metadata
license: apache-2.0
library_name: peft
tags:
- generated_from_trainer
base_model: tiiuae/falcon-7b-instruct
model-index:
- name: working
results: []
working
This model is a fine-tuned version of tiiuae/falcon-7b-instruct on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.6039
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: 0.0002
- train_batch_size: 6
- eval_batch_size: 6
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 24
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 2
- num_epochs: 50
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.7249 | 0.95 | 5 | 2.3111 |
| 2.4979 | 1.9 | 10 | 2.0912 |
| 2.291 | 2.86 | 15 | 1.8826 |
| 1.6598 | 4.0 | 21 | 1.6395 |
| 1.691 | 4.95 | 26 | 1.4690 |
| 1.499 | 5.9 | 31 | 1.3555 |
| 1.3066 | 6.86 | 36 | 1.2151 |
| 0.866 | 8.0 | 42 | 1.1219 |
| 0.8635 | 8.95 | 47 | 1.0441 |
| 0.6727 | 9.9 | 52 | 0.9735 |
| 0.6457 | 10.86 | 57 | 0.8954 |
| 0.3947 | 12.0 | 63 | 0.8019 |
| 0.3888 | 12.95 | 68 | 0.7246 |
| 0.3304 | 13.9 | 73 | 0.6799 |
| 0.283 | 14.86 | 78 | 0.6329 |
| 0.2018 | 16.0 | 84 | 0.6151 |
| 0.2111 | 16.95 | 89 | 0.6028 |
| 0.1869 | 17.9 | 94 | 0.5947 |
| 0.1878 | 18.86 | 99 | 0.5988 |
| 0.1492 | 20.0 | 105 | 0.6044 |
| 0.2301 | 20.95 | 110 | 0.5920 |
| 0.1697 | 21.9 | 115 | 0.5870 |
| 0.1693 | 22.86 | 120 | 0.5907 |
| 0.1312 | 24.0 | 126 | 0.5881 |
| 0.147 | 24.95 | 131 | 0.5789 |
| 0.1576 | 25.9 | 136 | 0.5820 |
| 0.1438 | 26.86 | 141 | 0.5897 |
| 0.1197 | 28.0 | 147 | 0.5965 |
| 0.1383 | 28.95 | 152 | 0.5937 |
| 0.1352 | 29.9 | 157 | 0.5904 |
| 0.1665 | 30.86 | 162 | 0.5907 |
| 0.1206 | 32.0 | 168 | 0.5899 |
| 0.1283 | 32.95 | 173 | 0.5996 |
| 0.1274 | 33.9 | 178 | 0.5972 |
| 0.1633 | 34.86 | 183 | 0.5933 |
| 0.1194 | 36.0 | 189 | 0.6019 |
| 0.1496 | 36.95 | 194 | 0.6017 |
| 0.1584 | 37.9 | 199 | 0.6002 |
| 0.1325 | 38.86 | 204 | 0.5975 |
| 0.1036 | 40.0 | 210 | 0.5976 |
| 0.128 | 40.95 | 215 | 0.5997 |
| 0.1234 | 41.9 | 220 | 0.6017 |
| 0.1234 | 42.86 | 225 | 0.6034 |
| 0.1056 | 44.0 | 231 | 0.6046 |
| 0.1326 | 44.95 | 236 | 0.6049 |
| 0.1301 | 45.9 | 241 | 0.6046 |
| 0.134 | 46.86 | 246 | 0.6036 |
| 0.0939 | 47.62 | 250 | 0.6039 |
Framework versions
- PEFT 0.10.0
- Transformers 4.38.2
- Pytorch 2.1.2
- Datasets 2.1.0
- Tokenizers 0.15.2