metadata
license: apache-2.0
base_model: EleutherAI/pythia-160m
tags:
- trl
- reward-trainer
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: rm1
results: []
rm1
This model is a fine-tuned version of EleutherAI/pythia-160m on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.5286
- Accuracy: 0.8456
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: 3e-06
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 32
- total_train_batch_size: 64
- 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 |
|---|---|---|---|---|
| 0.6883 | 0.0621 | 5 | 0.6846 | 0.7537 |
| 0.6723 | 0.1242 | 10 | 0.6717 | 0.7978 |
| 0.6761 | 0.1863 | 15 | 0.6559 | 0.8162 |
| 0.6627 | 0.2484 | 20 | 0.6379 | 0.8125 |
| 0.6156 | 0.3104 | 25 | 0.6175 | 0.8125 |
| 0.6232 | 0.3725 | 30 | 0.5937 | 0.8272 |
| 0.5985 | 0.4346 | 35 | 0.5711 | 0.8456 |
| 0.6024 | 0.4967 | 40 | 0.5549 | 0.8309 |
| 0.5906 | 0.5588 | 45 | 0.5449 | 0.8346 |
| 0.6184 | 0.6209 | 50 | 0.5383 | 0.8419 |
| 0.5379 | 0.6830 | 55 | 0.5338 | 0.8382 |
| 0.564 | 0.7451 | 60 | 0.5312 | 0.8456 |
| 0.5635 | 0.8071 | 65 | 0.5299 | 0.8456 |
| 0.5892 | 0.8692 | 70 | 0.5292 | 0.8493 |
| 0.5416 | 0.9313 | 75 | 0.5288 | 0.8456 |
| 0.5994 | 0.9934 | 80 | 0.5286 | 0.8456 |
Framework versions
- Transformers 4.40.1
- Pytorch 2.2.0+cu121
- Datasets 2.19.0
- Tokenizers 0.19.1