<!-- 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-1e-3
This model is a fine-tuned version of models/smolm-autoreg-bpe-babylm-1e-3/config.json on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 2.9702
- Accuracy: 0.4451
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 |
---|---|---|---|---|
3.5519 | 1.0 | 9209 | 3.4821 | 0.3833 |
3.2755 | 2.0 | 18418 | 3.2460 | 0.4063 |
3.1572 | 3.0 | 27627 | 3.1486 | 0.4173 |
3.0755 | 4.0 | 36836 | 3.0770 | 0.4251 |
2.9852 | 5.0 | 46045 | 3.0270 | 0.4311 |
2.93 | 6.0 | 55254 | 2.9953 | 0.4350 |
2.8816 | 7.0 | 64463 | 2.9793 | 0.4375 |
2.8418 | 8.0 | 73672 | 2.9675 | 0.4392 |
2.806 | 9.0 | 82881 | 2.9581 | 0.4413 |
2.7773 | 10.0 | 92090 | 2.9467 | 0.4427 |
2.751 | 11.0 | 101299 | 2.9482 | 0.4429 |
2.7239 | 12.0 | 110508 | 2.9498 | 0.4436 |
2.7029 | 13.0 | 119717 | 2.9448 | 0.4442 |
2.6753 | 14.0 | 128926 | 2.9497 | 0.4447 |
2.6561 | 15.0 | 138135 | 2.9491 | 0.4450 |
2.6324 | 16.0 | 147344 | 2.9510 | 0.4450 |
2.6107 | 17.0 | 156553 | 2.9549 | 0.4450 |
2.5889 | 18.0 | 165762 | 2.9600 | 0.4451 |
2.5673 | 19.0 | 174971 | 2.9665 | 0.4451 |
2.5506 | 20.0 | 184180 | 2.9702 | 0.4451 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.0.1+cu117
- Datasets 2.12.0
- Tokenizers 0.13.3