning_rate": 0.0002,
+      "loss": 0.5828,
+      "step": 1650
+    },
+    {
+      "epoch": 3.5698924731182795,
+      "grad_norm": 0.9782460331916809,
+      "learning_rate": 0.0002,
+      "loss": 0.546,
+      "step": 1660
+    },
+    {
+      "epoch": 3.5913978494623655,
+      "grad_norm": 0.7189919352531433,
+      "learning_rate": 0.0002,
+      "loss": 0.5424,
+      "step": 1670
+    },
+    {
+      "epoch": 3.6129032258064515,
+      "grad_norm": 0.9689221382141113,
+      "learning_rate": 0.0002,
+      "loss": 0.5514,
+      "step": 1680
+    },
+    {
+      "epoch": 3.6344086021505375,
+      "grad_norm": 0.9684675335884094,
+      "learning_rate": 0.0002,
+      "loss": 0.5379,
+      "step": 1690
+    },
+    {
+      "epoch": 3.6559139784946235,
+      "grad_norm": 0.8851472735404968,
+      "learning_rate": 0.0002,
+      "loss": 0.5748,
+      "step": 1700
+    },
+    {
+      "epoch": 3.6774193548387095,
+      "grad_norm": 0.7709833383560181,
+      "learning_rate": 0.0002,
+      "loss": 0.5412,
+      "step": 1710
+    },
+    {
+      "epoch": 3.698924731182796,
+      "grad_norm": 0.818236231803894,
+      "learning_rate": 0.0002,
+      "loss": 0.521,
+      "step": 1720
+    },
+    {
+      "epoch": 3.720430107526882,
+      "grad_norm": 0.870642364025116,
+      "learning_rate": 0.0002,
+      "loss": 0.5445,
+      "step": 1730
+    },
+    {
+      "epoch": 3.741935483870968,
+      "grad_norm": 1.0245511531829834,
+      "learning_rate": 0.0002,
+      "loss": 0.5307,
+      "step": 1740
+    },
+    {
+      "epoch": 3.763440860215054,
+      "grad_norm": 0.8607558608055115,
+      "learning_rate": 0.0002,
+      "loss": 0.5593,
+      "step": 1750
+    },
+    {
+      "epoch": 3.78494623655914,
+      "grad_norm": 0.8511829972267151,
+      "learning_rate": 0.0002,
+      "loss": 0.536,
+      "step": 1760
+    },
+    {
+      "epoch": 3.806451612903226,
+      "grad_norm": 0.7969087362289429,
+      "learning_rate": 0.0002,
+      "loss": 0.5193,
+      "step": 1770
+    },
+    {
+      "epoch": 3.827956989247312,
+      "grad_norm": 0.8457245826721191,
+      "learning_rate": 0.0002,
+      "loss": 0.5578,
+      "step": 1780
+    },
+    {
+      "epoch": 3.849462365591398,
+      "grad_norm": 0.8893467783927917,
+      "learning_rate": 0.0002,
+      "loss": 0.5337,
+      "step": 1790
+    },
+    {
+      "epoch": 3.870967741935484,
+      "grad_norm": 0.8593819737434387,
+      "learning_rate": 0.0002,
+      "loss": 0.5024,
+      "step": 1800
+    },
+    {
+      "epoch": 3.89247311827957,
+      "grad_norm": 0.7574560642242432,
+      "learning_rate": 0.0002,
+      "loss": 0.5134,
+      "step": 1810
+    },
+    {
+      "epoch": 3.913978494623656,
+      "grad_norm": 0.8681567311286926,
+      "learning_rate": 0.0002,
+      "loss": 0.5263,
+      "step": 1820
+    },
+    {
+      "epoch": 3.935483870967742,
+      "grad_norm": 0.9068132042884827,
+      "learning_rate": 0.0002,
+      "loss": 0.532,
+      "step": 1830
+    },
+    {
+      "epoch": 3.956989247311828,
+      "grad_norm": 0.8668948411941528,
+      "learning_rate": 0.0002,
+      "loss": 0.5427,
+      "step": 1840
+    },
+    {
+      "epoch": 3.978494623655914,
+      "grad_norm": 1.046032428741455,
+      "learning_rate": 0.0002,
+      "loss": 0.5349,
+      "step": 1850
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.904780387878418,
+      "learning_rate": 0.0002,
+      "loss": 0.5087,
+      "step": 1860
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 0.5737715363502502,
+      "eval_runtime": 21.4915,
+      "eval_samples_per_second": 15.401,
+      "eval_steps_per_second": 1.954,
+      "step": 1860
+    },
+    {
+      "epoch": 4.021505376344086,
+      "grad_norm": 0.8611752986907959,
+      "learning_rate": 0.0002,
+      "loss": 0.4843,
+      "step": 1870
+    },
+    {
+      "epoch": 4.043010752688172,
+      "grad_norm": 0.838782548904419,
+      "learning_rate": 0.0002,
+      "loss": 0.4814,
+      "step": 1880
+    },
+    {
+      "epoch": 4.064516129032258,
+      "grad_norm": 0.9119709134101868,
+      "learning_rate": 0.0002,
+      "loss": 0.474,
+      "step": 1890
+    },
+    {
+      "epoch": 4.086021505376344,
+      "grad_norm": 0.8026251196861267,
+      "learning_rate": 0.0002,
+      "loss": 0.4951,
+      "step": 1900
+    },
+    {
+      "epoch": 4.10752688172043,
+      "grad_norm": 0.8773705363273621,
+      "learning_rate": 0.0002,
+      "loss": 0.491,
+      "step": 1910
+    },
+    {
+      "epoch": 4.129032258064516,
+      "grad_norm": 0.8762255907058716,
+      "learning_rate": 0.0002,
+      "loss": 0.474,
+      "step": 1920
+    },
+    {
+      "epoch": 4.150537634408602,
+      "grad_norm": 0.8371861577033997,
+      "learning_rate": 0.0002,
+      "loss": 0.4816,
+      "step": 1930
+    },
+    {
+      "epoch": 4.172043010752688,
+      "grad_norm": 0.9703728556632996,
+      "learning_rate": 0.0002,
+      "loss": 0.472,
+      "step": 1940
+    },
+    {
+      "epoch": 4.193548387096774,
+      "grad_norm": 0.8802874684333801,
+      "learning_rate": 0.0002,
+      "loss": 0.4772,
+      "step": 1950
+    },
+    {
+      "epoch": 4.21505376344086,
+      "grad_norm": 1.0103057622909546,
+      "learning_rate": 0.0002,
+      "loss": 0.5032,
+      "step": 1960
+    },
+    {
+      "epoch": 4.236559139784946,
+      "grad_norm": 0.9212995171546936,
+      "learning_rate": 0.0002,
+      "loss": 0.4945,
+      "step": 1970
+    },
+    {
+      "epoch": 4.258064516129032,
+      "grad_norm": 1.009544849395752,
+      "learning_rate": 0.0002,
+      "loss": 0.4753,
+      "step": 1980
+    },
+    {
+      "epoch": 4.279569892473118,
+      "grad_norm": 0.8535077571868896,
+      "learning_rate": 0.0002,
+      "loss": 0.4789,
+      "step": 1990
+    },
+    {
+      "epoch": 4.301075268817204,
+      "grad_norm": 0.8363022804260254,
+      "learning_rate": 0.0002,
+      "loss": 0.4782,
+      "step": 2000
+    },
+    {
+      "epoch": 4.32258064516129,
+      "grad_norm": 0.9041762948036194,
+      "learning_rate": 0.0002,
+      "loss": 0.4875,
+      "step": 2010
+    },
+    {
+      "epoch": 4.344086021505376,
+      "grad_norm": 0.960790753364563,
+      "learning_rate": 0.0002,
+      "loss": 0.4779,
+      "step": 2020
+    },
+    {
+      "epoch": 4.365591397849462,
+      "grad_norm": 0.8823095560073853,
+      "learning_rate": 0.0002,
+      "loss": 0.4626,
+      "step": 2030
+    },
+    {
+      "epoch": 4.387096774193548,
+      "grad_norm": 0.952100396156311,
+      "learning_rate": 0.0002,
+      "loss": 0.4883,
+      "step": 2040
+    },
+    {
+      "epoch": 4.408602150537634,
+      "grad_norm": 1.0793498754501343,
+      "learning_rate": 0.0002,
+      "loss": 0.4789,
+      "step": 2050
+    },
+    {
+      "epoch": 4.43010752688172,
+      "grad_norm": 0.8987208008766174,
+      "learning_rate": 0.0002,
+      "loss": 0.4827,
+      "step": 2060
+    },
+    {
+      "epoch": 4.451612903225806,
+      "grad_norm": 0.8539772033691406,
+      "learning_rate": 0.0002,
+      "loss": 0.4594,
+      "step": 2070
+    },
+    {
+      "epoch": 4.473118279569892,
+      "grad_norm": 0.9160863757133484,
+      "learning_rate": 0.0002,
+      "loss": 0.4752,
+      "step": 2080
+    },
+    {
+      "epoch": 4.494623655913978,
+      "grad_norm": 0.9946850538253784,
+      "learning_rate": 0.0002,
+      "loss": 0.5033,
+      "step": 2090
+    },
+    {
+      "epoch": 4.516129032258064,
+      "grad_norm": 0.908039391040802,
+      "learning_rate": 0.0002,
+      "loss": 0.4842,
+      "step": 2100
+    },
+    {
+      "epoch": 4.53763440860215,
+      "grad_norm": 1.1462254524230957,
+      "learning_rate": 0.0002,
+      "loss": 0.4861,
+      "step": 2110
+    },
+    {
+      "epoch": 4.559139784946236,
+      "grad_norm": 0.8392056226730347,
+      "learning_rate": 0.0002,
+      "loss": 0.4892,
+      "step": 2120
+    },
+    {
+      "epoch": 4.580645161290323,
+      "grad_norm": 0.9673896431922913,
+      "learning_rate": 0.0002,
+      "loss": 0.4824,
+      "step": 2130
+    },
+    {
+      "epoch": 4.602150537634409,
+      "grad_norm": 0.9047091603279114,
+      "learning_rate": 0.0002,
+      "loss": 0.4665,
+      "step": 2140
+    },
+    {
+      "epoch": 4.623655913978495,
+      "grad_norm": 0.9013425707817078,
+      "learning_rate": 0.0002,
+      "loss": 0.4714,
+      "step": 2150
+    },
+    {
+      "epoch": 4.645161290322581,
+      "grad_norm": 0.8899165391921997,
+      "learning_rate": 0.0002,
+      "loss": 0.472,
+      "step": 2160
+    },
+    {
+      "epoch": 4.666666666666667,
+      "grad_norm": 0.748602569103241,
+      "learning_rate": 0.0002,
+      "loss": 0.4635,
+      "step": 2170
+    },
+    {
+      "epoch": 4.688172043010753,
+      "grad_norm": 0.8694155216217041,
+      "learning_rate": 0.0002,
+      "loss": 0.4695,
+      "step": 2180
+    },
+    {
+      "epoch": 4.709677419354839,
+      "grad_norm": 0.9134316444396973,
+      "learning_rate": 0.0002,
+      "loss": 0.4929,
+      "step": 2190
+    },
+    {
+      "epoch": 4.731182795698925,
+      "grad_norm": 0.8504763245582581,
+      "learning_rate": 0.0002,
+      "loss": 0.4855,
+      "step": 2200
+    },
+    {
+      "epoch": 4.752688172043011,
+      "grad_norm": 1.0321544408798218,
+      "learning_rate": 0.0002,
+      "loss": 0.4517,
+      "step": 2210
+    },
+    {
+      "epoch": 4.774193548387097,
+      "grad_norm": 0.9368237257003784,
+      "learning_rate": 0.0002,
+      "loss": 0.4796,
+      "step": 2220
+    },
+    {
+      "epoch": 4.795698924731183,
+      "grad_norm": 0.9319947361946106,
+      "learning_rate": 0.0002,
+      "loss": 0.4837,
+      "step": 2230
+    },
+    {
+      "epoch": 4.817204301075269,
+      "grad_norm": 0.904333770275116,
+      "learning_rate": 0.0002,
+      "loss": 0.4696,
+      "step": 2240
+    },
+    {
+      "epoch": 4.838709677419355,
+      "grad_norm": 0.8097078204154968,
+      "learning_rate": 0.0002,
+      "loss": 0.4746,
+      "step": 2250
+    },
+    {
+      "epoch": 4.860215053763441,
+      "grad_norm": 0.9128859043121338,
+      "learning_rate": 0.0002,
+      "loss": 0.4438,
+      "step": 2260
+    },
+    {
+      "epoch": 4.881720430107527,
+      "grad_norm": 0.883129894733429,
+      "learning_rate": 0.0002,
+      "loss": 0.4693,
+      "step": 2270
+    },
+    {
+      "epoch": 4.903225806451613,
+      "grad_norm": 0.85712730884552,
+      "learning_rate": 0.0002,
+      "loss": 0.4494,
+      "step": 2280
+    },
+    {
+      "epoch": 4.924731182795699,
+      "grad_norm": 1.2101863622665405,
+      "learning_rate": 0.0002,
+      "loss": 0.4593,
+      "step": 2290
+    },
+    {
+      "epoch": 4.946236559139785,
+      "grad_norm": 0.917966902256012,
+      "learning_rate": 0.0002,
+      "loss": 0.4779,
+      "step": 2300
+    },
+    {
+      "epoch": 4.967741935483871,
+      "grad_norm": 0.7740724086761475,
+      "learning_rate": 0.0002,
+      "loss": 0.4666,
+      "step": 2310
+    },
+    {
+      "epoch": 4.989247311827957,
+      "grad_norm": 1.0199906826019287,
+      "learning_rate": 0.0002,
+      "loss": 0.4629,
+      "step": 2320
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 0.5363914370536804,
+      "eval_runtime": 21.3941,
+      "eval_samples_per_second": 15.472,
+      "eval_steps_per_second": 1.963,
+      "step": 2325
+    },
+    {
+      "epoch": 5.010752688172043,
+      "grad_norm": 0.8580502271652222,
+      "learning_rate": 0.0002,
+      "loss": 0.4543,
+      "step": 2330
+    },
+    {
+      "epoch": 5.032258064516129,
+      "grad_norm": 0.7702704668045044,
+      "learning_rate": 0.0002,
+      "loss": 0.404,
+      "step": 2340
+    },
+    {
+      "epoch": 5.053763440860215,
+      "grad_norm": 0.9417401552200317,
+      "learning_rate": 0.0002,
+      "loss": 0.4408,
+      "step": 2350
+    },
+    {
+      "epoch": 5.075268817204301,
+      "grad_norm": 0.9461463689804077,
+      "learning_rate": 0.0002,
+      "loss": 0.4306,
+      "step": 2360
+    },
+    {
+      "epoch": 5.096774193548387,
+      "grad_norm": 0.8931282162666321,
+      "learning_rate": 0.0002,
+      "loss": 0.4251,
+      "step": 2370
+    },
+    {
+      "epoch": 5.118279569892473,
+      "grad_norm": 1.000909447669983,
+      "learning_rate": 0.0002,
+      "loss": 0.4249,
+      "step": 2380
+    },
+    {
+      "epoch": 5.139784946236559,
+      "grad_norm": 0.8640249967575073,
+      "learning_rate": 0.0002,
+      "loss": 0.4231,
+      "step": 2390
+    },
+    {
+      "epoch": 5.161290322580645,
+      "grad_norm": 1.0451020002365112,
+      "learning_rate": 0.0002,
+      "loss": 0.4272,
+      "step": 2400
+    },
+    {
+      "epoch": 5.182795698924731,
+      "grad_norm": 0.7896912097930908,
+      "learning_rate": 0.0002,
+      "loss": 0.4177,
+      "step": 2410
+    },
+    {
+      "epoch": 5.204301075268817,
+      "grad_norm": 0.8424463272094727,
+      "learning_rate": 0.0002,
+      "loss": 0.4116,
+      "step": 2420
+    },
+    {
+      "epoch": 5.225806451612903,
+      "grad_norm": 1.0852105617523193,
+      "learning_rate": 0.0002,
+      "loss": 0.4225,
+      "step": 2430
+    },
+    {
+      "epoch": 5.247311827956989,
+      "grad_norm": 0.9285983443260193,
+      "learning_rate": 0.0002,
+      "loss": 0.4352,
+      "step": 2440
+    },
+    {
+      "epoch": 5.268817204301075,
+      "grad_norm": 0.9119299054145813,
+      "learning_rate": 0.0002,
+      "loss": 0.4262,
+      "step": 2450
+    },
+    {
+      "epoch": 5.290322580645161,
+      "grad_norm": 0.8790456056594849,
+      "learning_rate": 0.0002,
+      "loss": 0.4494,
+      "step": 2460
+    },
+    {
+      "epoch": 5.311827956989247,
+      "grad_norm": 0.8726504445075989,
+      "learning_rate": 0.0002,
+      "loss": 0.4421,
+      "step": 2470
+    },
+    {
+      "epoch": 5.333333333333333,
+      "grad_norm": 0.9415227770805359,
+      "learning_rate": 0.0002,
+      "loss": 0.4372,
+      "step": 2480
+    },
+    {
+      "epoch": 5.354838709677419,
+      "grad_norm": 0.9133324027061462,
+      "learning_rate": 0.0002,
+      "loss": 0.4223,
+      "step": 2490
+    },
+    {
+      "epoch": 5.376344086021505,
+      "grad_norm": 0.9567879438400269,
+      "learning_rate": 0.0002,
+      "loss": 0.4401,
+      "step": 2500
+    },
+    {
+      "epoch": 5.397849462365591,
+      "grad_norm": 0.9239469766616821,
+      "learning_rate": 0.0002,
+      "loss": 0.4094,
+      "step": 2510
+    },
+    {
+      "epoch": 5.419354838709677,
+      "grad_norm": 1.0293527841567993,
+      "learning_rate": 0.0002,
+      "loss": 0.4416,
+      "step": 2520
+    },
+    {
+      "epoch": 5.440860215053763,
+      "grad_norm": 0.8618718981742859,
+      "learning_rate": 0.0002,
+      "loss": 0.4311,
+      "step": 2530
+    },
+    {
+      "epoch": 5.462365591397849,
+      "grad_norm": 0.740166187286377,
+      "learning_rate": 0.0002,
+      "loss": 0.462,
+      "step": 2540
+    },
+    {
+      "epoch": 5.483870967741936,
+      "grad_norm": 0.901566743850708,
+      "learning_rate": 0.0002,
+      "loss": 0.4172,
+      "step": 2550
+    },
+    {
+      "epoch": 5.505376344086022,
+      "grad_norm": 0.7957597970962524,
+      "learning_rate": 0.0002,
+      "loss": 0.4315,
+      "step": 2560
+    },
+    {
+      "epoch": 5.526881720430108,
+      "grad_norm": 1.1139343976974487,
+      "learning_rate": 0.0002,
+      "loss": 0.4263,
+      "step": 2570
+    },
+    {
+      "epoch": 5.548387096774194,
+      "grad_norm": 0.989765465259552,
+      "learning_rate": 0.0002,
+      "loss": 0.4056,
+      "step": 2580
+    },
+    {
+      "epoch": 5.56989247311828,
+      "grad_norm": 0.9416969418525696,
+      "learning_rate": 0.0002,
+      "loss": 0.4311,
+      "step": 2590
+    },
+    {
+      "epoch": 5.591397849462366,
+      "grad_norm": 0.9184830784797668,
+      "learning_rate": 0.0002,
+      "loss": 0.4363,
+      "step": 2600
+    },
+    {
+      "epoch": 5.612903225806452,
+      "grad_norm": 1.0512700080871582,
+      "learning_rate": 0.0002,
+      "loss": 0.432,
+      "step": 2610
+    },
+    {
+      "epoch": 5.634408602150538,
+      "grad_norm": 0.901462197303772,
+      "learning_rate": 0.0002,
+      "loss": 0.4227,
+      "step": 2620
+    },
+    {
+      "epoch": 5.655913978494624,
+      "grad_norm": 0.9732566475868225,
+      "learning_rate": 0.0002,
+      "loss": 0.4332,
+      "step": 2630
+    },
+    {
+      "epoch": 5.67741935483871,
+      "grad_norm": 0.8180275559425354,
+      "learning_rate": 0.0002,
+      "loss": 0.4223,
+      "step": 2640
+    },
+    {
+      "epoch": 5.698924731182796,
+      "grad_norm": 1.1354765892028809,
+      "learning_rate": 0.0002,
+      "loss": 0.4311,
+      "step": 2650
+    },
+    {
+      "epoch": 5.720430107526882,
+      "grad_norm": 0.9161503314971924,
+      "learning_rate": 0.0002,
+      "loss": 0.4409,
+      "step": 2660
+    },
+    {
+      "epoch": 5.741935483870968,
+      "grad_norm": 1.0561772584915161,
+      "learning_rate": 0.0002,
+      "loss": 0.4394,
+      "step": 2670
+    },
+    {
+      "epoch": 5.763440860215054,
+      "grad_norm": 0.7712787389755249,
+      "learning_rate": 0.0002,
+      "loss": 0.424,
+      "step": 2680
+    },
+    {
+      "epoch": 5.78494623655914,
+      "grad_norm": 0.9674550294876099,
+      "learning_rate": 0.0002,
+      "loss": 0.4326,
+      "step": 2690
+    },
+    {
+      "epoch": 5.806451612903226,
+      "grad_norm": 0.7531843781471252,
+      "learning_rate": 0.0002,
+      "loss": 0.4459,
+      "step": 2700
+    },
+    {
+      "epoch": 5.827956989247312,
+      "grad_norm": 1.1332131624221802,
+      "learning_rate": 0.0002,
+      "loss": 0.4276,
+      "step": 2710
+    },
+    {
+      "epoch": 5.849462365591398,
+      "grad_norm": 0.9367414116859436,
+      "learning_rate": 0.0002,
+      "loss": 0.4113,
+      "step": 2720
+    },
+    {
+      "epoch": 5.870967741935484,
+      "grad_norm": 0.8267706632614136,
+      "learning_rate": 0.0002,
+      "loss": 0.4227,
+      "step": 2730
+    },
+    {
+      "epoch": 5.89247311827957,
+      "grad_norm": 1.1040657758712769,
+      "learning_rate": 0.0002,
+      "loss": 0.4218,
+      "step": 2740
+    },
+    {
+      "epoch": 5.913978494623656,
+      "grad_norm": 0.8879582285881042,
+      "learning_rate": 0.0002,
+      "loss": 0.4129,
+      "step": 2750
+    },
+    {
+      "epoch": 5.935483870967742,
+      "grad_norm": 0.9264667630195618,
+      "learning_rate": 0.0002,
+      "loss": 0.4241,
+      "step": 2760
+    },
+    {
+      "epoch": 5.956989247311828,
+      "grad_norm": 0.9373905658721924,
+      "learning_rate": 0.0002,
+      "loss": 0.4318,
+      "step": 2770
+    },
+    {
+      "epoch": 5.978494623655914,
+      "grad_norm": 1.0063740015029907,
+      "learning_rate": 0.0002,
+      "loss": 0.423,
+      "step": 2780
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 0.8291367292404175,
+      "learning_rate": 0.0002,
+      "loss": 0.4382,
+      "step": 2790
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 0.5057176947593689,
+      "eval_runtime": 21.3206,
+      "eval_samples_per_second": 15.525,
+      "eval_steps_per_second": 1.97,
+      "step": 2790
+    },
+    {
+      "epoch": 6.021505376344086,
+      "grad_norm": 1.0137434005737305,
+      "learning_rate": 0.0002,
+      "loss": 0.3907,
+      "step": 2800
+    },
+    {
+      "epoch": 6.043010752688172,
+      "grad_norm": 0.7550579905509949,
+      "learning_rate": 0.0002,
+      "loss": 0.3793,
+      "step": 2810
+    },
+    {
+      "epoch": 6.064516129032258,
+      "grad_norm": 1.0664116144180298,
+      "learning_rate": 0.0002,
+      "loss": 0.4003,
+      "step": 2820
+    },
+    {
+      "epoch": 6.086021505376344,
+      "grad_norm": 0.7908814549446106,
+      "learning_rate": 0.0002,
+      "loss": 0.3876,
+      "step": 2830
+    },
+    {
+      "epoch": 6.10752688172043,
+      "grad_norm": 0.8101639747619629,
+      "learning_rate": 0.0002,
+      "loss": 0.3884,
+      "step": 2840
+    },
+    {
+      "epoch": 6.129032258064516,
+      "grad_norm": 0.7882567048072815,
+      "learning_rate": 0.0002,
+      "loss": 0.3835,
+      "step": 2850
+    },
+    {
+      "epoch": 6.150537634408602,
+      "grad_norm": 1.0134103298187256,
+      "learning_rate": 0.0002,
+      "loss": 0.3827,
+      "step": 2860
+    },
+    {
+      "epoch": 6.172043010752688,
+      "grad_norm": 0.9240215420722961,
+      "learning_rate": 0.0002,
+      "loss": 0.3963,
+      "step": 2870
+    },
+    {
+      "epoch": 6.193548387096774,
+      "grad_norm": 0.8322992920875549,
+      "learning_rate": 0.0002,
+      "loss": 0.4049,
+      "step": 2880
+    },
+    {
+      "epoch": 6.21505376344086,
+      "grad_norm": 0.9238720536231995,
+      "learning_rate": 0.0002,
+      "loss": 0.381,
+      "step": 2890
+    },
+    {
+      "epoch": 6.236559139784946,
+      "grad_norm": 0.9361863732337952,
+      "learning_rate": 0.0002,
+      "loss": 0.3852,
+      "step": 2900
+    },
+    {
+      "epoch": 6.258064516129032,
+      "grad_norm": 0.9670863747596741,
+      "learning_rate": 0.0002,
+      "loss": 0.3917,
+      "step": 2910
+    },
+    {
+      "epoch": 6.279569892473118,
+      "grad_norm": 0.7724685668945312,
+      "learning_rate": 0.0002,
+      "loss": 0.3826,
+      "step": 2920
+    },
+    {
+      "epoch": 6.301075268817204,
+      "grad_norm": 0.8125540614128113,
+      "learning_rate": 0.0002,
+      "loss": 0.3988,
+      "step": 2930
+    },
+    {
+      "epoch": 6.32258064516129,
+      "grad_norm": 0.9483002424240112,
+      "learning_rate": 0.0002,
+      "loss": 0.3778,
+      "step": 2940
+    },
+    {
+      "epoch": 6.344086021505376,
+      "grad_norm": 1.098374843597412,
+      "learning_rate": 0.0002,
+      "loss": 0.3823,
+      "step": 2950
+    },
+    {
+      "epoch": 6.365591397849462,
+      "grad_norm": 1.0169378519058228,
+      "learning_rate": 0.0002,
+      "loss": 0.3886,
+      "step": 2960
+    },
+    {
+      "epoch": 6.387096774193548,
+      "grad_norm": 0.8594151139259338,
+      "learning_rate": 0.0002,
+      "loss": 0.3936,
+      "step": 2970
+    },
+    {
+      "epoch": 6.408602150537634,
+      "grad_norm": 0.9507288336753845,
+      "learning_rate": 0.0002,
+      "loss": 0.3871,
+      "step": 2980
+    },
+    {
+      "epoch": 6.43010752688172,
+      "grad_norm": 0.9212459325790405,
+      "learning_rate": 0.0002,
+      "loss": 0.3852,
+      "step": 2990
+    },
+    {
+      "epoch": 6.451612903225806,
+      "grad_norm": 0.9696952104568481,
+      "learning_rate": 0.0002,
+      "loss": 0.3929,
+      "step": 3000
+    },
+    {
+      "epoch": 6.473118279569892,
+      "grad_norm": 0.8872610330581665,
+      "learning_rate": 0.0002,
+      "loss": 0.3933,
+      "step": 3010
+    },
+    {
+      "epoch": 6.494623655913978,
+      "grad_norm": 0.9207532405853271,
+      "learning_rate": 0.0002,
+      "loss": 0.393,
+      "step": 3020
+    },
+    {
+      "epoch": 6.516129032258064,
+      "grad_norm": 0.9116262793540955,
+      "learning_rate": 0.0002,
+      "loss": 0.3848,
+      "step": 3030
+    },
+    {
+      "epoch": 6.53763440860215,
+      "grad_norm": 0.83391934633255,
+      "learning_rate": 0.0002,
+      "loss": 0.3964,
+      "step": 3040
+    },
+    {
+      "epoch": 6.559139784946236,
+      "grad_norm": 0.890931248664856,
+      "learning_rate": 0.0002,
+      "loss": 0.3758,
+      "step": 3050
+    },
+    {
+      "epoch": 6.580645161290323,
+      "grad_norm": 1.0100581645965576,
+      "learning_rate": 0.0002,
+      "loss": 0.3944,
+      "step": 3060
+    },
+    {
+      "epoch": 6.602150537634409,
+      "grad_norm": 0.783526599407196,
+      "learning_rate": 0.0002,
+      "loss": 0.3992,
+      "step": 3070
+    },
+    {
+      "epoch": 6.623655913978495,
+      "grad_norm": 1.324326515197754,
+      "learning_rate": 0.0002,
+      "loss": 0.4144,
+      "step": 3080
+    },
+    {
+      "epoch": 6.645161290322581,
+      "grad_norm": 0.9102319478988647,
+      "learning_rate": 0.0002,
+      "loss": 0.3986,
+      "step": 3090
+    },
+    {
+      "epoch": 6.666666666666667,
+      "grad_norm": 0.96951824426651,
+      "learning_rate": 0.0002,
+      "loss": 0.3873,
+      "step": 3100
+    },
+    {
+      "epoch": 6.688172043010753,
+      "grad_norm": 0.9786809086799622,
+      "learning_rate": 0.0002,
+      "loss": 0.3931,
+      "step": 3110
+    },
+    {
+      "epoch": 6.709677419354839,
+      "grad_norm": 1.0301238298416138,
+      "learning_rate": 0.0002,
+      "loss": 0.3714,
+      "step": 3120
+    },
+    {
+      "epoch": 6.731182795698925,
+      "grad_norm": 1.1690906286239624,
+      "learning_rate": 0.0002,
+      "loss": 0.3823,
+      "step": 3130
+    },
+    {
+      "epoch": 6.752688172043011,
+      "grad_norm": 0.963306725025177,
+      "learning_rate": 0.0002,
+      "loss": 0.3936,
+      "step": 3140
+    },
+    {
+      "epoch": 6.774193548387097,
+      "grad_norm": 0.8565770983695984,
+      "learning_rate": 0.0002,
+      "loss": 0.3975,
+      "step": 3150
+    },
+    {
+      "epoch": 6.795698924731183,
+      "grad_norm": 0.8887158632278442,
+      "learning_rate": 0.0002,
+      "loss": 0.3903,
+      "step": 3160
+    },
+    {
+      "epoch": 6.817204301075269,
+      "grad_norm": 0.8234561085700989,
+      "learning_rate": 0.0002,
+      "loss": 0.4098,
+      "step": 3170
+    },
+    {
+      "epoch": 6.838709677419355,
+      "grad_norm": 0.9000219702720642,
+      "learning_rate": 0.0002,
+      "loss": 0.4041,
+      "step": 3180
+    },
+    {
+      "epoch": 6.860215053763441,
+      "grad_norm": 1.1366009712219238,
+      "learning_rate": 0.0002,
+      "loss": 0.3933,
+      "step": 3190
+    },
+    {
+      "epoch": 6.881720430107527,
+      "grad_norm": 0.8747097849845886,
+      "learning_rate": 0.0002,
+      "loss": 0.3972,
+      "step": 3200
+    },
+    {
+      "epoch": 6.903225806451613,
+      "grad_norm": 0.8533893823623657,
+      "learning_rate": 0.0002,
+      "loss": 0.404,
+      "step": 3210
+    },
+    {
+      "epoch": 6.924731182795699,
+      "grad_norm": 0.8127949237823486,
+      "learning_rate": 0.0002,
+      "loss": 0.3906,
+      "step": 3220
+    },
+    {
+      "epoch": 6.946236559139785,
+      "grad_norm": 0.8872477412223816,
+      "learning_rate": 0.0002,
+      "loss": 0.3747,
+      "step": 3230
+    },
+    {
+      "epoch": 6.967741935483871,
+      "grad_norm": 0.8541608452796936,
+      "learning_rate": 0.0002,
+      "loss": 0.3817,
+      "step": 3240
+    },
+    {
+      "epoch": 6.989247311827957,
+      "grad_norm": 0.8390752673149109,
+      "learning_rate": 0.0002,
+      "loss": 0.3863,
+      "step": 3250
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 0.48264747858047485,
+      "eval_runtime": 21.3942,
+      "eval_samples_per_second": 15.471,
+      "eval_steps_per_second": 1.963,
+      "step": 3255
+    },
+    {
+      "epoch": 7.010752688172043,
+      "grad_norm": 1.0476834774017334,
+      "learning_rate": 0.0002,
+      "loss": 0.391,
+      "step": 3260
+    },
+    {
+      "epoch": 7.032258064516129,
+      "grad_norm": 0.7501131296157837,
+      "learning_rate": 0.0002,
+      "loss": 0.3422,
+      "step": 3270
+    },
+    {
+      "epoch": 7.053763440860215,
+      "grad_norm": 0.9057435393333435,
+      "learning_rate": 0.0002,
+      "loss": 0.3542,
+      "step": 3280
+    },
+    {
+      "epoch": 7.075268817204301,
+      "grad_norm": 0.7058833241462708,
+      "learning_rate": 0.0002,
+      "loss": 0.3522,
+      "step": 3290
+    },
+    {
+      "epoch": 7.096774193548387,
+      "grad_norm": 0.9908691644668579,
+      "learning_rate": 0.0002,
+      "loss": 0.3575,
+      "step": 3300
+    },
+    {
+      "epoch": 7.118279569892473,
+      "grad_norm": 0.9515542984008789,
+      "learning_rate": 0.0002,
+      "loss": 0.3755,
+      "step": 3310
+    },
+    {
+      "epoch": 7.139784946236559,
+      "grad_norm": 0.997296154499054,
+      "learning_rate": 0.0002,
+      "loss": 0.3612,
+      "step": 3320
+    },
+    {
+      "epoch": 7.161290322580645,
+      "grad_norm": 0.9810499548912048,
+      "learning_rate": 0.0002,
+      "loss": 0.3616,
+      "step": 3330
+    },
+    {
+      "epoch": 7.182795698924731,
+      "grad_norm": 0.8133336901664734,
+      "learning_rate": 0.0002,
+      "loss": 0.3584,
+      "step": 3340
+    },
+    {
+      "epoch": 7.204301075268817,
+      "grad_norm": 1.0679855346679688,
+      "learning_rate": 0.0002,
+      "loss": 0.3644,
+      "step": 3350
+    },
+    {
+      "epoch": 7.225806451612903,
+      "grad_norm": 0.7656611204147339,
+      "learning_rate": 0.0002,
+      "loss": 0.358,
+      "step": 3360
+    },
+    {
+      "epoch": 7.247311827956989,
+      "grad_norm": 0.9478468298912048,
+      "learning_rate": 0.0002,
+      "loss": 0.3616,
+      "step": 3370
+    },
+    {
+      "epoch": 7.268817204301075,
+      "grad_norm": 0.8425832986831665,
+      "learning_rate": 0.0002,
+      "loss": 0.3631,
+      "step": 3380
+    },
+    {
+      "epoch": 7.290322580645161,
+      "grad_norm": 0.9573627710342407,
+      "learning_rate": 0.0002,
+      "loss": 0.3735,
+      "step": 3390
+    },
+    {
+      "epoch": 7.311827956989247,
+      "grad_norm": 0.9219972491264343,
+      "learning_rate": 0.0002,
+      "loss": 0.359,
+      "step": 3400
+    },
+    {
+      "epoch": 7.333333333333333,
+      "grad_norm": 0.876099705696106,
+      "learning_rate": 0.0002,
+      "loss": 0.3644,
+      "step": 3410
+    },
+    {
+      "epoch": 7.354838709677419,
+      "grad_norm": 1.0051969289779663,
+      "learning_rate": 0.0002,
+      "loss": 0.3747,
+      "step": 3420
+    },
+    {
+      "epoch": 7.376344086021505,
+      "grad_norm": 1.1347692012786865,
+      "learning_rate": 0.0002,
+      "loss": 0.3527,
+      "step": 3430
+    },
+    {
+      "epoch": 7.397849462365591,
+      "grad_norm": 0.9641520380973816,
+      "learning_rate": 0.0002,
+      "loss": 0.3644,
+      "step": 3440
+    },
+    {
+      "epoch": 7.419354838709677,
+      "grad_norm": 0.7777793407440186,
+      "learning_rate": 0.0002,
+      "loss": 0.3486,
+      "step": 3450
+    },
+    {
+      "epoch": 7.440860215053763,
+      "grad_norm": 0.9649308323860168,
+      "learning_rate": 0.0002,
+      "loss": 0.3593,
+      "step": 3460
+    },
+    {
+      "epoch": 7.462365591397849,
+      "grad_norm": 0.9245585203170776,
+      "learning_rate": 0.0002,
+      "loss": 0.3754,
+      "step": 3470
+    },
+    {
+      "epoch": 7.483870967741936,
+      "grad_norm": 0.8298666477203369,
+      "learning_rate": 0.0002,
+      "loss": 0.3732,
+      "step": 3480
+    },
+    {
+      "epoch": 7.505376344086022,
+      "grad_norm": 1.1579877138137817,
+      "learning_rate": 0.0002,
+      "loss": 0.3585,
+      "step": 3490
+    },
+    {
+      "epoch": 7.526881720430108,
+      "grad_norm": 0.8718803524971008,
+      "learning_rate": 0.0002,
+      "loss": 0.3505,
+      "step": 3500
+    },
+    {
+      "epoch": 7.548387096774194,
+      "grad_norm": 0.7785154581069946,
+      "learning_rate": 0.0002,
+      "loss": 0.3725,
+      "step": 3510
+    },
+    {
+      "epoch": 7.56989247311828,
+      "grad_norm": 0.753657877445221,
+      "learning_rate": 0.0002,
+      "loss": 0.3507,
+      "step": 3520
+    },
+    {
+      "epoch": 7.591397849462366,
+      "grad_norm": 0.834524929523468,
+      "learning_rate": 0.0002,
+      "loss": 0.3665,
+      "step": 3530
+    },
+    {
+      "epoch": 7.612903225806452,
+      "grad_norm": 0.9546446800231934,
+      "learning_rate": 0.0002,
+      "loss": 0.3686,
+      "step": 3540
+    },
+    {
+      "epoch": 7.634408602150538,
+      "grad_norm": 0.8275105357170105,
+      "learning_rate": 0.0002,
+      "loss": 0.3673,
+      "step": 3550
+    },
+    {
+      "epoch": 7.655913978494624,
+      "grad_norm": 0.9137991070747375,
+      "learning_rate": 0.0002,
+      "loss": 0.381,
+      "step": 3560
+    },
+    {
+      "epoch": 7.67741935483871,
+      "grad_norm": 0.993617057800293,
+      "learning_rate": 0.0002,
+      "loss": 0.3565,
+      "step": 3570
+    },
+    {
+      "epoch": 7.698924731182796,
+      "grad_norm": 1.0079665184020996,
+      "learning_rate": 0.0002,
+      "loss": 0.3701,
+      "step": 3580
+    },
+    {
+      "epoch": 7.720430107526882,
+      "grad_norm": 0.8295491337776184,
+      "learning_rate": 0.0002,
+      "loss": 0.3495,
+      "step": 3590
+    },
+    {
+      "epoch": 7.741935483870968,
+      "grad_norm": 0.814578115940094,
+      "learning_rate": 0.0002,
+      "loss": 0.374,
+      "step": 3600
+    },
+    {
+      "epoch": 7.763440860215054,
+      "grad_norm": 0.8422811031341553,
+      "learning_rate": 0.0002,
+      "loss": 0.3673,
+      "step": 3610
+    },
+    {
+      "epoch": 7.78494623655914,
+      "grad_norm": 1.0220918655395508,
+      "learning_rate": 0.0002,
+      "loss": 0.3452,
+      "step": 3620
+    },
+    {
+      "epoch": 7.806451612903226,
+      "grad_norm": 0.8065739870071411,
+      "learning_rate": 0.0002,
+      "loss": 0.3641,
+      "step": 3630
+    },
+    {
+      "epoch": 7.827956989247312,
+      "grad_norm": 0.8039169907569885,
+      "learning_rate": 0.0002,
+      "loss": 0.3699,
+      "step": 3640
+    },
+    {
+      "epoch": 7.849462365591398,
+      "grad_norm": 1.0766745805740356,
+      "learning_rate": 0.0002,
+      "loss": 0.3635,
+      "step": 3650
+    },
+    {
