File size: 2,938 Bytes
1b0fa9f |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 |
---
license: gemma
library_name: peft
tags:
- trl
- reward-trainer
- generated_from_trainer
base_model: google/gemma-2b
metrics:
- accuracy
model-index:
- name: RM-HH-AllMix_harmless_gpt3_20000_gemma2b_shuffleFalse_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-AllMix_harmless_gpt3_20000_gemma2b_shuffleFalse_extractchosenFalse
This model is a fine-tuned version of [google/gemma-2b](https://huggingface.co/google/gemma-2b) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0554
- Accuracy: 0.9722
## 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: 1.0
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 0.6806 | 0.04 | 250 | 0.4207 | 0.8447 |
| 0.6465 | 0.08 | 500 | 0.2075 | 0.9485 |
| 0.5999 | 0.13 | 750 | 0.1133 | 0.9725 |
| 0.5725 | 0.17 | 1000 | 0.0804 | 0.9786 |
| 0.6223 | 0.21 | 1250 | 0.0783 | 0.9778 |
| 0.5595 | 0.25 | 1500 | 0.0632 | 0.9789 |
| 0.5956 | 0.29 | 1750 | 0.0589 | 0.9778 |
| 0.557 | 0.33 | 2000 | 0.0599 | 0.9756 |
| 0.5641 | 0.38 | 2250 | 0.0558 | 0.9767 |
| 0.5871 | 0.42 | 2500 | 0.0589 | 0.9744 |
| 0.5512 | 0.46 | 2750 | 0.0568 | 0.9741 |
| 0.5775 | 0.5 | 3000 | 0.0529 | 0.9756 |
| 0.5923 | 0.54 | 3250 | 0.0555 | 0.9748 |
| 0.548 | 0.59 | 3500 | 0.0577 | 0.9722 |
| 0.564 | 0.63 | 3750 | 0.0579 | 0.9722 |
| 0.563 | 0.67 | 4000 | 0.0599 | 0.9718 |
| 0.5932 | 0.71 | 4250 | 0.0561 | 0.9729 |
| 0.5247 | 0.75 | 4500 | 0.0569 | 0.9725 |
| 0.5472 | 0.79 | 4750 | 0.0579 | 0.9718 |
| 0.5704 | 0.84 | 5000 | 0.0556 | 0.9729 |
| 0.5456 | 0.88 | 5250 | 0.0550 | 0.9725 |
| 0.5563 | 0.92 | 5500 | 0.0545 | 0.9729 |
| 0.552 | 0.96 | 5750 | 0.0554 | 0.9722 |
### Framework versions
- PEFT 0.10.0
- Transformers 4.38.2
- Pytorch 2.1.2+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2 |