<!-- 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-base-1e-3
This model is a fine-tuned version of models/smolm-autoreg-bpe-babylm-base-1e-3/config.json on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 2.8078
- Accuracy: 0.4772
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
- num_epochs: 40.0
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
3.2352 | 1.0 | 10468 | 3.1040 | 0.4394 |
3.0865 | 2.0 | 20936 | 3.0063 | 0.4510 |
3.0278 | 3.0 | 31404 | 2.9597 | 0.4567 |
2.9932 | 4.0 | 41872 | 2.9350 | 0.4596 |
2.9675 | 5.0 | 52340 | 2.9148 | 0.4622 |
2.9492 | 6.0 | 62808 | 2.9016 | 0.4643 |
2.9344 | 7.0 | 73276 | 2.8918 | 0.4652 |
2.9213 | 8.0 | 83744 | 2.8834 | 0.4666 |
2.912 | 9.0 | 94212 | 2.8747 | 0.4676 |
2.9028 | 10.0 | 104680 | 2.8695 | 0.4684 |
2.8964 | 11.0 | 115148 | 2.8627 | 0.4693 |
2.8901 | 12.0 | 125616 | 2.8589 | 0.4698 |
2.8839 | 13.0 | 136084 | 2.8543 | 0.4703 |
2.8807 | 14.0 | 146552 | 2.8507 | 0.4709 |
2.8738 | 15.0 | 157020 | 2.8475 | 0.4712 |
2.8706 | 16.0 | 167488 | 2.8422 | 0.4720 |
2.8645 | 17.0 | 177956 | 2.8448 | 0.4718 |
2.8595 | 18.0 | 188424 | 2.8365 | 0.4728 |
2.8542 | 19.0 | 198892 | 2.8366 | 0.4728 |
2.8523 | 20.0 | 209360 | 2.8318 | 0.4731 |
2.8479 | 21.0 | 219828 | 2.8316 | 0.4736 |
2.849 | 22.0 | 230296 | 2.8298 | 0.4739 |
2.8451 | 23.0 | 240764 | 2.8291 | 0.4739 |
2.8436 | 24.0 | 251232 | 2.8244 | 0.4744 |
2.8387 | 25.0 | 261700 | 2.8247 | 0.4745 |
2.8354 | 26.0 | 272168 | 2.8228 | 0.4750 |
2.8339 | 27.0 | 282636 | 2.8213 | 0.4751 |
2.8307 | 28.0 | 293104 | 2.8206 | 0.4752 |
2.8276 | 29.0 | 303572 | 2.8176 | 0.4756 |
2.8244 | 30.0 | 314040 | 2.8165 | 0.4757 |
2.8229 | 31.0 | 324508 | 2.8165 | 0.4757 |
2.8209 | 32.0 | 334976 | 2.8141 | 0.4761 |
2.8206 | 33.0 | 345444 | 2.8119 | 0.4763 |
2.8178 | 34.0 | 355912 | 2.8135 | 0.4764 |
2.8152 | 35.0 | 366380 | 2.8119 | 0.4766 |
2.813 | 36.0 | 376848 | 2.8106 | 0.4767 |
2.8107 | 37.0 | 387316 | 2.8081 | 0.4771 |
2.8113 | 38.0 | 397784 | 2.8091 | 0.4769 |
2.8073 | 39.0 | 408252 | 2.8078 | 0.4771 |
2.8061 | 40.0 | 418720 | 2.8078 | 0.4772 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.0.1+cu117
- Datasets 2.12.0
- Tokenizers 0.13.3