+      "epoch": 7.870967741935484,
+      "grad_norm": 1.0806103944778442,
+      "learning_rate": 0.0002,
+      "loss": 0.3633,
+      "step": 3660
+    },
+    {
+      "epoch": 7.89247311827957,
+      "grad_norm": 1.1005314588546753,
+      "learning_rate": 0.0002,
+      "loss": 0.3678,
+      "step": 3670
+    },
+    {
+      "epoch": 7.913978494623656,
+      "grad_norm": 0.9276911616325378,
+      "learning_rate": 0.0002,
+      "loss": 0.3583,
+      "step": 3680
+    },
+    {
+      "epoch": 7.935483870967742,
+      "grad_norm": 0.9914153814315796,
+      "learning_rate": 0.0002,
+      "loss": 0.374,
+      "step": 3690
+    },
+    {
+      "epoch": 7.956989247311828,
+      "grad_norm": 0.8128159046173096,
+      "learning_rate": 0.0002,
+      "loss": 0.3575,
+      "step": 3700
+    },
+    {
+      "epoch": 7.978494623655914,
+      "grad_norm": 0.8122950196266174,
+      "learning_rate": 0.0002,
+      "loss": 0.3491,
+      "step": 3710
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 1.0291857719421387,
+      "learning_rate": 0.0002,
+      "loss": 0.3723,
+      "step": 3720
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 0.46552130579948425,
+      "eval_runtime": 21.3919,
+      "eval_samples_per_second": 15.473,
+      "eval_steps_per_second": 1.963,
+      "step": 3720
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 3720,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4.546582636658688e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/training_args.bin b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..483b53d4fc1c568a5fc890fa850e3450f390b208
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3720/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e4c4fe2be590ed03492316230adb3a1edca3e4066c55f3716c0352d7134c564
+size 5560
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/README.md b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..830a14f7db2734beb59f320973504e45a3fe87f5
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/adapter_config.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e99bbcd43df1c19d98706c7e3be95c93844c5349
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/adapter_model.safetensors b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..6bbb58188844fc5715b13662e683c9bb2d19ebcc
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b25038edeef7fb2f5419a519ef6f7cecf0131ccd565c4322f9eda6c024a7da2
+size 29500848
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/optimizer.pt b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..924994f56601d458aa90608b55379cefe79b5dea
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b577877b637e44e84199f49726c667f72ebecd540dd62885a1b3388352ff61a
+size 15064314
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/rng_state.pth b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..bd9091c61ace4ca0a36fce7f975e446d2c6d7189
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab337f1f93edeb8702cf990e25b6fb77257b92ff4152411237615ef14d6f0160
+size 14244
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/scheduler.pt b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..6d7f87ba141fde557bb0d944880e0de0a4939c4c
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7e1590542588edd3aa2e2d5dc727769ef980fc339b8db6475867c3b5914d8e1e
+size 1064
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/special_tokens_map.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/tokenizer.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..f58963a682665634ab180c28667e4faa8cf02ba2
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/tokenizer.model b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/tokenizer_config.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1adb4796c13b8d975555ecec45876ee75d1ae8b7
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/trainer_state.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..04ee84a8a42117f9bddadfe1e3751ec5e6219b59
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/trainer_state.json
@@ -0,0 +1,363 @@
+{
+  "best_metric": 0.7740864157676697,
+  "best_model_checkpoint": "outputs-001/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465",
+  "epoch": 1.0,
+  "eval_steps": 10,
+  "global_step": 465,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.021505376344086023,
+      "grad_norm": 0.9075053930282593,
+      "learning_rate": 0.0002,
+      "loss": 3.4172,
+      "step": 10
+    },
+    {
+      "epoch": 0.043010752688172046,
+      "grad_norm": 1.4321208000183105,
+      "learning_rate": 0.0002,
+      "loss": 2.5888,
+      "step": 20
+    },
+    {
+      "epoch": 0.06451612903225806,
+      "grad_norm": 1.7500602006912231,
+      "learning_rate": 0.0002,
+      "loss": 2.1195,
+      "step": 30
+    },
+    {
+      "epoch": 0.08602150537634409,
+      "grad_norm": 0.7606641054153442,
+      "learning_rate": 0.0002,
+      "loss": 1.9303,
+      "step": 40
+    },
+    {
+      "epoch": 0.10752688172043011,
+      "grad_norm": 1.2754929065704346,
+      "learning_rate": 0.0002,
+      "loss": 1.6112,
+      "step": 50
+    },
+    {
+      "epoch": 0.12903225806451613,
+      "grad_norm": 1.0936230421066284,
+      "learning_rate": 0.0002,
+      "loss": 1.4319,
+      "step": 60
+    },
+    {
+      "epoch": 0.15053763440860216,
+      "grad_norm": 1.144593596458435,
+      "learning_rate": 0.0002,
+      "loss": 1.3568,
+      "step": 70
+    },
+    {
+      "epoch": 0.17204301075268819,
+      "grad_norm": 1.2181956768035889,
+      "learning_rate": 0.0002,
+      "loss": 1.2028,
+      "step": 80
+    },
+    {
+      "epoch": 0.1935483870967742,
+      "grad_norm": 1.1260095834732056,
+      "learning_rate": 0.0002,
+      "loss": 1.1534,
+      "step": 90
+    },
+    {
+      "epoch": 0.21505376344086022,
+      "grad_norm": 1.1155284643173218,
+      "learning_rate": 0.0002,
+      "loss": 1.1089,
+      "step": 100
+    },
+    {
+      "epoch": 0.23655913978494625,
+      "grad_norm": 1.089565396308899,
+      "learning_rate": 0.0002,
+      "loss": 1.0883,
+      "step": 110
+    },
+    {
+      "epoch": 0.25806451612903225,
+      "grad_norm": 0.9833471775054932,
+      "learning_rate": 0.0002,
+      "loss": 1.0814,
+      "step": 120
+    },
+    {
+      "epoch": 0.27956989247311825,
+      "grad_norm": 1.0265629291534424,
+      "learning_rate": 0.0002,
+      "loss": 1.0239,
+      "step": 130
+    },
+    {
+      "epoch": 0.3010752688172043,
+      "grad_norm": 0.9344286322593689,
+      "learning_rate": 0.0002,
+      "loss": 0.9888,
+      "step": 140
+    },
+    {
+      "epoch": 0.3225806451612903,
+      "grad_norm": 0.9883386492729187,
+      "learning_rate": 0.0002,
+      "loss": 1.0043,
+      "step": 150
+    },
+    {
+      "epoch": 0.34408602150537637,
+      "grad_norm": 0.9299277067184448,
+      "learning_rate": 0.0002,
+      "loss": 0.9338,
+      "step": 160
+    },
+    {
+      "epoch": 0.3655913978494624,
+      "grad_norm": 1.390045404434204,
+      "learning_rate": 0.0002,
+      "loss": 0.9432,
+      "step": 170
+    },
+    {
+      "epoch": 0.3870967741935484,
+      "grad_norm": 1.0313078165054321,
+      "learning_rate": 0.0002,
+      "loss": 0.9008,
+      "step": 180
+    },
+    {
+      "epoch": 0.40860215053763443,
+      "grad_norm": 1.1792205572128296,
+      "learning_rate": 0.0002,
+      "loss": 0.9434,
+      "step": 190
+    },
+    {
+      "epoch": 0.43010752688172044,
+      "grad_norm": 1.049809217453003,
+      "learning_rate": 0.0002,
+      "loss": 0.8761,
+      "step": 200
+    },
+    {
+      "epoch": 0.45161290322580644,
+      "grad_norm": 0.990111768245697,
+      "learning_rate": 0.0002,
+      "loss": 0.8709,
+      "step": 210
+    },
+    {
+      "epoch": 0.4731182795698925,
+      "grad_norm": 0.9870412349700928,
+      "learning_rate": 0.0002,
+      "loss": 0.905,
+      "step": 220
+    },
+    {
+      "epoch": 0.4946236559139785,
+      "grad_norm": 0.8557345867156982,
+      "learning_rate": 0.0002,
+      "loss": 0.9129,
+      "step": 230
+    },
+    {
+      "epoch": 0.5161290322580645,
+      "grad_norm": 0.9746861457824707,
+      "learning_rate": 0.0002,
+      "loss": 0.8836,
+      "step": 240
+    },
+    {
+      "epoch": 0.5376344086021505,
+      "grad_norm": 0.9010438323020935,
+      "learning_rate": 0.0002,
+      "loss": 0.873,
+      "step": 250
+    },
+    {
+      "epoch": 0.5591397849462365,
+      "grad_norm": 0.9061082005500793,
+      "learning_rate": 0.0002,
+      "loss": 0.8241,
+      "step": 260
+    },
+    {
+      "epoch": 0.5806451612903226,
+      "grad_norm": 0.9311846494674683,
+      "learning_rate": 0.0002,
+      "loss": 0.8652,
+      "step": 270
+    },
+    {
+      "epoch": 0.6021505376344086,
+      "grad_norm": 0.9140254855155945,
+      "learning_rate": 0.0002,
+      "loss": 0.8256,
+      "step": 280
+    },
+    {
+      "epoch": 0.6236559139784946,
+      "grad_norm": 0.9722253084182739,
+      "learning_rate": 0.0002,
+      "loss": 0.8441,
+      "step": 290
+    },
+    {
+      "epoch": 0.6451612903225806,
+      "grad_norm": 0.8539168238639832,
+      "learning_rate": 0.0002,
+      "loss": 0.8314,
+      "step": 300
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.9053162932395935,
+      "learning_rate": 0.0002,
+      "loss": 0.8528,
+      "step": 310
+    },
+    {
+      "epoch": 0.6881720430107527,
+      "grad_norm": 0.8444252610206604,
+      "learning_rate": 0.0002,
+      "loss": 0.8209,
+      "step": 320
+    },
+    {
+      "epoch": 0.7096774193548387,
+      "grad_norm": 0.8127437829971313,
+      "learning_rate": 0.0002,
+      "loss": 0.8101,
+      "step": 330
+    },
+    {
+      "epoch": 0.7311827956989247,
+      "grad_norm": 0.886555016040802,
+      "learning_rate": 0.0002,
+      "loss": 0.8223,
+      "step": 340
+    },
+    {
+      "epoch": 0.7526881720430108,
+      "grad_norm": 0.8458548784255981,
+      "learning_rate": 0.0002,
+      "loss": 0.8368,
+      "step": 350
+    },
+    {
+      "epoch": 0.7741935483870968,
+      "grad_norm": 0.8683297634124756,
+      "learning_rate": 0.0002,
+      "loss": 0.8295,
+      "step": 360
+    },
+    {
+      "epoch": 0.7956989247311828,
+      "grad_norm": 0.8308405876159668,
+      "learning_rate": 0.0002,
+      "loss": 0.8232,
+      "step": 370
+    },
+    {
+      "epoch": 0.8172043010752689,
+      "grad_norm": 0.8305579423904419,
+      "learning_rate": 0.0002,
+      "loss": 0.7752,
+      "step": 380
+    },
+    {
+      "epoch": 0.8387096774193549,
+      "grad_norm": 0.8545567393302917,
+      "learning_rate": 0.0002,
+      "loss": 0.8267,
+      "step": 390
+    },
+    {
+      "epoch": 0.8602150537634409,
+      "grad_norm": 0.8486055731773376,
+      "learning_rate": 0.0002,
+      "loss": 0.8212,
+      "step": 400
+    },
+    {
+      "epoch": 0.8817204301075269,
+      "grad_norm": 0.8126763105392456,
+      "learning_rate": 0.0002,
+      "loss": 0.743,
+      "step": 410
+    },
+    {
+      "epoch": 0.9032258064516129,
+      "grad_norm": 0.8494045734405518,
+      "learning_rate": 0.0002,
+      "loss": 0.7993,
+      "step": 420
+    },
+    {
+      "epoch": 0.9247311827956989,
+      "grad_norm": 0.7639183402061462,
+      "learning_rate": 0.0002,
+      "loss": 0.8213,
+      "step": 430
+    },
+    {
+      "epoch": 0.946236559139785,
+      "grad_norm": 0.858101487159729,
+      "learning_rate": 0.0002,
+      "loss": 0.8015,
+      "step": 440
+    },
+    {
+      "epoch": 0.967741935483871,
+      "grad_norm": 0.8141381740570068,
+      "learning_rate": 0.0002,
+      "loss": 0.7629,
+      "step": 450
+    },
+    {
+      "epoch": 0.989247311827957,
+      "grad_norm": 0.8072513937950134,
+      "learning_rate": 0.0002,
+      "loss": 0.7357,
+      "step": 460
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.7740864157676697,
+      "eval_runtime": 21.383,
+      "eval_samples_per_second": 15.48,
+      "eval_steps_per_second": 1.964,
+      "step": 465
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 3720,
+  "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": 5683228295823360.0,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/training_args.bin b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..483b53d4fc1c568a5fc890fa850e3450f390b208
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e4c4fe2be590ed03492316230adb3a1edca3e4066c55f3716c0352d7134c564
+size 5560
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/README.md b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..830a14f7db2734beb59f320973504e45a3fe87f5
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/adapter_config.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e99bbcd43df1c19d98706c7e3be95c93844c5349
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/adapter_model.safetensors b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..f940190d9795c50ad873dfd121e1a9ec33b167d5
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a08a9f9d94c71b4ae51ac490bf8084de74a3a512d3c57dbc41c48341ec460d59
+size 29500848
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/optimizer.pt b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..54a4f305b9b2db64f4657e04730cd6b3560d51f1
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dcdb8d97813ef09ab3f62cfcb1869ccc4eb3053b6d6147c01042dc6eb164df8b
+size 15064314
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/rng_state.pth b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..90073eaaf42985bbbdf288ae3c79ab3b7a859049
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d6cdf4123075a81389fb31c181195d1d3f878c02aba83d2601e972fdaf99df90
+size 14244
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/scheduler.pt b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..1ea512ee003a4a7388698db6b6ab095e5022a950
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cf9812a2e2f6e5ccc013a71db2228ca4973e2e4618013a6759f1a4c9ad6ca514
+size 1064
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/special_tokens_map.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/tokenizer.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..f58963a682665634ab180c28667e4faa8cf02ba2
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/tokenizer.model b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/tokenizer_config.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1adb4796c13b8d975555ecec45876ee75d1ae8b7
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/trainer_state.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..2024a1438f8d2f911191c5967d08712762fb118f
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/trainer_state.json
@@ -0,0 +1,700 @@
+{
+  "best_metric": 0.6885261535644531,
+  "best_model_checkpoint": "outputs-001/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930",
+  "epoch": 2.0,
+  "eval_steps": 10,
+  "global_step": 930,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.021505376344086023,
+      "grad_norm": 0.9075053930282593,
+      "learning_rate": 0.0002,
+      "loss": 3.4172,
+      "step": 10
+    },
+    {
+      "epoch": 0.043010752688172046,
+      "grad_norm": 1.4321208000183105,
+      "learning_rate": 0.0002,
+      "loss": 2.5888,
+      "step": 20
+    },
+    {
+      "epoch": 0.06451612903225806,
+      "grad_norm": 1.7500602006912231,
+      "learning_rate": 0.0002,
+      "loss": 2.1195,
+      "step": 30
+    },
+    {
+      "epoch": 0.08602150537634409,
+      "grad_norm": 0.7606641054153442,
+      "learning_rate": 0.0002,
+      "loss": 1.9303,
+      "step": 40
+    },
+    {
+      "epoch": 0.10752688172043011,
+      "grad_norm": 1.2754929065704346,
+      "learning_rate": 0.0002,
+      "loss": 1.6112,
+      "step": 50
+    },
+    {
+      "epoch": 0.12903225806451613,
+      "grad_norm": 1.0936230421066284,
+      "learning_rate": 0.0002,
+      "loss": 1.4319,
+      "step": 60
+    },
+    {
+      "epoch": 0.15053763440860216,
+      "grad_norm": 1.144593596458435,
+      "learning_rate": 0.0002,
+      "loss": 1.3568,
+      "step": 70
+    },
+    {
+      "epoch": 0.17204301075268819,
+      "grad_norm": 1.2181956768035889,
+      "learning_rate": 0.0002,
+      "loss": 1.2028,
+      "step": 80
+    },
+    {
+      "epoch": 0.1935483870967742,
+      "grad_norm": 1.1260095834732056,
+      "learning_rate": 0.0002,
+      "loss": 1.1534,
+      "step": 90
+    },
+    {
+      "epoch": 0.21505376344086022,
+      "grad_norm": 1.1155284643173218,
+      "learning_rate": 0.0002,
+      "loss": 1.1089,
+      "step": 100
+    },
+    {
+      "epoch": 0.23655913978494625,
+      "grad_norm": 1.089565396308899,
+      "learning_rate": 0.0002,
+      "loss": 1.0883,
+      "step": 110
+    },
+    {
+      "epoch": 0.25806451612903225,
+      "grad_norm": 0.9833471775054932,
+      "learning_rate": 0.0002,
+      "loss": 1.0814,
+      "step": 120
+    },
+    {
+      "epoch": 0.27956989247311825,
+      "grad_norm": 1.0265629291534424,
+      "learning_rate": 0.0002,
+      "loss": 1.0239,
+      "step": 130
+    },
+    {
+      "epoch": 0.3010752688172043,
+      "grad_norm": 0.9344286322593689,
+      "learning_rate": 0.0002,
+      "loss": 0.9888,
+      "step": 140
+    },
+    {
+      "epoch": 0.3225806451612903,
+      "grad_norm": 0.9883386492729187,
+      "learning_rate": 0.0002,
+      "loss": 1.0043,
+      "step": 150
+    },
+    {
+      "epoch": 0.34408602150537637,
+      "grad_norm": 0.9299277067184448,
+      "learning_rate": 0.0002,
+      "loss": 0.9338,
+      "step": 160
+    },
+    {
+      "epoch": 0.3655913978494624,
+      "grad_norm": 1.390045404434204,
+      "learning_rate": 0.0002,
+      "loss": 0.9432,
+      "step": 170
+    },
+    {
+      "epoch": 0.3870967741935484,
+      "grad_norm": 1.0313078165054321,
+      "learning_rate": 0.0002,
+      "loss": 0.9008,
+      "step": 180
+    },
+    {
+      "epoch": 0.40860215053763443,
+      "grad_norm": 1.1792205572128296,
+      "learning_rate": 0.0002,
+      "loss": 0.9434,
+      "step": 190
+    },
+    {
+      "epoch": 0.43010752688172044,
+      "grad_norm": 1.049809217453003,
+      "learning_rate": 0.0002,
+      "loss": 0.8761,
+      "step": 200
+    },
+    {
+      "epoch": 0.45161290322580644,
+      "grad_norm": 0.990111768245697,
+      "learning_rate": 0.0002,
+      "loss": 0.8709,
+      "step": 210
+    },
+    {
+      "epoch": 0.4731182795698925,
+      "grad_norm": 0.9870412349700928,
+      "learning_rate": 0.0002,
+      "loss": 0.905,
+      "step": 220
+    },
+    {
+      "epoch": 0.4946236559139785,
+      "grad_norm": 0.8557345867156982,
+      "learning_rate": 0.0002,
+      "loss": 0.9129,
+      "step": 230
+    },
+    {
+      "epoch": 0.5161290322580645,
+      "grad_norm": 0.9746861457824707,
+      "learning_rate": 0.0002,
+      "loss": 0.8836,
+      "step": 240
+    },
+    {
+      "epoch": 0.5376344086021505,
+      "grad_norm": 0.9010438323020935,
+      "learning_rate": 0.0002,
+      "loss": 0.873,
+      "step": 250
+    },
+    {
+      "epoch": 0.5591397849462365,
+      "grad_norm": 0.9061082005500793,
+      "learning_rate": 0.0002,
+      "loss": 0.8241,
+      "step": 260
+    },
+    {
+      "epoch": 0.5806451612903226,
+      "grad_norm": 0.9311846494674683,
+      "learning_rate": 0.0002,
+      "loss": 0.8652,
+      "step": 270
+    },
+    {
+      "epoch": 0.6021505376344086,
+      "grad_norm": 0.9140254855155945,
+      "learning_rate": 0.0002,
+      "loss": 0.8256,
+      "step": 280
+    },
+    {
+      "epoch": 0.6236559139784946,
+      "grad_norm": 0.9722253084182739,
+      "learning_rate": 0.0002,
+      "loss": 0.8441,
+      "step": 290
+    },
+    {
+      "epoch": 0.6451612903225806,
+      "grad_norm": 0.8539168238639832,
+      "learning_rate": 0.0002,
+      "loss": 0.8314,
+      "step": 300
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.9053162932395935,
+      "learning_rate": 0.0002,
+      "loss": 0.8528,
+      "step": 310
+    },
+    {
+      "epoch": 0.6881720430107527,
+      "grad_norm": 0.8444252610206604,
+      "learning_rate": 0.0002,
+      "loss": 0.8209,
+      "step": 320
+    },
+    {
+      "epoch": 0.7096774193548387,
+      "grad_norm": 0.8127437829971313,
+      "learning_rate": 0.0002,
+      "loss": 0.8101,
+      "step": 330
+    },
+    {
+      "epoch": 0.7311827956989247,
+      "grad_norm": 0.886555016040802,
+      "learning_rate": 0.0002,
+      "loss": 0.8223,
+      "step": 340
+    },
+    {
+      "epoch": 0.7526881720430108,
+      "grad_norm": 0.8458548784255981,
+      "learning_rate": 0.0002,
+      "loss": 0.8368,
+      "step": 350
+    },
+    {
+      "epoch": 0.7741935483870968,
+      "grad_norm": 0.8683297634124756,
+      "learning_rate": 0.0002,
+      "loss": 0.8295,
+      "step": 360
+    },
+    {
+      "epoch": 0.7956989247311828,
+      "grad_norm": 0.8308405876159668,
+      "learning_rate": 0.0002,
+      "loss": 0.8232,
+      "step": 370
+    },
+    {
+      "epoch": 0.8172043010752689,
+      "grad_norm": 0.8305579423904419,
+      "learning_rate": 0.0002,
+      "loss": 0.7752,
+      "step": 380
+    },
+    {
+      "epoch": 0.8387096774193549,
+      "grad_norm": 0.8545567393302917,
+      "learning_rate": 0.0002,
+      "loss": 0.8267,
+      "step": 390
+    },
+    {
+      "epoch": 0.8602150537634409,
+      "grad_norm": 0.8486055731773376,
+      "learning_rate": 0.0002,
+      "loss": 0.8212,
+      "step": 400
+    },
+    {
+      "epoch": 0.8817204301075269,
+      "grad_norm": 0.8126763105392456,
+      "learning_rate": 0.0002,
+      "loss": 0.743,
+      "step": 410
+    },
+    {
+      "epoch": 0.9032258064516129,
+      "grad_norm": 0.8494045734405518,
+      "learning_rate": 0.0002,
+      "loss": 0.7993,
+      "step": 420
+    },
+    {
+      "epoch": 0.9247311827956989,
+      "grad_norm": 0.7639183402061462,
+      "learning_rate": 0.0002,
+      "loss": 0.8213,
+      "step": 430
+    },
+    {
+      "epoch": 0.946236559139785,
+      "grad_norm": 0.858101487159729,
+      "learning_rate": 0.0002,
+      "loss": 0.8015,
+      "step": 440
+    },
+    {
+      "epoch": 0.967741935483871,
+      "grad_norm": 0.8141381740570068,
+      "learning_rate": 0.0002,
+      "loss": 0.7629,
+      "step": 450
+    },
+    {
+      "epoch": 0.989247311827957,
+      "grad_norm": 0.8072513937950134,
+      "learning_rate": 0.0002,
+      "loss": 0.7357,
+      "step": 460
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.7740864157676697,
+      "eval_runtime": 21.383,
+      "eval_samples_per_second": 15.48,
+      "eval_steps_per_second": 1.964,
+      "step": 465
+    },
+    {
+      "epoch": 1.010752688172043,
+      "grad_norm": 0.8269494771957397,
+      "learning_rate": 0.0002,
+      "loss": 0.7701,
+      "step": 470
+    },
+    {
+      "epoch": 1.032258064516129,
+      "grad_norm": 0.7814009189605713,
+      "learning_rate": 0.0002,
+      "loss": 0.7532,
+      "step": 480
+    },
+    {
+      "epoch": 1.053763440860215,
+      "grad_norm": 0.8183923363685608,
+      "learning_rate": 0.0002,
+      "loss": 0.7689,
+      "step": 490
+    },
+    {
+      "epoch": 1.075268817204301,
+      "grad_norm": 0.8146600723266602,
+      "learning_rate": 0.0002,
+      "loss": 0.765,
+      "step": 500
+    },
+    {
+      "epoch": 1.096774193548387,
+      "grad_norm": 0.8635126352310181,
+      "learning_rate": 0.0002,
+      "loss": 0.7358,
+      "step": 510
+    },
+    {
+      "epoch": 1.118279569892473,
+      "grad_norm": 0.8520359396934509,
+      "learning_rate": 0.0002,
+      "loss": 0.7302,
+      "step": 520
+    },
+    {
+      "epoch": 1.139784946236559,
+      "grad_norm": 0.8026443123817444,
+      "learning_rate": 0.0002,
+      "loss": 0.7492,
+      "step": 530
+    },
+    {
+      "epoch": 1.1612903225806452,
+      "grad_norm": 0.8157258629798889,
+      "learning_rate": 0.0002,
+      "loss": 0.7518,
+      "step": 540
+    },
+    {
+      "epoch": 1.1827956989247312,
+      "grad_norm": 0.9450796246528625,
+      "learning_rate": 0.0002,
+      "loss": 0.7461,
+      "step": 550
+    },
+    {
+      "epoch": 1.2043010752688172,
+      "grad_norm": 0.8859835863113403,
+      "learning_rate": 0.0002,
+      "loss": 0.7128,
+      "step": 560
+    },
+    {
+      "epoch": 1.2258064516129032,
+      "grad_norm": 0.7819921970367432,
+      "learning_rate": 0.0002,
+      "loss": 0.7067,
+      "step": 570
+    },
+    {
+      "epoch": 1.2473118279569892,
+      "grad_norm": 0.7823445796966553,
+      "learning_rate": 0.0002,
+      "loss": 0.7577,
+      "step": 580
+    },
+    {
+      "epoch": 1.2688172043010753,
+      "grad_norm": 0.7931883931159973,
+      "learning_rate": 0.0002,
+      "loss": 0.7358,
+      "step": 590
+    },
+    {
+      "epoch": 1.2903225806451613,
+      "grad_norm": 0.7495734095573425,
+      "learning_rate": 0.0002,
+      "loss": 0.723,
+      "step": 600
+    },
+    {
+      "epoch": 1.3118279569892473,
+      "grad_norm": 0.9272717237472534,
+      "learning_rate": 0.0002,
+      "loss": 0.7386,
+      "step": 610
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 0.7968398332595825,
+      "learning_rate": 0.0002,
+      "loss": 0.7498,
+      "step": 620
+    },
+    {
+      "epoch": 1.3548387096774195,
+      "grad_norm": 0.7813659310340881,
+      "learning_rate": 0.0002,
+      "loss": 0.7635,
+      "step": 630
+    },
+    {
+      "epoch": 1.3763440860215055,
+      "grad_norm": 0.730925977230072,
+      "learning_rate": 0.0002,
+      "loss": 0.6665,
+      "step": 640
+    },
+    {
+      "epoch": 1.3978494623655915,
+      "grad_norm": 0.8011482954025269,
+      "learning_rate": 0.0002,
+      "loss": 0.7037,
+      "step": 650
+    },
+    {
+      "epoch": 1.4193548387096775,
+      "grad_norm": 0.7770085334777832,
+      "learning_rate": 0.0002,
+      "loss": 0.6931,
+      "step": 660
+    },
+    {
+      "epoch": 1.4408602150537635,
+      "grad_norm": 0.7432682514190674,
+      "learning_rate": 0.0002,
+      "loss": 0.6949,
+      "step": 670
+    },
+    {
+      "epoch": 1.4623655913978495,
+      "grad_norm": 0.8820092678070068,
+      "learning_rate": 0.0002,
+      "loss": 0.7444,
+      "step": 680
+    },
+    {
+      "epoch": 1.4838709677419355,
+      "grad_norm": 0.7786208987236023,
+      "learning_rate": 0.0002,
+      "loss": 0.6758,
+      "step": 690
+    },
+    {
+      "epoch": 1.5053763440860215,
+      "grad_norm": 0.7467480301856995,
+      "learning_rate": 0.0002,
+      "loss": 0.6702,
+      "step": 700
+    },
+    {
+      "epoch": 1.5268817204301075,
+      "grad_norm": 0.8147122263908386,
+      "learning_rate": 0.0002,
+      "loss": 0.7107,
+      "step": 710
+    },
+    {
+      "epoch": 1.5483870967741935,
+      "grad_norm": 0.796030580997467,
+      "learning_rate": 0.0002,
+      "loss": 0.7144,
+      "step": 720
+    },
+    {
+      "epoch": 1.5698924731182795,
+      "grad_norm": 0.8776171207427979,
+      "learning_rate": 0.0002,
+      "loss": 0.6936,
+      "step": 730
+    },
+    {
+      "epoch": 1.5913978494623655,
+      "grad_norm": 0.8056126236915588,
+      "learning_rate": 0.0002,
+      "loss": 0.7101,
+      "step": 740
+    },
+    {
+      "epoch": 1.6129032258064515,
+      "grad_norm": 0.8141863346099854,
+      "learning_rate": 0.0002,
+      "loss": 0.7162,
+      "step": 750
+    },
+    {
+      "epoch": 1.6344086021505375,
+      "grad_norm": 0.8100557327270508,
+      "learning_rate": 0.0002,
+      "loss": 0.7088,
+      "step": 760
+    },
+    {
+      "epoch": 1.6559139784946235,
+      "grad_norm": 0.8283200860023499,
+      "learning_rate": 0.0002,
+      "loss": 0.7212,
+      "step": 770
+    },
+    {
+      "epoch": 1.6774193548387095,
+      "grad_norm": 0.800865113735199,
+      "learning_rate": 0.0002,
+      "loss": 0.694,
+      "step": 780
+    },
+    {
+      "epoch": 1.6989247311827957,
+      "grad_norm": 0.8052287697792053,
+      "learning_rate": 0.0002,
+      "loss": 0.7076,
+      "step": 790
+    },
+    {
+      "epoch": 1.7204301075268817,
+      "grad_norm": 0.8619674444198608,
+      "learning_rate": 0.0002,
+      "loss": 0.7257,
+      "step": 800
+    },
+    {
+      "epoch": 1.7419354838709677,
+      "grad_norm": 0.8907215595245361,
+      "learning_rate": 0.0002,
+      "loss": 0.7141,
+      "step": 810
+    },
+    {
+      "epoch": 1.7634408602150538,
+      "grad_norm": 0.6976316571235657,
+      "learning_rate": 0.0002,
+      "loss": 0.7035,
+      "step": 820
+    },
+    {
+      "epoch": 1.7849462365591398,
+      "grad_norm": 0.7533746957778931,
+      "learning_rate": 0.0002,
+      "loss": 0.6916,
+      "step": 830
+    },
+    {
+      "epoch": 1.8064516129032258,
+      "grad_norm": 0.7326804399490356,
+      "learning_rate": 0.0002,
+      "loss": 0.7094,
+      "step": 840
+    },
+    {
+      "epoch": 1.827956989247312,
+      "grad_norm": 0.7782683372497559,
+      "learning_rate": 0.0002,
+      "loss": 0.6891,
+      "step": 850
+    },
+    {
+      "epoch": 1.849462365591398,
+      "grad_norm": 0.7424806356430054,
+      "learning_rate": 0.0002,
+      "loss": 0.6931,
+      "step": 860
+    },
+    {
+      "epoch": 1.870967741935484,
+      "grad_norm": 1.172325611114502,
+      "learning_rate": 0.0002,
+      "loss": 0.7354,
+      "step": 870
+    },
+    {
+      "epoch": 1.89247311827957,
+      "grad_norm": 0.771058201789856,
+      "learning_rate": 0.0002,
+      "loss": 0.6866,
+      "step": 880
+    },
+    {
+      "epoch": 1.913978494623656,
+      "grad_norm": 0.8624904155731201,
+      "learning_rate": 0.0002,
+      "loss": 0.7296,
+      "step": 890
+    },
+    {
+      "epoch": 1.935483870967742,
+      "grad_norm": 0.7062820792198181,
+      "learning_rate": 0.0002,
+      "loss": 0.7233,
+      "step": 900
+    },
+    {
+      "epoch": 1.956989247311828,
+      "grad_norm": 0.7560103535652161,
+      "learning_rate": 0.0002,
+      "loss": 0.6966,
+      "step": 910
+    },
+    {
+      "epoch": 1.978494623655914,
+      "grad_norm": 0.788899838924408,
+      "learning_rate": 0.0002,
+      "loss": 0.69,
+      "step": 920
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.6562113761901855,
+      "learning_rate": 0.0002,
+      "loss": 0.6505,
+      "step": 930
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.6885261535644531,
+      "eval_runtime": 21.4291,
+      "eval_samples_per_second": 15.446,
+      "eval_steps_per_second": 1.96,
+      "step": 930
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 3720,
+  "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": 1.136645659164672e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/training_args.bin b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..483b53d4fc1c568a5fc890fa850e3450f390b208
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e4c4fe2be590ed03492316230adb3a1edca3e4066c55f3716c0352d7134c564
+size 5560
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/special_tokens_map.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/tokenizer.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..f58963a682665634ab180c28667e4faa8cf02ba2
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/tokenizer.model b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/tokenizer_config.json b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1adb4796c13b8d975555ecec45876ee75d1ae8b7
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/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_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/training_args.bin b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..483b53d4fc1c568a5fc890fa850e3450f390b208
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e4c4fe2be590ed03492316230adb3a1edca3e4066c55f3716c0352d7134c564
+size 5560
diff --git a/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/training_log.jsonl b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/training_log.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..04e083afc9c7ee4335afa1c19d4eea4fe4df4714
--- /dev/null
+++ b/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/training_log.jsonl
@@ -0,0 +1,8 @@
