---
license: gemma
library_name: peft
tags:
- trl
- reward-trainer
- generated_from_trainer
base_model: google/gemma-2b
metrics:
- accuracy
model-index:
- name: RM-HH-Gemma_helpful_human_loraR64_20000_gemma2b_shuffleTrue_extractchosenFalse
  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. -->

# RM-HH-Gemma_helpful_human_loraR64_20000_gemma2b_shuffleTrue_extractchosenFalse

This model is a fine-tuned version of [google/gemma-2b](https://huggingface.co/google/gemma-2b) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.6198
- Accuracy: 0.6540

## 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: 1.41e-05
- train_batch_size: 1
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 4
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 2.0

### Training results

| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 0.7745        | 0.06  | 250  | 0.7362          | 0.5088   |
| 0.6966        | 0.11  | 500  | 0.7087          | 0.5498   |
| 0.6929        | 0.17  | 750  | 0.6929          | 0.5814   |
| 0.702         | 0.22  | 1000 | 0.6814          | 0.5939   |
| 0.6633        | 0.28  | 1250 | 0.6735          | 0.6049   |
| 0.6529        | 0.33  | 1500 | 0.6669          | 0.6094   |
| 0.6487        | 0.39  | 1750 | 0.6610          | 0.6189   |
| 0.6737        | 0.45  | 2000 | 0.6536          | 0.6254   |
| 0.6314        | 0.5   | 2250 | 0.6501          | 0.6269   |
| 0.6474        | 0.56  | 2500 | 0.6454          | 0.6304   |
| 0.6225        | 0.61  | 2750 | 0.6429          | 0.6335   |
| 0.6338        | 0.67  | 3000 | 0.6393          | 0.6360   |
| 0.6268        | 0.72  | 3250 | 0.6360          | 0.6400   |
| 0.633         | 0.78  | 3500 | 0.6346          | 0.6425   |
| 0.641         | 0.83  | 3750 | 0.6305          | 0.6440   |
| 0.6439        | 0.89  | 4000 | 0.6286          | 0.6470   |
| 0.6123        | 0.95  | 4250 | 0.6274          | 0.6475   |
| 0.6082        | 1.0   | 4500 | 0.6277          | 0.6535   |
| 0.6275        | 1.06  | 4750 | 0.6267          | 0.6540   |
| 0.589         | 1.11  | 5000 | 0.6276          | 0.6535   |
| 0.588         | 1.17  | 5250 | 0.6297          | 0.6550   |
| 0.6126        | 1.22  | 5500 | 0.6305          | 0.6535   |
| 0.6216        | 1.28  | 5750 | 0.6286          | 0.6525   |
| 0.6071        | 1.34  | 6000 | 0.6269          | 0.6515   |
| 0.6063        | 1.39  | 6250 | 0.6271          | 0.6505   |
| 0.6166        | 1.45  | 6500 | 0.6246          | 0.6525   |
| 0.6076        | 1.5   | 6750 | 0.6230          | 0.6565   |
| 0.6007        | 1.56  | 7000 | 0.6233          | 0.6545   |
| 0.6452        | 1.61  | 7250 | 0.6205          | 0.6540   |
| 0.5932        | 1.67  | 7500 | 0.6207          | 0.6535   |
| 0.6093        | 1.72  | 7750 | 0.6207          | 0.6530   |
| 0.6183        | 1.78  | 8000 | 0.6206          | 0.6535   |
| 0.6244        | 1.84  | 8250 | 0.6200          | 0.6545   |
| 0.6183        | 1.89  | 8500 | 0.6199          | 0.6545   |
| 0.6281        | 1.95  | 8750 | 0.6198          | 0.6540   |


### Framework versions

- PEFT 0.9.0
- Transformers 4.38.2
- Pytorch 2.1.2
- Datasets 2.18.0
- Tokenizers 0.15.2