<!-- 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-seed_444
This model is a fine-tuned version of models/smolm-autoreg-bpe-seed_444/config.json on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 2.5604
- Accuracy: 0.4869
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: 0.003
- train_batch_size: 64
- eval_batch_size: 512
- seed: 444
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 24000
- num_epochs: 10.0
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
6.1051 | 1.0 | 732 | 3.5368 | 0.3949 |
3.4882 | 2.0 | 1464 | 3.1157 | 0.4337 |
2.9431 | 3.0 | 2196 | 2.9167 | 0.4502 |
2.7991 | 4.0 | 2928 | 2.7866 | 0.4619 |
2.6152 | 5.0 | 3660 | 2.6979 | 0.4708 |
2.5487 | 6.0 | 4392 | 2.6426 | 0.4759 |
2.4491 | 7.0 | 5124 | 2.6028 | 0.4819 |
2.4034 | 8.0 | 5856 | 2.5808 | 0.4835 |
2.364 | 9.0 | 6588 | 2.5657 | 0.4851 |
2.3269 | 10.0 | 7320 | 2.5604 | 0.4869 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.0.1+cu117
- Datasets 2.12.0
- Tokenizers 0.13.3