+{"epoch": 1.0, "step": 465, "epoch_duration": 452.59338760375977, "total_accumulated_duration": 452.59338760375977, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 3020.60888671875}, "peak_memory_usage": {"GPU_0": 5628.7490234375}, "avg_memory_reserved": {"GPU_0": 6182.0}, "peak_memory_reserved": {"GPU_0": 6182.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "N/A", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 3.4172, "grad_norm": 0.9075053930282593, "learning_rate": 0.0002, "epoch": 0.021505376344086023, "step": 10}, {"loss": 2.5888, "grad_norm": 1.4321208000183105, "learning_rate": 0.0002, "epoch": 0.043010752688172046, "step": 20}, {"loss": 2.1195, "grad_norm": 1.7500602006912231, "learning_rate": 0.0002, "epoch": 0.06451612903225806, "step": 30}, {"loss": 1.9303, "grad_norm": 0.7606641054153442, "learning_rate": 0.0002, "epoch": 0.08602150537634409, "step": 40}, {"loss": 1.6112, "grad_norm": 1.2754929065704346, "learning_rate": 0.0002, "epoch": 0.10752688172043011, "step": 50}, {"loss": 1.4319, "grad_norm": 1.0936230421066284, "learning_rate": 0.0002, "epoch": 0.12903225806451613, "step": 60}, {"loss": 1.3568, "grad_norm": 1.144593596458435, "learning_rate": 0.0002, "epoch": 0.15053763440860216, "step": 70}, {"loss": 1.2028, "grad_norm": 1.2181956768035889, "learning_rate": 0.0002, "epoch": 0.17204301075268819, "step": 80}, {"loss": 1.1534, "grad_norm": 1.1260095834732056, "learning_rate": 0.0002, "epoch": 0.1935483870967742, "step": 90}, {"loss": 1.1089, "grad_norm": 1.1155284643173218, "learning_rate": 0.0002, "epoch": 0.21505376344086022, "step": 100}, {"loss": 1.0883, "grad_norm": 1.089565396308899, "learning_rate": 0.0002, "epoch": 0.23655913978494625, "step": 110}, {"loss": 1.0814, "grad_norm": 0.9833471775054932, "learning_rate": 0.0002, "epoch": 0.25806451612903225, "step": 120}, {"loss": 1.0239, "grad_norm": 1.0265629291534424, "learning_rate": 0.0002, "epoch": 0.27956989247311825, "step": 130}, {"loss": 0.9888, "grad_norm": 0.9344286322593689, "learning_rate": 0.0002, "epoch": 0.3010752688172043, "step": 140}, {"loss": 1.0043, "grad_norm": 0.9883386492729187, "learning_rate": 0.0002, "epoch": 0.3225806451612903, "step": 150}, {"loss": 0.9338, "grad_norm": 0.9299277067184448, "learning_rate": 0.0002, "epoch": 0.34408602150537637, "step": 160}, {"loss": 0.9432, "grad_norm": 1.390045404434204, "learning_rate": 0.0002, "epoch": 0.3655913978494624, "step": 170}, {"loss": 0.9008, "grad_norm": 1.0313078165054321, "learning_rate": 0.0002, "epoch": 0.3870967741935484, "step": 180}, {"loss": 0.9434, "grad_norm": 1.1792205572128296, "learning_rate": 0.0002, "epoch": 0.40860215053763443, "step": 190}, {"loss": 0.8761, "grad_norm": 1.049809217453003, "learning_rate": 0.0002, "epoch": 0.43010752688172044, "step": 200}, {"loss": 0.8709, "grad_norm": 0.990111768245697, "learning_rate": 0.0002, "epoch": 0.45161290322580644, "step": 210}, {"loss": 0.905, "grad_norm": 0.9870412349700928, "learning_rate": 0.0002, "epoch": 0.4731182795698925, "step": 220}, {"loss": 0.9129, "grad_norm": 0.8557345867156982, "learning_rate": 0.0002, "epoch": 0.4946236559139785, "step": 230}, {"loss": 0.8836, "grad_norm": 0.9746861457824707, "learning_rate": 0.0002, "epoch": 0.5161290322580645, "step": 240}, {"loss": 0.873, "grad_norm": 0.9010438323020935, "learning_rate": 0.0002, "epoch": 0.5376344086021505, "step": 250}, {"loss": 0.8241, "grad_norm": 0.9061082005500793, "learning_rate": 0.0002, "epoch": 0.5591397849462365, "step": 260}, {"loss": 0.8652, "grad_norm": 0.9311846494674683, "learning_rate": 0.0002, "epoch": 0.5806451612903226, "step": 270}, {"loss": 0.8256, "grad_norm": 0.9140254855155945, "learning_rate": 0.0002, "epoch": 0.6021505376344086, "step": 280}, {"loss": 0.8441, "grad_norm": 0.9722253084182739, "learning_rate": 0.0002, "epoch": 0.6236559139784946, "step": 290}, {"loss": 0.8314, "grad_norm": 0.8539168238639832, "learning_rate": 0.0002, "epoch": 0.6451612903225806, "step": 300}, {"loss": 0.8528, "grad_norm": 0.9053162932395935, "learning_rate": 0.0002, "epoch": 0.6666666666666666, "step": 310}, {"loss": 0.8209, "grad_norm": 0.8444252610206604, "learning_rate": 0.0002, "epoch": 0.6881720430107527, "step": 320}, {"loss": 0.8101, "grad_norm": 0.8127437829971313, "learning_rate": 0.0002, "epoch": 0.7096774193548387, "step": 330}, {"loss": 0.8223, "grad_norm": 0.886555016040802, "learning_rate": 0.0002, "epoch": 0.7311827956989247, "step": 340}, {"loss": 0.8368, "grad_norm": 0.8458548784255981, "learning_rate": 0.0002, "epoch": 0.7526881720430108, "step": 350}, {"loss": 0.8295, "grad_norm": 0.8683297634124756, "learning_rate": 0.0002, "epoch": 0.7741935483870968, "step": 360}, {"loss": 0.8232, "grad_norm": 0.8308405876159668, "learning_rate": 0.0002, "epoch": 0.7956989247311828, "step": 370}, {"loss": 0.7752, "grad_norm": 0.8305579423904419, "learning_rate": 0.0002, "epoch": 0.8172043010752689, "step": 380}, {"loss": 0.8267, "grad_norm": 0.8545567393302917, "learning_rate": 0.0002, "epoch": 0.8387096774193549, "step": 390}, {"loss": 0.8212, "grad_norm": 0.8486055731773376, "learning_rate": 0.0002, "epoch": 0.8602150537634409, "step": 400}, {"loss": 0.743, "grad_norm": 0.8126763105392456, "learning_rate": 0.0002, "epoch": 0.8817204301075269, "step": 410}, {"loss": 0.7993, "grad_norm": 0.8494045734405518, "learning_rate": 0.0002, "epoch": 0.9032258064516129, "step": 420}, {"loss": 0.8213, "grad_norm": 0.7639183402061462, "learning_rate": 0.0002, "epoch": 0.9247311827956989, "step": 430}, {"loss": 0.8015, "grad_norm": 0.858101487159729, "learning_rate": 0.0002, "epoch": 0.946236559139785, "step": 440}, {"loss": 0.7629, "grad_norm": 0.8141381740570068, "learning_rate": 0.0002, "epoch": 0.967741935483871, "step": 450}, {"loss": 0.7357, "grad_norm": 0.8072513937950134, "learning_rate": 0.0002, "epoch": 0.989247311827957, "step": 460}]}
+{"epoch": 2.0, "step": 930, "epoch_duration": 451.6676411628723, "total_accumulated_duration": 904.2610287666321, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 3020.60888671875}, "peak_memory_usage": {"GPU_0": 15051.17431640625}, "avg_memory_reserved": {"GPU_0": 20170.0}, "peak_memory_reserved": {"GPU_0": 20170.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-465", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 3.4172, "grad_norm": 0.9075053930282593, "learning_rate": 0.0002, "epoch": 0.021505376344086023, "step": 10}, {"loss": 2.5888, "grad_norm": 1.4321208000183105, "learning_rate": 0.0002, "epoch": 0.043010752688172046, "step": 20}, {"loss": 2.1195, "grad_norm": 1.7500602006912231, "learning_rate": 0.0002, "epoch": 0.06451612903225806, "step": 30}, {"loss": 1.9303, "grad_norm": 0.7606641054153442, "learning_rate": 0.0002, "epoch": 0.08602150537634409, "step": 40}, {"loss": 1.6112, "grad_norm": 1.2754929065704346, "learning_rate": 0.0002, "epoch": 0.10752688172043011, "step": 50}, {"loss": 1.4319, "grad_norm": 1.0936230421066284, "learning_rate": 0.0002, "epoch": 0.12903225806451613, "step": 60}, {"loss": 1.3568, "grad_norm": 1.144593596458435, "learning_rate": 0.0002, "epoch": 0.15053763440860216, "step": 70}, {"loss": 1.2028, "grad_norm": 1.2181956768035889, "learning_rate": 0.0002, "epoch": 0.17204301075268819, "step": 80}, {"loss": 1.1534, "grad_norm": 1.1260095834732056, "learning_rate": 0.0002, "epoch": 0.1935483870967742, "step": 90}, {"loss": 1.1089, "grad_norm": 1.1155284643173218, "learning_rate": 0.0002, "epoch": 0.21505376344086022, "step": 100}, {"loss": 1.0883, "grad_norm": 1.089565396308899, "learning_rate": 0.0002, "epoch": 0.23655913978494625, "step": 110}, {"loss": 1.0814, "grad_norm": 0.9833471775054932, "learning_rate": 0.0002, "epoch": 0.25806451612903225, "step": 120}, {"loss": 1.0239, "grad_norm": 1.0265629291534424, "learning_rate": 0.0002, "epoch": 0.27956989247311825, "step": 130}, {"loss": 0.9888, "grad_norm": 0.9344286322593689, "learning_rate": 0.0002, "epoch": 0.3010752688172043, "step": 140}, {"loss": 1.0043, "grad_norm": 0.9883386492729187, "learning_rate": 0.0002, "epoch": 0.3225806451612903, "step": 150}, {"loss": 0.9338, "grad_norm": 0.9299277067184448, "learning_rate": 0.0002, "epoch": 0.34408602150537637, "step": 160}, {"loss": 0.9432, "grad_norm": 1.390045404434204, "learning_rate": 0.0002, "epoch": 0.3655913978494624, "step": 170}, {"loss": 0.9008, "grad_norm": 1.0313078165054321, "learning_rate": 0.0002, "epoch": 0.3870967741935484, "step": 180}, {"loss": 0.9434, "grad_norm": 1.1792205572128296, "learning_rate": 0.0002, "epoch": 0.40860215053763443, "step": 190}, {"loss": 0.8761, "grad_norm": 1.049809217453003, "learning_rate": 0.0002, "epoch": 0.43010752688172044, "step": 200}, {"loss": 0.8709, "grad_norm": 0.990111768245697, "learning_rate": 0.0002, "epoch": 0.45161290322580644, "step": 210}, {"loss": 0.905, "grad_norm": 0.9870412349700928, "learning_rate": 0.0002, "epoch": 0.4731182795698925, "step": 220}, {"loss": 0.9129, "grad_norm": 0.8557345867156982, "learning_rate": 0.0002, "epoch": 0.4946236559139785, "step": 230}, {"loss": 0.8836, "grad_norm": 0.9746861457824707, "learning_rate": 0.0002, "epoch": 0.5161290322580645, "step": 240}, {"loss": 0.873, "grad_norm": 0.9010438323020935, "learning_rate": 0.0002, "epoch": 0.5376344086021505, "step": 250}, {"loss": 0.8241, "grad_norm": 0.9061082005500793, "learning_rate": 0.0002, "epoch": 0.5591397849462365, "step": 260}, {"loss": 0.8652, "grad_norm": 0.9311846494674683, "learning_rate": 0.0002, "epoch": 0.5806451612903226, "step": 270}, {"loss": 0.8256, "grad_norm": 0.9140254855155945, "learning_rate": 0.0002, "epoch": 0.6021505376344086, "step": 280}, {"loss": 0.8441, "grad_norm": 0.9722253084182739, "learning_rate": 0.0002, "epoch": 0.6236559139784946, "step": 290}, {"loss": 0.8314, "grad_norm": 0.8539168238639832, "learning_rate": 0.0002, "epoch": 0.6451612903225806, "step": 300}, {"loss": 0.8528, "grad_norm": 0.9053162932395935, "learning_rate": 0.0002, "epoch": 0.6666666666666666, "step": 310}, {"loss": 0.8209, "grad_norm": 0.8444252610206604, "learning_rate": 0.0002, "epoch": 0.6881720430107527, "step": 320}, {"loss": 0.8101, "grad_norm": 0.8127437829971313, "learning_rate": 0.0002, "epoch": 0.7096774193548387, "step": 330}, {"loss": 0.8223, "grad_norm": 0.886555016040802, "learning_rate": 0.0002, "epoch": 0.7311827956989247, "step": 340}, {"loss": 0.8368, "grad_norm": 0.8458548784255981, "learning_rate": 0.0002, "epoch": 0.7526881720430108, "step": 350}, {"loss": 0.8295, "grad_norm": 0.8683297634124756, "learning_rate": 0.0002, "epoch": 0.7741935483870968, "step": 360}, {"loss": 0.8232, "grad_norm": 0.8308405876159668, "learning_rate": 0.0002, "epoch": 0.7956989247311828, "step": 370}, {"loss": 0.7752, "grad_norm": 0.8305579423904419, "learning_rate": 0.0002, "epoch": 0.8172043010752689, "step": 380}, {"loss": 0.8267, "grad_norm": 0.8545567393302917, "learning_rate": 0.0002, "epoch": 0.8387096774193549, "step": 390}, {"loss": 0.8212, "grad_norm": 0.8486055731773376, "learning_rate": 0.0002, "epoch": 0.8602150537634409, "step": 400}, {"loss": 0.743, "grad_norm": 0.8126763105392456, "learning_rate": 0.0002, "epoch": 0.8817204301075269, "step": 410}, {"loss": 0.7993, "grad_norm": 0.8494045734405518, "learning_rate": 0.0002, "epoch": 0.9032258064516129, "step": 420}, {"loss": 0.8213, "grad_norm": 0.7639183402061462, "learning_rate": 0.0002, "epoch": 0.9247311827956989, "step": 430}, {"loss": 0.8015, "grad_norm": 0.858101487159729, "learning_rate": 0.0002, "epoch": 0.946236559139785, "step": 440}, {"loss": 0.7629, "grad_norm": 0.8141381740570068, "learning_rate": 0.0002, "epoch": 0.967741935483871, "step": 450}, {"loss": 0.7357, "grad_norm": 0.8072513937950134, "learning_rate": 0.0002, "epoch": 0.989247311827957, "step": 460}, {"eval_loss": 0.7740864157676697, "eval_runtime": 21.383, "eval_samples_per_second": 15.48, "eval_steps_per_second": 1.964, "epoch": 1.0, "step": 465}, {"loss": 0.7701, "grad_norm": 0.8269494771957397, "learning_rate": 0.0002, "epoch": 1.010752688172043, "step": 470}, {"loss": 0.7532, "grad_norm": 0.7814009189605713, "learning_rate": 0.0002, "epoch": 1.032258064516129, "step": 480}, {"loss": 0.7689, "grad_norm": 0.8183923363685608, "learning_rate": 0.0002, "epoch": 1.053763440860215, "step": 490}, {"loss": 0.765, "grad_norm": 0.8146600723266602, "learning_rate": 0.0002, "epoch": 1.075268817204301, "step": 500}, {"loss": 0.7358, "grad_norm": 0.8635126352310181, "learning_rate": 0.0002, "epoch": 1.096774193548387, "step": 510}, {"loss": 0.7302, "grad_norm": 0.8520359396934509, "learning_rate": 0.0002, "epoch": 1.118279569892473, "step": 520}, {"loss": 0.7492, "grad_norm": 0.8026443123817444, "learning_rate": 0.0002, "epoch": 1.139784946236559, "step": 530}, {"loss": 0.7518, "grad_norm": 0.8157258629798889, "learning_rate": 0.0002, "epoch": 1.1612903225806452, "step": 540}, {"loss": 0.7461, "grad_norm": 0.9450796246528625, "learning_rate": 0.0002, "epoch": 1.1827956989247312, "step": 550}, {"loss": 0.7128, "grad_norm": 0.8859835863113403, "learning_rate": 0.0002, "epoch": 1.2043010752688172, "step": 560}, {"loss": 0.7067, "grad_norm": 0.7819921970367432, "learning_rate": 0.0002, "epoch": 1.2258064516129032, "step": 570}, {"loss": 0.7577, "grad_norm": 0.7823445796966553, "learning_rate": 0.0002, "epoch": 1.2473118279569892, "step": 580}, {"loss": 0.7358, "grad_norm": 0.7931883931159973, "learning_rate": 0.0002, "epoch": 1.2688172043010753, "step": 590}, {"loss": 0.723, "grad_norm": 0.7495734095573425, "learning_rate": 0.0002, "epoch": 1.2903225806451613, "step": 600}, {"loss": 0.7386, "grad_norm": 0.9272717237472534, "learning_rate": 0.0002, "epoch": 1.3118279569892473, "step": 610}, {"loss": 0.7498, "grad_norm": 0.7968398332595825, "learning_rate": 0.0002, "epoch": 1.3333333333333333, "step": 620}, {"loss": 0.7635, "grad_norm": 0.7813659310340881, "learning_rate": 0.0002, "epoch": 1.3548387096774195, "step": 630}, {"loss": 0.6665, "grad_norm": 0.730925977230072, "learning_rate": 0.0002, "epoch": 1.3763440860215055, "step": 640}, {"loss": 0.7037, "grad_norm": 0.8011482954025269, "learning_rate": 0.0002, "epoch": 1.3978494623655915, "step": 650}, {"loss": 0.6931, "grad_norm": 0.7770085334777832, "learning_rate": 0.0002, "epoch": 1.4193548387096775, "step": 660}, {"loss": 0.6949, "grad_norm": 0.7432682514190674, "learning_rate": 0.0002, "epoch": 1.4408602150537635, "step": 670}, {"loss": 0.7444, "grad_norm": 0.8820092678070068, "learning_rate": 0.0002, "epoch": 1.4623655913978495, "step": 680}, {"loss": 0.6758, "grad_norm": 0.7786208987236023, "learning_rate": 0.0002, "epoch": 1.4838709677419355, "step": 690}, {"loss": 0.6702, "grad_norm": 0.7467480301856995, "learning_rate": 0.0002, "epoch": 1.5053763440860215, "step": 700}, {"loss": 0.7107, "grad_norm": 0.8147122263908386, "learning_rate": 0.0002, "epoch": 1.5268817204301075, "step": 710}, {"loss": 0.7144, "grad_norm": 0.796030580997467, "learning_rate": 0.0002, "epoch": 1.5483870967741935, "step": 720}, {"loss": 0.6936, "grad_norm": 0.8776171207427979, "learning_rate": 0.0002, "epoch": 1.5698924731182795, "step": 730}, {"loss": 0.7101, "grad_norm": 0.8056126236915588, "learning_rate": 0.0002, "epoch": 1.5913978494623655, "step": 740}, {"loss": 0.7162, "grad_norm": 0.8141863346099854, "learning_rate": 0.0002, "epoch": 1.6129032258064515, "step": 750}, {"loss": 0.7088, "grad_norm": 0.8100557327270508, "learning_rate": 0.0002, "epoch": 1.6344086021505375, "step": 760}, {"loss": 0.7212, "grad_norm": 0.8283200860023499, "learning_rate": 0.0002, "epoch": 1.6559139784946235, "step": 770}, {"loss": 0.694, "grad_norm": 0.800865113735199, "learning_rate": 0.0002, "epoch": 1.6774193548387095, "step": 780}, {"loss": 0.7076, "grad_norm": 0.8052287697792053, "learning_rate": 0.0002, "epoch": 1.6989247311827957, "step": 790}, {"loss": 0.7257, "grad_norm": 0.8619674444198608, "learning_rate": 0.0002, "epoch": 1.7204301075268817, "step": 800}, {"loss": 0.7141, "grad_norm": 0.8907215595245361, "learning_rate": 0.0002, "epoch": 1.7419354838709677, "step": 810}, {"loss": 0.7035, "grad_norm": 0.6976316571235657, "learning_rate": 0.0002, "epoch": 1.7634408602150538, "step": 820}, {"loss": 0.6916, "grad_norm": 0.7533746957778931, "learning_rate": 0.0002, "epoch": 1.7849462365591398, "step": 830}, {"loss": 0.7094, "grad_norm": 0.7326804399490356, "learning_rate": 0.0002, "epoch": 1.8064516129032258, "step": 840}, {"loss": 0.6891, "grad_norm": 0.7782683372497559, "learning_rate": 0.0002, "epoch": 1.827956989247312, "step": 850}, {"loss": 0.6931, "grad_norm": 0.7424806356430054, "learning_rate": 0.0002, "epoch": 1.849462365591398, "step": 860}, {"loss": 0.7354, "grad_norm": 1.172325611114502, "learning_rate": 0.0002, "epoch": 1.870967741935484, "step": 870}, {"loss": 0.6866, "grad_norm": 0.771058201789856, "learning_rate": 0.0002, "epoch": 1.89247311827957, "step": 880}, {"loss": 0.7296, "grad_norm": 0.8624904155731201, "learning_rate": 0.0002, "epoch": 1.913978494623656, "step": 890}, {"loss": 0.7233, "grad_norm": 0.7062820792198181, "learning_rate": 0.0002, "epoch": 1.935483870967742, "step": 900}, {"loss": 0.6966, "grad_norm": 0.7560103535652161, "learning_rate": 0.0002, "epoch": 1.956989247311828, "step": 910}, {"loss": 0.69, "grad_norm": 0.788899838924408, "learning_rate": 0.0002, "epoch": 1.978494623655914, "step": 920}, {"loss": 0.6505, "grad_norm": 0.6562113761901855, "learning_rate": 0.0002, "epoch": 2.0, "step": 930}]}
+{"epoch": 3.0, "step": 1395, "epoch_duration": 448.89671635627747, "total_accumulated_duration": 1353.1577451229095, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 3020.60888671875}, "peak_memory_usage": {"GPU_0": 15051.17431640625}, "avg_memory_reserved": {"GPU_0": 20170.0}, "peak_memory_reserved": {"GPU_0": 20170.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-930", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 3.4172, "grad_norm": 0.9075053930282593, "learning_rate": 0.0002, "epoch": 0.021505376344086023, "step": 10}, {"loss": 2.5888, "grad_norm": 1.4321208000183105, "learning_rate": 0.0002, "epoch": 0.043010752688172046, "step": 20}, {"loss": 2.1195, "grad_norm": 1.7500602006912231, "learning_rate": 0.0002, "epoch": 0.06451612903225806, "step": 30}, {"loss": 1.9303, "grad_norm": 0.7606641054153442, "learning_rate": 0.0002, "epoch": 0.08602150537634409, "step": 40}, {"loss": 1.6112, "grad_norm": 1.2754929065704346, "learning_rate": 0.0002, "epoch": 0.10752688172043011, "step": 50}, {"loss": 1.4319, "grad_norm": 1.0936230421066284, "learning_rate": 0.0002, "epoch": 0.12903225806451613, "step": 60}, {"loss": 1.3568, "grad_norm": 1.144593596458435, "learning_rate": 0.0002, "epoch": 0.15053763440860216, "step": 70}, {"loss": 1.2028, "grad_norm": 1.2181956768035889, "learning_rate": 0.0002, "epoch": 0.17204301075268819, "step": 80}, {"loss": 1.1534, "grad_norm": 1.1260095834732056, "learning_rate": 0.0002, "epoch": 0.1935483870967742, "step": 90}, {"loss": 1.1089, "grad_norm": 1.1155284643173218, "learning_rate": 0.0002, "epoch": 0.21505376344086022, "step": 100}, {"loss": 1.0883, "grad_norm": 1.089565396308899, "learning_rate": 0.0002, "epoch": 0.23655913978494625, "step": 110}, {"loss": 1.0814, "grad_norm": 0.9833471775054932, "learning_rate": 0.0002, "epoch": 0.25806451612903225, "step": 120}, {"loss": 1.0239, "grad_norm": 1.0265629291534424, "learning_rate": 0.0002, "epoch": 0.27956989247311825, "step": 130}, {"loss": 0.9888, "grad_norm": 0.9344286322593689, "learning_rate": 0.0002, "epoch": 0.3010752688172043, "step": 140}, {"loss": 1.0043, "grad_norm": 0.9883386492729187, "learning_rate": 0.0002, "epoch": 0.3225806451612903, "step": 150}, {"loss": 0.9338, "grad_norm": 0.9299277067184448, "learning_rate": 0.0002, "epoch": 0.34408602150537637, "step": 160}, {"loss": 0.9432, "grad_norm": 1.390045404434204, "learning_rate": 0.0002, "epoch": 0.3655913978494624, "step": 170}, {"loss": 0.9008, "grad_norm": 1.0313078165054321, "learning_rate": 0.0002, "epoch": 0.3870967741935484, "step": 180}, {"loss": 0.9434, "grad_norm": 1.1792205572128296, "learning_rate": 0.0002, "epoch": 0.40860215053763443, "step": 190}, {"loss": 0.8761, "grad_norm": 1.049809217453003, "learning_rate": 0.0002, "epoch": 0.43010752688172044, "step": 200}, {"loss": 0.8709, "grad_norm": 0.990111768245697, "learning_rate": 0.0002, "epoch": 0.45161290322580644, "step": 210}, {"loss": 0.905, "grad_norm": 0.9870412349700928, "learning_rate": 0.0002, "epoch": 0.4731182795698925, "step": 220}, {"loss": 0.9129, "grad_norm": 0.8557345867156982, "learning_rate": 0.0002, "epoch": 0.4946236559139785, "step": 230}, {"loss": 0.8836, "grad_norm": 0.9746861457824707, "learning_rate": 0.0002, "epoch": 0.5161290322580645, "step": 240}, {"loss": 0.873, "grad_norm": 0.9010438323020935, "learning_rate": 0.0002, "epoch": 0.5376344086021505, "step": 250}, {"loss": 0.8241, "grad_norm": 0.9061082005500793, "learning_rate": 0.0002, "epoch": 0.5591397849462365, "step": 260}, {"loss": 0.8652, "grad_norm": 0.9311846494674683, "learning_rate": 0.0002, "epoch": 0.5806451612903226, "step": 270}, {"loss": 0.8256, "grad_norm": 0.9140254855155945, "learning_rate": 0.0002, "epoch": 0.6021505376344086, "step": 280}, {"loss": 0.8441, "grad_norm": 0.9722253084182739, "learning_rate": 0.0002, "epoch": 0.6236559139784946, "step": 290}, {"loss": 0.8314, "grad_norm": 0.8539168238639832, "learning_rate": 0.0002, "epoch": 0.6451612903225806, "step": 300}, {"loss": 0.8528, "grad_norm": 0.9053162932395935, "learning_rate": 0.0002, "epoch": 0.6666666666666666, "step": 310}, {"loss": 0.8209, "grad_norm": 0.8444252610206604, "learning_rate": 0.0002, "epoch": 0.6881720430107527, "step": 320}, {"loss": 0.8101, "grad_norm": 0.8127437829971313, "learning_rate": 0.0002, "epoch": 0.7096774193548387, "step": 330}, {"loss": 0.8223, "grad_norm": 0.886555016040802, "learning_rate": 0.0002, "epoch": 0.7311827956989247, "step": 340}, {"loss": 0.8368, "grad_norm": 0.8458548784255981, "learning_rate": 0.0002, "epoch": 0.7526881720430108, "step": 350}, {"loss": 0.8295, "grad_norm": 0.8683297634124756, "learning_rate": 0.0002, "epoch": 0.7741935483870968, "step": 360}, {"loss": 0.8232, "grad_norm": 0.8308405876159668, "learning_rate": 0.0002, "epoch": 0.7956989247311828, "step": 370}, {"loss": 0.7752, "grad_norm": 0.8305579423904419, "learning_rate": 0.0002, "epoch": 0.8172043010752689, "step": 380}, {"loss": 0.8267, "grad_norm": 0.8545567393302917, "learning_rate": 0.0002, "epoch": 0.8387096774193549, "step": 390}, {"loss": 0.8212, "grad_norm": 0.8486055731773376, "learning_rate": 0.0002, "epoch": 0.8602150537634409, "step": 400}, {"loss": 0.743, "grad_norm": 0.8126763105392456, "learning_rate": 0.0002, "epoch": 0.8817204301075269, "step": 410}, {"loss": 0.7993, "grad_norm": 0.8494045734405518, "learning_rate": 0.0002, "epoch": 0.9032258064516129, "step": 420}, {"loss": 0.8213, "grad_norm": 0.7639183402061462, "learning_rate": 0.0002, "epoch": 0.9247311827956989, "step": 430}, {"loss": 0.8015, "grad_norm": 0.858101487159729, "learning_rate": 0.0002, "epoch": 0.946236559139785, "step": 440}, {"loss": 0.7629, "grad_norm": 0.8141381740570068, "learning_rate": 0.0002, "epoch": 0.967741935483871, "step": 450}, {"loss": 0.7357, "grad_norm": 0.8072513937950134, "learning_rate": 0.0002, "epoch": 0.989247311827957, "step": 460}, {"eval_loss": 0.7740864157676697, "eval_runtime": 21.383, "eval_samples_per_second": 15.48, "eval_steps_per_second": 1.964, "epoch": 1.0, "step": 465}, {"loss": 0.7701, "grad_norm": 0.8269494771957397, "learning_rate": 0.0002, "epoch": 1.010752688172043, "step": 470}, {"loss": 0.7532, "grad_norm": 0.7814009189605713, "learning_rate": 0.0002, "epoch": 1.032258064516129, "step": 480}, {"loss": 0.7689, "grad_norm": 0.8183923363685608, "learning_rate": 0.0002, "epoch": 1.053763440860215, "step": 490}, {"loss": 0.765, "grad_norm": 0.8146600723266602, "learning_rate": 0.0002, "epoch": 1.075268817204301, "step": 500}, {"loss": 0.7358, "grad_norm": 0.8635126352310181, "learning_rate": 0.0002, "epoch": 1.096774193548387, "step": 510}, {"loss": 0.7302, "grad_norm": 0.8520359396934509, "learning_rate": 0.0002, "epoch": 1.118279569892473, "step": 520}, {"loss": 0.7492, "grad_norm": 0.8026443123817444, "learning_rate": 0.0002, "epoch": 1.139784946236559, "step": 530}, {"loss": 0.7518, "grad_norm": 0.8157258629798889, "learning_rate": 0.0002, "epoch": 1.1612903225806452, "step": 540}, {"loss": 0.7461, "grad_norm": 0.9450796246528625, "learning_rate": 0.0002, "epoch": 1.1827956989247312, "step": 550}, {"loss": 0.7128, "grad_norm": 0.8859835863113403, "learning_rate": 0.0002, "epoch": 1.2043010752688172, "step": 560}, {"loss": 0.7067, "grad_norm": 0.7819921970367432, "learning_rate": 0.0002, "epoch": 1.2258064516129032, "step": 570}, {"loss": 0.7577, "grad_norm": 0.7823445796966553, "learning_rate": 0.0002, "epoch": 1.2473118279569892, "step": 580}, {"loss": 0.7358, "grad_norm": 0.7931883931159973, "learning_rate": 0.0002, "epoch": 1.2688172043010753, "step": 590}, {"loss": 0.723, "grad_norm": 0.7495734095573425, "learning_rate": 0.0002, "epoch": 1.2903225806451613, "step": 600}, {"loss": 0.7386, "grad_norm": 0.9272717237472534, "learning_rate": 0.0002, "epoch": 1.3118279569892473, "step": 610}, {"loss": 0.7498, "grad_norm": 0.7968398332595825, "learning_rate": 0.0002, "epoch": 1.3333333333333333, "step": 620}, {"loss": 0.7635, "grad_norm": 0.7813659310340881, "learning_rate": 0.0002, "epoch": 1.3548387096774195, "step": 630}, {"loss": 0.6665, "grad_norm": 0.730925977230072, "learning_rate": 0.0002, "epoch": 1.3763440860215055, "step": 640}, {"loss": 0.7037, "grad_norm": 0.8011482954025269, "learning_rate": 0.0002, "epoch": 1.3978494623655915, "step": 650}, {"loss": 0.6931, "grad_norm": 0.7770085334777832, "learning_rate": 0.0002, "epoch": 1.4193548387096775, "step": 660}, {"loss": 0.6949, "grad_norm": 0.7432682514190674, "learning_rate": 0.0002, "epoch": 1.4408602150537635, "step": 670}, {"loss": 0.7444, "grad_norm": 0.8820092678070068, "learning_rate": 0.0002, "epoch": 1.4623655913978495, "step": 680}, {"loss": 0.6758, "grad_norm": 0.7786208987236023, "learning_rate": 0.0002, "epoch": 1.4838709677419355, "step": 690}, {"loss": 0.6702, "grad_norm": 0.7467480301856995, "learning_rate": 0.0002, "epoch": 1.5053763440860215, "step": 700}, {"loss": 0.7107, "grad_norm": 0.8147122263908386, "learning_rate": 0.0002, "epoch": 1.5268817204301075, "step": 710}, {"loss": 0.7144, "grad_norm": 0.796030580997467, "learning_rate": 0.0002, "epoch": 1.5483870967741935, "step": 720}, {"loss": 0.6936, "grad_norm": 0.8776171207427979, "learning_rate": 0.0002, "epoch": 1.5698924731182795, "step": 730}, {"loss": 0.7101, "grad_norm": 0.8056126236915588, "learning_rate": 0.0002, "epoch": 1.5913978494623655, "step": 740}, {"loss": 0.7162, "grad_norm": 0.8141863346099854, "learning_rate": 0.0002, "epoch": 1.6129032258064515, "step": 750}, {"loss": 0.7088, "grad_norm": 0.8100557327270508, "learning_rate": 0.0002, "epoch": 1.6344086021505375, "step": 760}, {"loss": 0.7212, "grad_norm": 0.8283200860023499, "learning_rate": 0.0002, "epoch": 1.6559139784946235, "step": 770}, {"loss": 0.694, "grad_norm": 0.800865113735199, "learning_rate": 0.0002, "epoch": 1.6774193548387095, "step": 780}, {"loss": 0.7076, "grad_norm": 0.8052287697792053, "learning_rate": 0.0002, "epoch": 1.6989247311827957, "step": 790}, {"loss": 0.7257, "grad_norm": 0.8619674444198608, "learning_rate": 0.0002, "epoch": 1.7204301075268817, "step": 800}, {"loss": 0.7141, "grad_norm": 0.8907215595245361, "learning_rate": 0.0002, "epoch": 1.7419354838709677, "step": 810}, {"loss": 0.7035, "grad_norm": 0.6976316571235657, "learning_rate": 0.0002, "epoch": 1.7634408602150538, "step": 820}, {"loss": 0.6916, "grad_norm": 0.7533746957778931, "learning_rate": 0.0002, "epoch": 1.7849462365591398, "step": 830}, {"loss": 0.7094, "grad_norm": 0.7326804399490356, "learning_rate": 0.0002, "epoch": 1.8064516129032258, "step": 840}, {"loss": 0.6891, "grad_norm": 0.7782683372497559, "learning_rate": 0.0002, "epoch": 1.827956989247312, "step": 850}, {"loss": 0.6931, "grad_norm": 0.7424806356430054, "learning_rate": 0.0002, "epoch": 1.849462365591398, "step": 860}, {"loss": 0.7354, "grad_norm": 1.172325611114502, "learning_rate": 0.0002, "epoch": 1.870967741935484, "step": 870}, {"loss": 0.6866, "grad_norm": 0.771058201789856, "learning_rate": 0.0002, "epoch": 1.89247311827957, "step": 880}, {"loss": 0.7296, "grad_norm": 0.8624904155731201, "learning_rate": 0.0002, "epoch": 1.913978494623656, "step": 890}, {"loss": 0.7233, "grad_norm": 0.7062820792198181, "learning_rate": 0.0002, "epoch": 1.935483870967742, "step": 900}, {"loss": 0.6966, "grad_norm": 0.7560103535652161, "learning_rate": 0.0002, "epoch": 1.956989247311828, "step": 910}, {"loss": 0.69, "grad_norm": 0.788899838924408, "learning_rate": 0.0002, "epoch": 1.978494623655914, "step": 920}, {"loss": 0.6505, "grad_norm": 0.6562113761901855, "learning_rate": 0.0002, "epoch": 2.0, "step": 930}, {"eval_loss": 0.6885261535644531, "eval_runtime": 21.4291, "eval_samples_per_second": 15.446, "eval_steps_per_second": 1.96, "epoch": 2.0, "step": 930}, {"loss": 0.6625, "grad_norm": 0.8216531872749329, "learning_rate": 0.0002, "epoch": 2.021505376344086, "step": 940}, {"loss": 0.6398, "grad_norm": 0.8317142724990845, "learning_rate": 0.0002, "epoch": 2.043010752688172, "step": 950}, {"loss": 0.649, "grad_norm": 0.8446708917617798, "learning_rate": 0.0002, "epoch": 2.064516129032258, "step": 960}, {"loss": 0.657, "grad_norm": 0.735055148601532, "learning_rate": 0.0002, "epoch": 2.086021505376344, "step": 970}, {"loss": 0.649, "grad_norm": 0.7487243413925171, "learning_rate": 0.0002, "epoch": 2.10752688172043, "step": 980}, {"loss": 0.6419, "grad_norm": 0.8573887944221497, "learning_rate": 0.0002, "epoch": 2.129032258064516, "step": 990}, {"loss": 0.6431, "grad_norm": 0.6284521818161011, "learning_rate": 0.0002, "epoch": 2.150537634408602, "step": 1000}, {"loss": 0.6128, "grad_norm": 0.754183292388916, "learning_rate": 0.0002, "epoch": 2.172043010752688, "step": 1010}, {"loss": 0.6253, "grad_norm": 0.9445359110832214, "learning_rate": 0.0002, "epoch": 2.193548387096774, "step": 1020}, {"loss": 0.605, "grad_norm": 0.808508038520813, "learning_rate": 0.0002, "epoch": 2.21505376344086, "step": 1030}, {"loss": 0.6786, "grad_norm": 0.9394679665565491, "learning_rate": 0.0002, "epoch": 2.236559139784946, "step": 1040}, {"loss": 0.6176, "grad_norm": 0.8151357769966125, "learning_rate": 0.0002, "epoch": 2.258064516129032, "step": 1050}, {"loss": 0.66, "grad_norm": 0.7909848093986511, "learning_rate": 0.0002, "epoch": 2.279569892473118, "step": 1060}, {"loss": 0.6254, "grad_norm": 0.7506507039070129, "learning_rate": 0.0002, "epoch": 2.3010752688172045, "step": 1070}, {"loss": 0.6608, "grad_norm": 0.8240520358085632, "learning_rate": 0.0002, "epoch": 2.3225806451612905, "step": 1080}, {"loss": 0.6207, "grad_norm": 0.9342400431632996, "learning_rate": 0.0002, "epoch": 2.3440860215053765, "step": 1090}, {"loss": 0.6029, "grad_norm": 1.0598735809326172, "learning_rate": 0.0002, "epoch": 2.3655913978494625, "step": 1100}, {"loss": 0.6035, "grad_norm": 0.7907650470733643, "learning_rate": 0.0002, "epoch": 2.3870967741935485, "step": 1110}, {"loss": 0.6237, "grad_norm": 0.9388798475265503, "learning_rate": 0.0002, "epoch": 2.4086021505376345, "step": 1120}, {"loss": 0.6207, "grad_norm": 0.8985419869422913, "learning_rate": 0.0002, "epoch": 2.4301075268817205, "step": 1130}, {"loss": 0.5902, "grad_norm": 0.7471932768821716, "learning_rate": 0.0002, "epoch": 2.4516129032258065, "step": 1140}, {"loss": 0.6446, "grad_norm": 0.761131763458252, "learning_rate": 0.0002, "epoch": 2.4731182795698925, "step": 1150}, {"loss": 0.6088, "grad_norm": 0.7901819348335266, "learning_rate": 0.0002, "epoch": 2.4946236559139785, "step": 1160}, {"loss": 0.6142, "grad_norm": 0.9932922720909119, "learning_rate": 0.0002, "epoch": 2.5161290322580645, "step": 1170}, {"loss": 0.6407, "grad_norm": 0.7414287328720093, "learning_rate": 0.0002, "epoch": 2.5376344086021505, "step": 1180}, {"loss": 0.6161, "grad_norm": 0.8111771941184998, "learning_rate": 0.0002, "epoch": 2.5591397849462365, "step": 1190}, {"loss": 0.6006, "grad_norm": 0.7520156502723694, "learning_rate": 0.0002, "epoch": 2.5806451612903225, "step": 1200}, {"loss": 0.615, "grad_norm": 0.9022907018661499, "learning_rate": 0.0002, "epoch": 2.6021505376344085, "step": 1210}, {"loss": 0.6211, "grad_norm": 0.7746260166168213, "learning_rate": 0.0002, "epoch": 2.6236559139784945, "step": 1220}, {"loss": 0.616, "grad_norm": 0.8482862114906311, "learning_rate": 0.0002, "epoch": 2.6451612903225805, "step": 1230}, {"loss": 0.6417, "grad_norm": 0.7925458550453186, "learning_rate": 0.0002, "epoch": 2.6666666666666665, "step": 1240}, {"loss": 0.6187, "grad_norm": 0.8369929194450378, "learning_rate": 0.0002, "epoch": 2.688172043010753, "step": 1250}, {"loss": 0.6138, "grad_norm": 0.8311542868614197, "learning_rate": 0.0002, "epoch": 2.709677419354839, "step": 1260}, {"loss": 0.5894, "grad_norm": 0.7204853296279907, "learning_rate": 0.0002, "epoch": 2.731182795698925, "step": 1270}, {"loss": 0.6325, "grad_norm": 0.8447284698486328, "learning_rate": 0.0002, "epoch": 2.752688172043011, "step": 1280}, {"loss": 0.5946, "grad_norm": 0.7738404273986816, "learning_rate": 0.0002, "epoch": 2.774193548387097, "step": 1290}, {"loss": 0.5678, "grad_norm": 0.8393287062644958, "learning_rate": 0.0002, "epoch": 2.795698924731183, "step": 1300}, {"loss": 0.6092, "grad_norm": 0.79121994972229, "learning_rate": 0.0002, "epoch": 2.817204301075269, "step": 1310}, {"loss": 0.5889, "grad_norm": 0.7331557869911194, "learning_rate": 0.0002, "epoch": 2.838709677419355, "step": 1320}, {"loss": 0.6048, "grad_norm": 0.9593998193740845, "learning_rate": 0.0002, "epoch": 2.860215053763441, "step": 1330}, {"loss": 0.6108, "grad_norm": 0.7215158343315125, "learning_rate": 0.0002, "epoch": 2.881720430107527, "step": 1340}, {"loss": 0.5897, "grad_norm": 0.840404212474823, "learning_rate": 0.0002, "epoch": 2.903225806451613, "step": 1350}, {"loss": 0.6056, "grad_norm": 0.870659351348877, "learning_rate": 0.0002, "epoch": 2.924731182795699, "step": 1360}, {"loss": 0.6205, "grad_norm": 0.8744975328445435, "learning_rate": 0.0002, "epoch": 2.946236559139785, "step": 1370}, {"loss": 0.5966, "grad_norm": 0.8030612468719482, "learning_rate": 0.0002, "epoch": 2.967741935483871, "step": 1380}, {"loss": 0.6004, "grad_norm": 0.825814962387085, "learning_rate": 0.0002, "epoch": 2.989247311827957, "step": 1390}]}
