---
library_name: peft
license: gemma
base_model: google/gemma-2-2b
tags:
- generated_from_trainer
model-index:
- name: gemma-2-2b-lora-finetuned-30k
  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. -->

# gemma-2-2b-lora-finetuned-30k

This model is a fine-tuned version of [google/gemma-2-2b](https://huggingface.co/google/gemma-2-2b) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 1.8385

## 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: 0.0005
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- 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
- num_epochs: 6
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step  | Validation Loss |
|:-------------:|:-----:|:-----:|:---------------:|
| 1.9208        | 1.0   | 6750  | 1.9144          |
| 1.8575        | 2.0   | 13500 | 1.8821          |
| 1.8024        | 3.0   | 20250 | 1.8586          |
| 1.7602        | 4.0   | 27000 | 1.8430          |
| 1.7283        | 5.0   | 33750 | 1.8370          |
| 1.6777        | 6.0   | 40500 | 1.8385          |


### Framework versions

- PEFT 0.14.0
- Transformers 4.47.0
- Pytorch 2.5.1+cu118
- Datasets 3.1.0
- Tokenizers 0.21.0