<!-- 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. -->
smolm-autoreg-bpe-babylm-no_aann-infilling-1e-3
This model is a fine-tuned version of models/smolm-autoreg-bpe-babylm-no_aann-infilling-1e-3/config.json on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 3.1612
- Accuracy: 0.4318
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: 0.001
- train_batch_size: 64
- eval_batch_size: 256
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 32000
- num_epochs: 20.0
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
3.5506 | 1.0 | 9180 | 3.6452 | 0.3697 |
3.2731 | 2.0 | 18360 | 3.4078 | 0.3940 |
3.1564 | 3.0 | 27540 | 3.3062 | 0.4059 |
3.0714 | 4.0 | 36720 | 3.2481 | 0.4127 |
2.9886 | 5.0 | 45900 | 3.1881 | 0.4192 |
2.9236 | 6.0 | 55080 | 3.1613 | 0.4228 |
2.8816 | 7.0 | 64260 | 3.1470 | 0.4256 |
2.8446 | 8.0 | 73440 | 3.1379 | 0.4270 |
2.8065 | 9.0 | 82620 | 3.1353 | 0.4285 |
2.7774 | 10.0 | 91800 | 3.1329 | 0.4293 |
2.7501 | 11.0 | 100980 | 3.1256 | 0.4306 |
2.7231 | 12.0 | 110160 | 3.1269 | 0.4309 |
2.701 | 13.0 | 119340 | 3.1334 | 0.4309 |
2.6756 | 14.0 | 128520 | 3.1308 | 0.4317 |
2.6533 | 15.0 | 137700 | 3.1365 | 0.4319 |
2.6306 | 16.0 | 146880 | 3.1389 | 0.4318 |
2.6082 | 17.0 | 156060 | 3.1452 | 0.4317 |
2.5874 | 18.0 | 165240 | 3.1518 | 0.4320 |
2.5684 | 19.0 | 174420 | 3.1557 | 0.4318 |
2.548 | 20.0 | 183600 | 3.1612 | 0.4318 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.0.1+cu117
- Datasets 2.12.0
- Tokenizers 0.13.3