+{"epoch": 4.0, "step": 1860, "epoch_duration": 449.0518436431885, "total_accumulated_duration": 1802.209588766098, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 3020.60888671875}, "peak_memory_usage": {"GPU_0": 15051.17431640625}, "avg_memory_reserved": {"GPU_0": 20170.0}, "peak_memory_reserved": {"GPU_0": 20170.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1395", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 3.4172, "grad_norm": 0.9075053930282593, "learning_rate": 0.0002, "epoch": 0.021505376344086023, "step": 10}, {"loss": 2.5888, "grad_norm": 1.4321208000183105, "learning_rate": 0.0002, "epoch": 0.043010752688172046, "step": 20}, {"loss": 2.1195, "grad_norm": 1.7500602006912231, "learning_rate": 0.0002, "epoch": 0.06451612903225806, "step": 30}, {"loss": 1.9303, "grad_norm": 0.7606641054153442, "learning_rate": 0.0002, "epoch": 0.08602150537634409, "step": 40}, {"loss": 1.6112, "grad_norm": 1.2754929065704346, "learning_rate": 0.0002, "epoch": 0.10752688172043011, "step": 50}, {"loss": 1.4319, "grad_norm": 1.0936230421066284, "learning_rate": 0.0002, "epoch": 0.12903225806451613, "step": 60}, {"loss": 1.3568, "grad_norm": 1.144593596458435, "learning_rate": 0.0002, "epoch": 0.15053763440860216, "step": 70}, {"loss": 1.2028, "grad_norm": 1.2181956768035889, "learning_rate": 0.0002, "epoch": 0.17204301075268819, "step": 80}, {"loss": 1.1534, "grad_norm": 1.1260095834732056, "learning_rate": 0.0002, "epoch": 0.1935483870967742, "step": 90}, {"loss": 1.1089, "grad_norm": 1.1155284643173218, "learning_rate": 0.0002, "epoch": 0.21505376344086022, "step": 100}, {"loss": 1.0883, "grad_norm": 1.089565396308899, "learning_rate": 0.0002, "epoch": 0.23655913978494625, "step": 110}, {"loss": 1.0814, "grad_norm": 0.9833471775054932, "learning_rate": 0.0002, "epoch": 0.25806451612903225, "step": 120}, {"loss": 1.0239, "grad_norm": 1.0265629291534424, "learning_rate": 0.0002, "epoch": 0.27956989247311825, "step": 130}, {"loss": 0.9888, "grad_norm": 0.9344286322593689, "learning_rate": 0.0002, "epoch": 0.3010752688172043, "step": 140}, {"loss": 1.0043, "grad_norm": 0.9883386492729187, "learning_rate": 0.0002, "epoch": 0.3225806451612903, "step": 150}, {"loss": 0.9338, "grad_norm": 0.9299277067184448, "learning_rate": 0.0002, "epoch": 0.34408602150537637, "step": 160}, {"loss": 0.9432, "grad_norm": 1.390045404434204, "learning_rate": 0.0002, "epoch": 0.3655913978494624, "step": 170}, {"loss": 0.9008, "grad_norm": 1.0313078165054321, "learning_rate": 0.0002, "epoch": 0.3870967741935484, "step": 180}, {"loss": 0.9434, "grad_norm": 1.1792205572128296, "learning_rate": 0.0002, "epoch": 0.40860215053763443, "step": 190}, {"loss": 0.8761, "grad_norm": 1.049809217453003, "learning_rate": 0.0002, "epoch": 0.43010752688172044, "step": 200}, {"loss": 0.8709, "grad_norm": 0.990111768245697, "learning_rate": 0.0002, "epoch": 0.45161290322580644, "step": 210}, {"loss": 0.905, "grad_norm": 0.9870412349700928, "learning_rate": 0.0002, "epoch": 0.4731182795698925, "step": 220}, {"loss": 0.9129, "grad_norm": 0.8557345867156982, "learning_rate": 0.0002, "epoch": 0.4946236559139785, "step": 230}, {"loss": 0.8836, "grad_norm": 0.9746861457824707, "learning_rate": 0.0002, "epoch": 0.5161290322580645, "step": 240}, {"loss": 0.873, "grad_norm": 0.9010438323020935, "learning_rate": 0.0002, "epoch": 0.5376344086021505, "step": 250}, {"loss": 0.8241, "grad_norm": 0.9061082005500793, "learning_rate": 0.0002, "epoch": 0.5591397849462365, "step": 260}, {"loss": 0.8652, "grad_norm": 0.9311846494674683, "learning_rate": 0.0002, "epoch": 0.5806451612903226, "step": 270}, {"loss": 0.8256, "grad_norm": 0.9140254855155945, "learning_rate": 0.0002, "epoch": 0.6021505376344086, "step": 280}, {"loss": 0.8441, "grad_norm": 0.9722253084182739, "learning_rate": 0.0002, "epoch": 0.6236559139784946, "step": 290}, {"loss": 0.8314, "grad_norm": 0.8539168238639832, "learning_rate": 0.0002, "epoch": 0.6451612903225806, "step": 300}, {"loss": 0.8528, "grad_norm": 0.9053162932395935, "learning_rate": 0.0002, "epoch": 0.6666666666666666, "step": 310}, {"loss": 0.8209, "grad_norm": 0.8444252610206604, "learning_rate": 0.0002, "epoch": 0.6881720430107527, "step": 320}, {"loss": 0.8101, "grad_norm": 0.8127437829971313, "learning_rate": 0.0002, "epoch": 0.7096774193548387, "step": 330}, {"loss": 0.8223, "grad_norm": 0.886555016040802, "learning_rate": 0.0002, "epoch": 0.7311827956989247, "step": 340}, {"loss": 0.8368, "grad_norm": 0.8458548784255981, "learning_rate": 0.0002, "epoch": 0.7526881720430108, "step": 350}, {"loss": 0.8295, "grad_norm": 0.8683297634124756, "learning_rate": 0.0002, "epoch": 0.7741935483870968, "step": 360}, {"loss": 0.8232, "grad_norm": 0.8308405876159668, "learning_rate": 0.0002, "epoch": 0.7956989247311828, "step": 370}, {"loss": 0.7752, "grad_norm": 0.8305579423904419, "learning_rate": 0.0002, "epoch": 0.8172043010752689, "step": 380}, {"loss": 0.8267, "grad_norm": 0.8545567393302917, "learning_rate": 0.0002, "epoch": 0.8387096774193549, "step": 390}, {"loss": 0.8212, "grad_norm": 0.8486055731773376, "learning_rate": 0.0002, "epoch": 0.8602150537634409, "step": 400}, {"loss": 0.743, "grad_norm": 0.8126763105392456, "learning_rate": 0.0002, "epoch": 0.8817204301075269, "step": 410}, {"loss": 0.7993, "grad_norm": 0.8494045734405518, "learning_rate": 0.0002, "epoch": 0.9032258064516129, "step": 420}, {"loss": 0.8213, "grad_norm": 0.7639183402061462, "learning_rate": 0.0002, "epoch": 0.9247311827956989, "step": 430}, {"loss": 0.8015, "grad_norm": 0.858101487159729, "learning_rate": 0.0002, "epoch": 0.946236559139785, "step": 440}, {"loss": 0.7629, "grad_norm": 0.8141381740570068, "learning_rate": 0.0002, "epoch": 0.967741935483871, "step": 450}, {"loss": 0.7357, "grad_norm": 0.8072513937950134, "learning_rate": 0.0002, "epoch": 0.989247311827957, "step": 460}, {"eval_loss": 0.7740864157676697, "eval_runtime": 21.383, "eval_samples_per_second": 15.48, "eval_steps_per_second": 1.964, "epoch": 1.0, "step": 465}, {"loss": 0.7701, "grad_norm": 0.8269494771957397, "learning_rate": 0.0002, "epoch": 1.010752688172043, "step": 470}, {"loss": 0.7532, "grad_norm": 0.7814009189605713, "learning_rate": 0.0002, "epoch": 1.032258064516129, "step": 480}, {"loss": 0.7689, "grad_norm": 0.8183923363685608, "learning_rate": 0.0002, "epoch": 1.053763440860215, "step": 490}, {"loss": 0.765, "grad_norm": 0.8146600723266602, "learning_rate": 0.0002, "epoch": 1.075268817204301, "step": 500}, {"loss": 0.7358, "grad_norm": 0.8635126352310181, "learning_rate": 0.0002, "epoch": 1.096774193548387, "step": 510}, {"loss": 0.7302, "grad_norm": 0.8520359396934509, "learning_rate": 0.0002, "epoch": 1.118279569892473, "step": 520}, {"loss": 0.7492, "grad_norm": 0.8026443123817444, "learning_rate": 0.0002, "epoch": 1.139784946236559, "step": 530}, {"loss": 0.7518, "grad_norm": 0.8157258629798889, "learning_rate": 0.0002, "epoch": 1.1612903225806452, "step": 540}, {"loss": 0.7461, "grad_norm": 0.9450796246528625, "learning_rate": 0.0002, "epoch": 1.1827956989247312, "step": 550}, {"loss": 0.7128, "grad_norm": 0.8859835863113403, "learning_rate": 0.0002, "epoch": 1.2043010752688172, "step": 560}, {"loss": 0.7067, "grad_norm": 0.7819921970367432, "learning_rate": 0.0002, "epoch": 1.2258064516129032, "step": 570}, {"loss": 0.7577, "grad_norm": 0.7823445796966553, "learning_rate": 0.0002, "epoch": 1.2473118279569892, "step": 580}, {"loss": 0.7358, "grad_norm": 0.7931883931159973, "learning_rate": 0.0002, "epoch": 1.2688172043010753, "step": 590}, {"loss": 0.723, "grad_norm": 0.7495734095573425, "learning_rate": 0.0002, "epoch": 1.2903225806451613, "step": 600}, {"loss": 0.7386, "grad_norm": 0.9272717237472534, "learning_rate": 0.0002, "epoch": 1.3118279569892473, "step": 610}, {"loss": 0.7498, "grad_norm": 0.7968398332595825, "learning_rate": 0.0002, "epoch": 1.3333333333333333, "step": 620}, {"loss": 0.7635, "grad_norm": 0.7813659310340881, "learning_rate": 0.0002, "epoch": 1.3548387096774195, "step": 630}, {"loss": 0.6665, "grad_norm": 0.730925977230072, "learning_rate": 0.0002, "epoch": 1.3763440860215055, "step": 640}, {"loss": 0.7037, "grad_norm": 0.8011482954025269, "learning_rate": 0.0002, "epoch": 1.3978494623655915, "step": 650}, {"loss": 0.6931, "grad_norm": 0.7770085334777832, "learning_rate": 0.0002, "epoch": 1.4193548387096775, "step": 660}, {"loss": 0.6949, "grad_norm": 0.7432682514190674, "learning_rate": 0.0002, "epoch": 1.4408602150537635, "step": 670}, {"loss": 0.7444, "grad_norm": 0.8820092678070068, "learning_rate": 0.0002, "epoch": 1.4623655913978495, "step": 680}, {"loss": 0.6758, "grad_norm": 0.7786208987236023, "learning_rate": 0.0002, "epoch": 1.4838709677419355, "step": 690}, {"loss": 0.6702, "grad_norm": 0.7467480301856995, "learning_rate": 0.0002, "epoch": 1.5053763440860215, "step": 700}, {"loss": 0.7107, "grad_norm": 0.8147122263908386, "learning_rate": 0.0002, "epoch": 1.5268817204301075, "step": 710}, {"loss": 0.7144, "grad_norm": 0.796030580997467, "learning_rate": 0.0002, "epoch": 1.5483870967741935, "step": 720}, {"loss": 0.6936, "grad_norm": 0.8776171207427979, "learning_rate": 0.0002, "epoch": 1.5698924731182795, "step": 730}, {"loss": 0.7101, "grad_norm": 0.8056126236915588, "learning_rate": 0.0002, "epoch": 1.5913978494623655, "step": 740}, {"loss": 0.7162, "grad_norm": 0.8141863346099854, "learning_rate": 0.0002, "epoch": 1.6129032258064515, "step": 750}, {"loss": 0.7088, "grad_norm": 0.8100557327270508, "learning_rate": 0.0002, "epoch": 1.6344086021505375, "step": 760}, {"loss": 0.7212, "grad_norm": 0.8283200860023499, "learning_rate": 0.0002, "epoch": 1.6559139784946235, "step": 770}, {"loss": 0.694, "grad_norm": 0.800865113735199, "learning_rate": 0.0002, "epoch": 1.6774193548387095, "step": 780}, {"loss": 0.7076, "grad_norm": 0.8052287697792053, "learning_rate": 0.0002, "epoch": 1.6989247311827957, "step": 790}, {"loss": 0.7257, "grad_norm": 0.8619674444198608, "learning_rate": 0.0002, "epoch": 1.7204301075268817, "step": 800}, {"loss": 0.7141, "grad_norm": 0.8907215595245361, "learning_rate": 0.0002, "epoch": 1.7419354838709677, "step": 810}, {"loss": 0.7035, "grad_norm": 0.6976316571235657, "learning_rate": 0.0002, "epoch": 1.7634408602150538, "step": 820}, {"loss": 0.6916, "grad_norm": 0.7533746957778931, "learning_rate": 0.0002, "epoch": 1.7849462365591398, "step": 830}, {"loss": 0.7094, "grad_norm": 0.7326804399490356, "learning_rate": 0.0002, "epoch": 1.8064516129032258, "step": 840}, {"loss": 0.6891, "grad_norm": 0.7782683372497559, "learning_rate": 0.0002, "epoch": 1.827956989247312, "step": 850}, {"loss": 0.6931, "grad_norm": 0.7424806356430054, "learning_rate": 0.0002, "epoch": 1.849462365591398, "step": 860}, {"loss": 0.7354, "grad_norm": 1.172325611114502, "learning_rate": 0.0002, "epoch": 1.870967741935484, "step": 870}, {"loss": 0.6866, "grad_norm": 0.771058201789856, "learning_rate": 0.0002, "epoch": 1.89247311827957, "step": 880}, {"loss": 0.7296, "grad_norm": 0.8624904155731201, "learning_rate": 0.0002, "epoch": 1.913978494623656, "step": 890}, {"loss": 0.7233, "grad_norm": 0.7062820792198181, "learning_rate": 0.0002, "epoch": 1.935483870967742, "step": 900}, {"loss": 0.6966, "grad_norm": 0.7560103535652161, "learning_rate": 0.0002, "epoch": 1.956989247311828, "step": 910}, {"loss": 0.69, "grad_norm": 0.788899838924408, "learning_rate": 0.0002, "epoch": 1.978494623655914, "step": 920}, {"loss": 0.6505, "grad_norm": 0.6562113761901855, "learning_rate": 0.0002, "epoch": 2.0, "step": 930}, {"eval_loss": 0.6885261535644531, "eval_runtime": 21.4291, "eval_samples_per_second": 15.446, "eval_steps_per_second": 1.96, "epoch": 2.0, "step": 930}, {"loss": 0.6625, "grad_norm": 0.8216531872749329, "learning_rate": 0.0002, "epoch": 2.021505376344086, "step": 940}, {"loss": 0.6398, "grad_norm": 0.8317142724990845, "learning_rate": 0.0002, "epoch": 2.043010752688172, "step": 950}, {"loss": 0.649, "grad_norm": 0.8446708917617798, "learning_rate": 0.0002, "epoch": 2.064516129032258, "step": 960}, {"loss": 0.657, "grad_norm": 0.735055148601532, "learning_rate": 0.0002, "epoch": 2.086021505376344, "step": 970}, {"loss": 0.649, "grad_norm": 0.7487243413925171, "learning_rate": 0.0002, "epoch": 2.10752688172043, "step": 980}, {"loss": 0.6419, "grad_norm": 0.8573887944221497, "learning_rate": 0.0002, "epoch": 2.129032258064516, "step": 990}, {"loss": 0.6431, "grad_norm": 0.6284521818161011, "learning_rate": 0.0002, "epoch": 2.150537634408602, "step": 1000}, {"loss": 0.6128, "grad_norm": 0.754183292388916, "learning_rate": 0.0002, "epoch": 2.172043010752688, "step": 1010}, {"loss": 0.6253, "grad_norm": 0.9445359110832214, "learning_rate": 0.0002, "epoch": 2.193548387096774, "step": 1020}, {"loss": 0.605, "grad_norm": 0.808508038520813, "learning_rate": 0.0002, "epoch": 2.21505376344086, "step": 1030}, {"loss": 0.6786, "grad_norm": 0.9394679665565491, "learning_rate": 0.0002, "epoch": 2.236559139784946, "step": 1040}, {"loss": 0.6176, "grad_norm": 0.8151357769966125, "learning_rate": 0.0002, "epoch": 2.258064516129032, "step": 1050}, {"loss": 0.66, "grad_norm": 0.7909848093986511, "learning_rate": 0.0002, "epoch": 2.279569892473118, "step": 1060}, {"loss": 0.6254, "grad_norm": 0.7506507039070129, "learning_rate": 0.0002, "epoch": 2.3010752688172045, "step": 1070}, {"loss": 0.6608, "grad_norm": 0.8240520358085632, "learning_rate": 0.0002, "epoch": 2.3225806451612905, "step": 1080}, {"loss": 0.6207, "grad_norm": 0.9342400431632996, "learning_rate": 0.0002, "epoch": 2.3440860215053765, "step": 1090}, {"loss": 0.6029, "grad_norm": 1.0598735809326172, "learning_rate": 0.0002, "epoch": 2.3655913978494625, "step": 1100}, {"loss": 0.6035, "grad_norm": 0.7907650470733643, "learning_rate": 0.0002, "epoch": 2.3870967741935485, "step": 1110}, {"loss": 0.6237, "grad_norm": 0.9388798475265503, "learning_rate": 0.0002, "epoch": 2.4086021505376345, "step": 1120}, {"loss": 0.6207, "grad_norm": 0.8985419869422913, "learning_rate": 0.0002, "epoch": 2.4301075268817205, "step": 1130}, {"loss": 0.5902, "grad_norm": 0.7471932768821716, "learning_rate": 0.0002, "epoch": 2.4516129032258065, "step": 1140}, {"loss": 0.6446, "grad_norm": 0.761131763458252, "learning_rate": 0.0002, "epoch": 2.4731182795698925, "step": 1150}, {"loss": 0.6088, "grad_norm": 0.7901819348335266, "learning_rate": 0.0002, "epoch": 2.4946236559139785, "step": 1160}, {"loss": 0.6142, "grad_norm": 0.9932922720909119, "learning_rate": 0.0002, "epoch": 2.5161290322580645, "step": 1170}, {"loss": 0.6407, "grad_norm": 0.7414287328720093, "learning_rate": 0.0002, "epoch": 2.5376344086021505, "step": 1180}, {"loss": 0.6161, "grad_norm": 0.8111771941184998, "learning_rate": 0.0002, "epoch": 2.5591397849462365, "step": 1190}, {"loss": 0.6006, "grad_norm": 0.7520156502723694, "learning_rate": 0.0002, "epoch": 2.5806451612903225, "step": 1200}, {"loss": 0.615, "grad_norm": 0.9022907018661499, "learning_rate": 0.0002, "epoch": 2.6021505376344085, "step": 1210}, {"loss": 0.6211, "grad_norm": 0.7746260166168213, "learning_rate": 0.0002, "epoch": 2.6236559139784945, "step": 1220}, {"loss": 0.616, "grad_norm": 0.8482862114906311, "learning_rate": 0.0002, "epoch": 2.6451612903225805, "step": 1230}, {"loss": 0.6417, "grad_norm": 0.7925458550453186, "learning_rate": 0.0002, "epoch": 2.6666666666666665, "step": 1240}, {"loss": 0.6187, "grad_norm": 0.8369929194450378, "learning_rate": 0.0002, "epoch": 2.688172043010753, "step": 1250}, {"loss": 0.6138, "grad_norm": 0.8311542868614197, "learning_rate": 0.0002, "epoch": 2.709677419354839, "step": 1260}, {"loss": 0.5894, "grad_norm": 0.7204853296279907, "learning_rate": 0.0002, "epoch": 2.731182795698925, "step": 1270}, {"loss": 0.6325, "grad_norm": 0.8447284698486328, "learning_rate": 0.0002, "epoch": 2.752688172043011, "step": 1280}, {"loss": 0.5946, "grad_norm": 0.7738404273986816, "learning_rate": 0.0002, "epoch": 2.774193548387097, "step": 1290}, {"loss": 0.5678, "grad_norm": 0.8393287062644958, "learning_rate": 0.0002, "epoch": 2.795698924731183, "step": 1300}, {"loss": 0.6092, "grad_norm": 0.79121994972229, "learning_rate": 0.0002, "epoch": 2.817204301075269, "step": 1310}, {"loss": 0.5889, "grad_norm": 0.7331557869911194, "learning_rate": 0.0002, "epoch": 2.838709677419355, "step": 1320}, {"loss": 0.6048, "grad_norm": 0.9593998193740845, "learning_rate": 0.0002, "epoch": 2.860215053763441, "step": 1330}, {"loss": 0.6108, "grad_norm": 0.7215158343315125, "learning_rate": 0.0002, "epoch": 2.881720430107527, "step": 1340}, {"loss": 0.5897, "grad_norm": 0.840404212474823, "learning_rate": 0.0002, "epoch": 2.903225806451613, "step": 1350}, {"loss": 0.6056, "grad_norm": 0.870659351348877, "learning_rate": 0.0002, "epoch": 2.924731182795699, "step": 1360}, {"loss": 0.6205, "grad_norm": 0.8744975328445435, "learning_rate": 0.0002, "epoch": 2.946236559139785, "step": 1370}, {"loss": 0.5966, "grad_norm": 0.8030612468719482, "learning_rate": 0.0002, "epoch": 2.967741935483871, "step": 1380}, {"loss": 0.6004, "grad_norm": 0.825814962387085, "learning_rate": 0.0002, "epoch": 2.989247311827957, "step": 1390}, {"eval_loss": 0.6257933378219604, "eval_runtime": 21.3692, "eval_samples_per_second": 15.49, "eval_steps_per_second": 1.965, "epoch": 3.0, "step": 1395}, {"loss": 0.5696, "grad_norm": 0.8650677800178528, "learning_rate": 0.0002, "epoch": 3.010752688172043, "step": 1400}, {"loss": 0.5483, "grad_norm": 0.8364197015762329, "learning_rate": 0.0002, "epoch": 3.032258064516129, "step": 1410}, {"loss": 0.5606, "grad_norm": 0.8278448581695557, "learning_rate": 0.0002, "epoch": 3.053763440860215, "step": 1420}, {"loss": 0.5572, "grad_norm": 0.8806642889976501, "learning_rate": 0.0002, "epoch": 3.075268817204301, "step": 1430}, {"loss": 0.585, "grad_norm": 0.8180029988288879, "learning_rate": 0.0002, "epoch": 3.096774193548387, "step": 1440}, {"loss": 0.5667, "grad_norm": 0.8561782836914062, "learning_rate": 0.0002, "epoch": 3.118279569892473, "step": 1450}, {"loss": 0.5246, "grad_norm": 0.8377029299736023, "learning_rate": 0.0002, "epoch": 3.139784946236559, "step": 1460}, {"loss": 0.5464, "grad_norm": 0.885779082775116, "learning_rate": 0.0002, "epoch": 3.161290322580645, "step": 1470}, {"loss": 0.541, "grad_norm": 0.9388518333435059, "learning_rate": 0.0002, "epoch": 3.182795698924731, "step": 1480}, {"loss": 0.5447, "grad_norm": 0.8816235661506653, "learning_rate": 0.0002, "epoch": 3.204301075268817, "step": 1490}, {"loss": 0.5466, "grad_norm": 0.9885783791542053, "learning_rate": 0.0002, "epoch": 3.225806451612903, "step": 1500}, {"loss": 0.5455, "grad_norm": 0.8635850548744202, "learning_rate": 0.0002, "epoch": 3.247311827956989, "step": 1510}, {"loss": 0.5419, "grad_norm": 0.829853355884552, "learning_rate": 0.0002, "epoch": 3.268817204301075, "step": 1520}, {"loss": 0.54, "grad_norm": 0.9037486910820007, "learning_rate": 0.0002, "epoch": 3.2903225806451615, "step": 1530}, {"loss": 0.5375, "grad_norm": 0.8173713684082031, "learning_rate": 0.0002, "epoch": 3.3118279569892475, "step": 1540}, {"loss": 0.5405, "grad_norm": 0.796953022480011, "learning_rate": 0.0002, "epoch": 3.3333333333333335, "step": 1550}, {"loss": 0.5505, "grad_norm": 0.7894400358200073, "learning_rate": 0.0002, "epoch": 3.3548387096774195, "step": 1560}, {"loss": 0.5395, "grad_norm": 0.9434949159622192, "learning_rate": 0.0002, "epoch": 3.3763440860215055, "step": 1570}, {"loss": 0.5271, "grad_norm": 0.8666760325431824, "learning_rate": 0.0002, "epoch": 3.3978494623655915, "step": 1580}, {"loss": 0.5439, "grad_norm": 0.7782467007637024, "learning_rate": 0.0002, "epoch": 3.4193548387096775, "step": 1590}, {"loss": 0.5161, "grad_norm": 0.8849126696586609, "learning_rate": 0.0002, "epoch": 3.4408602150537635, "step": 1600}, {"loss": 0.5353, "grad_norm": 0.7863831520080566, "learning_rate": 0.0002, "epoch": 3.4623655913978495, "step": 1610}, {"loss": 0.5308, "grad_norm": 1.0403116941452026, "learning_rate": 0.0002, "epoch": 3.4838709677419355, "step": 1620}, {"loss": 0.5339, "grad_norm": 0.8307499289512634, "learning_rate": 0.0002, "epoch": 3.5053763440860215, "step": 1630}, {"loss": 0.5361, "grad_norm": 0.9132118821144104, "learning_rate": 0.0002, "epoch": 3.5268817204301075, "step": 1640}, {"loss": 0.5828, "grad_norm": 0.9322578310966492, "learning_rate": 0.0002, "epoch": 3.5483870967741935, "step": 1650}, {"loss": 0.546, "grad_norm": 0.9782460331916809, "learning_rate": 0.0002, "epoch": 3.5698924731182795, "step": 1660}, {"loss": 0.5424, "grad_norm": 0.7189919352531433, "learning_rate": 0.0002, "epoch": 3.5913978494623655, "step": 1670}, {"loss": 0.5514, "grad_norm": 0.9689221382141113, "learning_rate": 0.0002, "epoch": 3.6129032258064515, "step": 1680}, {"loss": 0.5379, "grad_norm": 0.9684675335884094, "learning_rate": 0.0002, "epoch": 3.6344086021505375, "step": 1690}, {"loss": 0.5748, "grad_norm": 0.8851472735404968, "learning_rate": 0.0002, "epoch": 3.6559139784946235, "step": 1700}, {"loss": 0.5412, "grad_norm": 0.7709833383560181, "learning_rate": 0.0002, "epoch": 3.6774193548387095, "step": 1710}, {"loss": 0.521, "grad_norm": 0.818236231803894, "learning_rate": 0.0002, "epoch": 3.698924731182796, "step": 1720}, {"loss": 0.5445, "grad_norm": 0.870642364025116, "learning_rate": 0.0002, "epoch": 3.720430107526882, "step": 1730}, {"loss": 0.5307, "grad_norm": 1.0245511531829834, "learning_rate": 0.0002, "epoch": 3.741935483870968, "step": 1740}, {"loss": 0.5593, "grad_norm": 0.8607558608055115, "learning_rate": 0.0002, "epoch": 3.763440860215054, "step": 1750}, {"loss": 0.536, "grad_norm": 0.8511829972267151, "learning_rate": 0.0002, "epoch": 3.78494623655914, "step": 1760}, {"loss": 0.5193, "grad_norm": 0.7969087362289429, "learning_rate": 0.0002, "epoch": 3.806451612903226, "step": 1770}, {"loss": 0.5578, "grad_norm": 0.8457245826721191, "learning_rate": 0.0002, "epoch": 3.827956989247312, "step": 1780}, {"loss": 0.5337, "grad_norm": 0.8893467783927917, "learning_rate": 0.0002, "epoch": 3.849462365591398, "step": 1790}, {"loss": 0.5024, "grad_norm": 0.8593819737434387, "learning_rate": 0.0002, "epoch": 3.870967741935484, "step": 1800}, {"loss": 0.5134, "grad_norm": 0.7574560642242432, "learning_rate": 0.0002, "epoch": 3.89247311827957, "step": 1810}, {"loss": 0.5263, "grad_norm": 0.8681567311286926, "learning_rate": 0.0002, "epoch": 3.913978494623656, "step": 1820}, {"loss": 0.532, "grad_norm": 0.9068132042884827, "learning_rate": 0.0002, "epoch": 3.935483870967742, "step": 1830}, {"loss": 0.5427, "grad_norm": 0.8668948411941528, "learning_rate": 0.0002, "epoch": 3.956989247311828, "step": 1840}, {"loss": 0.5349, "grad_norm": 1.046032428741455, "learning_rate": 0.0002, "epoch": 3.978494623655914, "step": 1850}, {"loss": 0.5087, "grad_norm": 0.904780387878418, "learning_rate": 0.0002, "epoch": 4.0, "step": 1860}]}
+{"epoch": 5.0, "step": 2325, "epoch_duration": 451.0625660419464, "total_accumulated_duration": 2253.2721548080444, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 3020.60888671875}, "peak_memory_usage": {"GPU_0": 15051.17431640625}, "avg_memory_reserved": {"GPU_0": 20170.0}, "peak_memory_reserved": {"GPU_0": 20170.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-1860", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 3.4172, "grad_norm": 0.9075053930282593, "learning_rate": 0.0002, "epoch": 0.021505376344086023, "step": 10}, {"loss": 2.5888, "grad_norm": 1.4321208000183105, "learning_rate": 0.0002, "epoch": 0.043010752688172046, "step": 20}, {"loss": 2.1195, "grad_norm": 1.7500602006912231, "learning_rate": 0.0002, "epoch": 0.06451612903225806, "step": 30}, {"loss": 1.9303, "grad_norm": 0.7606641054153442, "learning_rate": 0.0002, "epoch": 0.08602150537634409, "step": 40}, {"loss": 1.6112, "grad_norm": 1.2754929065704346, "learning_rate": 0.0002, "epoch": 0.10752688172043011, "step": 50}, {"loss": 1.4319, "grad_norm": 1.0936230421066284, "learning_rate": 0.0002, "epoch": 0.12903225806451613, "step": 60}, {"loss": 1.3568, "grad_norm": 1.144593596458435, "learning_rate": 0.0002, "epoch": 0.15053763440860216, "step": 70}, {"loss": 1.2028, "grad_norm": 1.2181956768035889, "learning_rate": 0.0002, "epoch": 0.17204301075268819, "step": 80}, {"loss": 1.1534, "grad_norm": 1.1260095834732056, "learning_rate": 0.0002, "epoch": 0.1935483870967742, "step": 90}, {"loss": 1.1089, "grad_norm": 1.1155284643173218, "learning_rate": 0.0002, "epoch": 0.21505376344086022, "step": 100}, {"loss": 1.0883, "grad_norm": 1.089565396308899, "learning_rate": 0.0002, "epoch": 0.23655913978494625, "step": 110}, {"loss": 1.0814, "grad_norm": 0.9833471775054932, "learning_rate": 0.0002, "epoch": 0.25806451612903225, "step": 120}, {"loss": 1.0239, "grad_norm": 1.0265629291534424, "learning_rate": 0.0002, "epoch": 0.27956989247311825, "step": 130}, {"loss": 0.9888, "grad_norm": 0.9344286322593689, "learning_rate": 0.0002, "epoch": 0.3010752688172043, "step": 140}, {"loss": 1.0043, "grad_norm": 0.9883386492729187, "learning_rate": 0.0002, "epoch": 0.3225806451612903, "step": 150}, {"loss": 0.9338, "grad_norm": 0.9299277067184448, "learning_rate": 0.0002, "epoch": 0.34408602150537637, "step": 160}, {"loss": 0.9432, "grad_norm": 1.390045404434204, "learning_rate": 0.0002, "epoch": 0.3655913978494624, "step": 170}, {"loss": 0.9008, "grad_norm": 1.0313078165054321, "learning_rate": 0.0002, "epoch": 0.3870967741935484, "step": 180}, {"loss": 0.9434, "grad_norm": 1.1792205572128296, "learning_rate": 0.0002, "epoch": 0.40860215053763443, "step": 190}, {"loss": 0.8761, "grad_norm": 1.049809217453003, "learning_rate": 0.0002, "epoch": 0.43010752688172044, "step": 200}, {"loss": 0.8709, "grad_norm": 0.990111768245697, "learning_rate": 0.0002, "epoch": 0.45161290322580644, "step": 210}, {"loss": 0.905, "grad_norm": 0.9870412349700928, "learning_rate": 0.0002, "epoch": 0.4731182795698925, "step": 220}, {"loss": 0.9129, "grad_norm": 0.8557345867156982, "learning_rate": 0.0002, "epoch": 0.4946236559139785, "step": 230}, {"loss": 0.8836, "grad_norm": 0.9746861457824707, "learning_rate": 0.0002, "epoch": 0.5161290322580645, "step": 240}, {"loss": 0.873, "grad_norm": 0.9010438323020935, "learning_rate": 0.0002, "epoch": 0.5376344086021505, "step": 250}, {"loss": 0.8241, "grad_norm": 0.9061082005500793, "learning_rate": 0.0002, "epoch": 0.5591397849462365, "step": 260}, {"loss": 0.8652, "grad_norm": 0.9311846494674683, "learning_rate": 0.0002, "epoch": 0.5806451612903226, "step": 270}, {"loss": 0.8256, "grad_norm": 0.9140254855155945, "learning_rate": 0.0002, "epoch": 0.6021505376344086, "step": 280}, {"loss": 0.8441, "grad_norm": 0.9722253084182739, "learning_rate": 0.0002, "epoch": 0.6236559139784946, "step": 290}, {"loss": 0.8314, "grad_norm": 0.8539168238639832, "learning_rate": 0.0002, "epoch": 0.6451612903225806, "step": 300}, {"loss": 0.8528, "grad_norm": 0.9053162932395935, "learning_rate": 0.0002, "epoch": 0.6666666666666666, "step": 310}, {"loss": 0.8209, "grad_norm": 0.8444252610206604, "learning_rate": 0.0002, "epoch": 0.6881720430107527, "step": 320}, {"loss": 0.8101, "grad_norm": 0.8127437829971313, "learning_rate": 0.0002, "epoch": 0.7096774193548387, "step": 330}, {"loss": 0.8223, "grad_norm": 0.886555016040802, "learning_rate": 0.0002, "epoch": 0.7311827956989247, "step": 340}, {"loss": 0.8368, "grad_norm": 0.8458548784255981, "learning_rate": 0.0002, "epoch": 0.7526881720430108, "step": 350}, {"loss": 0.8295, "grad_norm": 0.8683297634124756, "learning_rate": 0.0002, "epoch": 0.7741935483870968, "step": 360}, {"loss": 0.8232, "grad_norm": 0.8308405876159668, "learning_rate": 0.0002, "epoch": 0.7956989247311828, "step": 370}, {"loss": 0.7752, "grad_norm": 0.8305579423904419, "learning_rate": 0.0002, "epoch": 0.8172043010752689, "step": 380}, {"loss": 0.8267, "grad_norm": 0.8545567393302917, "learning_rate": 0.0002, "epoch": 0.8387096774193549, "step": 390}, {"loss": 0.8212, "grad_norm": 0.8486055731773376, "learning_rate": 0.0002, "epoch": 0.8602150537634409, "step": 400}, {"loss": 0.743, "grad_norm": 0.8126763105392456, "learning_rate": 0.0002, "epoch": 0.8817204301075269, "step": 410}, {"loss": 0.7993, "grad_norm": 0.8494045734405518, "learning_rate": 0.0002, "epoch": 0.9032258064516129, "step": 420}, {"loss": 0.8213, "grad_norm": 0.7639183402061462, "learning_rate": 0.0002, "epoch": 0.9247311827956989, "step": 430}, {"loss": 0.8015, "grad_norm": 0.858101487159729, "learning_rate": 0.0002, "epoch": 0.946236559139785, "step": 440}, {"loss": 0.7629, "grad_norm": 0.8141381740570068, "learning_rate": 0.0002, "epoch": 0.967741935483871, "step": 450}, {"loss": 0.7357, "grad_norm": 0.8072513937950134, "learning_rate": 0.0002, "epoch": 0.989247311827957, "step": 460}, {"eval_loss": 0.7740864157676697, "eval_runtime": 21.383, "eval_samples_per_second": 15.48, "eval_steps_per_second": 1.964, "epoch": 1.0, "step": 465}, {"loss": 0.7701, "grad_norm": 0.8269494771957397, "learning_rate": 0.0002, "epoch": 1.010752688172043, "step": 470}, {"loss": 0.7532, "grad_norm": 0.7814009189605713, "learning_rate": 0.0002, "epoch": 1.032258064516129, "step": 480}, {"loss": 0.7689, "grad_norm": 0.8183923363685608, "learning_rate": 0.0002, "epoch": 1.053763440860215, "step": 490}, {"loss": 0.765, "grad_norm": 0.8146600723266602, "learning_rate": 0.0002, "epoch": 1.075268817204301, "step": 500}, {"loss": 0.7358, "grad_norm": 0.8635126352310181, "learning_rate": 0.0002, "epoch": 1.096774193548387, "step": 510}, {"loss": 0.7302, "grad_norm": 0.8520359396934509, "learning_rate": 0.0002, "epoch": 1.118279569892473, "step": 520}, {"loss": 0.7492, "grad_norm": 0.8026443123817444, "learning_rate": 0.0002, "epoch": 1.139784946236559, "step": 530}, {"loss": 0.7518, "grad_norm": 0.8157258629798889, "learning_rate": 0.0002, "epoch": 1.1612903225806452, "step": 540}, {"loss": 0.7461, "grad_norm": 0.9450796246528625, "learning_rate": 0.0002, "epoch": 1.1827956989247312, "step": 550}, {"loss": 0.7128, "grad_norm": 0.8859835863113403, "learning_rate": 0.0002, "epoch": 1.2043010752688172, "step": 560}, {"loss": 0.7067, "grad_norm": 0.7819921970367432, "learning_rate": 0.0002, "epoch": 1.2258064516129032, "step": 570}, {"loss": 0.7577, "grad_norm": 0.7823445796966553, "learning_rate": 0.0002, "epoch": 1.2473118279569892, "step": 580}, {"loss": 0.7358, "grad_norm": 0.7931883931159973, "learning_rate": 0.0002, "epoch": 1.2688172043010753, "step": 590}, {"loss": 0.723, "grad_norm": 0.7495734095573425, "learning_rate": 0.0002, "epoch": 1.2903225806451613, "step": 600}, {"loss": 0.7386, "grad_norm": 0.9272717237472534, "learning_rate": 0.0002, "epoch": 1.3118279569892473, "step": 610}, {"loss": 0.7498, "grad_norm": 0.7968398332595825, "learning_rate": 0.0002, "epoch": 1.3333333333333333, "step": 620}, {"loss": 0.7635, "grad_norm": 0.7813659310340881, "learning_rate": 0.0002, "epoch": 1.3548387096774195, "step": 630}, {"loss": 0.6665, "grad_norm": 0.730925977230072, "learning_rate": 0.0002, "epoch": 1.3763440860215055, "step": 640}, {"loss": 0.7037, "grad_norm": 0.8011482954025269, "learning_rate": 0.0002, "epoch": 1.3978494623655915, "step": 650}, {"loss": 0.6931, "grad_norm": 0.7770085334777832, "learning_rate": 0.0002, "epoch": 1.4193548387096775, "step": 660}, {"loss": 0.6949, "grad_norm": 0.7432682514190674, "learning_rate": 0.0002, "epoch": 1.4408602150537635, "step": 670}, {"loss": 0.7444, "grad_norm": 0.8820092678070068, "learning_rate": 0.0002, "epoch": 1.4623655913978495, "step": 680}, {"loss": 0.6758, "grad_norm": 0.7786208987236023, "learning_rate": 0.0002, "epoch": 1.4838709677419355, "step": 690}, {"loss": 0.6702, "grad_norm": 0.7467480301856995, "learning_rate": 0.0002, "epoch": 1.5053763440860215, "step": 700}, {"loss": 0.7107, "grad_norm": 0.8147122263908386, "learning_rate": 0.0002, "epoch": 1.5268817204301075, "step": 710}, {"loss": 0.7144, "grad_norm": 0.796030580997467, "learning_rate": 0.0002, "epoch": 1.5483870967741935, "step": 720}, {"loss": 0.6936, "grad_norm": 0.8776171207427979, "learning_rate": 0.0002, "epoch": 1.5698924731182795, "step": 730}, {"loss": 0.7101, "grad_norm": 0.8056126236915588, "learning_rate": 0.0002, "epoch": 1.5913978494623655, "step": 740}, {"loss": 0.7162, "grad_norm": 0.8141863346099854, "learning_rate": 0.0002, "epoch": 1.6129032258064515, "step": 750}, {"loss": 0.7088, "grad_norm": 0.8100557327270508, "learning_rate": 0.0002, "epoch": 1.6344086021505375, "step": 760}, {"loss": 0.7212, "grad_norm": 0.8283200860023499, "learning_rate": 0.0002, "epoch": 1.6559139784946235, "step": 770}, {"loss": 0.694, "grad_norm": 0.800865113735199, "learning_rate": 0.0002, "epoch": 1.6774193548387095, "step": 780}, {"loss": 0.7076, "grad_norm": 0.8052287697792053, "learning_rate": 0.0002, "epoch": 1.6989247311827957, "step": 790}, {"loss": 0.7257, "grad_norm": 0.8619674444198608, "learning_rate": 0.0002, "epoch": 1.7204301075268817, "step": 800}, {"loss": 0.7141, "grad_norm": 0.8907215595245361, "learning_rate": 0.0002, "epoch": 1.7419354838709677, "step": 810}, {"loss": 0.7035, "grad_norm": 0.6976316571235657, "learning_rate": 0.0002, "epoch": 1.7634408602150538, "step": 820}, {"loss": 0.6916, "grad_norm": 0.7533746957778931, "learning_rate": 0.0002, "epoch": 1.7849462365591398, "step": 830}, {"loss": 0.7094, "grad_norm": 0.7326804399490356, "learning_rate": 0.0002, "epoch": 1.8064516129032258, "step": 840}, {"loss": 0.6891, "grad_norm": 0.7782683372497559, "learning_rate": 0.0002, "epoch": 1.827956989247312, "step": 850}, {"loss": 0.6931, "grad_norm": 0.7424806356430054, "learning_rate": 0.0002, "epoch": 1.849462365591398, "step": 860}, {"loss": 0.7354, "grad_norm": 1.172325611114502, "learning_rate": 0.0002, "epoch": 1.870967741935484, "step": 870}, {"loss": 0.6866, "grad_norm": 0.771058201789856, "learning_rate": 0.0002, "epoch": 1.89247311827957, "step": 880}, {"loss": 0.7296, "grad_norm": 0.8624904155731201, "learning_rate": 0.0002, "epoch": 1.913978494623656, "step": 890}, {"loss": 0.7233, "grad_norm": 0.7062820792198181, "learning_rate": 0.0002, "epoch": 1.935483870967742, "step": 900}, {"loss": 0.6966, "grad_norm": 0.7560103535652161, "learning_rate": 0.0002, "epoch": 1.956989247311828, "step": 910}, {"loss": 0.69, "grad_norm": 0.788899838924408, "learning_rate": 0.0002, "epoch": 1.978494623655914, "step": 920}, {"loss": 0.6505, "grad_norm": 0.6562113761901855, "learning_rate": 0.0002, "epoch": 2.0, "step": 930}, {"eval_loss": 0.6885261535644531, "eval_runtime": 21.4291, "eval_samples_per_second": 15.446, "eval_steps_per_second": 1.96, "epoch": 2.0, "step": 930}, {"loss": 0.6625, "grad_norm": 0.8216531872749329, "learning_rate": 0.0002, "epoch": 2.021505376344086, "step": 940}, {"loss": 0.6398, "grad_norm": 0.8317142724990845, "learning_rate": 0.0002, "epoch": 2.043010752688172, "step": 950}, {"loss": 0.649, "grad_norm": 0.8446708917617798, "learning_rate": 0.0002, "epoch": 