<!-- 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. -->
Llama-2-7b-hf-finetuned-mrpc
This model is a fine-tuned version of meta-llama/Llama-2-7b-hf on the glue dataset. It achieves the following results on the evaluation set:
- Accuracy: 0.7941
- F1: 0.8571
- Loss: 0.4479
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: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Accuracy | F1 | Validation Loss |
|---|---|---|---|---|---|
| No log | 1.0 | 230 | 0.7206 | 0.8155 | 0.6045 |
| No log | 2.0 | 460 | 0.6912 | 0.8158 | 0.6488 |
| 0.6326 | 3.0 | 690 | 0.7279 | 0.8235 | 0.5236 |
| 0.6326 | 4.0 | 920 | 0.7255 | 0.8282 | 0.5273 |
| 0.5602 | 5.0 | 1150 | 0.7402 | 0.8044 | 0.5246 |
| 0.5602 | 6.0 | 1380 | 0.75 | 0.8311 | 0.4893 |
| 0.5139 | 7.0 | 1610 | 0.7623 | 0.8289 | 0.4884 |
| 0.5139 | 8.0 | 1840 | 0.7402 | 0.8307 | 0.4989 |
| 0.4754 | 9.0 | 2070 | 0.7745 | 0.8435 | 0.4732 |
| 0.4754 | 10.0 | 2300 | 0.7672 | 0.8403 | 0.4716 |
| 0.5407 | 11.0 | 2530 | 0.7598 | 0.8393 | 0.4823 |
| 0.5407 | 12.0 | 2760 | 0.7451 | 0.8333 | 0.4782 |
| 0.5407 | 13.0 | 2990 | 0.7451 | 0.8333 | 0.4713 |
| 0.4951 | 14.0 | 3220 | 0.7819 | 0.8489 | 0.4553 |
| 0.4951 | 15.0 | 3450 | 0.7745 | 0.8506 | 0.4591 |
| 0.4724 | 16.0 | 3680 | 0.7770 | 0.8423 | 0.4631 |
| 0.4724 | 17.0 | 3910 | 0.8015 | 0.8576 | 0.4581 |
| 0.4455 | 18.0 | 4140 | 0.7819 | 0.8468 | 0.4548 |
| 0.4455 | 19.0 | 4370 | 0.7819 | 0.8484 | 0.4511 |
| 0.4354 | 20.0 | 4600 | 0.7941 | 0.8571 | 0.4479 |
Framework versions
- Transformers 4.31.0
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.13.3
