---
library_name: peft
license: llama3.2
base_model: meta-llama/Llama-3.2-3B
tags:
- trl
- sft
- generated_from_trainer
datasets:
- generator
model-index:
- name: llama3.1-3b-coding-gpt4o-100k2
  results: []
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# llama3.1-3b-coding-gpt4o-100k2

This model is a fine-tuned version of [meta-llama/Llama-3.2-3B](https://huggingface.co/meta-llama/Llama-3.2-3B) on the generator dataset.
It achieves the following results on the evaluation set:
- Loss: 1.6301

## Model description

More information needed

## Intended uses & limitations

More information needed

## Training and evaluation data

More information needed

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 0.002
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- distributed_type: multi-GPU
- num_devices: 8
- gradient_accumulation_steps: 2
- total_train_batch_size: 256
- total_eval_batch_size: 128
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10

### Training results

| Training Loss | Epoch  | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 1.0031        | 1.0    | 68   | 1.5510          |
| 0.9546        | 2.0    | 136  | 1.5149          |
| 0.936         | 3.0    | 204  | 1.5085          |
| 0.9186        | 4.0    | 272  | 1.5175          |
| 0.8948        | 5.0    | 340  | 1.5302          |
| 0.8742        | 6.0    | 408  | 1.5502          |
| 0.8556        | 7.0    | 476  | 1.5617          |
| 0.8428        | 8.0    | 544  | 1.5965          |
| 0.8168        | 9.0    | 612  | 1.6217          |
| 0.8191        | 9.8593 | 670  | 1.6301          |


### Framework versions

- PEFT 0.15.1
- Transformers 4.50.3
- Pytorch 2.6.0+cu124
- Datasets 3.5.0
- Tokenizers 0.21.1