2.064516129032258, "step": 960}, {"loss": 0.657, "grad_norm": 0.735055148601532, "learning_rate": 0.0002, "epoch": 2.086021505376344, "step": 970}, {"loss": 0.649, "grad_norm": 0.7487243413925171, "learning_rate": 0.0002, "epoch": 2.10752688172043, "step": 980}, {"loss": 0.6419, "grad_norm": 0.8573887944221497, "learning_rate": 0.0002, "epoch": 2.129032258064516, "step": 990}, {"loss": 0.6431, "grad_norm": 0.6284521818161011, "learning_rate": 0.0002, "epoch": 2.150537634408602, "step": 1000}, {"loss": 0.6128, "grad_norm": 0.754183292388916, "learning_rate": 0.0002, "epoch": 2.172043010752688, "step": 1010}, {"loss": 0.6253, "grad_norm": 0.9445359110832214, "learning_rate": 0.0002, "epoch": 2.193548387096774, "step": 1020}, {"loss": 0.605, "grad_norm": 0.808508038520813, "learning_rate": 0.0002, "epoch": 2.21505376344086, "step": 1030}, {"loss": 0.6786, "grad_norm": 0.9394679665565491, "learning_rate": 0.0002, "epoch": 2.236559139784946, "step": 1040}, {"loss": 0.6176, "grad_norm": 0.8151357769966125, "learning_rate": 0.0002, "epoch": 2.258064516129032, "step": 1050}, {"loss": 0.66, "grad_norm": 0.7909848093986511, "learning_rate": 0.0002, "epoch": 2.279569892473118, "step": 1060}, {"loss": 0.6254, "grad_norm": 0.7506507039070129, "learning_rate": 0.0002, "epoch": 2.3010752688172045, "step": 1070}, {"loss": 0.6608, "grad_norm": 0.8240520358085632, "learning_rate": 0.0002, "epoch": 2.3225806451612905, "step": 1080}, {"loss": 0.6207, "grad_norm": 0.9342400431632996, "learning_rate": 0.0002, "epoch": 2.3440860215053765, "step": 1090}, {"loss": 0.6029, "grad_norm": 1.0598735809326172, "learning_rate": 0.0002, "epoch": 2.3655913978494625, "step": 1100}, {"loss": 0.6035, "grad_norm": 0.7907650470733643, "learning_rate": 0.0002, "epoch": 2.3870967741935485, "step": 1110}, {"loss": 0.6237, "grad_norm": 0.9388798475265503, "learning_rate": 0.0002, "epoch": 2.4086021505376345, "step": 1120}, {"loss": 0.6207, "grad_norm": 0.8985419869422913, "learning_rate": 0.0002, "epoch": 2.4301075268817205, "step": 1130}, {"loss": 0.5902, "grad_norm": 0.7471932768821716, "learning_rate": 0.0002, "epoch": 2.4516129032258065, "step": 1140}, {"loss": 0.6446, "grad_norm": 0.761131763458252, "learning_rate": 0.0002, "epoch": 2.4731182795698925, "step": 1150}, {"loss": 0.6088, "grad_norm": 0.7901819348335266, "learning_rate": 0.0002, "epoch": 2.4946236559139785, "step": 1160}, {"loss": 0.6142, "grad_norm": 0.9932922720909119, "learning_rate": 0.0002, "epoch": 2.5161290322580645, "step": 1170}, {"loss": 0.6407, "grad_norm": 0.7414287328720093, "learning_rate": 0.0002, "epoch": 2.5376344086021505, "step": 1180}, {"loss": 0.6161, "grad_norm": 0.8111771941184998, "learning_rate": 0.0002, "epoch": 2.5591397849462365, "step": 1190}, {"loss": 0.6006, "grad_norm": 0.7520156502723694, "learning_rate": 0.0002, "epoch": 2.5806451612903225, "step": 1200}, {"loss": 0.615, "grad_norm": 0.9022907018661499, "learning_rate": 0.0002, "epoch": 2.6021505376344085, "step": 1210}, {"loss": 0.6211, "grad_norm": 0.7746260166168213, "learning_rate": 0.0002, "epoch": 2.6236559139784945, "step": 1220}, {"loss": 0.616, "grad_norm": 0.8482862114906311, "learning_rate": 0.0002, "epoch": 2.6451612903225805, "step": 1230}, {"loss": 0.6417, "grad_norm": 0.7925458550453186, "learning_rate": 0.0002, "epoch": 2.6666666666666665, "step": 1240}, {"loss": 0.6187, "grad_norm": 0.8369929194450378, "learning_rate": 0.0002, "epoch": 2.688172043010753, "step": 1250}, {"loss": 0.6138, "grad_norm": 0.8311542868614197, "learning_rate": 0.0002, "epoch": 2.709677419354839, "step": 1260}, {"loss": 0.5894, "grad_norm": 0.7204853296279907, "learning_rate": 0.0002, "epoch": 2.731182795698925, "step": 1270}, {"loss": 0.6325, "grad_norm": 0.8447284698486328, "learning_rate": 0.0002, "epoch": 2.752688172043011, "step": 1280}, {"loss": 0.5946, "grad_norm": 0.7738404273986816, "learning_rate": 0.0002, "epoch": 2.774193548387097, "step": 1290}, {"loss": 0.5678, "grad_norm": 0.8393287062644958, "learning_rate": 0.0002, "epoch": 2.795698924731183, "step": 1300}, {"loss": 0.6092, "grad_norm": 0.79121994972229, "learning_rate": 0.0002, "epoch": 2.817204301075269, "step": 1310}, {"loss": 0.5889, "grad_norm": 0.7331557869911194, "learning_rate": 0.0002, "epoch": 2.838709677419355, "step": 1320}, {"loss": 0.6048, "grad_norm": 0.9593998193740845, "learning_rate": 0.0002, "epoch": 2.860215053763441, "step": 1330}, {"loss": 0.6108, "grad_norm": 0.7215158343315125, "learning_rate": 0.0002, "epoch": 2.881720430107527, "step": 1340}, {"loss": 0.5897, "grad_norm": 0.840404212474823, "learning_rate": 0.0002, "epoch": 2.903225806451613, "step": 1350}, {"loss": 0.6056, "grad_norm": 0.870659351348877, "learning_rate": 0.0002, "epoch": 2.924731182795699, "step": 1360}, {"loss": 0.6205, "grad_norm": 0.8744975328445435, "learning_rate": 0.0002, "epoch": 2.946236559139785, "step": 1370}, {"loss": 0.5966, "grad_norm": 0.8030612468719482, "learning_rate": 0.0002, "epoch": 2.967741935483871, "step": 1380}, {"loss": 0.6004, "grad_norm": 0.825814962387085, "learning_rate": 0.0002, "epoch": 2.989247311827957, "step": 1390}, {"eval_loss": 0.6257933378219604, "eval_runtime": 21.3692, "eval_samples_per_second": 15.49, "eval_steps_per_second": 1.965, "epoch": 3.0, "step": 1395}, {"loss": 0.5696, "grad_norm": 0.8650677800178528, "learning_rate": 0.0002, "epoch": 3.010752688172043, "step": 1400}, {"loss": 0.5483, "grad_norm": 0.8364197015762329, "learning_rate": 0.0002, "epoch": 3.032258064516129, "step": 1410}, {"loss": 0.5606, "grad_norm": 0.8278448581695557, "learning_rate": 0.0002, "epoch": 3.053763440860215, "step": 1420}, {"loss": 0.5572, "grad_norm": 0.8806642889976501, "learning_rate": 0.0002, "epoch": 3.075268817204301, "step": 1430}, {"loss": 0.585, "grad_norm": 0.8180029988288879, "learning_rate": 0.0002, "epoch": 3.096774193548387, "step": 1440}, {"loss": 0.5667, "grad_norm": 0.8561782836914062, "learning_rate": 0.0002, "epoch": 3.118279569892473, "step": 1450}, {"loss": 0.5246, "grad_norm": 0.8377029299736023, "learning_rate": 0.0002, "epoch": 3.139784946236559, "step": 1460}, {"loss": 0.5464, "grad_norm": 0.885779082775116, "learning_rate": 0.0002, "epoch": 3.161290322580645, "step": 1470}, {"loss": 0.541, "grad_norm": 0.9388518333435059, "learning_rate": 0.0002, "epoch": 3.182795698924731, "step": 1480}, {"loss": 0.5447, "grad_norm": 0.8816235661506653, "learning_rate": 0.0002, "epoch": 3.204301075268817, "step": 1490}, {"loss": 0.5466, "grad_norm": 0.9885783791542053, "learning_rate": 0.0002, "epoch": 3.225806451612903, "step": 1500}, {"loss": 0.5455, "grad_norm": 0.8635850548744202, "learning_rate": 0.0002, "epoch": 3.247311827956989, "step": 1510}, {"loss": 0.5419, "grad_norm": 0.829853355884552, "learning_rate": 0.0002, "epoch": 3.268817204301075, "step": 1520}, {"loss": 0.54, "grad_norm": 0.9037486910820007, "learning_rate": 0.0002, "epoch": 3.2903225806451615, "step": 1530}, {"loss": 0.5375, "grad_norm": 0.8173713684082031, "learning_rate": 0.0002, "epoch": 3.3118279569892475, "step": 1540}, {"loss": 0.5405, "grad_norm": 0.796953022480011, "learning_rate": 0.0002, "epoch": 3.3333333333333335, "step": 1550}, {"loss": 0.5505, "grad_norm": 0.7894400358200073, "learning_rate": 0.0002, "epoch": 3.3548387096774195, "step": 1560}, {"loss": 0.5395, "grad_norm": 0.9434949159622192, "learning_rate": 0.0002, "epoch": 3.3763440860215055, "step": 1570}, {"loss": 0.5271, "grad_norm": 0.8666760325431824, "learning_rate": 0.0002, "epoch": 3.3978494623655915, "step": 1580}, {"loss": 0.5439, "grad_norm": 0.7782467007637024, "learning_rate": 0.0002, "epoch": 3.4193548387096775, "step": 1590}, {"loss": 0.5161, "grad_norm": 0.8849126696586609, "learning_rate": 0.0002, "epoch": 3.4408602150537635, "step": 1600}, {"loss": 0.5353, "grad_norm": 0.7863831520080566, "learning_rate": 0.0002, "epoch": 3.4623655913978495, "step": 1610}, {"loss": 0.5308, "grad_norm": 1.0403116941452026, "learning_rate": 0.0002, "epoch": 3.4838709677419355, "step": 1620}, {"loss": 0.5339, "grad_norm": 0.8307499289512634, "learning_rate": 0.0002, "epoch": 3.5053763440860215, "step": 1630}, {"loss": 0.5361, "grad_norm": 0.9132118821144104, "learning_rate": 0.0002, "epoch": 3.5268817204301075, "step": 1640}, {"loss": 0.5828, "grad_norm": 0.9322578310966492, "learning_rate": 0.0002, "epoch": 3.5483870967741935, "step": 1650}, {"loss": 0.546, "grad_norm": 0.9782460331916809, "learning_rate": 0.0002, "epoch": 3.5698924731182795, "step": 1660}, {"loss": 0.5424, "grad_norm": 0.7189919352531433, "learning_rate": 0.0002, "epoch": 3.5913978494623655, "step": 1670}, {"loss": 0.5514, "grad_norm": 0.9689221382141113, "learning_rate": 0.0002, "epoch": 3.6129032258064515, "step": 1680}, {"loss": 0.5379, "grad_norm": 0.9684675335884094, "learning_rate": 0.0002, "epoch": 3.6344086021505375, "step": 1690}, {"loss": 0.5748, "grad_norm": 0.8851472735404968, "learning_rate": 0.0002, "epoch": 3.6559139784946235, "step": 1700}, {"loss": 0.5412, "grad_norm": 0.7709833383560181, "learning_rate": 0.0002, "epoch": 3.6774193548387095, "step": 1710}, {"loss": 0.521, "grad_norm": 0.818236231803894, "learning_rate": 0.0002, "epoch": 3.698924731182796, "step": 1720}, {"loss": 0.5445, "grad_norm": 0.870642364025116, "learning_rate": 0.0002, "epoch": 3.720430107526882, "step": 1730}, {"loss": 0.5307, "grad_norm": 1.0245511531829834, "learning_rate": 0.0002, "epoch": 3.741935483870968, "step": 1740}, {"loss": 0.5593, "grad_norm": 0.8607558608055115, "learning_rate": 0.0002, "epoch": 3.763440860215054, "step": 1750}, {"loss": 0.536, "grad_norm": 0.8511829972267151, "learning_rate": 0.0002, "epoch": 3.78494623655914, "step": 1760}, {"loss": 0.5193, "grad_norm": 0.7969087362289429, "learning_rate": 0.0002, "epoch": 3.806451612903226, "step": 1770}, {"loss": 0.5578, "grad_norm": 0.8457245826721191, "learning_rate": 0.0002, "epoch": 3.827956989247312, "step": 1780}, {"loss": 0.5337, "grad_norm": 0.8893467783927917, "learning_rate": 0.0002, "epoch": 3.849462365591398, "step": 1790}, {"loss": 0.5024, "grad_norm": 0.8593819737434387, "learning_rate": 0.0002, "epoch": 3.870967741935484, "step": 1800}, {"loss": 0.5134, "grad_norm": 0.7574560642242432, "learning_rate": 0.0002, "epoch": 3.89247311827957, "step": 1810}, {"loss": 0.5263, "grad_norm": 0.8681567311286926, "learning_rate": 0.0002, "epoch": 3.913978494623656, "step": 1820}, {"loss": 0.532, "grad_norm": 0.9068132042884827, "learning_rate": 0.0002, "epoch": 3.935483870967742, "step": 1830}, {"loss": 0.5427, "grad_norm": 0.8668948411941528, "learning_rate": 0.0002, "epoch": 3.956989247311828, "step": 1840}, {"loss": 0.5349, "grad_norm": 1.046032428741455, "learning_rate": 0.0002, "epoch": 3.978494623655914, "step": 1850}, {"loss": 0.5087, "grad_norm": 0.904780387878418, "learning_rate": 0.0002, "epoch": 4.0, "step": 1860}, {"eval_loss": 0.5737715363502502, "eval_runtime": 21.4915, "eval_samples_per_second": 15.401, "eval_steps_per_second": 1.954, "epoch": 4.0, "step": 1860}, {"loss": 0.4843, "grad_norm": 0.8611752986907959, "learning_rate": 0.0002, "epoch": 4.021505376344086, "step": 1870}, {"loss": 0.4814, "grad_norm": 0.838782548904419, "learning_rate": 0.0002, "epoch": 4.043010752688172, "step": 1880}, {"loss": 0.474, "grad_norm": 0.9119709134101868, "learning_rate": 0.0002, "epoch": 4.064516129032258, "step": 1890}, {"loss": 0.4951, "grad_norm": 0.8026251196861267, "learning_rate": 0.0002, "epoch": 4.086021505376344, "step": 1900}, {"loss": 0.491, "grad_norm": 0.8773705363273621, "learning_rate": 0.0002, "epoch": 4.10752688172043, "step": 1910}, {"loss": 0.474, "grad_norm": 0.8762255907058716, "learning_rate": 0.0002, "epoch": 4.129032258064516, "step": 1920}, {"loss": 0.4816, "grad_norm": 0.8371861577033997, "learning_rate": 0.0002, "epoch": 4.150537634408602, "step": 1930}, {"loss": 0.472, "grad_norm": 0.9703728556632996, "learning_rate": 0.0002, "epoch": 4.172043010752688, "step": 1940}, {"loss": 0.4772, "grad_norm": 0.8802874684333801, "learning_rate": 0.0002, "epoch": 4.193548387096774, "step": 1950}, {"loss": 0.5032, "grad_norm": 1.0103057622909546, "learning_rate": 0.0002, "epoch": 4.21505376344086, "step": 1960}, {"loss": 0.4945, "grad_norm": 0.9212995171546936, "learning_rate": 0.0002, "epoch": 4.236559139784946, "step": 1970}, {"loss": 0.4753, "grad_norm": 1.009544849395752, "learning_rate": 0.0002, "epoch": 4.258064516129032, "step": 1980}, {"loss": 0.4789, "grad_norm": 0.8535077571868896, "learning_rate": 0.0002, "epoch": 4.279569892473118, "step": 1990}, {"loss": 0.4782, "grad_norm": 0.8363022804260254, "learning_rate": 0.0002, "epoch": 4.301075268817204, "step": 2000}, {"loss": 0.4875, "grad_norm": 0.9041762948036194, "learning_rate": 0.0002, "epoch": 4.32258064516129, "step": 2010}, {"loss": 0.4779, "grad_norm": 0.960790753364563, "learning_rate": 0.0002, "epoch": 4.344086021505376, "step": 2020}, {"loss": 0.4626, "grad_norm": 0.8823095560073853, "learning_rate": 0.0002, "epoch": 4.365591397849462, "step": 2030}, {"loss": 0.4883, "grad_norm": 0.952100396156311, "learning_rate": 0.0002, "epoch": 4.387096774193548, "step": 2040}, {"loss": 0.4789, "grad_norm": 1.0793498754501343, "learning_rate": 0.0002, "epoch": 4.408602150537634, "step": 2050}, {"loss": 0.4827, "grad_norm": 0.8987208008766174, "learning_rate": 0.0002, "epoch": 4.43010752688172, "step": 2060}, {"loss": 0.4594, "grad_norm": 0.8539772033691406, "learning_rate": 0.0002, "epoch": 4.451612903225806, "step": 2070}, {"loss": 0.4752, "grad_norm": 0.9160863757133484, "learning_rate": 0.0002, "epoch": 4.473118279569892, "step": 2080}, {"loss": 0.5033, "grad_norm": 0.9946850538253784, "learning_rate": 0.0002, "epoch": 4.494623655913978, "step": 2090}, {"loss": 0.4842, "grad_norm": 0.908039391040802, "learning_rate": 0.0002, "epoch": 4.516129032258064, "step": 2100}, {"loss": 0.4861, "grad_norm": 1.1462254524230957, "learning_rate": 0.0002, "epoch": 4.53763440860215, "step": 2110}, {"loss": 0.4892, "grad_norm": 0.8392056226730347, "learning_rate": 0.0002, "epoch": 4.559139784946236, "step": 2120}, {"loss": 0.4824, "grad_norm": 0.9673896431922913, "learning_rate": 0.0002, "epoch": 4.580645161290323, "step": 2130}, {"loss": 0.4665, "grad_norm": 0.9047091603279114, "learning_rate": 0.0002, "epoch": 4.602150537634409, "step": 2140}, {"loss": 0.4714, "grad_norm": 0.9013425707817078, "learning_rate": 0.0002, "epoch": 4.623655913978495, "step": 2150}, {"loss": 0.472, "grad_norm": 0.8899165391921997, "learning_rate": 0.0002, "epoch": 4.645161290322581, "step": 2160}, {"loss": 0.4635, "grad_norm": 0.748602569103241, "learning_rate": 0.0002, "epoch": 4.666666666666667, "step": 2170}, {"loss": 0.4695, "grad_norm": 0.8694155216217041, "learning_rate": 0.0002, "epoch": 4.688172043010753, "step": 2180}, {"loss": 0.4929, "grad_norm": 0.9134316444396973, "learning_rate": 0.0002, "epoch": 4.709677419354839, "step": 2190}, {"loss": 0.4855, "grad_norm": 0.8504763245582581, "learning_rate": 0.0002, "epoch": 4.731182795698925, "step": 2200}, {"loss": 0.4517, "grad_norm": 1.0321544408798218, "learning_rate": 0.0002, "epoch": 4.752688172043011, "step": 2210}, {"loss": 0.4796, "grad_norm": 0.9368237257003784, "learning_rate": 0.0002, "epoch": 4.774193548387097, "step": 2220}, {"loss": 0.4837, "grad_norm": 0.9319947361946106, "learning_rate": 0.0002, "epoch": 4.795698924731183, "step": 2230}, {"loss": 0.4696, "grad_norm": 0.904333770275116, "learning_rate": 0.0002, "epoch": 4.817204301075269, "step": 2240}, {"loss": 0.4746, "grad_norm": 0.8097078204154968, "learning_rate": 0.0002, "epoch": 4.838709677419355, "step": 2250}, {"loss": 0.4438, "grad_norm": 0.9128859043121338, "learning_rate": 0.0002, "epoch": 4.860215053763441, "step": 2260}, {"loss": 0.4693, "grad_norm": 0.883129894733429, "learning_rate": 0.0002, "epoch": 4.881720430107527, "step": 2270}, {"loss": 0.4494, "grad_norm": 0.85712730884552, "learning_rate": 0.0002, "epoch": 4.903225806451613, "step": 2280}, {"loss": 0.4593, "grad_norm": 1.2101863622665405, "learning_rate": 0.0002, "epoch": 4.924731182795699, "step": 2290}, {"loss": 0.4779, "grad_norm": 0.917966902256012, "learning_rate": 0.0002, "epoch": 4.946236559139785, "step": 2300}, {"loss": 0.4666, "grad_norm": 0.7740724086761475, "learning_rate": 0.0002, "epoch": 4.967741935483871, "step": 2310}, {"loss": 0.4629, "grad_norm": 1.0199906826019287, "learning_rate": 0.0002, "epoch": 4.989247311827957, "step": 2320}]}
+{"epoch": 6.0, "step": 2790, "epoch_duration": 449.90328192710876, "total_accumulated_duration": 2703.175436735153, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 3020.60888671875}, "peak_memory_usage": {"GPU_0": 15051.17431640625}, "avg_memory_reserved": {"GPU_0": 20170.0}, "peak_memory_reserved": {"GPU_0": 20170.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2325", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 3.4172, "grad_norm": 0.9075053930282593, "learning_rate": 0.0002, "epoch": 0.021505376344086023, "step": 10}, {"loss": 2.5888, "grad_norm": 1.4321208000183105, "learning_rate": 0.0002, "epoch": 0.043010752688172046, "step": 20}, {"loss": 2.1195, "grad_norm": 1.7500602006912231, "learning_rate": 0.0002, "epoch": 0.06451612903225806, "step": 30}, {"loss": 1.9303, "grad_norm": 0.7606641054153442, "learning_rate": 0.0002, "epoch": 0.08602150537634409, "step": 40}, {"loss": 1.6112, "grad_norm": 1.2754929065704346, "learning_rate": 0.0002, "epoch": 0.10752688172043011, "step": 50}, {"loss": 1.4319, "grad_norm": 1.0936230421066284, "learning_rate": 0.0002, "epoch": 0.12903225806451613, "step": 60}, {"loss": 1.3568, "grad_norm": 1.144593596458435, "learning_rate": 0.0002, "epoch": 0.15053763440860216, "step": 70}, {"loss": 1.2028, "grad_norm": 1.2181956768035889, "learning_rate": 0.0002, "epoch": 0.17204301075268819, "step": 80}, {"loss": 1.1534, "grad_norm": 1.1260095834732056, "learning_rate": 0.0002, "epoch": 0.1935483870967742, "step": 90}, {"loss": 1.1089, "grad_norm": 1.1155284643173218, "learning_rate": 0.0002, "epoch": 0.21505376344086022, "step": 100}, {"loss": 1.0883, "grad_norm": 1.089565396308899, "learning_rate": 0.0002, "epoch": 0.23655913978494625, "step": 110}, {"loss": 1.0814, "grad_norm": 0.9833471775054932, "learning_rate": 0.0002, "epoch": 0.25806451612903225, "step": 120}, {"loss": 1.0239, "grad_norm": 1.0265629291534424, "learning_rate": 0.0002, "epoch": 0.27956989247311825, "step": 130}, {"loss": 0.9888, "grad_norm": 0.9344286322593689, "learning_rate": 0.0002, "epoch": 0.3010752688172043, "step": 140}, {"loss": 1.0043, "grad_norm": 0.9883386492729187, "learning_rate": 0.0002, "epoch": 0.3225806451612903, "step": 150}, {"loss": 0.9338, "grad_norm": 0.9299277067184448, "learning_rate": 0.0002, "epoch": 0.34408602150537637, "step": 160}, {"loss": 0.9432, "grad_norm": 1.390045404434204, "learning_rate": 0.0002, "epoch": 0.3655913978494624, "step": 170}, {"loss": 0.9008, "grad_norm": 1.0313078165054321, "learning_rate": 0.0002, "epoch": 0.3870967741935484, "step": 180}, {"loss": 0.9434, "grad_norm": 1.1792205572128296, "learning_rate": 0.0002, "epoch": 0.40860215053763443, "step": 190}, {"loss": 0.8761, "grad_norm": 1.049809217453003, "learning_rate": 0.0002, "epoch": 0.43010752688172044, "step": 200}, {"loss": 0.8709, "grad_norm": 0.990111768245697, "learning_rate": 0.0002, "epoch": 0.45161290322580644, "step": 210}, {"loss": 0.905, "grad_norm": 0.9870412349700928, "learning_rate": 0.0002, "epoch": 0.4731182795698925, "step": 220}, {"loss": 0.9129, "grad_norm": 0.8557345867156982, "learning_rate": 0.0002, "epoch": 0.4946236559139785, "step": 230}, {"loss": 0.8836, "grad_norm": 0.9746861457824707, "learning_rate": 0.0002, "epoch": 0.5161290322580645, "step": 240}, {"loss": 0.873, "grad_norm": 0.9010438323020935, "learning_rate": 0.0002, "epoch": 0.5376344086021505, "step": 250}, {"loss": 0.8241, "grad_norm": 0.9061082005500793, "learning_rate": 0.0002, "epoch": 0.5591397849462365, "step": 260}, {"loss": 0.8652, "grad_norm": 0.9311846494674683, "learning_rate": 0.0002, "epoch": 0.5806451612903226, "step": 270}, {"loss": 0.8256, "grad_norm": 0.9140254855155945, "learning_rate": 0.0002, "epoch": 0.6021505376344086, "step": 280}, {"loss": 0.8441, "grad_norm": 0.9722253084182739, "learning_rate": 0.0002, "epoch": 0.6236559139784946, "step": 290}, {"loss": 0.8314, "grad_norm": 0.8539168238639832, "learning_rate": 0.0002, "epoch": 0.6451612903225806, "step": 300}, {"loss": 0.8528, "grad_norm": 0.9053162932395935, "learning_rate": 0.0002, "epoch": 0.6666666666666666, "step": 310}, {"loss": 0.8209, "grad_norm": 0.8444252610206604, "learning_rate": 0.0002, "epoch": 0.6881720430107527, "step": 320}, {"loss": 0.8101, "grad_norm": 0.8127437829971313, "learning_rate": 0.0002, "epoch": 0.7096774193548387, "step": 330}, {"loss": 0.8223, "grad_norm": 0.886555016040802, "learning_rate": 0.0002, "epoch": 0.7311827956989247, "step": 340}, {"loss": 0.8368, "grad_norm": 0.8458548784255981, "learning_rate": 0.0002, "epoch": 0.7526881720430108, "step": 350}, {"loss": 0.8295, "grad_norm": 0.8683297634124756, "learning_rate": 0.0002, "epoch": 0.7741935483870968, "step": 360}, {"loss": 0.8232, "grad_norm": 0.8308405876159668, "learning_rate": 0.0002, "epoch": 0.7956989247311828, "step": 370}, {"loss": 0.7752, "grad_norm": 0.8305579423904419, "learning_rate": 0.0002, "epoch": 0.8172043010752689, "step": 380}, {"loss": 0.8267, "grad_norm": 0.8545567393302917, "learning_rate": 0.0002, "epoch": 0.8387096774193549, "step": 390}, {"loss": 0.8212, "grad_norm": 0.8486055731773376, "learning_rate": 0.0002, "epoch": 0.8602150537634409, "step": 400}, {"loss": 0.743, "grad_norm": 0.8126763105392456, "learning_rate": 0.0002, "epoch": 0.8817204301075269, "step": 410}, {"loss": 0.7993, "grad_norm": 0.8494045734405518, "learning_rate": 0.0002, "epoch": 0.9032258064516129, "step": 420}, {"loss": 0.8213, "grad_norm": 0.7639183402061462, "learning_rate": 0.0002, "epoch": 0.9247311827956989, "step": 430}, {"loss": 0.8015, "grad_norm": 0.858101487159729, "learning_rate": 0.0002, "epoch": 0.946236559139785, "step": 440}, {"loss": 0.7629, "grad_norm": 0.8141381740570068, "learning_rate": 0.0002, "epoch": 0.967741935483871, "step": 450}, {"loss": 0.7357, "grad_norm": 0.8072513937950134, "learning_rate": 0.0002, "epoch": 0.989247311827957, "step": 460}, {"eval_loss": 0.7740864157676697, "eval_runtime": 21.383, "eval_samples_per_second": 15.48, "eval_steps_per_second": 1.964, "epoch": 1.0, "step": 465}, {"loss": 0.7701, "grad_norm": 0.8269494771957397, "learning_rate": 0.0002, "epoch": 1.010752688172043, "step": 470}, {"loss": 0.7532, "grad_norm": 0.7814009189605713, "learning_rate": 0.0002, "epoch": 1.032258064516129, "step": 480}, {"loss": 0.7689, "grad_norm": 0.8183923363685608, "learning_rate": 0.0002, "epoch": 1.053763440860215, "step": 490}, {"loss": 0.765, "grad_norm": 0.8146600723266602, "learning_rate": 0.0002, "epoch": 1.075268817204301, "step": 500}, {"loss": 0.7358, "grad_norm": 0.8635126352310181, "learning_rate": 0.0002, "epoch": 1.096774193548387, "step": 510}, {"loss": 0.7302, "grad_norm": 0.8520359396934509, "learning_rate": 0.0002, "epoch": 1.118279569892473, "step": 520}, {"loss": 0.7492, "grad_norm": 0.8026443123817444, "learning_rate": 0.0002, "epoch": 1.139784946236559, "step": 530}, {"loss": 0.7518, "grad_norm": 0.8157258629798889, "learning_rate": 0.0002, "epoch": 1.1612903225806452, "step": 540}, {"loss": 0.7461, "grad_norm": 0.9450796246528625, "learning_rate": 0.0002, "epoch": 1.1827956989247312, "step": 550}, {"loss": 0.7128, "grad_norm": 0.8859835863113403, "learning_rate": 0.0002, "epoch": 1.2043010752688172, "step": 560}, {"loss": 0.7067, "grad_norm": 0.7819921970367432, "learning_rate": 0.0002, "epoch": 1.2258064516129032, "step": 570}, {"loss": 0.7577, "grad_norm": 0.7823445796966553, "learning_rate": 0.0002, "epoch": 1.2473118279569892, "step": 580}, {"loss": 0.7358, "grad_norm": 0.7931883931159973, "learning_rate": 0.0002, "epoch": 1.2688172043010753, "step": 590}, {"loss": 0.723, "grad_norm": 0.7495734095573425, "learning_rate": 0.0002, "epoch": 1.2903225806451613, "step": 600}, {"loss": 0.7386, "grad_norm": 0.9272717237472534, "learning_rate": 0.0002, "epoch": 1.3118279569892473, "step": 610}, {"loss": 0.7498, "grad_norm": 0.7968398332595825, "learning_rate": 0.0002, "epoch": 1.3333333333333333, "step": 620}, {"loss": 0.7635, "grad_norm": 0.7813659310340881, "learning_rate": 0.0002, "epoch": 1.3548387096774195, "step": 630}, {"loss": 0.6665, "grad_norm": 0.730925977230072, "learning_rate": 0.0002, "epoch": 1.3763440860215055, "step": 640}, {"loss": 0.7037, "grad_norm": 0.8011482954025269, "learning_rate": 0.0002, "epoch": 1.3978494623655915, "step": 650}, {"loss": 0.6931, "grad_norm": 0.7770085334777832, "learning_rate": 0.0002, "epoch": 1.4193548387096775, "step": 660}, {"loss": 0.6949, "grad_norm": 0.7432682514190674, "learning_rate": 0.0002, "epoch": 1.4408602150537635, "step": 670}, {"loss": 0.7444, "grad_norm": 0.8820092678070068, "learning_rate": 0.0002, "epoch": 1.4623655913978495, "step": 680}, {"loss": 0.6758, "grad_norm": 0.7786208987236023, "learning_rate": 0.0002, "epoch": 1.4838709677419355, "step": 690}, {"loss": 0.6702, "grad_norm": 0.7467480301856995, "learning_rate": 0.0002, "epoch": 1.5053763440860215, "step": 700}, {"loss": 0.7107, "grad_norm": 0.8147122263908386, "learning_rate": 0.0002, "epoch": 1.5268817204301075, "step": 710}, {"loss": 0.7144, "grad_norm": 0.796030580997467, "learning_rate": 0.0002, "epoch": 1.5483870967741935, "step": 720}, {"loss": 0.6936, "grad_norm": 0.8776171207427979, "learning_rate": 0.0002, "epoch": 1.5698924731182795, "step": 730}, {"loss": 0.7101, "grad_norm": 0.8056126236915588, "learning_rate": 0.0002, "epoch": 1.5913978494623655, "step": 740}, {"loss": 0.7162, "grad_norm": 0.8141863346099854, "learning_rate": 0.0002, "epoch": 1.6129032258064515, "step": 750}, {"loss": 0.7088, "grad_norm": 0.8100557327270508, "learning_rate": 0.0002, "epoch": 1.6344086021505375, "step": 760}, {"loss": 0.7212, "grad_norm": 0.8283200860023499, "learning_rate": 0.0002, "epoch": 1.6559139784946235, "step": 770}, {"loss": 0.694, "grad_norm": 0.800865113735199, "learning_rate": 0.0002, "epoch": 1.6774193548387095, "step": 780}, {"loss": 0.7076, "grad_norm": 0.8052287697792053, "learning_rate": 0.0002, "epoch": 1.6989247311827957, "step": 790}, {"loss": 0.7257, "grad_norm": 0.8619674444198608, "learning_rate": 0.0002, "epoch": 1.7204301075268817, "step": 800}, {"loss": 0.7141, "grad_norm": 0.8907215595245361, "learning_rate": 0.0002, "epoch": 1.7419354838709677, "step": 810}, {"loss": 0.7035, "grad_norm": 0.6976316571235657, "learning_rate": 0.0002, "epoch": 1.7634408602150538, "step": 820}, {"loss": 0.6916, "grad_norm": 0.7533746957778931, "learning_rate": 0.0002, "epoch": 1.7849462365591398, "step": 830}, {"loss": 0.7094, "grad_norm": 0.7326804399490356, "learning_rate": 0.0002, "epoch": 1.8064516129032258, "step": 840}, {"loss": 0.6891, "grad_norm": 0.7782683372497559, "learning_rate": 0.0002, "epoch": 1.827956989247312, "step": 850}, {"loss": 0.6931, "grad_norm": 0.7424806356430054, "learning_rate": 0.0002, "epoch": 1.849462365591398, "step": 860}, {"loss": 0.7354, "grad_norm": 1.172325611114502, "learning_rate": 0.0002, "epoch": 1.870967741935484, "step": 870}, {"loss": 0.6866, "grad_norm": 0.771058201789856, "learning_rate": 0.0002, "epoch": 1.89247311827957, "step": 880}, {"loss": 0.7296, "grad_norm": 0.8624904155731201, "learning_rate": 0.0002, "epoch": 1.913978494623656, "step": 890}, {"loss": 0.7233, "grad_norm": 0.7062820792198181, "learning_rate": 0.0002, "epoch": 1.935483870967742, "step": 900}, {"loss": 0.6966, "grad_norm": 0.7560103535652161, "learning_rate": 0.0002, "epoch": 1.956989247311828, "step": 910}, {"loss": 0.69, "grad_norm": 0.788899838924408, "learning_rate": 0.0002, "epoch": 1.978494623655914, "step": 920}, {"loss": 0.6505, "grad_norm": 0.6562113761901855, "learning_rate": 0.0002, "epoch": 2.0, "step": 930}, {"eval_loss": 0.6885261535644531, "eval_runtime": 21.4291, "eval_samples_per_second": 15.446, "eval_steps_per_second": 1.96, "epoch": 2.0, "step": 930}, {"loss": 0.6625, "grad_norm": 0.8216531872749329, "learning_rate": 0.0002, "epoch": 2.021505376344086, "step": 940}, {"loss": 0.6398, "grad_norm": 0.8317142724990845, "learning_rate": 0.0002, "epoch": 2.043010752688172, "step": 950}, {"loss": 0.649, "grad_norm": 0.8446708917617798, "learning_rate": 0.0002, "epoch": 2.064516129032258, "step": 960}, {"loss": 0.657, "grad_norm": 0.735055148601532, "learning_rate": 0.0002, "epoch": 2.086021505376344, "step": 970}, {"loss": 0.649, "grad_norm": 0.7487243413925171, "learning_rate": 0.0002, "epoch": 2.10752688172043, "step": 980}, {"loss": 0.6419, "grad_norm": 0.8573887944221497, "learning_rate": 0.0002, "epoch": 2.129032258064516, "step": 990}, {"loss": 0.6431, "grad_norm": 0.6284521818161011, "learning_rate": 0.0002, "epoch": 2.150537634408602, "step": 1000}, {"loss": 0.6128, "grad_norm": 0.754183292388916, "learning_rate": 0.0002, "epoch": 2.172043010752688, "step": 1010}, {"loss": 0.6253, "grad_norm": 0.9445359110832214, "learning_rate": 0.0002, "epoch": 2.193548387096774, "step": 1020}, {"loss": 0.605, "grad_norm": 0.808508038520813, "learning_rate": 0.0002, "epoch": 2.21505376344086, "step": 1030}, {"loss": 0.6786, "grad_norm": 0.9394679665565491, "learning_rate": 0.0002, "epoch": 2.236559139784946, "step": 1040}, {"loss": 0.6176, "grad_norm": 0.8151357769966125, "learning_rate": 0.0002, "epoch": 2.258064516129032, "step": 1050}, {"loss": 0.66, "grad_norm": 0.7909848093986511, "learning_rate": 0.0002, "epoch": 2.279569892473118, "step": 1060}, {"loss": 0.6254, "grad_norm": 0.7506507039070129, "learning_rate": 0.0002, "epoch": 2.3010752688172045, "step": 1070}, {"loss": 0.6608, "grad_norm": 0.8240520358085632, "learning_rate": 0.0002, "epoch": 2.3225806451612905, "step": 1080}, {"loss": 0.6207, "grad_norm": 0.9342400431632996, "learning_rate": 0.0002, "epoch": 2.3440860215053765, "step": 1090}, {"loss": 0.6029, "grad_norm": 1.0598735809326172, "learning_rate": 0.0002, "epoch": 2.3655913978494625, "step": 1100}, {"loss": 0.6035, "grad_norm": 0.7907650470733643, "learning_rate": 0.0002, "epoch": 2.3870967741935485, "step": 1110}, {"loss": 0.6237, "grad_norm": 0.9388798475265503, "learning_rate": 0.0002, "epoch": 2.4086021505376345, "step": 1120}, {"loss": 0.6207, "grad_norm": 0.8985419869422913, "learning_rate": 0.0002, "epoch": 2.4301075268817205, "step": 1130}, {"loss": 0.5902, "grad_norm": 0.7471932768821716, "learning_rate": 0.0002, "epoch": 2.4516129032258065, "step": 1140}, {"loss": 0.6446, "grad_norm": 0.761131763458252, "learning_rate": 0.0002, "epoch": 2.4731182795698925, "step": 1150}, {"loss": 0.6088, "grad_norm": 0.7901819348335266, "learning_rate": 0.0002, "epoch": 2.4946236559139785, "step": 1160}, {"loss": 0.6142, "grad_norm": 0.9932922720909119, "learning_rate": 0.0002, "epoch": 2.5161290322580645, "step": 1170}, {"loss": 0.6407, "grad_norm": 0.7414287328720093, "learning_rate": 0.0002, "epoch": 2.5376344086021505, "step": 1180}, {"loss": 0.6161, "grad_norm": 0.8111771941184998, "learning_rate": 0.0002, "epoch": 2.5591397849462365, "step": 1190}, {"loss": 0.6006, "grad_norm": 0.7520156502723694, "learning_rate": 0.0002, "epoch": 2.5806451612903225, "step": 1200}, {"loss": 0.615, "grad_norm": 0.9022907018661499, "learning_rate": 0.0002, "epoch": 2.6021505376344085, "step": 1210}, {"loss": 0.6211, "grad_norm": 0.7746260166168213, "learning_rate": 0.0002, "epoch": 2.6236559139784945, "step": 1220}, {"loss": 0.616, "grad_norm": 0.8482862114906311, "learning_rate": 0.0002, "epoch": 2.6451612903225805, "step": 1230}, {"loss": 0.6417, "grad_norm": 0.7925458550453186, "learning_rate": 0.0002, "epoch": 2.6666666666666665, "step": 1240}, {"loss": 0.6187, "grad_norm": 0.8369929194450378, "learning_rate": 0.0002, "epoch": 2.688172043010753, "step": 1250}, {"loss": 0.6138, "grad_norm": 0.8311542868614197, "learning_rate": 0.0002, "epoch": 2.709677419354839, "step": 1260}, {"loss": 0.5894, "grad_norm": 0.7204853296279907, "learning_rate": 0.0002, "epoch": 2.731182795698925, "step": 1270}, {"loss": 0.6325, "grad_norm": 0.8447284698486328, "learning_rate": 0.0002, "epoch": 2.752688172043011, "step": 1280}, {"loss": 0.5946, "grad_norm": 0.7738404273986816, "learning_rate": 0.0002, "epoch": 2.774193548387097, "step": 1290}, {"loss": 0.5678, "grad_norm": 0.8393287062644958, "learning_rate": 0.0002, "epoch": 2.795698924731183, "step": 1300}, {"loss": 0.6092, "grad_norm": 0.79121994972229, "learning_rate": 0.0002, "epoch": 2.817204301075269, "step": 1310}, {"loss": 0.5889, "grad_norm": 0.7331557869911194, "learning_rate": 0.0002, "epoch": 2.838709677419355, "step": 1320}, {"loss": 0.6048, "grad_norm": 0.9593998193740845, "learning_rate": 0.0002, "epoch": 2.860215053763441, "step": 1330}, {"loss": 0.6108, "grad_norm": 0.7215158343315125, "learning_rate": 0.0002, "epoch": 2.881720430107527, "step": 1340}, {"loss": 0.5897, "grad_norm": 0.840404212474823, "learning_rate": 0.0002, "epoch": 2.903225806451613, "step": 1350}, {"loss": 0.6056, "grad_norm": 0.870659351348877, "learning_rate": 0.0002, "epoch": 2.924731182795699, "step": 1360}, {"loss": 0.6205, "grad_norm": 0.8744975328445435, "learning_rate": 0.0002, "epoch": 2.946236559139785, "step": 1370}, {"loss": 0.5966, "grad_norm": 0.8030612468719482, "learning_rate": 0.0002, "epoch": 2.967741935483871, "step": 1380}, {"loss": 0.6004, "grad_norm": 0.825814962387085, "learning_rate": 0.0002, "epoch": 2.989247311827957, "step": 1390}, {"eval_loss": 0.6257933378219604, "eval_runtime": 21.3692, "eval_samples_per_second": 15.49, "eval_steps_per_second": 1.965, "epoch": 3.0, "step": 1395}, {"loss": 0.5696, "grad_norm": 0.8650677800178528, "learning_rate": 0.0002, "epoch": 3.010752688172043, "step": 1400}, {"loss": 0.5483, "grad_norm": 0.8364197015762329, "learning_rate": 0.0002, "epoch": 3.032258064516129, "step": 