metadata
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- facebook/anli
model-index:
- name: sft-trl-no-claim
results: []
sft-trl-no-claim
This model is a fine-tuned version of EleutherAI/gpt-j-6b on the anli dataset. It achieves the following results on the evaluation set:
- Loss: 0.6203
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: 5e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- num_epochs: 3.0
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.1604 | 0.2 | 1000 | 1.8712 |
| 1.0077 | 0.4 | 2000 | 1.4700 |
| 0.8175 | 0.6 | 3000 | 1.0772 |
| 0.7621 | 0.8 | 4000 | 0.7959 |
| 0.3346 | 1.0 | 5000 | 0.7570 |
| 0.2903 | 1.2 | 6000 | 0.6882 |
| 0.2313 | 1.4 | 7000 | 0.6471 |
| 0.3038 | 1.6 | 8000 | 0.6203 |
| 0.2292 | 1.8 | 9000 | 0.6203 |
| 0.193 | 2.0 | 10000 | 0.6203 |
| 0.1198 | 2.2 | 11000 | 0.6203 |
| 0.2185 | 2.4 | 12000 | 0.6203 |
| 0.1038 | 2.6 | 13000 | 0.6203 |
| 0.1469 | 2.8 | 14000 | 0.6203 |
| 0.1857 | 3.0 | 15000 | 0.6203 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu117
- Datasets 2.13.1
- Tokenizers 0.13.3