1410}, {"loss": 0.5606, "grad_norm": 0.8278448581695557, "learning_rate": 0.0002, "epoch": 3.053763440860215, "step": 1420}, {"loss": 0.5572, "grad_norm": 0.8806642889976501, "learning_rate": 0.0002, "epoch": 3.075268817204301, "step": 1430}, {"loss": 0.585, "grad_norm": 0.8180029988288879, "learning_rate": 0.0002, "epoch": 3.096774193548387, "step": 1440}, {"loss": 0.5667, "grad_norm": 0.8561782836914062, "learning_rate": 0.0002, "epoch": 3.118279569892473, "step": 1450}, {"loss": 0.5246, "grad_norm": 0.8377029299736023, "learning_rate": 0.0002, "epoch": 3.139784946236559, "step": 1460}, {"loss": 0.5464, "grad_norm": 0.885779082775116, "learning_rate": 0.0002, "epoch": 3.161290322580645, "step": 1470}, {"loss": 0.541, "grad_norm": 0.9388518333435059, "learning_rate": 0.0002, "epoch": 3.182795698924731, "step": 1480}, {"loss": 0.5447, "grad_norm": 0.8816235661506653, "learning_rate": 0.0002, "epoch": 3.204301075268817, "step": 1490}, {"loss": 0.5466, "grad_norm": 0.9885783791542053, "learning_rate": 0.0002, "epoch": 3.225806451612903, "step": 1500}, {"loss": 0.5455, "grad_norm": 0.8635850548744202, "learning_rate": 0.0002, "epoch": 3.247311827956989, "step": 1510}, {"loss": 0.5419, "grad_norm": 0.829853355884552, "learning_rate": 0.0002, "epoch": 3.268817204301075, "step": 1520}, {"loss": 0.54, "grad_norm": 0.9037486910820007, "learning_rate": 0.0002, "epoch": 3.2903225806451615, "step": 1530}, {"loss": 0.5375, "grad_norm": 0.8173713684082031, "learning_rate": 0.0002, "epoch": 3.3118279569892475, "step": 1540}, {"loss": 0.5405, "grad_norm": 0.796953022480011, "learning_rate": 0.0002, "epoch": 3.3333333333333335, "step": 1550}, {"loss": 0.5505, "grad_norm": 0.7894400358200073, "learning_rate": 0.0002, "epoch": 3.3548387096774195, "step": 1560}, {"loss": 0.5395, "grad_norm": 0.9434949159622192, "learning_rate": 0.0002, "epoch": 3.3763440860215055, "step": 1570}, {"loss": 0.5271, "grad_norm": 0.8666760325431824, "learning_rate": 0.0002, "epoch": 3.3978494623655915, "step": 1580}, {"loss": 0.5439, "grad_norm": 0.7782467007637024, "learning_rate": 0.0002, "epoch": 3.4193548387096775, "step": 1590}, {"loss": 0.5161, "grad_norm": 0.8849126696586609, "learning_rate": 0.0002, "epoch": 3.4408602150537635, "step": 1600}, {"loss": 0.5353, "grad_norm": 0.7863831520080566, "learning_rate": 0.0002, "epoch": 3.4623655913978495, "step": 1610}, {"loss": 0.5308, "grad_norm": 1.0403116941452026, "learning_rate": 0.0002, "epoch": 3.4838709677419355, "step": 1620}, {"loss": 0.5339, "grad_norm": 0.8307499289512634, "learning_rate": 0.0002, "epoch": 3.5053763440860215, "step": 1630}, {"loss": 0.5361, "grad_norm": 0.9132118821144104, "learning_rate": 0.0002, "epoch": 3.5268817204301075, "step": 1640}, {"loss": 0.5828, "grad_norm": 0.9322578310966492, "learning_rate": 0.0002, "epoch": 3.5483870967741935, "step": 1650}, {"loss": 0.546, "grad_norm": 0.9782460331916809, "learning_rate": 0.0002, "epoch": 3.5698924731182795, "step": 1660}, {"loss": 0.5424, "grad_norm": 0.7189919352531433, "learning_rate": 0.0002, "epoch": 3.5913978494623655, "step": 1670}, {"loss": 0.5514, "grad_norm": 0.9689221382141113, "learning_rate": 0.0002, "epoch": 3.6129032258064515, "step": 1680}, {"loss": 0.5379, "grad_norm": 0.9684675335884094, "learning_rate": 0.0002, "epoch": 3.6344086021505375, "step": 1690}, {"loss": 0.5748, "grad_norm": 0.8851472735404968, "learning_rate": 0.0002, "epoch": 3.6559139784946235, "step": 1700}, {"loss": 0.5412, "grad_norm": 0.7709833383560181, "learning_rate": 0.0002, "epoch": 3.6774193548387095, "step": 1710}, {"loss": 0.521, "grad_norm": 0.818236231803894, "learning_rate": 0.0002, "epoch": 3.698924731182796, "step": 1720}, {"loss": 0.5445, "grad_norm": 0.870642364025116, "learning_rate": 0.0002, "epoch": 3.720430107526882, "step": 1730}, {"loss": 0.5307, "grad_norm": 1.0245511531829834, "learning_rate": 0.0002, "epoch": 3.741935483870968, "step": 1740}, {"loss": 0.5593, "grad_norm": 0.8607558608055115, "learning_rate": 0.0002, "epoch": 3.763440860215054, "step": 1750}, {"loss": 0.536, "grad_norm": 0.8511829972267151, "learning_rate": 0.0002, "epoch": 3.78494623655914, "step": 1760}, {"loss": 0.5193, "grad_norm": 0.7969087362289429, "learning_rate": 0.0002, "epoch": 3.806451612903226, "step": 1770}, {"loss": 0.5578, "grad_norm": 0.8457245826721191, "learning_rate": 0.0002, "epoch": 3.827956989247312, "step": 1780}, {"loss": 0.5337, "grad_norm": 0.8893467783927917, "learning_rate": 0.0002, "epoch": 3.849462365591398, "step": 1790}, {"loss": 0.5024, "grad_norm": 0.8593819737434387, "learning_rate": 0.0002, "epoch": 3.870967741935484, "step": 1800}, {"loss": 0.5134, "grad_norm": 0.7574560642242432, "learning_rate": 0.0002, "epoch": 3.89247311827957, "step": 1810}, {"loss": 0.5263, "grad_norm": 0.8681567311286926, "learning_rate": 0.0002, "epoch": 3.913978494623656, "step": 1820}, {"loss": 0.532, "grad_norm": 0.9068132042884827, "learning_rate": 0.0002, "epoch": 3.935483870967742, "step": 1830}, {"loss": 0.5427, "grad_norm": 0.8668948411941528, "learning_rate": 0.0002, "epoch": 3.956989247311828, "step": 1840}, {"loss": 0.5349, "grad_norm": 1.046032428741455, "learning_rate": 0.0002, "epoch": 3.978494623655914, "step": 1850}, {"loss": 0.5087, "grad_norm": 0.904780387878418, "learning_rate": 0.0002, "epoch": 4.0, "step": 1860}, {"eval_loss": 0.5737715363502502, "eval_runtime": 21.4915, "eval_samples_per_second": 15.401, "eval_steps_per_second": 1.954, "epoch": 4.0, "step": 1860}, {"loss": 0.4843, "grad_norm": 0.8611752986907959, "learning_rate": 0.0002, "epoch": 4.021505376344086, "step": 1870}, {"loss": 0.4814, "grad_norm": 0.838782548904419, "learning_rate": 0.0002, "epoch": 4.043010752688172, "step": 1880}, {"loss": 0.474, "grad_norm": 0.9119709134101868, "learning_rate": 0.0002, "epoch": 4.064516129032258, "step": 1890}, {"loss": 0.4951, "grad_norm": 0.8026251196861267, "learning_rate": 0.0002, "epoch": 4.086021505376344, "step": 1900}, {"loss": 0.491, "grad_norm": 0.8773705363273621, "learning_rate": 0.0002, "epoch": 4.10752688172043, "step": 1910}, {"loss": 0.474, "grad_norm": 0.8762255907058716, "learning_rate": 0.0002, "epoch": 4.129032258064516, "step": 1920}, {"loss": 0.4816, "grad_norm": 0.8371861577033997, "learning_rate": 0.0002, "epoch": 4.150537634408602, "step": 1930}, {"loss": 0.472, "grad_norm": 0.9703728556632996, "learning_rate": 0.0002, "epoch": 4.172043010752688, "step": 1940}, {"loss": 0.4772, "grad_norm": 0.8802874684333801, "learning_rate": 0.0002, "epoch": 4.193548387096774, "step": 1950}, {"loss": 0.5032, "grad_norm": 1.0103057622909546, "learning_rate": 0.0002, "epoch": 4.21505376344086, "step": 1960}, {"loss": 0.4945, "grad_norm": 0.9212995171546936, "learning_rate": 0.0002, "epoch": 4.236559139784946, "step": 1970}, {"loss": 0.4753, "grad_norm": 1.009544849395752, "learning_rate": 0.0002, "epoch": 4.258064516129032, "step": 1980}, {"loss": 0.4789, "grad_norm": 0.8535077571868896, "learning_rate": 0.0002, "epoch": 4.279569892473118, "step": 1990}, {"loss": 0.4782, "grad_norm": 0.8363022804260254, "learning_rate": 0.0002, "epoch": 4.301075268817204, "step": 2000}, {"loss": 0.4875, "grad_norm": 0.9041762948036194, "learning_rate": 0.0002, "epoch": 4.32258064516129, "step": 2010}, {"loss": 0.4779, "grad_norm": 0.960790753364563, "learning_rate": 0.0002, "epoch": 4.344086021505376, "step": 2020}, {"loss": 0.4626, "grad_norm": 0.8823095560073853, "learning_rate": 0.0002, "epoch": 4.365591397849462, "step": 2030}, {"loss": 0.4883, "grad_norm": 0.952100396156311, "learning_rate": 0.0002, "epoch": 4.387096774193548, "step": 2040}, {"loss": 0.4789, "grad_norm": 1.0793498754501343, "learning_rate": 0.0002, "epoch": 4.408602150537634, "step": 2050}, {"loss": 0.4827, "grad_norm": 0.8987208008766174, "learning_rate": 0.0002, "epoch": 4.43010752688172, "step": 2060}, {"loss": 0.4594, "grad_norm": 0.8539772033691406, "learning_rate": 0.0002, "epoch": 4.451612903225806, "step": 2070}, {"loss": 0.4752, "grad_norm": 0.9160863757133484, "learning_rate": 0.0002, "epoch": 4.473118279569892, "step": 2080}, {"loss": 0.5033, "grad_norm": 0.9946850538253784, "learning_rate": 0.0002, "epoch": 4.494623655913978, "step": 2090}, {"loss": 0.4842, "grad_norm": 0.908039391040802, "learning_rate": 0.0002, "epoch": 4.516129032258064, "step": 2100}, {"loss": 0.4861, "grad_norm": 1.1462254524230957, "learning_rate": 0.0002, "epoch": 4.53763440860215, "step": 2110}, {"loss": 0.4892, "grad_norm": 0.8392056226730347, "learning_rate": 0.0002, "epoch": 4.559139784946236, "step": 2120}, {"loss": 0.4824, "grad_norm": 0.9673896431922913, "learning_rate": 0.0002, "epoch": 4.580645161290323, "step": 2130}, {"loss": 0.4665, "grad_norm": 0.9047091603279114, "learning_rate": 0.0002, "epoch": 4.602150537634409, "step": 2140}, {"loss": 0.4714, "grad_norm": 0.9013425707817078, "learning_rate": 0.0002, "epoch": 4.623655913978495, "step": 2150}, {"loss": 0.472, "grad_norm": 0.8899165391921997, "learning_rate": 0.0002, "epoch": 4.645161290322581, "step": 2160}, {"loss": 0.4635, "grad_norm": 0.748602569103241, "learning_rate": 0.0002, "epoch": 4.666666666666667, "step": 2170}, {"loss": 0.4695, "grad_norm": 0.8694155216217041, "learning_rate": 0.0002, "epoch": 4.688172043010753, "step": 2180}, {"loss": 0.4929, "grad_norm": 0.9134316444396973, "learning_rate": 0.0002, "epoch": 4.709677419354839, "step": 2190}, {"loss": 0.4855, "grad_norm": 0.8504763245582581, "learning_rate": 0.0002, "epoch": 4.731182795698925, "step": 2200}, {"loss": 0.4517, "grad_norm": 1.0321544408798218, "learning_rate": 0.0002, "epoch": 4.752688172043011, "step": 2210}, {"loss": 0.4796, "grad_norm": 0.9368237257003784, "learning_rate": 0.0002, "epoch": 4.774193548387097, "step": 2220}, {"loss": 0.4837, "grad_norm": 0.9319947361946106, "learning_rate": 0.0002, "epoch": 4.795698924731183, "step": 2230}, {"loss": 0.4696, "grad_norm": 0.904333770275116, "learning_rate": 0.0002, "epoch": 4.817204301075269, "step": 2240}, {"loss": 0.4746, "grad_norm": 0.8097078204154968, "learning_rate": 0.0002, "epoch": 4.838709677419355, "step": 2250}, {"loss": 0.4438, "grad_norm": 0.9128859043121338, "learning_rate": 0.0002, "epoch": 4.860215053763441, "step": 2260}, {"loss": 0.4693, "grad_norm": 0.883129894733429, "learning_rate": 0.0002, "epoch": 4.881720430107527, "step": 2270}, {"loss": 0.4494, "grad_norm": 0.85712730884552, "learning_rate": 0.0002, "epoch": 4.903225806451613, "step": 2280}, {"loss": 0.4593, "grad_norm": 1.2101863622665405, "learning_rate": 0.0002, "epoch": 4.924731182795699, "step": 2290}, {"loss": 0.4779, "grad_norm": 0.917966902256012, "learning_rate": 0.0002, "epoch": 4.946236559139785, "step": 2300}, {"loss": 0.4666, "grad_norm": 0.7740724086761475, "learning_rate": 0.0002, "epoch": 4.967741935483871, "step": 2310}, {"loss": 0.4629, "grad_norm": 1.0199906826019287, "learning_rate": 0.0002, "epoch": 4.989247311827957, "step": 2320}, {"eval_loss": 0.5363914370536804, "eval_runtime": 21.3941, "eval_samples_per_second": 15.472, "eval_steps_per_second": 1.963, "epoch": 5.0, "step": 2325}, {"loss": 0.4543, "grad_norm": 0.8580502271652222, "learning_rate": 0.0002, "epoch": 5.010752688172043, "step": 2330}, {"loss": 0.404, "grad_norm": 0.7702704668045044, "learning_rate": 0.0002, "epoch": 5.032258064516129, "step": 2340}, {"loss": 0.4408, "grad_norm": 0.9417401552200317, "learning_rate": 0.0002, "epoch": 5.053763440860215, "step": 2350}, {"loss": 0.4306, "grad_norm": 0.9461463689804077, "learning_rate": 0.0002, "epoch": 5.075268817204301, "step": 2360}, {"loss": 0.4251, "grad_norm": 0.8931282162666321, "learning_rate": 0.0002, "epoch": 5.096774193548387, "step": 2370}, {"loss": 0.4249, "grad_norm": 1.000909447669983, "learning_rate": 0.0002, "epoch": 5.118279569892473, "step": 2380}, {"loss": 0.4231, "grad_norm": 0.8640249967575073, "learning_rate": 0.0002, "epoch": 5.139784946236559, "step": 2390}, {"loss": 0.4272, "grad_norm": 1.0451020002365112, "learning_rate": 0.0002, "epoch": 5.161290322580645, "step": 2400}, {"loss": 0.4177, "grad_norm": 0.7896912097930908, "learning_rate": 0.0002, "epoch": 5.182795698924731, "step": 2410}, {"loss": 0.4116, "grad_norm": 0.8424463272094727, "learning_rate": 0.0002, "epoch": 5.204301075268817, "step": 2420}, {"loss": 0.4225, "grad_norm": 1.0852105617523193, "learning_rate": 0.0002, "epoch": 5.225806451612903, "step": 2430}, {"loss": 0.4352, "grad_norm": 0.9285983443260193, "learning_rate": 0.0002, "epoch": 5.247311827956989, "step": 2440}, {"loss": 0.4262, "grad_norm": 0.9119299054145813, "learning_rate": 0.0002, "epoch": 5.268817204301075, "step": 2450}, {"loss": 0.4494, "grad_norm": 0.8790456056594849, "learning_rate": 0.0002, "epoch": 5.290322580645161, "step": 2460}, {"loss": 0.4421, "grad_norm": 0.8726504445075989, "learning_rate": 0.0002, "epoch": 5.311827956989247, "step": 2470}, {"loss": 0.4372, "grad_norm": 0.9415227770805359, "learning_rate": 0.0002, "epoch": 5.333333333333333, "step": 2480}, {"loss": 0.4223, "grad_norm": 0.9133324027061462, "learning_rate": 0.0002, "epoch": 5.354838709677419, "step": 2490}, {"loss": 0.4401, "grad_norm": 0.9567879438400269, "learning_rate": 0.0002, "epoch": 5.376344086021505, "step": 2500}, {"loss": 0.4094, "grad_norm": 0.9239469766616821, "learning_rate": 0.0002, "epoch": 5.397849462365591, "step": 2510}, {"loss": 0.4416, "grad_norm": 1.0293527841567993, "learning_rate": 0.0002, "epoch": 5.419354838709677, "step": 2520}, {"loss": 0.4311, "grad_norm": 0.8618718981742859, "learning_rate": 0.0002, "epoch": 5.440860215053763, "step": 2530}, {"loss": 0.462, "grad_norm": 0.740166187286377, "learning_rate": 0.0002, "epoch": 5.462365591397849, "step": 2540}, {"loss": 0.4172, "grad_norm": 0.901566743850708, "learning_rate": 0.0002, "epoch": 5.483870967741936, "step": 2550}, {"loss": 0.4315, "grad_norm": 0.7957597970962524, "learning_rate": 0.0002, "epoch": 5.505376344086022, "step": 2560}, {"loss": 0.4263, "grad_norm": 1.1139343976974487, "learning_rate": 0.0002, "epoch": 5.526881720430108, "step": 2570}, {"loss": 0.4056, "grad_norm": 0.989765465259552, "learning_rate": 0.0002, "epoch": 5.548387096774194, "step": 2580}, {"loss": 0.4311, "grad_norm": 0.9416969418525696, "learning_rate": 0.0002, "epoch": 5.56989247311828, "step": 2590}, {"loss": 0.4363, "grad_norm": 0.9184830784797668, "learning_rate": 0.0002, "epoch": 5.591397849462366, "step": 2600}, {"loss": 0.432, "grad_norm": 1.0512700080871582, "learning_rate": 0.0002, "epoch": 5.612903225806452, "step": 2610}, {"loss": 0.4227, "grad_norm": 0.901462197303772, "learning_rate": 0.0002, "epoch": 5.634408602150538, "step": 2620}, {"loss": 0.4332, "grad_norm": 0.9732566475868225, "learning_rate": 0.0002, "epoch": 5.655913978494624, "step": 2630}, {"loss": 0.4223, "grad_norm": 0.8180275559425354, "learning_rate": 0.0002, "epoch": 5.67741935483871, "step": 2640}, {"loss": 0.4311, "grad_norm": 1.1354765892028809, "learning_rate": 0.0002, "epoch": 5.698924731182796, "step": 2650}, {"loss": 0.4409, "grad_norm": 0.9161503314971924, "learning_rate": 0.0002, "epoch": 5.720430107526882, "step": 2660}, {"loss": 0.4394, "grad_norm": 1.0561772584915161, "learning_rate": 0.0002, "epoch": 5.741935483870968, "step": 2670}, {"loss": 0.424, "grad_norm": 0.7712787389755249, "learning_rate": 0.0002, "epoch": 5.763440860215054, "step": 2680}, {"loss": 0.4326, "grad_norm": 0.9674550294876099, "learning_rate": 0.0002, "epoch": 5.78494623655914, "step": 2690}, {"loss": 0.4459, "grad_norm": 0.7531843781471252, "learning_rate": 0.0002, "epoch": 5.806451612903226, "step": 2700}, {"loss": 0.4276, "grad_norm": 1.1332131624221802, "learning_rate": 0.0002, "epoch": 5.827956989247312, "step": 2710}, {"loss": 0.4113, "grad_norm": 0.9367414116859436, "learning_rate": 0.0002, "epoch": 5.849462365591398, "step": 2720}, {"loss": 0.4227, "grad_norm": 0.8267706632614136, "learning_rate": 0.0002, "epoch": 5.870967741935484, "step": 2730}, {"loss": 0.4218, "grad_norm": 1.1040657758712769, "learning_rate": 0.0002, "epoch": 5.89247311827957, "step": 2740}, {"loss": 0.4129, "grad_norm": 0.8879582285881042, "learning_rate": 0.0002, "epoch": 5.913978494623656, "step": 2750}, {"loss": 0.4241, "grad_norm": 0.9264667630195618, "learning_rate": 0.0002, "epoch": 5.935483870967742, "step": 2760}, {"loss": 0.4318, "grad_norm": 0.9373905658721924, "learning_rate": 0.0002, "epoch": 5.956989247311828, "step": 2770}, {"loss": 0.423, "grad_norm": 1.0063740015029907, "learning_rate": 0.0002, "epoch": 5.978494623655914, "step": 2780}, {"loss": 0.4382, "grad_norm": 0.8291367292404175, "learning_rate": 0.0002, "epoch": 6.0, "step": 2790}]}
+{"epoch": 7.0, "step": 3255, "epoch_duration": 449.4360945224762, "total_accumulated_duration": 3152.6115312576294, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 3020.60888671875}, "peak_memory_usage": {"GPU_0": 15051.17431640625}, "avg_memory_reserved": {"GPU_0": 20170.0}, "peak_memory_reserved": {"GPU_0": 20170.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-2790", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 3.4172, "grad_norm": 0.9075053930282593, "learning_rate": 0.0002, "epoch": 0.021505376344086023, "step": 10}, {"loss": 2.5888, "grad_norm": 1.4321208000183105, "learning_rate": 0.0002, "epoch": 0.043010752688172046, "step": 20}, {"loss": 2.1195, "grad_norm": 1.7500602006912231, "learning_rate": 0.0002, "epoch": 0.06451612903225806, "step": 30}, {"loss": 1.9303, "grad_norm": 0.7606641054153442, "learning_rate": 0.0002, "epoch": 0.08602150537634409, "step": 40}, {"loss": 1.6112, "grad_norm": 1.2754929065704346, "learning_rate": 0.0002, "epoch": 0.10752688172043011, "step": 50}, {"loss": 1.4319, "grad_norm": 1.0936230421066284, "learning_rate": 0.0002, "epoch": 0.12903225806451613, "step": 60}, {"loss": 1.3568, "grad_norm": 1.144593596458435, "learning_rate": 0.0002, "epoch": 0.15053763440860216, "step": 70}, {"loss": 1.2028, "grad_norm": 1.2181956768035889, "learning_rate": 0.0002, "epoch": 0.17204301075268819, "step": 80}, {"loss": 1.1534, "grad_norm": 1.1260095834732056, "learning_rate": 0.0002, "epoch": 0.1935483870967742, "step": 90}, {"loss": 1.1089, "grad_norm": 1.1155284643173218, "learning_rate": 0.0002, "epoch": 0.21505376344086022, "step": 100}, {"loss": 1.0883, "grad_norm": 1.089565396308899, "learning_rate": 0.0002, "epoch": 0.23655913978494625, "step": 110}, {"loss": 1.0814, "grad_norm": 0.9833471775054932, "learning_rate": 0.0002, "epoch": 0.25806451612903225, "step": 120}, {"loss": 1.0239, "grad_norm": 1.0265629291534424, "learning_rate": 0.0002, "epoch": 0.27956989247311825, "step": 130}, {"loss": 0.9888, "grad_norm": 0.9344286322593689, "learning_rate": 0.0002, "epoch": 0.3010752688172043, "step": 140}, {"loss": 1.0043, "grad_norm": 0.9883386492729187, "learning_rate": 0.0002, "epoch": 0.3225806451612903, "step": 150}, {"loss": 0.9338, "grad_norm": 0.9299277067184448, "learning_rate": 0.0002, "epoch": 0.34408602150537637, "step": 160}, {"loss": 0.9432, "grad_norm": 1.390045404434204, "learning_rate": 0.0002, "epoch": 0.3655913978494624, "step": 170}, {"loss": 0.9008, "grad_norm": 1.0313078165054321, "learning_rate": 0.0002, "epoch": 0.3870967741935484, "step": 180}, {"loss": 0.9434, "grad_norm": 1.1792205572128296, "learning_rate": 0.0002, "epoch": 0.40860215053763443, "step": 190}, {"loss": 0.8761, "grad_norm": 1.049809217453003, "learning_rate": 0.0002, "epoch": 0.43010752688172044, "step": 200}, {"loss": 0.8709, "grad_norm": 0.990111768245697, "learning_rate": 0.0002, "epoch": 0.45161290322580644, "step": 210}, {"loss": 0.905, "grad_norm": 0.9870412349700928, "learning_rate": 0.0002, "epoch": 0.4731182795698925, "step": 220}, {"loss": 0.9129, "grad_norm": 0.8557345867156982, "learning_rate": 0.0002, "epoch": 0.4946236559139785, "step": 230}, {"loss": 0.8836, "grad_norm": 0.9746861457824707, "learning_rate": 0.0002, "epoch": 0.5161290322580645, "step": 240}, {"loss": 0.873, "grad_norm": 0.9010438323020935, "learning_rate": 0.0002, "epoch": 0.5376344086021505, "step": 250}, {"loss": 0.8241, "grad_norm": 0.9061082005500793, "learning_rate": 0.0002, "epoch": 0.5591397849462365, "step": 260}, {"loss": 0.8652, "grad_norm": 0.9311846494674683, "learning_rate": 0.0002, "epoch": 0.5806451612903226, "step": 270}, {"loss": 0.8256, "grad_norm": 0.9140254855155945, "learning_rate": 0.0002, "epoch": 0.6021505376344086, "step": 280}, {"loss": 0.8441, "grad_norm": 0.9722253084182739, "learning_rate": 0.0002, "epoch": 0.6236559139784946, "step": 290}, {"loss": 0.8314, "grad_norm": 0.8539168238639832, "learning_rate": 0.0002, "epoch": 0.6451612903225806, "step": 300}, {"loss": 0.8528, "grad_norm": 0.9053162932395935, "learning_rate": 0.0002, "epoch": 0.6666666666666666, "step": 310}, {"loss": 0.8209, "grad_norm": 0.8444252610206604, "learning_rate": 0.0002, "epoch": 0.6881720430107527, "step": 320}, {"loss": 0.8101, "grad_norm": 0.8127437829971313, "learning_rate": 0.0002, "epoch": 0.7096774193548387, "step": 330}, {"loss": 0.8223, "grad_norm": 0.886555016040802, "learning_rate": 0.0002, "epoch": 0.7311827956989247, "step": 340}, {"loss": 0.8368, "grad_norm": 0.8458548784255981, "learning_rate": 0.0002, "epoch": 0.7526881720430108, "step": 350}, {"loss": 0.8295, "grad_norm": 0.8683297634124756, "learning_rate": 0.0002, "epoch": 0.7741935483870968, "step": 360}, {"loss": 0.8232, "grad_norm": 0.8308405876159668, "learning_rate": 0.0002, "epoch": 0.7956989247311828, "step": 370}, {"loss": 0.7752, "grad_norm": 0.8305579423904419, "learning_rate": 0.0002, "epoch": 0.8172043010752689, "step": 380}, {"loss": 0.8267, "grad_norm": 0.8545567393302917, "learning_rate": 0.0002, "epoch": 0.8387096774193549, "step": 390}, {"loss": 0.8212, "grad_norm": 0.8486055731773376, "learning_rate": 0.0002, "epoch": 0.8602150537634409, "step": 400}, {"loss": 0.743, "grad_norm": 0.8126763105392456, "learning_rate": 0.0002, "epoch": 0.8817204301075269, "step": 410}, {"loss": 0.7993, "grad_norm": 0.8494045734405518, "learning_rate": 0.0002, "epoch": 0.9032258064516129, "step": 420}, {"loss": 0.8213, "grad_norm": 0.7639183402061462, "learning_rate": 0.0002, "epoch": 0.9247311827956989, "step": 430}, {"loss": 0.8015, "grad_norm": 0.858101487159729, "learning_rate": 0.0002, "epoch": 0.946236559139785, "step": 440}, {"loss": 0.7629, "grad_norm": 0.8141381740570068, "learning_rate": 0.0002, "epoch": 0.967741935483871, "step": 450}, {"loss": 0.7357, "grad_norm": 0.8072513937950134, "learning_rate": 0.0002, "epoch": 0.989247311827957, "step": 460}, {"eval_loss": 0.7740864157676697, "eval_runtime": 21.383, "eval_samples_per_second": 15.48, "eval_steps_per_second": 1.964, "epoch": 1.0, "step": 465}, {"loss": 0.7701, "grad_norm": 0.8269494771957397, "learning_rate": 0.0002, "epoch": 1.010752688172043, "step": 470}, {"loss": 0.7532, "grad_norm": 0.7814009189605713, "learning_rate": 0.0002, "epoch": 1.032258064516129, "step": 480}, {"loss": 0.7689, "grad_norm": 0.8183923363685608, "learning_rate": 0.0002, "epoch": 1.053763440860215, "step": 490}, {"loss": 0.765, "grad_norm": 0.8146600723266602, "learning_rate": 0.0002, "epoch": 1.075268817204301, "step": 500}, {"loss": 0.7358, "grad_norm": 0.8635126352310181, "learning_rate": 0.0002, "epoch": 1.096774193548387, "step": 510}, {"loss": 0.7302, "grad_norm": 0.8520359396934509, "learning_rate": 0.0002, "epoch": 1.118279569892473, "step": 520}, {"loss": 0.7492, "grad_norm": 0.8026443123817444, "learning_rate": 0.0002, "epoch": 1.139784946236559, "step": 530}, {"loss": 0.7518, "grad_norm": 0.8157258629798889, "learning_rate": 0.0002, "epoch": 1.1612903225806452, "step": 540}, {"loss": 0.7461, "grad_norm": 0.9450796246528625, "learning_rate": 0.0002, "epoch": 1.1827956989247312, "step": 550}, {"loss": 0.7128, "grad_norm": 0.8859835863113403, "learning_rate": 0.0002, "epoch": 1.2043010752688172, "step": 560}, {"loss": 0.7067, "grad_norm": 0.7819921970367432, "learning_rate": 0.0002, "epoch": 1.2258064516129032, "step": 570}, {"loss": 0.7577, "grad_norm": 0.7823445796966553, "learning_rate": 0.0002, "epoch": 1.2473118279569892, "step": 580}, {"loss": 0.7358, "grad_norm": 0.7931883931159973, "learning_rate": 0.0002, "epoch": 1.2688172043010753, "step": 590}, {"loss": 0.723, "grad_norm": 0.7495734095573425, "learning_rate": 0.0002, "epoch": 1.2903225806451613, "step": 600}, {"loss": 0.7386, "grad_norm": 0.9272717237472534, "learning_rate": 0.0002, "epoch": 1.3118279569892473, "step": 610}, {"loss": 0.7498, "grad_norm": 0.7968398332595825, "learning_rate": 0.0002, "epoch": 1.3333333333333333, "step": 620}, {"loss": 0.7635, "grad_norm": 0.7813659310340881, "learning_rate": 0.0002, "epoch": 1.3548387096774195, "step": 630}, {"loss": 0.6665, "grad_norm": 0.730925977230072, "learning_rate": 0.0002, "epoch": 1.3763440860215055, "step": 640}, {"loss": 0.7037, "grad_norm": 0.8011482954025269, "learning_rate": 0.0002, "epoch": 1.3978494623655915, "step": 650}, {"loss": 0.6931, "grad_norm": 0.7770085334777832, "learning_rate": 0.0002, "epoch": 1.4193548387096775, "step": 660}, {"loss": 0.6949, "grad_norm": 0.7432682514190674, "learning_rate": 0.0002, "epoch": 1.4408602150537635, "step": 670}, {"loss": 0.7444, "grad_norm": 0.8820092678070068, "learning_rate": 0.0002, "epoch": 1.4623655913978495, "step": 680}, {"loss": 0.6758, "grad_norm": 0.7786208987236023, "learning_rate": 0.0002, "epoch": 1.4838709677419355, "step": 690}, {"loss": 0.6702, "grad_norm": 0.7467480301856995, "learning_rate": 0.0002, "epoch": 1.5053763440860215, "step": 700}, {"loss": 0.7107, "grad_norm": 0.8147122263908386, "learning_rate": 0.0002, "epoch": 1.5268817204301075, "step": 710}, {"loss": 0.7144, "grad_norm": 0.796030580997467, "learning_rate": 0.0002, "epoch": 1.5483870967741935, "step": 720}, {"loss": 0.6936, "grad_norm": 0.8776171207427979, "learning_rate": 0.0002, "epoch": 1.5698924731182795, "step": 730}, {"loss": 0.7101, "grad_norm": 0.8056126236915588, "learning_rate": 0.0002, "epoch": 1.5913978494623655, "step": 740}, {"loss": 0.7162, "grad_norm": 0.8141863346099854, "learning_rate": 0.0002, "epoch": 1.6129032258064515, "step": 750}, {"loss": 0.7088, "grad_norm": 0.8100557327270508, "learning_rate": 0.0002, "epoch": 1.6344086021505375, "step": 760}, {"loss": 0.7212, "grad_norm": 0.8283200860023499, "learning_rate": 0.0002, "epoch": 1.6559139784946235, "step": 770}, {"loss": 0.694, "grad_norm": 0.800865113735199, "learning_rate": 0.0002, "epoch": 1.6774193548387095, "step": 780}, {"loss": 0.7076, "grad_norm": 0.8052287697792053, "learning_rate": 0.0002, "epoch": 1.6989247311827957, "step": 790}, {"loss": 0.7257, "grad_norm": 0.8619674444198608, "learning_rate": 0.0002, "epoch": 1.7204301075268817, "step": 800}, {"loss": 0.7141, "grad_norm": 0.8907215595245361, "learning_rate": 0.0002, "epoch": 1.7419354838709677, "step": 810}, {"loss": 0.7035, "grad_norm": 0.6976316571235657, "learning_rate": 0.0002, "epoch": 1.7634408602150538, "step": 820}, {"loss": 0.6916, "grad_norm": 0.7533746957778931, "learning_rate": 0.0002, "epoch": 1.7849462365591398, "step": 830}, {"loss": 0.7094, "grad_norm": 0.7326804399490356, "learning_rate": 0.0002, "epoch": 1.8064516129032258, "step": 840}, {"loss": 0.6891, "grad_norm": 0.7782683372497559, "learning_rate": 0.0002, "epoch": 1.827956989247312, "step": 850}, {"loss": 0.6931, "grad_norm": 0.7424806356430054, "learning_rate": 0.0002, "epoch": 1.849462365591398, "step": 860}, {"loss": 0.7354, "grad_norm": 1.172325611114502, "learning_rate": 0.0002, "epoch": 1.870967741935484, "step": 870}, {"loss": 0.6866, "grad_norm": 0.771058201789856, "learning_rate": 0.0002, "epoch": 1.89247311827957, "step": 880}, {"loss": 0.7296, "grad_norm": 0.8624904155731201, "learning_rate": 0.0002, "epoch": 1.913978494623656, "step": 890}, {"loss": 0.7233, "grad_norm": 0.7062820792198181, "learning_rate": 0.0002, "epoch": 1.935483870967742, "step": 900}, {"loss": 0.6966, "grad_norm": 0.7560103535652161, "learning_rate": 0.0002, "epoch": 1.956989247311828, "step": 910}, {"loss": 0.69, "grad_norm": 0.788899838924408, "learning_rate": 0.0002, "epoch": 1.978494623655914, "step": 920}, {"loss": 0.6505, "grad_norm": 0.6562113761901855, "learning_rate": 0.0002, "epoch": 2.0, "step": 930}, {"eval_loss": 0.6885261535644531, "eval_runtime": 21.4291, "eval_samples_per_second": 15.446, "eval_steps_per_second": 1.96, "epoch": 2.0, "step": 930}, {"loss": 0.6625, "grad_norm": 0.8216531872749329, "learning_rate": 0.0002, "epoch": 2.021505376344086, "step": 940}, {"loss": 0.6398, "grad_norm": 0.8317142724990845, "learning_rate": 0.0002, "epoch": 2.043010752688172, "step": 950}, {"loss": 0.649, "grad_norm": 0.8446708917617798, "learning_rate": 0.0002, "epoch": 2.064516129032258, "step": 960}, {"loss": 0.657, "grad_norm": 0.735055148601532, "learning_rate": 0.0002, "epoch": 2.086021505376344, "step": 970}, {"loss": 0.649, "grad_norm": 0.7487243413925171, "learning_rate": 0.0002, "epoch": 2.10752688172043, "step": 980}, {"loss": 0.6419, "grad_norm": 0.8573887944221497, "learning_rate": 0.0002, "epoch": 2.129032258064516, "step": 990}, {"loss": 0.6431, "grad_norm": 0.6284521818161011, "learning_rate": 0.0002, "epoch": 2.150537634408602, "step": 1000}, {"loss": 0.6128, "grad_norm": 0.754183292388916, "learning_rate": 0.0002, "epoch": 2.172043010752688, "step": 1010}, {"loss": 0.6253, "grad_norm": 0.9445359110832214, "learning_rate": 0.0002, "epoch": 2.193548387096774, "step": 1020}, {"loss": 0.605, "grad_norm": 0.808508038520813, "learning_rate": 0.0002, "epoch": 2.21505376344086, "step": 1030}, {"loss": 0.6786, "grad_norm": 0.9394679665565491, "learning_rate": 0.0002, "epoch": 2.236559139784946, "step": 1040}, {"loss": 0.6176, "grad_norm": 0.8151357769966125, "learning_rate": 0.0002, "epoch": 2.258064516129032, "step": 1050}, {"loss": 0.66, "grad_norm": 0.7909848093986511, "learning_rate": 0.0002, "epoch": 2.279569892473118, "step": 1060}, {"loss": 0.6254, "grad_norm": 0.7506507039070129, "learning_rate": 0.0002, "epoch": 2.3010752688172045, "step": 1070}, {"loss": 0.6608, "grad_norm": 0.8240520358085632, "learning_rate": 0.0002, "epoch": 2.3225806451612905, "step": 1080}, {"loss": 0.6207, "grad_norm": 0.9342400431632996, "learning_rate": 0.0002, "epoch": 2.3440860215053765, "step": 1090}, {"loss": 0.6029, "grad_norm": 1.0598735809326172, "learning_rate": 0.0002, "epoch": 2.3655913978494625, "step": 1100}, {"loss": 0.6035, "grad_norm": 0.7907650470733643, "learning_rate": 0.0002, "epoch": 2.3870967741935485, "step": 1110}, {"loss": 0.6237, "grad_norm": 0.9388798475265503, "learning_rate": 0.0002, "epoch": 2.4086021505376345, "step": 1120}, {"loss": 0.6207, "grad_norm": 0.8985419869422913, "learning_rate": 0.0002, "epoch": 2.4301075268817205, "step": 1130}, {"loss": 0.5902, "grad_norm": 0.7471932768821716, "learning_rate": 0.0002, "epoch": 2.4516129032258065, "step": 1140}, {"loss": 0.6446, "grad_norm": 0.761131763458252, "learning_rate": 0.0002, "epoch": 2.4731182795698925, "step": 1150}, {"loss": 0.6088, "grad_norm": 0.7901819348335266, "learning_rate": 0.0002, "epoch": 2.4946236559139785, "step": 1160}, {"loss": 0.6142, "grad_norm": 0.9932922720909119, "learning_rate": 0.0002, "epoch": 2.5161290322580645, "step": 1170}, {"loss": 0.6407, "grad_norm": 0.7414287328720093, "learning_rate": 0.0002, "epoch": 2.5376344086021505, "step": 1180}, {"loss": 0.6161, "grad_norm": 0.8111771941184998, "learning_rate": 0.0002, "epoch": 2.5591397849462365, "step": 1190}, {"loss": 0.6006, "grad_norm": 0.7520156502723694, "learning_rate": 0.0002, "epoch": 2.5806451612903225, "step": 1200}, {"loss": 0.615, "grad_norm": 0.9022907018661499, "learning_rate": 0.0002, "epoch": 2.6021505376344085, "step": 1210}, {"loss": 0.6211, "grad_norm": 0.7746260166168213, "learning_rate": 0.0002, "epoch": 2.6236559139784945, "step": 1220}, {"loss": 0.616, "grad_norm": 0.8482862114906311, "learning_rate": 0.0002, "epoch": 2.6451612903225805, "step": 1230}, {"loss": 0.6417, "grad_norm": 0.7925458550453186, "learning_rate": 0.0002, "epoch": 2.6666666666666665, "step": 1240}, {"loss": 0.6187, "grad_norm": 0.8369929194450378, "learning_rate": 0.0002, "epoch": 2.688172043010753, "step": 1250}, {"loss": 0.6138, "grad_norm": 0.8311542868614197, "learning_rate": 0.0002, "epoch": 2.709677419354839, "step": 1260}, {"loss": 0.5894, "grad_norm": 0.7204853296279907, "learning_rate": 0.0002, "epoch": 2.731182795698925, "step": 1270}, {"loss": 0.6325, "grad_norm": 0.8447284698486328, "learning_rate": 0.0002, "epoch": 2.752688172043011, "step": 1280}, {"loss": 0.5946, "grad_norm": 0.7738404273986816, "learning_rate": 0.0002, "epoch": 2.774193548387097, "step": 1290}, {"loss": 0.5678, "grad_norm": 0.8393287062644958, "learning_rate": 0.0002, "epoch": 2.795698924731183, "step": 1300}, {"loss": 0.6092, "grad_norm": 0.79121994972229, "learning_rate": 0.0002, "epoch": 2.817204301075269, "step": 1310}, {"loss": 0.5889, "grad_norm": 0.7331557869911194, "learning_rate": 0.0002, "epoch": 2.838709677419355, "step": 1320}, {"loss": 0.6048, "grad_norm": 0.9593998193740845, "learning_rate": 0.0002, "epoch": 2.860215053763441, "step": 1330}, {"loss": 0.6108, "grad_norm": 0.7215158343315125, "learning_rate": 0.0002, "epoch": 2.881720430107527, "step": 1340}, {"loss": 0.5897, "grad_norm": 0.840404212474823, "learning_rate": 0.0002, "epoch": 2.903225806451613, "step": 1350}, {"loss": 0.6056, "grad_norm": 0.870659351348877, "learning_rate": 0.0002, "epoch": 2.924731182795699, "step": 1360}, {"loss": 0.6205, "grad_norm": 0.8744975328445435, "learning_rate": 0.0002, "epoch": 2.946236559139785, "step": 1370}, {"loss": 0.5966, "grad_norm": 0.8030612468719482, "learning_rate": 0.0002, "epoch": 2.967741935483871, "step": 1380}, {"loss": 0.6004, "grad_norm": 0.825814962387085, "learning_rate": 0.0002, "epoch": 2.989247311827957, "step": 1390}, {"eval_loss": 0.6257933378219604, "eval_runtime": 21.3692, "eval_samples_per_second": 15.49, "eval_steps_per_second": 1.965, "epoch": 3.0, "step": 1395}, {"loss": 0.5696, "grad_norm": 0.8650677800178528, "learning_rate": 0.0002, "epoch": 3.010752688172043, "step": 1400}, {"loss": 0.5483, "grad_norm": 0.8364197015762329, "learning_rate": 0.0002, "epoch": 3.032258064516129, "step": 1410}, {"loss": 0.5606, "grad_norm": 0.8278448581695557, "learning_rate": 0.0002, "epoch": 3.053763440860215, "step": 1420}, {"loss": 0.5572, "grad_norm": 0.8806642889976501, "learning_rate": 0.0002, "epoch": 3.075268817204301, "step": 1430}, {"loss": 0.585, "grad_norm": 0.8180029988288879, "learning_rate": 0.0002, "epoch": 3.096774193548387, "step": 1440}, {"loss": 0.5667, "grad_norm": 0.8561782836914062, "learning_rate": 0.0002, "epoch": 3.118279569892473, "step": 1450}, {"loss": 0.5246, "grad_norm": 0.8377029299736023, "learning_rate": 0.0002, "epoch": 3.139784946236559, "step": 1460}, {"loss": 0.5464, "grad_norm": 0.885779082775116, "learning_rate": 0.0002, "epoch": 3.161290322580645, "step": 1470}, {"loss": 0.541, "grad_norm": 0.9388518333435059, "learning_rate": 0.0002, "epoch": 3.182795698924731, "step": 1480}, {"loss": 0.5447, "grad_norm": 0.8816235661506653, "learning_rate": 0.0002, "epoch": 3.204301075268817, "step": 1490}, {"loss": 0.5466, "grad_norm": 0.9885783791542053, "learning_rate": 0.0002, "epoch": 3.225806451612903, "step": 1500}, {"loss": 0.5455, "grad_norm": 0.8635850548744202, "learning_rate": 0.0002, "epoch": 3.247311827956989, "step": 1510}, {"loss": 0.5419, "grad_norm": 0.829853355884552, "learning_rate": 0.0002, "epoch": 3.268817204301075, "step": 1520}, {"loss": 0.54, "grad_norm": 0.9037486910820007, "learning_rate": 0.0002, "epoch": 3.2903225806451615, "step": 1530}, {"loss": 0.5375, "grad_norm": 0.8173713684082031, "learning_rate": 0.0002, "epoch": 3.3118279569892475, "step": 1540}, {"loss": 0.5405, "grad_norm": 0.796953022480011, "learning_rate": 0.0002, "epoch": 3.3333333333333335, "step": 1550}, {"loss": 0.5505, "grad_norm": 0.7894400358200073, "learning_rate": 0.0002, "epoch": 3.3548387096774195, "step": 1560}, {"loss": 0.5395, "grad_norm": 0.9434949159622192, "learning_rate": 0.0002, "epoch": 3.3763440860215055, "step": 1570}, {"loss": 0.5271, "grad_norm": 0.8666760325431824, "learning_rate": 0.0002, "epoch": 3.3978494623655915, "step": 1580}, {"loss": 0.5439, "grad_norm": 0.7782467007637024, "learning_rate": 0.0002, "epoch": 3.4193548387096775, "step": 1590}, {"loss": 0.5161, "grad_norm": 0.8849126696586609, "learning_rate": 0.0002, "epoch": 3.4408602150537635, "step": 1600}, {"loss": 0.5353, "grad_norm": 0.7863831520080566, "learning_rate": 0.0002, "epoch": 3.4623655913978495, "step": 1610}, {"loss": 0.5308, "grad_norm": 1.0403116941452026, "learning_rate": 0.0002, "epoch": 3.4838709677419355, "step": 1620}, {"loss": 0.5339, "grad_norm": 0.8307499289512634, "learning_rate": 0.0002, "epoch": 3.5053763440860215, "step": 1630}, {"loss": 0.5361, "grad_norm": 0.9132118821144104, "learning_rate": 0.0002, "epoch": 3.5268817204301075, "step": 1640}, {"loss": 0.5828, "grad_norm": 0.9322578310966492, "learning_rate": 0.0002, "epoch": 3.5483870967741935, "step": 1650}, {"loss": 0.546, "grad_norm": 0.9782460331916809, "learning_rate": 0.0002, "epoch": 3.5698924731182795, "step": 1660}, {"loss": 0.5424, "grad_norm": 0.7189919352531433, "learning_rate": 0.0002, "epoch": 3.5913978494623655, "step": 1670}, {"loss": 0.5514, "grad_norm": 0.9689221382141113, "learning_rate": 0.0002, "epoch": 3.6129032258064515, "step": 1680}, {"loss": 0.5379, "grad_norm": 0.9684675335884094, "learning_rate": 0.0002, "epoch": 3.6344086021505375, "step": 1690}, {"loss": 0.5748, "grad_norm": 0.8851472735404968, "learning_rate": 0.0002, "epoch": 3.6559139784946235, "step": 1700}, {"loss": 0.5412, "grad_norm": 0.7709833383560181, "learning_rate": 0.0002, "epoch": 3.6774193548387095, "step": 1710}, {"loss": 0.521, "grad_norm": 0.818236231803894, "learning_rate": 0.0002, "epoch": 3.698924731182796, "step": 1720}, {"loss": 0.5445, "grad_norm": 0.870642364025116, "learning_rate": 0.0002, "epoch": 3.720430107526882, "step": 1730}, {"loss": 0.5307, "grad_norm": 1.0245511531829834, "learning_rate": 0.0002, "epoch": 3.741935483870968, "step": 1740}, {"loss": 0.5593, "grad_norm": 0.8607558608055115, "learning_rate": 0.0002, "epoch": 3.763440860215054, "step": 1750}, {"loss": 0.536, "grad_norm": 0.8511829972267151, "learning_rate": 0.0002, "epoch": 3.78494623655914, "step": 1760}, {"loss": 0.5193, "grad_norm": 0.7969087362289429, "learning_rate": 0.0002, "epoch": 3.806451612903226, "step": 1770}, {"loss": 0.5578, "grad_norm": 0.8457245826721191, "learning_rate": 0.0002, "epoch": 3.827956989247312, "step": 1780}, {"loss": 0.5337, "grad_norm": 0.8893467783927917, "learning_rate": 0.0002, "epoch": 3.849462365591398, "step": 1790}, {"loss": 0.5024, "grad_norm": 0.8593819737434387, "learning_rate": 0.0002, "epoch": 3.870967741935484, "step": 1800}, {"loss": 0.5134, "grad_norm": 0.7574560642242432, "learning_rate": 0.0002, "epoch": 3.89247311827957, "step": 1810}, {"loss": 0.5263, "grad_norm": 0.8681567311286926, "learning_rate": 0.0002, "epoch": 3.913978494623656, "step": 1820}, {"loss": 0.532, "grad_norm": 0.9068132042884827, "learning_rate": 0.0002, "epoch": 3.935483870967742, "step": 1830}, {"loss": 0.5427, "grad_norm": 0.8668948411941528, "learning_rate": 0.0002, "epoch": 3.956989247311828, "step": 1840}, {"loss": 0.5349, "grad_norm": 1.046032428741455, "learning_rate": 0.0002, "epoch": 3.978494623655914, "step": 1850}, {"loss": 0.5087, "grad_norm": 0.904780387878418, "learning_rate": 0.0002, "epoch": 4.0, "step": 1860}, {"eval_loss": 0.5737715363502502, "eval_runtime": 21.4915, "eval_samples_per_second": 15.401, "eval_steps_per_second": 1.954, "epoch": 4.0, "step": 1860}, {"loss": 0.4843, "grad_norm": 0.8611752986907959, "learning_rate": 0.0002, "epoch": 4.021505376344086, "step": 1870}, {"loss": 0.4814, "grad_norm": 0.838782548904419, "learning_rate": 0.0002, "epoch": 4.043010752688172, "step": 1880}, {"loss": 0.474, "grad_norm": 0.9119709134101868, "learning_rate": 0.0002, "epoch": 4.064516129032258, "step": 1890}, {"loss": 0.4951, "grad_norm": 0.8026251196861267, "learning_rate": 0.0002, "epoch": 4.086021505376344, "step": 1900}, {"loss": 0.491, "grad_norm": 0.8773705363273621, "learning_rate": 0.0002, "epoch": 4.10752688172043, "step": 1910}, {"loss": 0.474, "grad_norm": 0.8762255907058716, "learning_rate": 0.0002, "epoch": 4.129032258064516, "step": 1920}, {"loss": 0.4816, "grad_norm": 0.8371861577033997, "learning_rate": 0.0002, "epoch": 4.150537634408602, "step": 1930}, {"loss": 0.472, "grad_norm": 0.9703728556632996, "learning_rate": 0.0002, "epoch": 4.172043010752688, "step": 1940}, {"loss": 0.4772, "grad_norm": 0.8802874684333801, "learning_rate": 0.0002, "epoch": 4.193548387096774, "step": 1950}, {"loss": 0.5032, "grad_norm": 1.0103057622909546, "learning_rate": 0.0002, "epoch": 4.21505376344086, "step": 1960}, {"loss": 0.4945, "grad_norm": 0.9212995171546936, "learning_rate": 0.0002, "epoch": 4.236559139784946, "step": 1970}, {"loss": 0.4753, "grad_norm": 1.009544849395752, "learning_rate": 0.0002, "epoch": 4.258064516129032, "step": 1980}, {"loss": 0.4789, "grad_norm": 0.8535077571868896, "learning_rate": 0.0002, "epoch": 4.279569892473118, "step": 1990}, {"loss": 0.4782, "grad_norm": 0.8363022804260254, "learning_rate": 0.0002, "epoch": 4.301075268817204, "step": 2000}, {"loss": 0.4875, "grad_norm": 0.9041762948036194, "learning_rate": 0.0002, "epoch": 4.32258064516129, "step": 2010}, {"loss": 0.4779, "grad_norm": 0.960790753364563, "learning_rate": 0.0002, "epoch": 4.344086021505376, "step": 2020}, {"loss": 0.4626, "grad_norm": 0.8823095560073853, "learning_rate": 0.0002, "epoch": 4.365591397849462, "step": 2030}, {"loss": 0.4883, "grad_norm": 0.952100396156311, "learning_rate": 0.0002, "epoch": 4.387096774193548, "step": 2040}, {"loss": 0.4789, "grad_norm": 1.0793498754501343, "learning_rate": 0.0002, "epoch": 4.408602150537634, "step": 2050}, {"loss": 0.4827, "grad_norm": 0.8987208008766174, "learning_rate": 0.0002, "epoch": 4.43010752688172, "step": 2060}, {"loss": 0.4594, "grad_norm": 0.8539772033691406, "learning_rate": 0.0002, "epoch": 4.451612903225806, "step": 2070}, {"loss": 0.4752, "grad_norm": 0.9160863757133484, "learning_rate": 0.0002, "epoch": 4.473118279569892, "step": 2080}, {"loss": 0.5033, "grad_norm": 0.9946850538253784, "learning_rate": 0.0002, "epoch": 4.494623655913978, "step": 2090}, {"loss": 0.4842, "grad_norm": 0.908039391040802, "learning_rate": 0.0002, "epoch": 4.516129032258064, "step": 2100}, {"loss": 0.4861, "grad_norm": 1.1462254524230957, "learning_rate": 0.0002, "epoch": 4.53763440860215, "step": 2110}, {"loss": 0.4892, "grad_norm": 0.8392056226730347, "learning_rate": 0.0002, "epoch": 4.559139784946236, "step": 2120}, {"loss": 0.4824, "grad_norm": 0.9673896431922913, "learning_rate": 0.0002, "epoch": 4.580645161290323, "step": 2130}, {"loss": 0.4665, "grad_norm": 0.9047091603279114, "learning_rate": 0.0002, "epoch": 4.602150537634409, "step": 2140}, {"loss": 0.4714, "grad_norm": 0.9013425707817078, "learning_rate": 0.0002, "epoch": 4.623655913978495, "step": 2150}, {"loss": 0.472, "grad_norm": 0.8899165391921997, "learning_rate": 0.0002, "epoch": 4.645161290322581, "step": 2160}, {"loss": 0.4635, "grad_norm": 0.748602569103241, "learning_rate": 0.0002, "epoch": 4.666666666666667, "step": 2170}, {"loss": 0.4695, "grad_norm": 0.8694155216217041, "learning_rate": 0.0002, "epoch": 4.688172043010753, "step": 2180}, {"loss": 0.4929, "grad_norm": 0.9134316444396973, "learning_rate": 0.0002, "epoch": 4.709677419354839, "step": 2190}, {"loss": 0.4855, "grad_norm": 0.8504763245582581, "learning_rate": 0.0002, "epoch": 4.731182795698925, "step": 2200}, {"loss": 0.4517, "grad_norm": 1.0321544408798218, "learning_rate": 0.0002, "epoch": 4.752688172043011, "step": 2210}, {"loss": 0.4796, "grad_norm": 0.9368237257003784, "learning_rate": 0.0002, "epoch": 4.774193548387097, "step": 2220}, {"loss": 0.4837, "grad_norm": 0.9319947361946106, "learning_rate": 0.0002, "epoch": 4.795698924731183, "step": 2230}, {"loss": 0.4696, "grad_norm": 0.904333770275116, "learning_rate": 0.0002, "epoch": 4.817204301075269, "step": 2240}, {"loss": 0.4746, "grad_norm": 0.8097078204154968, "learning_rate": 0.0002, "epoch": 4.838709677419355, "step": 2250}, {"loss": 0.4438, "grad_norm": 0.9128859043121338, "learning_rate": 0.0002, "epoch": 4.860215053763441, "step": 2260}, {"loss": 0.4693, "grad_norm": 0.883129894733429, "learning_rate": 0.0002, "epoch": 4.881720430107527, "step": 2270}, {"loss": 0.4494, "grad_norm": 0.85712730884552, "learning_rate": 0.0002, "epoch": 4.903225806451613, "step": 2280}, {"loss": 0.4593, "grad_norm": 1.2101863622665405, "learning_rate": 0.0002, "epoch": 4.924731182795699, "step": 2290}, {"loss": 0.4779, "grad_norm": 0.917966902256012, "learning_rate": 0.0002, "epoch": 4.946236559139785, "step": 2300}, {"loss": 0.4666, "grad_norm": 0.7740724086761475, "learning_rate": 0.0002, "epoch": 4.967741935483871, "step": 2310}, {"loss": 0.4629, "grad_norm": 1.0199906826019287, "learning_rate": 0.0002, "epoch": 4.989247311827957, "step": 2320}, {"eval_loss": 0.5363914370536804, "eval_runtime": 21.3941, "eval_samples_per_second": 15.472, "eval_steps_per_second": 1.963, "epoch": 5.0, "step": 2325}, {"loss": 0.4543, "grad_norm": 0.8580502271652222, "learning_rate": 0.0002, "epoch": 5.010752688172043, "step": 2330}, {"loss": 0.404, "grad_norm": 0.7702704668045044, "learning_rate": 0.0002, "epoch": 5.032258064516129, "step": 2340}, {"loss": 0.4408, "grad_norm": 0.9417401552200317, "learning_rate": 0.0002, "epoch": 5.053763440860215, "step": 2350}, {"loss": 0.4306, "grad_norm": 0.9461463689804077, "learning_rate": 0.0002, "epoch": 5.075268817204301, "step": 2360}, {"loss": 0.4251, "grad_norm": 0.8931282162666321, "learning_rate": 0.0002, "epoch": 5.096774193548387, "step": 2370}, {"loss": 0.4249, "grad_norm": 1.000909447669983, "learning_rate": 0.0002, "epoch": 5.118279569892473, "step": 2380}, {"loss": 0.4231, "grad_norm": 0.8640249967575073, "learning_rate": 0.0002, "epoch": 5.139784946236559, "step": 2390}, {"loss": 0.4272, "grad_norm": 1.0451020002365112, "learning_rate": 0.0002, "epoch": 5.161290322580645, "step": 2400}, {"loss": 0.4177, "grad_norm": 0.7896912097930908, "learning_rate": 0.0002, "epoch": 5.182795698924731, "step": 2410}, {"loss": 0.4116, "grad_norm": 0.8424463272094727, "learning_rate": 0.0002, "epoch": 5.204301075268817, "step": 2420}, {"loss": 0.4225, "grad_norm": 1.0852105617523193, "learning_rate": 0.0002, "epoch": 5.225806451612903, "step": 2430}, {"loss": 0.4352, "grad_norm": 0.9285983443260193, "learning_rate": 0.0002, "epoch": 5.247311827956989, "step": 2440}, {"loss": 0.4262, "grad_norm": 0.9119299054145813, "learning_rate": 0.0002, "epoch": 5.268817204301075, "step": 2450}, {"loss": 0.4494, "grad_norm": 0.8790456056594849, "learning_rate": 0.0002, "epoch": 5.290322580645161, "step": 2460}, {"loss": 0.4421, "grad_norm": 0.8726504445075989, "learning_rate": 0.0002, "epoch": 5.311827956989247, "step": 2470}, {"loss": 0.4372, "grad_norm": 0.9415227770805359, "learning_rate": 0.0002, "epoch": 5.333333333333333, "step": 2480}, {"loss": 0.4223, "grad_norm": 0.9133324027061462, "learning_rate": 0.0002, "epoch": 5.354838709677419, "step": 2490}, {"loss": 0.4401, "grad_norm": 0.9567879438400269, "learning_rate": 0.0002, "epoch": 5.376344086021505, "step": 2500}, {"loss": 0.4094, "grad_norm": 0.9239469766616821, "learning_rate": 0.0002, "epoch": 5.397849462365591, "step": 2510}, {"loss": 0.4416, "grad_norm": 1.0293527841567993, "learning_rate": 0.0002, "epoch": 5.419354838709677, "step": 2520}, {"loss": 0.4311, "grad_norm": 0.8618718981742859, "learning_rate": 0.0002, "epoch": 5.440860215053763, "step": 2530}, {"loss": 0.462, "grad_norm": 0.740166187286377, "learning_rate": 0.0002, "epoch": 5.462365591397849, "step": 2540}, {"loss": 0.4172, "grad_norm": 0.901566743850708, "learning_rate": 0.0002, "epoch": 5.483870967741936, "step": 2550}, {"loss": 0.4315, "grad_norm": 0.7957597970962524, "learning_rate": 0.0002, "epoch": 5.505376344086022, "step": 2560}, {"loss": 0.4263, "grad_norm": 1.1139343976974487, "learning_rate": 0.0002, "epoch": 5.526881720430108, "step": 2570}, {"loss": 0.4056, "grad_norm": 0.989765465259552, "learning_rate": 0.0002, "epoch": 5.548387096774194, "step": 2580}, {"loss": 0.4311, "grad_norm": 0.9416969418525696, "learning_rate": 0.0002, "epoch": 5.56989247311828, "step": 2590}, {"loss": 0.4363, "grad_norm": 0.9184830784797668, "learning_rate": 0.0002, "epoch": 5.591397849462366, "step": 2600}, {"loss": 0.432, "grad_norm": 1.0512700080871582, "learning_rate": 0.0002, "epoch": 5.612903225806452, "step": 2610}, {"loss": 0.4227, "grad_norm": 0.901462197303772, "learning_rate": 0.0002, "epoch": 5.634408602150538, "step": 2620}, {"loss": 0.4332, "grad_norm": 0.9732566475868225, "learning_rate": 0.0002, "epoch": 5.655913978494624, "step": 2630}, {"loss": 0.4223, "grad_norm": 0.8180275559425354, "learning_rate": 0.0002, "epoch": 5.67741935483871, "step": 2640}, {"loss": 0.4311, "grad_norm": 1.1354765892028809, "learning_rate": 0.0002, "epoch": 5.698924731182796, "step": 2650}, {"loss": 0.4409, "grad_norm": 0.9161503314971924, "learning_rate": 0.0002, "epoch": 5.720430107526882, "step": 2660}, {"loss": 0.4394, "grad_norm": 1.0561772584915161, "learning_rate": 0.0002, "epoch": 5.741935483870968, "step": 2670}, {"loss": 0.424, "grad_norm": 0.7712787389755249, "learning_rate": 0.0002, "epoch": 5.763440860215054, "step": 2680}, {"loss": 0.4326, "grad_norm": 0.9674550294876099, "learning_rate": 0.0002, "epoch": 5.78494623655914, "step": 2690}, {"loss": 0.4459, "grad_norm": 0.7531843781471252, "learning_rate": 0.0002, "epoch": 5.806451612903226, "step": 2700}, {"loss": 0.4276, "grad_norm": 1.1332131624221802, "learning_rate": 0.0002, "epoch": 5.827956989247312, "step": 2710}, {"loss": 0.4113, "grad_norm": 0.9367414116859436, "learning_rate": 0.0002, "epoch": 5.849462365591398, "step": 2720}, {"loss": 0.4227, "grad_norm": 0.8267706632614136, "learning_rate": 0.0002, "epoch": 5.870967741935484, "step": 2730}, {"loss": 0.4218, "grad_norm": 1.1040657758712769, "learning_rate": 0.0002, "epoch": 5.89247311827957, "step": 2740}, {"loss": 0.4129, "grad_norm": 0.8879582285881042, "learning_rate": 0.0002, "epoch": 5.913978494623656, "step": 2750}, {"loss": 0.4241, "grad_norm": 0.9264667630195618, "learning_rate": 0.0002, "epoch": 5.935483870967742, "step": 2760}, {"loss": 0.4318, "grad_norm": 0.9373905658721924, "learning_rate": 0.0002, "epoch": 5.956989247311828, "step": 2770}, {"loss": 0.423, "grad_norm": 1.0063740015029907, "learning_rate": 0.0002, "epoch": 5.978494623655914, "step": 2780}, {"loss": 0.4382, "grad_norm": 0.8291367292404175, "learning_rate": 0.0002, "epoch": 6.0, "step": 2790}, {"eval_loss": 0.5057176947593689, "eval_runtime": 21.3206, "eval_samples_per_second": 15.525, "eval_steps_per_second": 1.97, "epoch": 6.0, "step": 2790}, {"loss": 0.3907, "grad_norm": 1.0137434005737305, "learning_rate": 0.0002, "epoch": 6.021505376344086, "step": 2800}, {"loss": 0.3793, "grad_norm": 0.7550579905509949, "learning_rate": 0.0002, "epoch": 6.043010752688172, "step": 2810}, {"loss": 0.4003, "grad_norm": 1.0664116144180298, "learning_rate": 0.0002, "epoch": 6.064516129032258, "step": 2820}, {"loss": 0.3876, "grad_norm": 0.7908814549446106, "learning_rate": 0.0002, "epoch": 6.086021505376344, "step": 2830}, {"loss": 0.3884, "grad_norm": 0.8101639747619629, "learning_rate": 0.0002, "epoch": 6.10752688172043, "step": 2840}, {"loss": 0.3835, "grad_norm": 0.7882567048072815, "learning_rate": 0.0002, "epoch": 6.129032258064516, "step": 2850}, {"loss": 0.3827, "grad_norm": 1.0134103298187256, "learning_rate": 0.0002, "epoch": 6.150537634408602, "step": 2860}, {"loss": 0.3963, "grad_norm": 0.9240215420722961, "learning_rate": 0.0002, "epoch": 6.172043010752688, "step": 2870}, {"loss": 0.4049, "grad_norm": 0.8322992920875549, "learning_rate": 0.0002, "epoch": 6.193548387096774, "step": 2880}, {"loss": 0.381, "grad_norm": 0.9238720536231995, "learning_rate": 0.0002, "epoch": 6.21505376344086, "step": 2890}, {"loss": 0.3852, "grad_norm": 0.9361863732337952, "learning_rate": 0.0002, "epoch": 6.236559139784946, "step": 2900}, {"loss": 0.3917, "grad_norm": 0.9670863747596741, "learning_rate": 0.0002, "epoch": 6.258064516129032, "step": 2910}, {"loss": 0.3826, "grad_norm": 0.7724685668945312, "learning_rate": 0.0002, "epoch": 6.279569892473118, "step": 2920}, {"loss": 0.3988, "grad_norm": 0.8125540614128113, "learning_rate": 0.0002, "epoch": 6.301075268817204, "step": 2930}, {"loss": 0.3778, "grad_norm": 0.9483002424240112, "learning_rate": 0.0002, "epoch": 6.32258064516129, "step": 2940}, {"loss": 0.3823, "grad_norm": 1.098374843597412, "learning_rate": 0.0002, "epoch": 6.344086021505376, "step": 2950}, {"loss": 0.3886, "grad_norm": 1.0169378519058228, "learning_rate": 0.0002, "epoch": 6.365591397849462, "step": 2960}, {"loss": 0.3936, "grad_norm": 0.8594151139259338, "learning_rate": 0.0002, "epoch": 6.387096774193548, "step": 2970}, {"loss": 0.3871, "grad_norm": 0.9507288336753845, "learning_rate": 0.0002, "epoch": 6.408602150537634, "step": 2980}, {"loss": 0.3852, "grad_norm": 0.9212459325790405, "learning_rate": 0.0002, "epoch": 6.43010752688172, "step": 2990}, {"loss": 0.3929, "grad_norm": 0.9696952104568481, "learning_rate": 0.0002, "epoch": 6.451612903225806, "step": 3000}, {"loss": 0.3933, "grad_norm": 0.8872610330581665, "learning_rate": 0.0002, "epoch": 6.473118279569892, "step": 3010}, {"loss": 0.393, "grad_norm": 0.9207532405853271, "learning_rate": 0.0002, "epoch": 6.494623655913978, "step": 3020}, {"loss": 0.3848, "grad_norm": 0.9116262793540955, "learning_rate": 0.0002, "epoch": 6.516129032258064, "step": 3030}, {"loss": 0.3964, "grad_norm": 0.83391934633255, "learning_rate": 0.0002, "epoch": 6.53763440860215, "step": 3040}, {"loss": 0.3758, "grad_norm": 0.890931248664856, "learning_rate": 0.0002, "epoch": 6.559139784946236, "step": 3050}, {"loss": 0.3944, "grad_norm": 1.0100581645965576, "learning_rate": 0.0002, "epoch": 6.580645161290323, "step": 3060}, {"loss": 0.3992, "grad_norm": 0.783526599407196, "learning_rate": 0.0002, "epoch": 6.602150537634409, "step": 3070}, {"loss": 0.4144, "grad_norm": 1.324326515197754, "learning_rate": 0.0002, "epoch": 6.623655913978495, "step": 3080}, {"loss": 0.3986, "grad_norm": 0.9102319478988647, "learning_rate": 0.0002, "epoch": 6.645161290322581, "step": 3090}, {"loss": 0.3873, "grad_norm": 0.96951824426651, "learning_rate": 0.0002, "epoch": 6.666666666666667, "step": 3100}, {"loss": 0.3931, "grad_norm": 0.9786809086799622, "learning_rate": 0.0002, "epoch": 6.688172043010753, "step": 3110}, {"loss": 0.3714, "grad_norm": 1.0301238298416138, "learning_rate": 0.0002, "epoch": 6.709677419354839, "step": 3120}, {"loss": 0.3823, "grad_norm": 1.1690906286239624, "learning_rate": 0.0002, "epoch": 6.731182795698925, "step": 3130}, {"loss": 0.3936, "grad_norm": 0.963306725025177, "learning_rate": 0.0002, "epoch": 6.752688172043011, "step": 3140}, {"loss": 0.3975, "grad_norm": 0.8565770983695984, "learning_rate": 0.0002, "epoch": 6.774193548387097, "step": 3150}, {"loss": 0.3903, "grad_norm": 0.8887158632278442, "learning_rate": 0.0002, "epoch": 6.795698924731183, "step": 3160}, {"loss": 0.4098, "grad_norm": 0.8234561085700989, "learning_rate": 0.0002, "epoch": 6.817204301075269, "step": 3170}, {"loss": 0.4041, "grad_norm": 0.9000219702720642, "learning_rate": 0.0002, "epoch": 6.838709677419355, "step": 3180}, {"loss": 0.3933, "grad_norm": 1.1366009712219238, "learning_rate": 0.0002, "epoch": 6.860215053763441, "step": 3190}, {"loss": 0.3972, "grad_norm": 0.8747097849845886, "learning_rate": 0.0002, "epoch": 6.881720430107527, "step": 3200}, {"loss": 0.404, "grad_norm": 0.8533893823623657, "learning_rate": 0.0002, "epoch": 6.903225806451613, "step": 3210}, {"loss": 0.3906, "grad_norm": 0.8127949237823486, "learning_rate": 0.0002, "epoch": 6.924731182795699, "step": 3220}, {"loss": 0.3747, "grad_norm": 0.8872477412223816, "learning_rate": 0.0002, "epoch": 6.946236559139785, "step": 3230}, {"loss": 0.3817, "grad_norm": 0.8541608452796936, "learning_rate": 0.0002, "epoch": 6.967741935483871, "step": 3240}, {"loss": 0.3863, "grad_norm": 0.8390752673149109, "learning_rate": 0.0002, "epoch": 6.989247311827957, "step": 3250}]}
+{"epoch": 8.0, "step": 3720, "epoch_duration": 451.2651696205139, "total_accumulated_duration": 3603.8767008781433, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 3020.60888671875}, "peak_memory_usage": {"GPU_0": 15051.17431640625}, "avg_memory_reserved": {"GPU_0": 20170.0}, "peak_memory_reserved": {"GPU_0": 20170.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2b-it_int4_flare-headlines_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.4-num-8628-sd-4/checkpoint-3255", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 3.4172, "grad_norm": 0.9075053930282593, "learning_rate": 0.0002, "epoch": 0.021505376344086023, "step": 10}, {"loss": 2.5888, "grad_norm": 1.4321208000183105, "learning_rate": 0.0002, "epoch": 0.043010752688172046, "step": 20}, {"loss": 2.1195, "grad_norm": 1.7500602006912231, "learning_rate": 0.0002, "epoch": 0.06451612903225806, "step": 30}, {"loss": 1.9303, "grad_norm": 0.7606641054153442, "learning_rate": 0.0002, "epoch": 0.08602150537634409, "step": 40}, {"loss": 1.6112, "grad_norm": 1.2754929065704346, "learning_rate": 0.0002, "epoch": 0.10752688172043011, "step": 50}, {"loss": 1.4319, "grad_norm": 1.0936230421066284, "learning_rate": 0.0002, "epoch": 0.12903225806451613, "step": 60}, {"loss": 1.3568, "grad_norm": 1.144593596458435, "learning_rate": 0.0002, "epoch": 0.15053763440860216, "step": 70}, {"loss": 1.2028, "grad_norm": 1.2181956768035889, "learning_rate": 0.0002, "epoch": 0.17204301075268819, "step": 80}, {"loss": 1.1534, "grad_norm": 1.1260095834732056, "learning_rate": 0.0002, "epoch": 0.1935483870967742, "step": 90}, {"loss": 1.1089, "grad_norm": 1.1155284643173218, "learning_rate": 0.0002, "epoch": 0.21505376344086022, "step": 100}, {"loss": 1.0883, "grad_norm": 1.089565396308899, "learning_rate": 0.0002, "epoch": 0.23655913978494625, "step": 110}, {"loss": 1.0814, "grad_norm": 0.9833471775054932, "learning_rate": 0.0002, "epoch": 0.25806451612903225, "step": 120}, {"loss": 1.0239, "grad_norm": 1.0265629291534424, "learning_rate": 0.0002, "epoch": 0.27956989247311825, "step": 130}, {"loss": 0.9888, "grad_norm": 0.9344286322593689, "learning_rate": 0.0002, "epoch": 0.3010752688172043, "step": 140}, {"loss": 1.0043, "grad_norm": 0.9883386492729187, "learning_rate": 0.0002, "epoch": 0.3225806451612903, "step": 150}, {"loss": 0.9338, "grad_norm": 0.9299277067184448, "learning_rate": 0.0002, "epoch": 0.34408602150537637, "step": 160}, {"loss": 0.9432, "grad_norm": 1.390045404434204, "learning_rate": 0.0002, "epoch": 0.3655913978494624, "step": 170}, {"loss": 0.9008, "grad_norm": 1.0313078165054321, "learning_rate": 0.0002, "epoch": 0.3870967741935484, "step": 180}, {"loss": 0.9434, "grad_norm": 1.1792205572128296, "learning_rate": 0.0002, "epoch": 0.40860215053763443, "step": 190}, {"loss": 0.8761, "grad_norm": 1.049809217453003, "learning_rate": 0.0002, "epoch": 0.43010752688172044, "step": 200}, {"loss": 0.8709, "grad_norm": 0.990111768245697, "learning_rate": 0.0002, "epoch": 0.45161290322580644, "step": 210}, {"loss": 0.905, "grad_norm": 0.9870412349700928, "learning_rate": 0.0002, "epoch": 0.4731182795698925, "step": 220}, {"loss": 0.9129, "grad_norm": 0.8557345867156982, "learning_rate": 0.0002, "epoch": 0.4946236559139785, "step": 230}, {"loss": 0.8836, "grad_norm": 0.9746861457824707, "learning_rate": 0.0002, "epoch": 0.5161290322580645, "step": 240}, {"loss": 0.873, "grad_norm": 0.9010438323020935, "learning_rate": 0.0002, "epoch": 0.5376344086021505, "step": 250}, {"loss": 0.8241, "grad_norm": 0.9061082005500793, "learning_rate": 0.0002, "epoch": 0.5591397849462365, "step": 260}, {"loss": 0.8652, "grad_norm": 0.9311846494674683, "learning_rate": 0.0002, "epoch": 0.5806451612903226, "step": 270}, {"loss": 0.8256, "grad_norm": 0.9140254855155945, "learning_rate": 0.0002, "epoch": 0.6021505376344086, "step": 280}, {"loss": 0.8441, "grad_norm": 0.9722253084182739, "learning_rate": 0.0002, "epoch": 0.6236559139784946, "step": 290}, {"loss": 0.8314, "grad_norm": 0.8539168238639832, "learning_rate": 0.0002, "epoch": 0.6451612903225806, "step": 300}, {"loss": 0.8528, "grad_norm": 0.9053162932395935, "learning_rate": 0.0002, "epoch": 0.6666666666666666, "step": 310}, {"loss": 0.8209, "grad_norm": 0.8444252610206604, "learning_rate": 0.0002, "epoch": 0.6881720430107527, "step": 320}, {"loss": 0.8101, "grad_norm": 0.8127437829971313, "learning_rate": 0.0002, "epoch": 0.7096774193548387, "step": 330}, {"loss": 0.8223, "grad_norm": 0.886555016040802, "learning_rate": 0.0002, "epoch": 0.7311827956989247, "step": 340}, {"loss": 0.8368, "grad_norm": 0.8458548784255981, "learning_rate": 0.0002, "epoch": 0.7526881720430108, "step": 350}, {"loss": 0.8295, "grad_norm": 0.8683297634124756, "learning_rate": 0.0002, "epoch": 0.7741935483870968, "step": 360}, {"loss": 0.8232, "grad_norm": 0.8308405876159668, "learning_rate": 0.0002, "epoch": 0.7956989247311828, "step": 370}, {"loss": 0.7752, "grad_norm": 0.8305579423904419, "learning_rate": 0.0002, "epoch": 0.8172043010752689, "step": 380}, {"loss": 0.8267, "grad_norm": 0.8545567393302917, "learning_rate": 0.0002, "epoch": 0.8387096774193549, "step": 390}, {"loss": 0.8212, "grad_norm": 0.8486055731773376, "learning_rate": 0.0002, "epoch": 0.8602150537634409, "step": 400}, {"loss": 0.743, "grad_norm": 0.8126763105392456, "learning_rate": 0.0002, "epoch": 0.8817204301075269, "step": 410}, {"loss": 0.7993, "grad_norm": 0.8494045734405518, "learning_rate": 0.0002, "epoch": 0.9032258064516129, "step": 420}, {"loss": 0.8213, "grad_norm": 0.7639183402061462, "learning_rate": 0.0002, "epoch": 0.9247311827956989, "step": 430}, {"loss": 0.8015, "grad_norm": 0.858101487159729, "learning_rate": 0.0002, "epoch": 0.946236559139785, "step": 440}, {"loss": 0.7629, "grad_norm": 0.8141381740570068, "learning_rate": 0.0002, "epoch": 0.967741935483871, "step": 450}, {"loss": 0.7357, "grad_norm": 0.8072513937950134, "learning_rate": 0.0002, "epoch": 0.989247311827957, "step": 460}, {"eval_loss": 0.7740864157676697, "eval_runtime": 21.383, "eval_samples_per_second": 15.48, "eval_steps_per_second": 1.964, "epoch": 1.0, "step": 465}, {"loss": 0.7701, "grad_norm": 0.8269494771957397, "learning_rate": 0.0002, "epoch": 1.010752688172043, "step": 470}, {"loss": 0.7532, "grad_norm": 0.7814009189605713, "learning_rate": 0.0002, "epoch": 1.032258064516129, "step": 480}, {"loss": 0.7689, "grad_norm": 0.8183923363685608, "learning_rate": 0.0002, "epoch": 1.053763440860215, "step": 490}, {"loss": 0.765, "grad_norm": 0.8146600723266602, "learning_rate": 0.0002, "epoch": 1.075268817204301, "step": 500}, {"loss": 0.7358, "grad_norm": 0.8635126352310181, "learning_rate": 0.0002, "epoch": 1.096774193548387, "step": 510}, {"loss": 0.7302, "grad_norm": 0.8520359396934509, "learning_rate": 0.0002, "epoch": 1.118279569892473, "step": 520}, {"loss": 0.7492, "grad_norm": 0.8026443123817444, "learning_rate": 0.0002, "epoch": 1.139784946236559, "step": 530}, {"loss": 0.7518, "grad_norm": 0.8157258629798889, "learning_rate": 0.0002, "epoch": 1.1612903225806452, "step": 540}, {"loss": 0.7461, "grad_norm": 0.9450796246528625, "learning_rate": 0.0002, "epoch": 1.1827956989247312, "step": 550}, {"loss": 0.7128, "grad_norm": 0.8859835863113403, "learning_rate": 0.0002, "epoch": 1.2043010752688172, "step": 560}, {"loss": 0.7067, "grad_norm": 0.7819921970367432, "learning_rate": 0.0002, "epoch": 1.2258064516129032, "step": 570}, {"loss": 0.7577, "grad_norm": 0.7823445796966553, "learning_rate": 0.0002, "epoch": 1.2473118279569892, "step": 580}, {"loss": 0.7358, "grad_norm": 0.7931883931159973, "learning_rate": 0.0002, "epoch": 1.2688172043010753, "step": 590}, {"loss": 0.723, "grad_norm": 0.7495734095573425, "learning_rate": 0.0002, "epoch": 1.2903225806451613, "step": 600}, {"loss": 0.7386, "grad_norm": 0.9272717237472534, "learning_rate": 0.0002, "epoch": 1.3118279569892473, "step": 610}, {"loss": 0.7498, "grad_norm": 0.7968398332595825, "learning_rate": 0.0002, "epoch": 1.3333333333333333, "step": 620}, {"loss": 0.7635, "grad_norm": 0.7813659310340881, "learning_rate": 0.0002, "epoch": 1.3548387096774195, "step": 630}, {"loss": 0.6665, "grad_norm": 0.730925977230072, "learning_rate": 0.0002, "epoch": 1.3763440860215055, "step": 640}, {"loss": 0.7037, "grad_norm": 0.8011482954025269, "learning_rate": 0.0002, "epoch": 1.3978494623655915, "step": 650}, {"loss": 0.6931, "grad_norm": 0.7770085334777832, "learning_rate": 0.0002, "epoch": 1.4193548387096775, "step": 660}, {"loss": 0.6949, "grad_norm": 0.7432682514190674, "learning_rate": 0.0002, "epoch": 1.4408602150537635, "step": 670}, {"loss": 0.7444, "grad_norm": 0.8820092678070068, "learning_rate": 0.0002, "epoch": 1.4623655913978495, "step": 680}, {"loss": 0.6758, "grad_norm": 0.7786208987236023, "learning_rate": 0.0002, "epoch": 1.4838709677419355, "step": 690}, {"loss": 0.6702, "grad_norm": 0.7467480301856995, "learning_rate": 0.0002, "epoch": 1.5053763440860215, "step": 700}, {"loss": 0.7107, "grad_norm": 0.8147122263908386, "learning_rate": 0.0002, "epoch": 1.5268817204301075, "step": 710}, {"loss": 0.7144, "grad_norm": 0.796030580997467, "learning_rate": 0.0002, "epoch": 1.5483870967741935, "step": 720}, {"loss": 0.6936, "grad_norm": 0.8776171207427979, "learning_rate": 0.0002, "epoch": 1.5698924731182795, "step": 730}, {"loss": 0.7101, "grad_norm": 0.8056126236915588, "learning_rate": 0.0002, "epoch": 1.5913978494623655, "step": 740}, {"loss": 0.7162, "grad_norm": 0.8141863346099854, "learning_rate": 0.0002, "epoch": 1.6129032258064515, "step": 750}, {"loss": 0.7088, "grad_norm": 0.8100557327270508, "learning_rate": 0.0002, "epoch": 1.6344086021505375, "step": 760}, {"loss": 0.7212, "grad_norm": 0.8283200860023499, "learning_rate": 0.0002, "epoch": 1.6559139784946235, "step": 770}, {"loss": 0.694, "grad_norm": 0.800865113735199, "learning_rate": 0.0002, "epoch": 1.6774193548387095, "step": 780}, {"loss": 0.7076, "grad_norm": 0.8052287697792053, "learning_rate": 0.0002, "epoch": 1.6989247311827957, "step": 790}, {"loss": 0.7257, "grad_norm": 0.8619674444198608, "learning_rate": 0.0002, "epoch": 1.7204301075268817, "step": 800}, {"loss": 0.7141, "grad_norm": 0.8907215595245361, "learning_rate": 0.0002, "epoch": 1.7419354838709677, "step": 810}, {"loss": 0.7035, "grad_norm": 0.6976316571235657, "learning_rate": 0.0002, "epoch": 1.7634408602150538, "step": 820}, {"loss": 0.6916, "grad_norm": 0.7533746957778931, "learning_rate": 0.0002, "epoch": 1.7849462365591398, "step": 830}, {"loss": 0.7094, "grad_norm": 0.7326804399490356, "learning_rate": 0.0002, "epoch": 1.8064516129032258, "step": 840}, {"loss": 0.6891, "grad_norm": 0.7782683372497559, "learning_rate": 0.0002, "epoch": 1.827956989247312, "step": 850}, {"loss": 0.6931, "grad_norm": 0.7424806356430054, "learning_rate": 0.0002, "epoch": 1.849462365591398, "step": 860}, {"loss": 0.7354, "grad_norm": 1.172325611114502, "learning_rate": 0.0002, "epoch": 1.870967741935484, "step": 870}, {"loss": 0.6866, "grad_norm": 0.771058201789856, "learning_rate": 0.0002, "epoch": 1.89247311827957, "step": 880}, {"loss": 0.7296, "grad_norm": 0.8624904155731201, "learning_rate": 0.0002, "epoch": 1.913978494623656, "step": 890}, {"loss": 0.7233, "grad_norm": 0.7062820792198181, "learning_rate": 0.0002, "epoch": 1.935483870967742, "step": 900}, {"loss": 0.6966, "grad_norm": 0.7560103535652161, "learning_rate": 0.0002, "epoch": 1.956989247311828, "step": 910}, {"loss": 0.69, "grad_norm": 0.788899838924408, "learning_rate": 0.0002, "epoch": 1.978494623655914, "step": 920}, {"loss": 0.6505, "grad_norm": 0.6562113761901855, "learning_rate": 0.0002, "epoch": 2.0, "step": 930}, {"eval_loss": 0.6885261535644531, "eval_runtime": 21.4291, "eval_samples_per_second": 15.446, "eval_steps_per_second": 1.96, "epoch": 2.0, "step": 930}, {"loss": 0.6625, "grad_norm": 0.8216531872749329, "learning_rate": 0.0002, "epoch": 2.021505376344086, "step": 940}, {"loss": 0.6398, "grad_norm": 0.8317142724990845, "learning_rate": 0.0002, "epoch": 2.043010752688172, "step": 950}, {"loss": 0.649, "grad_norm": 0.8446708917617798, "learning_rate": 0.0002, "epoch": 2.064516129032258, "step": 960}, {"loss": 0.657, "grad_norm": 0.735055148601532, "learning_rate": 0.0002, "epoch": 2.086021505376344, "step": 970}, {"loss": 0.649, "grad_norm": 0.7487243413925171, "learning_rate": 0.0002, "epoch": 2.10752688172043, "step": 980}, {"loss": 0.6419, "grad_norm": 0.8573887944221497, "learning_rate": 0.0002, "epoch": 2.129032258064516, "step": 990}, {"loss": 0.6431, "grad_norm": 0.6284521818161011, "learning_rate": 0.0002, "epoch": 2.150537634408602, "step": 1000}, {"loss": 0.6128, "grad_norm": 0.754183292388916, "learning_rate": 0.0002, "epoch": 2.172043010752688, "step": 1010}, {"loss": 0.6253, "grad_norm": 0.9445359110832214, "learning_rate": 0.0002, "epoch": 2.193548387096774, "step": 1020}, {"loss": 0.605, "grad_norm": 0.808508038520813, "learning_rate": 0.0002, "epoch": 2.21505376344086, "step": 1030}, {"loss": 0.6786, "grad_norm": 0.9394679665565491, "learning_rate": 0.0002, "epoch": 2.236559139784946, "step": 1040}, {"loss": 0.6176, "grad_norm": 0.8151357769966125, "learning_rate": 0.0002, "epoch": 2.258064516129032, "step": 1050}, {"loss": 0.66, "grad_norm": 0.7909848093986511, "learning_rate": 0.0002, "epoch": 2.279569892473118, "step": 1060}, {"loss": 0.6254, "grad_norm": 0.7506507039070129, "learning_rate": 0.0002, "epoch": 2.3010752688172045, "step": 1070}, {"loss": 0.6608, "grad_norm": 0.8240520358085632, "learning_rate": 0.0002, "epoch": 2.3225806451612905, "step": 1080}, {"loss": 0.6207, "grad_norm": 0.9342400431632996, "learning_rate": 0.0002, "epoch": 2.3440860215053765, "step": 1090}, {"loss": 0.6029, "grad_norm": 1.0598735809326172, "learning_rate": 0.0002, "epoch": 2.3655913978494625, "step": 1100}, {"loss": 0.6035, "grad_norm": 0.7907650470733643, "learning_rate": 0.0002, "epoch": 2.3870967741935485, "step": 1110}, {"loss": 0.6237, "grad_norm": 0.9388798475265503, "learning_rate": 0.0002, "epoch": 2.4086021505376345, "step": 1120}, {"loss": 0.6207, "grad_norm": 0.8985419869422913, "learning_rate": 0.0002, "epoch": 2.4301075268817205, "step": 1130}, {"loss": 0.5902, "grad_norm": 0.7471932768821716, "learning_rate": 0.0002, "epoch": 2.4516129032258065, "step": 1140}, {"loss": 0.6446, "grad_norm": 0.761131763458252, "learning_rate": 0.0002, "epoch": 2.4731182795698925, "step": 1150}, {"loss": 0.6088, "grad_norm": 0.7901819348335266, "learning_rate": 0.0002, "epoch": 2.4946236559139785, "step": 1160}, {"loss": 0.6142, "grad_norm": 0.9932922720909119, "learning_rate": 0.0002, "epoch": 2.5161290322580645, "step": 1170}, {"loss": 0.6407, "grad_norm": 0.7414287328720093, "learning_rate": 0.0002, "epoch": 2.5376344086021505, "step": 1180}, {"loss": 0.6161, "grad_norm": 0.8111771941184998, "learning_rate": 0.0002, "epoch": 2.5591397849462365, "step": 1190}, {"loss": 0.6006, "grad_norm": 0.7520156502723694, "learning_rate": 0.0002, "epoch": 2.5806451612903225, "step": 1200}, {"loss": 0.615, "grad_norm": 0.9022907018661499, "learning_rate": 0.0002, "epoch": 2.6021505376344085, "step": 1210}, {"loss": 0.6211, "grad_norm": 0.7746260166168213, "learning_rate": 0.0002, "epoch": 2.6236559139784945, "step": 1220}, {"loss": 0.616, "grad_norm": 0.8482862114906311, "learning_rate": 0.0002, "epoch": 2.6451612903225805, "step": 1230}, {"loss": 0.6417, "grad_norm": 0.7925458550453186, "learning_rate": 0.0002, "epoch": 2.6666666666666665, "step": 1240}, {"loss": 0.6187, "grad_norm": 0.8369929194450378, "learning_rate": 0.0002, "epoch": 2.688172043010753, "step": 1250}, {"loss": 0.6138, "grad_norm": 0.8311542868614197, "learning_rate": 0.0002, "epoch": 2.709677419354839, "step": 1260}, {"loss": 0.5894, "grad_norm": 0.7204853296279907, "learning_rate": 0.0002, "epoch": 2.731182795698925, "step": 1270}, {"loss": 0.6325, "grad_norm": 0.8447284698486328, "learning_rate": 0.0002, "epoch": 2.752688172043011, "step": 1280}, {"loss": 0.5946, "grad_norm": 0.7738404273986816, "learning_rate": 0.0002, "epoch": 2.774193548387097, "step": 1290}, {"loss": 0.5678, "grad_norm": 0.8393287062644958, "learning_rate": 0.0002, "epoch": 2.795698924731183, "step": 1300}, {"loss": 0.6092, "grad_norm": 0.79121994972229, "learning_rate": 0.0002, "epoch": 2.817204301075269, "step": 1310}, {"loss": 0.5889, "grad_norm": 0.7331557869911194, "learning_rate": 0.0002, "epoch": 2.838709677419355, "step": 1320}, {"loss": 0.6048, "grad_norm": 0.9593998193740845, "learning_rate": 0.0002, "epoch": 2.860215053763441, "step": 1330}, {"loss": 0.6108, "grad_norm": 0.7215158343315125, "learning_rate": 0.0002, "epoch": 2.881720430107527, "step": 1340}, {"loss": 0.5897, "grad_norm": 0.840404212474823, "learning_rate": 0.0002, "epoch": 2.903225806451613, "step": 1350}, {"loss": 0.6056, "grad_norm": 0.870659351348877, "learning_rate": 0.0002, "epoch": 2.924731182795699, "step": 1360}, {"loss": 0.6205, "grad_norm": 0.8744975328445435, "learning_rate": 0.0002, "epoch": 2.946236559139785, "step": 1370}, {"loss": 0.5966, "grad_norm": 0.8030612468719482, "learning_rate": 0.0002, "epoch": 2.967741935483871, "step": 1380}, {"loss": 0.6004, "grad_norm": 0.825814962387085, "learning_rate": 0.0002, "epoch": 2.989247311827957, "step": 1390}, {"eval_loss": 0.6257933378219604, "eval_runtime": 21.3692, "eval_samples_per_second": 15.49, "eval_steps_per_second": 1.965, "epoch": 3.0, "step": 1395}, {"loss": 0.5696, "grad_norm": 0.8650677800178528, "learning_rate": 0.0002, "epoch": 3.010752688172043, "step": 1400}, {"loss": 0.5483, "grad_norm": 0.8364197015762329, "learning_rate": 0.0002, "epoch": 3.032258064516129, "step": 1410}, {"loss": 0.5606, "grad_norm": 0.8278448581695557, "learning_rate": 0.0002, "epoch": 3.053763440860215, "step": 1420}, {"loss": 0.5572, "grad_norm": 0.8806642889976501, "learning_rate": 0.0002, "epoch": 3.075268817204301, "step": 1430}, {"loss": 0.585, "grad_norm": 0.8180029988288879, "learning_rate": 0.0002, "epoch": 3.096774193548387, "step": 1440}, {"loss": 0.5667, "grad_norm": 0.8561782836914062, "learning_rate": 0.0002, "epoch": 3.118279569892473, "step": 1450}, {"loss": 0.5246, "grad_norm": 0.8377029299736023, "learning_rate": 0.0002, "epoch": 3.139784946236559, "step": 1460}, {"loss": 0.5464, "grad_norm": 0.885779082775116, "learning_rate": 0.0002, "epoch": 3.161290322580645, "step": 1470}, {"loss": 0.541, "grad_norm": 0.9388518333435059, "learning_rate": 0.0002, "epoch": 3.182795698924731, "step": 1480}, {"loss": 0.5447, "grad_norm": 0.8816235661506653, "learning_rate": 0.0002, "epoch": 3.204301075268817, "step": 1490}, {"loss": 0.5466, "grad_norm": 0.9885783791542053, "learning_rate": 0.0002, "epoch": 3.225806451612903, "step": 1500}, {"loss": 0.5455, "grad_norm": 0.8635850548744202, "learning_rate": 0.0002, "epoch": 3.247311827956989, "step": 1510}, {"loss": 0.5419, "grad_norm": 0.829853355884552, "learning_rate": 0.0002, "epoch": 3.268817204301075, "step": 1520}, {"loss": 0.54, "grad_norm": 0.9037486910820007, "learning_rate": 0.0002, "epoch": 3.2903225806451615, "step": 1530}, {"loss": 0.5375, "grad_norm": 0.8173713684082031, "learning_rate": 0.0002, "epoch": 3.3118279569892475, "step": 1540}, {"loss": 0.5405, "grad_norm": 0.796953022480011, "learning_rate": 0.0002, "epoch": 3.3333333333333335, "step": 1550}, {"loss": 0.5505, "grad_norm": 0.7894400358200073, "learning_rate": 0.0002, "epoch": 3.3548387096774195, "step": 1560}, {"loss": 0.5395, "grad_norm": 0.9434949159622192, "learning_rate": 0.0002, "epoch": 3.3763440860215055, "step": 1570}, {"loss": 0.5271, "grad_norm": 0.8666760325431824, "learning_rate": 0.0002, "epoch": 3.3978494623655915, "step": 1580}, {"loss": 0.5439, "grad_norm": 0.7782467007637024, "learning_rate": 0.0002, "epoch": 3.4193548387096775, "step": 1590}, {"loss": 0.5161, "grad_norm": 0.8849126696586609, "learning_rate": 0.0002, "epoch": 3.4408602150537635, "step": 1600}, {"loss": 0.5353, "grad_norm": 0.7863831520080566, "learning_rate": 0.0002, "epoch": 3.4623655913978495, "step": 1610}, {"loss": 0.5308, "grad_norm": 1.0403116941452026, "learning_rate": 0.0002, "epoch": 3.4838709677419355, "step": 1620}, {"loss": 0.5339, "grad_norm": 0.8307499289512634, "learning_rate": 0.0002, "epoch": 3.5053763440860215, "step": 1630}, {"loss": 0.5361, "grad_norm": 0.9132118821144104, "learning_rate": 0.0002, "epoch": 3.5268817204301075, "step": 1640}, {"loss": 0.5828, "grad_norm": 0.9322578310966492, "learning_rate": 0.0002, "epoch": 3.5483870967741935, "step": 1650}, {"loss": 0.546, "grad_norm": 0.9782460331916809, "learning_rate": 0.0002, "epoch": 3.5698924731182795, "step": 1660}, {"loss": 0.5424, "grad_norm": 0.7189919352531433, "learning_rate": 0.0002, "epoch": 3.5913978494623655, "step": 1670}, {"loss": 0.5514, "grad_norm": 0.9689221382141113, "learning_rate": 0.0002, "epoch": 3.6129032258064515, "step": 1680}, {"loss": 0.5379, "grad_norm": 0.9684675335884094, "learning_rate": 0.0002, "epoch": 3.6344086021505375, "step": 1690}, {"loss": 0.5748, "grad_norm": 0.8851472735404968, "learning_rate": 0.0002, "epoch": 3.6559139784946235, "step": 1700}, {"loss": 0.5412, "grad_norm": 0.7709833383560181, "learning_rate": 0.0002, "epoch": 3.6774193548387095, "step": 1710}, {"loss": 0.521, "grad_norm": 0.818236231803894, "learning_rate": 0.0002, "epoch": 3.698924731182796, "step": 1720}, {"loss": 0.5445, "grad_norm": 0.870642364025116, "learning_rate": 0.0002, "epoch": 3.720430107526882, "step": 1730}, {"loss": 0.5307, "grad_norm": 1.0245511531829834, "learning_rate": 0.0002, "epoch": 3.741935483870968, "step": 1740}, {"loss": 0.5593, "grad_norm": 0.8607558608055115, "learning_rate": 0.0002, "epoch": 3.763440860215054, "step": 1750}, {"loss": 0.536, "grad_norm": 0.8511829972267151, "learning_rate": 0.0002, "epoch": 3.78494623655914, "step": 1760}, {"loss": 0.5193, "grad_norm": 0.7969087362289429, "learning_rate": 0.0002, "epoch": 3.806451612903226, "step": 1770}, {"loss": 0.5578, "grad_norm": 0.8457245826721191, "learning_rate": 0.0002, "epoch": 3.827956989247312, "step": 1780}, {"loss": 0.5337, "grad_norm": 0.8893467783927917, "learning_rate": 0.0002, "epoch": 3.849462365591398, "step": 1790}, {"loss": 0.5024, "grad_norm": 0.8593819737434387, "learning_rate": 0.0002, "epoch": 3.870967741935484, "step": 1800}, {"loss": 0.5134, "grad_norm": 0.7574560642242432, "learning_rate": 0.0002, "epoch": 3.89247311827957, "step": 1810}, {"loss": 0.5263, "grad_norm": 0.8681567311286926, "learning_rate": 0.0002, "epoch": 3.913978494623656, "step": 1820}, {"loss": 0.532, "grad_norm": 0.9068132042884827, "learning_rate": 0.0002, "epoch": 3.935483870967742, "step": 1830}, {"loss": 0.5427, "grad_norm": 0.8668948411941528, "learning_rate": 0.0002, "epoch": 3.956989247311828, "step": 1840}, {"loss": 0.5349, "grad_norm": 1.046032428741455, "learning_rate": 0.0002, "epoch": 3.978494623655914, "step": 1850}, {"loss": 0.5087, "grad_norm": 0.904780387878418, "learning_rate": 0.0002, "epoch": 4.0, "step": 1860}, {"eval_loss": 0.5737715363502502, "eval_runtime": 21.4915, "eval_samples_per_second": 15.401, "eval_steps_per_second": 1.954, "epoch": 4.0, "step": 1860}, {"loss": 0.4843, "grad_norm": 0.8611752986907959, "learning_rate": 0.0002, "epoch": 4.021505376344086, "step": 1870}, {"loss": 0.4814, "grad_norm": 0.838782548904419, "learning_rate": 0.0002, "epoch": 4.043010752688172, "step": 1880}, {"loss": 0.474, "grad_norm": 0.9119709134101868, "learning_rate": 0.0002, "epoch": 4.064516129032258, "step": 1890}, {"loss": 0.4951, "grad_norm": 0.8026251196861267, "learning_rate": 0.0002, "epoch": 4.086021505376344, "step": 1900}, {"loss": 0.491, "grad_norm": 0.8773705363273621, "learning_rate": 0.0002, "epoch": 4.10752688172043, "step": 1910}, {"loss": 0.474, "grad_norm": 0.8762255907058716, "learning_rate": 0.0002, "epoch": 4.129032258064516, "step": 1920}, {"loss": 0.4816, "grad_norm": 0.8371861577033997, "learning_rate": 0.0002, "epoch": 4.150537634408602, "step": 1930}, {"loss": 0.472, "grad_norm": 0.9703728556632996, "learning_rate": 0.0002, "epoch": 4.172043010752688, "step": 1940}, {"loss": 0.4772, "grad_norm": 0.8802874684333801, "learning_rate": 0.0002, "epoch": 4.193548387096774, "step": 1950}, {"loss": 0.5032, "grad_norm": 1.0103057622909546, "learning_rate": 0.0002, "epoch": 4.21505376344086, "step": 1960}, {"loss": 0.4945, "grad_norm": 0.9212995171546936, "learning_rate": 0.0002, "epoch": 4.236559139784946, "step": 1970}, {"loss": 0.4753, "grad_norm": 1.009544849395752, "learning_rate": 0.0002, "epoch": 4.258064516129032, "step": 1980}, {"loss": 0.4789, "grad_norm": 0.8535077571868896, "learning_rate": 0.0002, "epoch": 4.279569892473118, "step": 1990}, {"loss": 0.4782, "grad_norm": 0.8363022804260254, "learning_rate": 0.0002, "epoch": 4.301075268817204, "step": 2000}, {"loss": 0.4875, "grad_norm": 0.9041762948036194, "learning_rate": 0.0002, "epoch": 4.32258064516129, "step": 2010}, {"loss": 0.4779, "grad_norm": 0.960790753364563, "learning_rate": 0.0002, "epoch": 4.344086021505376, "step": 2020}, {"loss": 0.4626, "grad_norm": 0.8823095560073853, "learning_rate": 0.0002, "epoch": 4.365591397849462, "step": 2030}, {"loss": 0.4883, "grad_norm": 0.952100396156311, "learning_rate": 0.0002, "epoch": 4.387096774193548, "step": 2040}, {"loss": 0.4789, "grad_norm": 1.0793498754501343, "learning_rate": 0.0002, "epoch": 4.408602150537634, "step": 2050}, {"loss": 0.4827, "grad_norm": 0.8987208008766174, "learning_rate": 0.0002, "epoch": 4.43010752688172, "step": 2060}, {"loss": 0.4594, "grad_norm": 0.8539772033691406, "learning_rate": 0.0002, "epoch": 4.451612903225806, "step": 2070}, {"loss": 0.4752, "grad_norm": 0.9160863757133484, "learning_rate": 0.0002, "epoch": 4.473118279569892, "step": 2080}, {"loss": 0.5033, "grad_norm": 0.9946850538253784, "learning_rate": 0.0002, "epoch": 4.494623655913978, "step": 2090}, {"loss": 0.4842, "grad_norm": 0.908039391040802, "learning_rate": 0.0002, "epoch": 4.516129032258064, "step": 2100}, {"loss": 0.4861, "grad_norm": 1.1462254524230957, "learning_rate": 0.0002, "epoch": 4.53763440860215, "step": 2110}, {"loss": 0.4892, "grad_norm": 0.8392056226730347, "learning_rate": 0.0002, "epoch": 4.559139784946236, "step": 2120}, {"loss": 0.4824, "grad_norm": 0.9673896431922913, "learning_rate": 0.0002, "epoch": 4.580645161290323, "step": 2130}, {"loss": 0.4665, "grad_norm": 0.9047091603279114, "learning_rate": 0.0002, "epoch": 4.602150537634409, "step": 2140}, {"loss": 0.4714, "grad_norm": 0.9013425707817078, "learning_rate": 0.0002, "epoch": 4.623655913978495, "step": 2150}, {"loss": 0.472, "grad_norm": 0.8899165391921997, "learning_rate": 0.0002, "epoch": 4.645161290322581, "step": 2160}, {"loss": 0.4635, "grad_norm": 0.748602569103241, "learning_rate": 0.0002, "epoch": 4.666666666666667, "step": 2170}, {"loss": 0.4695, "grad_norm": 0.8694155216217041, "learning_rate": 0.0002, "epoch": 4.688172043010753, "step": 2180}, {"loss": 0.4929, "grad_norm": 0.9134316444396973, "learning_rate": 0.0002, "epoch": 4.709677419354839, "step": 2190}, {"loss": 0.4855, "grad_norm": 0.8504763245582581, "learning_rate": 0.0002, "epoch": 4.731182795698925, "step": 2200}, {"loss": 0.4517, "grad_norm": 1.0321544408798218, "learning_rate": 0.0002, "epoch": 4.752688172043011, "step": 2210}, {"loss": 0.4796, "grad_norm": 0.9368237257003784, "learning_rate": 0.0002, "epoch": 4.774193548387097, "step": 2220}, {"loss": 0.4837, "grad_norm": 0.9319947361946106, "learning_rate": 0.0002, "epoch": 4.795698924731183, "step": 2230}, {"loss": 0.4696, "grad_norm": 0.904333770275116, "learning_rate": 0.0002, "epoch": 4.817204301075269, "step": 2240}, {"loss": 0.4746, "grad_norm": 0.8097078204154968, "learning_rate": 0.0002, "epoch": 4.838709677419355, "step": 2250}, {"loss": 0.4438, "grad_norm": 0.9128859043121338, "learning_rate": 0.0002, "epoch": 4.860215053763441, "step": 2260}, {"loss": 0.4693, "grad_norm": 0.883129894733429, "learning_rate": 0.0002, "epoch": 4.881720430107527, "step": 2270}, {"loss": 0.4494, "grad_norm": 0.85712730884552, "learning_rate": 0.0002, "epoch": 4.903225806451613, "step": 2280}, {"loss": 0.4593, "grad_norm": 1.2101863622665405, "learning_rate": 0.0002, "epoch": 4.924731182795699, "step": 2290}, {"loss": 0.4779, "grad_norm": 0.917966902256012, "learning_rate": 0.0002, "epoch": 4.946236559139785, "step": 2300}, {"loss": 0.4666, "grad_norm": 0.7740724086761475, "learning_rate": 0.0002, "epoch": 4.967741935483871, "step": 2310}, {"loss": 0.4629, "grad_norm": 1.0199906826019287, "learning_rate": 0.0002, "epoch": 4.989247311827957, "step": 2320}, {"eval_loss": 0.5363914370536804, "eval_runtime": 21.3941, "eval_samples_per_second": 15.472, "eval_steps_per_second": 1.963, "epoch": 5.0, "step": 2325}, {"loss": 0.4543, "grad_norm": 0.8580502271652222, "learning_rate": 0.0002, "epoch": 5.010752688172043, "step": 2330}, {"loss": 0.404, "grad_norm": 0.7702704668045044, "learning_rate": 0.0002, "epoch": 5.032258064516129, "step": 2340}, {"loss": 0.4408, "grad_norm": 0.9417401552200317, "learning_rate": 0.0002, "epoch": 5.053763440860215, "step": 2350}, {"loss": 0.4306, "grad_norm": 0.9461463689804077, "learning_rate": 0.0002, "epoch": 5.075268817204301, "step": 2360}, {"loss": 0.4251, "grad_norm": 0.8931282162666321, "learning_rate": 0.0002, "epoch": 5.096774193548387, "step": 2370}, {"loss": 0.4249, "grad_norm": 1.000909447669983, "learning_rate": 0.0002, "epoch": 5.118279569892473, "step": 2380}, {"loss": 0.4231, "grad_norm": 0.8640249967575073, "learning_rate": 0.0002, "epoch": 5.139784946236559, "step": 2390}, {"loss": 0.4272, "grad_norm": 1.0451020002365112, "learning_rate": 0.0002, "epoch": 5.161290322580645, "step": 2400}, {"loss": 0.4177, "grad_norm": 0.7896912097930908, "learning_rate": 0.0002, "epoch": 5.182795698924731, "step": 2410}, {"loss": 0.4116, "grad_norm": 0.8424463272094727, "learning_rate": 0.0002, "epoch": 5.204301075268817, "step": 2420}, {"loss": 0.4225, "grad_norm": 1.0852105617523193, "learning_rate": 0.0002, "epoch": 5.225806451612903, "step": 2430}, {"loss": 0.4352, "grad_norm": 0.9285983443260193, "learning_rate": 0.0002, "epoch": 5.247311827956989, "step": 2440}, {"loss": 0.4262, "grad_norm": 0.9119299054145813, "learning_rate": 0.0002, "epoch": 5.268817204301075, "step": 2450}, {"loss": 0.4494, "grad_norm": 0.8790456056594849, "learning_rate": 0.0002, "epoch": 5.290322580645161, "step": 2460}, {"loss": 0.4421, "grad_norm": 0.8726504445075989, "learning_rate": 0.0002, "epoch": 5.311827956989247, "step": 2470}, {"loss": 0.4372, "grad_norm": 0.9415227770805359, "learning_rate": 0.0002, "epoch": 5.333333333333333, "step": 2480}, {"loss": 0.4223, "grad_norm": 0.9133324027061462, "learning_rate": 0.0002, "epoch": 5.354838709677419, "step": 2490}, {"loss": 0.4401, "grad_norm": 0.9567879438400269, "learning_rate": 0.0002, "epoch": 5.376344086021505, "step": 2500}, {"loss": 0.4094, "grad_norm": 0.9239469766616821, "learning_rate": 0.0002, "epoch": 5.397849462365591, "step": 2510}, {"loss": 0.4416, "grad_norm": 1.0293527841567993, "learning_rate": 0.0002, "epoch": 5.419354838709677, "step": 2520}, {"loss": 0.4311, "grad_norm": 0.8618718981742859, "learning_rate": 0.0002, "epoch": 5.440860215053763, "step": 2530}, {"loss": 0.462, "grad_norm": 0.740166187286377, "learning_rate": 0.0002, "epoch": 5.462365591397849, "step": 2540}, {"loss": 0.4172, "grad_norm": 0.901566743850708, "learning_rate": 0.0002, "epoch": 5.483870967741936, "step": 2550}, {"loss": 0.4315, "grad_norm": 0.7957597970962524, "learning_rate": 0.0002, "epoch": 5.505376344086022, "step": 2560}, {"loss": 0.4263, "grad_norm": 1.1139343976974487, "learning_rate": 0.0002, "epoch": 5.526881720430108, "step": 2570}, {"loss": 0.4056, "grad_norm": 0.989765465259552, "learning_rate": 0.0002, "epoch": 5.548387096774194, "step": 2580}, {"loss": 0.4311, "grad_norm": 0.9416969418525696, "learning_rate": 0.0002, "epoch": 5.56989247311828, "step": 2590}, {"loss": 0.4363, "grad_norm": 0.9184830784797668, "learning_rate": 0.0002, "epoch": 5.591397849462366, "step": 2600}, {"loss": 0.432, "grad_norm": 1.0512700080871582, "learning_rate": 0.0002, "epoch": 5.612903225806452, "step": 2610}, {"loss": 0.4227, "grad_norm": 0.901462197303772, "learning_rate": 0.0002, "epoch": 5.634408602150538, "step": 2620}, {"loss": 0.4332, "grad_norm": 0.9732566475868225, "learning_rate": 0.0002, "epoch": 5.655913978494624, "step": 2630}, {"loss": 0.4223, "grad_norm": 0.8180275559425354, "learning_rate": 0.0002, "epoch": 5.67741935483871, "step": 2640}, {"loss": 0.4311, "grad_norm": 1.1354765892028809, "learning_rate": 0.0002, "epoch": 5.698924731182796, "step": 2650}, {"loss": 0.4409, "grad_norm": 0.9161503314971924, "learning_rate": 0.0002, "epoch": 5.720430107526882, "step": 2660}, {"loss": 0.4394, "grad_norm": 1.0561772584915161, "learning_rate": 0.0002, "epoch": 5.741935483870968, "step": 2670}, {"loss": 0.424, "grad_norm": 0.7712787389755249, "learning_rate": 0.0002, "epoch": 5.763440860215054, "step": 2680}, {"loss": 0.4326, "grad_norm": 0.9674550294876099, "learning_rate": 0.0002, "epoch": 5.78494623655914, "step": 2690}, {"loss": 0.4459, "grad_norm": 0.7531843781471252, "learning_rate": 0.0002, "epoch": 5.806451612903226, "step": 2700}, {"loss": 0.4276, "grad_norm": 1.1332131624221802, "learning_rate": 0.0002, "epoch": 5.827956989247312, "step": 2710}, {"loss": 0.4113, "grad_norm": 0.9367414116859436, "learning_rate": 0.0002, "epoch": 5.849462365591398, "step": 2720}, {"loss": 0.4227, "grad_norm": 0.8267706632614136, "learning_rate": 0.0002, "epoch": 5.870967741935484, "step": 2730}, {"loss": 0.4218, "grad_norm": 1.1040657758712769, "learning_rate": 0.0002, "epoch": 5.89247311827957, "step": 2740}, {"loss": 0.4129, "grad_norm": 0.8879582285881042, "learning_rate": 0.0002, "epoch": 5.913978494623656, "step": 2750}, {"loss": 0.4241, "grad_norm": 0.9264667630195618, "learning_rate": 0.0002, "epoch": 5.935483870967742, "step": 2760}, {"loss": 0.4318, "grad_norm": 0.9373905658721924, "learning_rate": 0.0002, "epoch": 5.956989247311828, "step": 2770}, {"loss": 0.423, "grad_norm": 1.0063740015029907, "learning_rate": 0.0002, "epoch": 5.978494623655914, "step": 2780}, {"loss": 0.4382, "grad_norm": 0.8291367292404175, "learning_rate": 0.0002, "epoch": 6.0, "step": 2790}, {"eval_loss": 0.5057176947593689, "eval_runtime": 21.3206, "eval_samples_per_second": 15.525, "eval_steps_per_second": 1.97, "epoch": 6.0, "step": 2790}, {"loss": 0.3907, "grad_norm": 1.0137434005737305, "learning_rate": 0.0002, "epoch": 6.021505376344086, "step": 2800}, {"loss": 0.3793, "grad_norm": 0.7550579905509949, "learning_rate": 0.0002, "epoch": 6.043010752688172, "step": 2810}, {"loss": 0.4003, "grad_norm": 1.0664116144180298, "learning_rate": 0.0002, "epoch": 6.064516129032258, "step": 2820}, {"loss": 0.3876, "grad_norm": 0.7908814549446106, "learning_rate": 0.0002, "epoch": 6.086021505376344, "step": 2830}, {"loss": 0.3884, "grad_norm": 0.8101639747619629, "learning_rate": 0.0002, "epoch": 6.10752688172043, "step": 2840}, {"loss": 0.3835, "grad_norm": 0.7882567048072815, "learning_rate": 0.0002, "epoch": 6.129032258064516, "step": 2850}, {"loss": 0.3827, "grad_norm": 1.0134103298187256, "learning_rate": 0.0002, "epoch": 6.150537634408602, "step": 2860}, {"loss": 0.3963, "grad_norm": 0.9240215420722961, "learning_rate": 0.0002, "epoch": 6.172043010752688, "step": 2870}, {"loss": 0.4049, "grad_norm": 0.8322992920875549, "learning_rate": 0.0002, "epoch": 6.193548387096774, "step": 2880}, {"loss": 0.381, "grad_norm": 0.9238720536231995, "learning_rate": 0.0002, "epoch": 6.21505376344086, "step": 2890}, {"loss": 0.3852, "grad_norm": 0.9361863732337952, "learning_rate": 0.0002, "epoch": 6.236559139784946, "step": 2900}, {"loss": 0.3917, "grad_norm": 0.9670863747596741, "learning_rate": 0.0002, "epoch": 6.258064516129032, "step": 2910}, {"loss": 0.3826, "grad_norm": 0.7724685668945312, "learning_rate": 0.0002, "epoch": 6.279569892473118, "step": 2920}, {"loss": 0.3988, "grad_norm": 0.8125540614128113, "learning_rate": 0.0002, "epoch": 6.301075268817204, "step": 2930}, {"loss": 0.3778, "grad_norm": 0.9483002424240112, "learning_rate": 0.0002, "epoch": 6.32258064516129, "step": 2940}, {"loss": 0.3823, "grad_norm": 1.098374843597412, "learning_rate": 0.0002, "epoch": 6.344086021505376, "step": 2950}, {"loss": 0.3886, "grad_norm": 1.0169378519058228, "learning_rate": 0.0002, "epoch": 6.365591397849462, "step": 2960}, {"loss": 0.3936, "grad_norm": 0.8594151139259338, "learning_rate": 0.0002, "epoch": 6.387096774193548, "step": 2970}, {"loss": 0.3871, "grad_norm": 0.9507288336753845, "learning_rate": 0.0002, "epoch": 6.408602150537634, "step": 2980}, {"loss": 0.3852, "grad_norm": 0.9212459325790405, "learning_rate": 0.0002, "epoch": 6.43010752688172, "step": 2990}, {"loss": 0.3929, "grad_norm": 0.9696952104568481, "learning_rate": 0.0002, "epoch": 6.451612903225806, "step": 3000}, {"loss": 0.3933, "grad_norm": 0.8872610330581665, "learning_rate": 0.0002, "epoch": 6.473118279569892, "step": 3010}, {"loss": 0.393, "grad_norm": 0.9207532405853271, "learning_rate": 0.0002, "epoch": 6.494623655913978, "step": 3020}, {"loss": 0.3848, "grad_norm": 0.9116262793540955, "learning_rate": 0.0002, "epoch": 6.516129032258064, "step": 3030}, {"loss": 0.3964, "grad_norm": 0.83391934633255, "learning_rate": 0.0002, "epoch": 6.53763440860215, "step": 3040}, {"loss": 0.3758, "grad_norm": 0.890931248664856, "learning_rate": 0.0002, "epoch": 6.559139784946236, "step": 3050}, {"loss": 0.3944, "grad_norm": 1.0100581645965576, "learning_rate": 0.0002, "epoch": 6.580645161290323, "step": 3060}, {"loss": 0.3992, "grad_norm": 0.783526599407196, "learning_rate": 0.0002, "epoch": 6.602150537634409, "step": 3070}, {"loss": 0.4144, "grad_norm": 1.324326515197754, "learning_rate": 0.0002, "epoch": 6.623655913978495, "step": 3080}, {"loss": 0.3986, "grad_norm": 0.9102319478988647, "learning_rate": 0.0002, "epoch": 6.645161290322581, "step": 3090}, {"loss": 0.3873, "grad_norm": 0.96951824426651, "learning_rate": 0.0002, "epoch": 6.666666666666667, "step": 3100}, {"loss": 0.3931, "grad_norm": 0.9786809086799622, "learning_rate": 0.0002, "epoch": 6.688172043010753, "step": 3110}, {"loss": 0.3714, "grad_norm": 1.0301238298416138, "learning_rate": 0.0002, "epoch": 6.709677419354839, "step": 3120}, {"loss": 0.3823, "grad_norm": 1.1690906286239624, "learning_rate": 0.0002, "epoch": 6.731182795698925, "step": 3130}, {"loss": 0.3936, "grad_norm": 0.963306725025177, "learning_rate": 0.0002, "epoch": 6.752688172043011, "step": 3140}, {"loss": 0.3975, "grad_norm": 0.8565770983695984, "learning_rate": 0.0002, "epoch": 6.774193548387097, "step": 3150}, {"loss": 0.3903, "grad_norm": 0.8887158632278442, "learning_rate": 0.0002, "epoch": 6.795698924731183, "step": 3160}, {"loss": 0.4098, "grad_norm": 0.8234561085700989, "learning_rate": 0.0002, "epoch": 6.817204301075269, "step": 3170}, {"loss": 0.4041, "grad_norm": 0.9000219702720642, "learning_rate": 0.0002, "epoch": 6.838709677419355, "step": 3180}, {"loss": 0.3933, "grad_norm": 1.1366009712219238, "learning_rate": 0.0002, "epoch": 6.860215053763441, "step": 3190}, {"loss": 0.3972, "grad_norm": 0.8747097849845886, "learning_rate": 0.0002, "epoch": 6.881720430107527, "step": 3200}, {"loss": 0.404, "grad_norm": 0.8533893823623657, "learning_rate": 0.0002, "epoch": 6.903225806451613, "step": 3210}, {"loss": 0.3906, "grad_norm": 0.8127949237823486, "learning_rate": 0.0002, "epoch": 6.924731182795699, "step": 3220}, {"loss": 0.3747, "grad_norm": 0.8872477412223816, "learning_rate": 0.0002, "epoch": 6.946236559139785, "step": 3230}, {"loss": 0.3817, "grad_norm": 0.8541608452796936, "learning_rate": 0.0002, "epoch": 6.967741935483871, "step": 3240}, {"loss": 0.3863, "grad_norm": 0.8390752673149109, "learning_rate": 0.0002, "epoch": 6.989247311827957, "step": 3250}, {"eval_loss": 0.48264747858047485, "eval_runtime": 21.3942, "eval_samples_per_second": 15.471, "eval_steps_per_second": 1.963, "epoch": 7.0, "step": 3255}, {"loss": 0.391, "grad_norm": 1.0476834774017334, "learning_rate": 0.0002, "epoch": 7.010752688172043, "step": 3260}, {"loss": 0.3422, "grad_norm": 0.7501131296157837, "learning_rate": 0.0002, "epoch": 7.032258064516129, "step": 3270}, {"loss": 0.3542, "grad_norm": 0.9057435393333435, "learning_rate": 0.0002, "epoch": 7.053763440860215, "step": 3280}, {"loss": 0.3522, "grad_norm": 0.7058833241462708, "learning_rate": 0.0002, "epoch": 7.075268817204301, "step": 3290}, {"loss": 0.3575, "grad_norm": 0.9908691644668579, "learning_rate": 0.0002, "epoch": 7.096774193548387, "step": 3300}, {"loss": 0.3755, "grad_norm": 0.9515542984008789, "learning_rate": 0.0002, "epoch": 7.118279569892473, "step": 3310}, {"loss": 0.3612, "grad_norm": 0.997296154499054, "learning_rate": 0.0002, "epoch": 7.139784946236559, "step": 3320}, {"loss": 0.3616, "grad_norm": 0.9810499548912048, "learning_rate": 0.0002, "epoch": 7.161290322580645, "step": 3330}, {"loss": 0.3584, "grad_norm": 0.8133336901664734, "learning_rate": 0.0002, "epoch": 7.182795698924731, "step": 3340}, {"loss": 0.3644, "grad_norm": 1.0679855346679688, "learning_rate": 0.0002, "epoch": 7.204301075268817, "step": 3350}, {"loss": 0.358, "grad_norm": 0.7656611204147339, "learning_rate": 0.0002, "epoch": 7.225806451612903, "step": 3360}, {"loss": 0.3616, "grad_norm": 0.9478468298912048, "learning_rate": 0.0002, "epoch": 7.247311827956989, "step": 3370}, {"loss": 0.3631, "grad_norm": 0.8425832986831665, "learning_rate": 0.0002, "epoch": 7.268817204301075, "step": 3380}, {"loss": 0.3735, "grad_norm": 0.9573627710342407, "learning_rate": 0.0002, "epoch": 7.290322580645161, "step": 3390}, {"loss": 0.359, "grad_norm": 0.9219972491264343, "learning_rate": 0.0002, "epoch": 7.311827956989247, "step": 3400}, {"loss": 0.3644, "grad_norm": 0.876099705696106, "learning_rate": 0.0002, "epoch": 7.333333333333333, "step": 3410}, {"loss": 0.3747, "grad_norm": 1.0051969289779663, "learning_rate": 0.0002, "epoch": 7.354838709677419, "step": 3420}, {"loss": 0.3527, "grad_norm": 1.1347692012786865, "learning_rate": 0.0002, "epoch": 7.376344086021505, "step": 3430}, {"loss": 0.3644, "grad_norm": 0.9641520380973816, "learning_rate": 0.0002, "epoch": 7.397849462365591, "step": 3440}, {"loss": 0.3486, "grad_norm": 0.7777793407440186, "learning_rate": 0.0002, "epoch": 7.419354838709677, "step": 3450}, {"loss": 0.3593, "grad_norm": 0.9649308323860168, "learning_rate": 0.0002, "epoch": 7.440860215053763, "step": 3460}, {"loss": 0.3754, "grad_norm": 0.9245585203170776, "learning_rate": 0.0002, "epoch": 7.462365591397849, "step": 3470}, {"loss": 0.3732, "grad_norm": 0.8298666477203369, "learning_rate": 0.0002, "epoch": 7.483870967741936, "step": 3480}, {"loss": 0.3585, "grad_norm": 1.1579877138137817, "learning_rate": 0.0002, "epoch": 7.505376344086022, "step": 3490}, {"loss": 0.3505, "grad_norm": 0.8718803524971008, "learning_rate": 0.0002, "epoch": 7.526881720430108, "step": 3500}, {"loss": 0.3725, "grad_norm": 0.7785154581069946, "learning_rate": 0.0002, "epoch": 7.548387096774194, "step": 3510}, {"loss": 0.3507, "grad_norm": 0.753657877445221, "learning_rate": 0.0002, "epoch": 7.56989247311828, "step": 3520}, {"loss": 0.3665, "grad_norm": 0.834524929523468, "learning_rate": 0.0002, "epoch": 7.591397849462366, "step": 3530}, {"loss": 0.3686, "grad_norm": 0.9546446800231934, "learning_rate": 0.0002, "epoch": 7.612903225806452, "step": 3540}, {"loss": 0.3673, "grad_norm": 0.8275105357170105, "learning_rate": 0.0002, "epoch": 7.634408602150538, "step": 3550}, {"loss": 0.381, "grad_norm": 0.9137991070747375, "learning_rate": 0.0002, "epoch": 7.655913978494624, "step": 3560}, {"loss": 0.3565, "grad_norm": 0.993617057800293, "learning_rate": 0.0002, "epoch": 7.67741935483871, "step": 3570}, {"loss": 0.3701, "grad_norm": 1.0079665184020996, "learning_rate": 0.0002, "epoch": 7.698924731182796, "step": 3580}, {"loss": 0.3495, "grad_norm": 0.8295491337776184, "learning_rate": 0.0002, "epoch": 7.720430107526882, "step": 3590}, {"loss": 0.374, "grad_norm": 0.814578115940094, "learning_rate": 0.0002, "epoch": 7.741935483870968, "step": 3600}, {"loss": 0.3673, "grad_norm": 0.8422811031341553, "learning_rate": 0.0002, "epoch": 7.763440860215054, "step": 3610}, {"loss": 0.3452, "grad_norm": 1.0220918655395508, "learning_rate": 0.0002, "epoch": 7.78494623655914, "step": 3620}, {"loss": 0.3641, "grad_norm": 0.8065739870071411, "learning_rate": 0.0002, "epoch": 7.806451612903226, "step": 3630}, {"loss": 0.3699, "grad_norm": 0.8039169907569885, "learning_rate": 0.0002, "epoch": 7.827956989247312, "step": 3640}, {"loss": 0.3635, "grad_norm": 1.0766745805740356, "learning_rate": 0.0002, "epoch": 7.849462365591398, "step": 3650}, {"loss": 0.3633, "grad_norm": 1.0806103944778442, "learning_rate": 0.0002, "epoch": 7.870967741935484, "step": 3660}, {"loss": 0.3678, "grad_norm": 1.1005314588546753, "learning_rate": 0.0002, "epoch": 7.89247311827957, "step": 3670}, {"loss": 0.3583, "grad_norm": 0.9276911616325378, "learning_rate": 0.0002, "epoch": 7.913978494623656, "step": 3680}, {"loss": 0.374, "grad_norm": 0.9914153814315796, "learning_rate": 0.0002, "epoch": 7.935483870967742, "step": 3690}, {"loss": 0.3575, "grad_norm": 0.8128159046173096, "learning_rate": 0.0002, "epoch": 7.956989247311828, "step": 3700}, {"loss": 0.3491, "grad_norm": 0.8122950196266174, "learning_rate": 0.0002, "epoch": 7.978494623655914, "step": 3710}, {"loss": 0.3723, "grad_norm": 1.0291857719421387, "learning_rate": 0.0002, "epoch": 8.0, "step": 3720}]}