metadata
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: >-
smolm-autoreg-bpe-counterfactual-babylm-measure_nps_as_singular-seed_211-1e-4
results: []
smolm-autoreg-bpe-counterfactual-babylm-measure_nps_as_singular-seed_211-1e-4
This model was trained from scratch on the None dataset. It achieves the following results on the evaluation set:
- Loss: 3.4453
- Accuracy: 0.4059
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: 0.0001
- train_batch_size: 32
- eval_batch_size: 64
- seed: 211
- 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
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 4.0492 | 1.0 | 18601 | 4.2903 | 0.3076 |
| 3.5701 | 2.0 | 37202 | 3.7600 | 0.3619 |
| 3.3901 | 3.0 | 55803 | 3.5643 | 0.3811 |
| 3.2877 | 4.0 | 74404 | 3.4952 | 0.3893 |
| 3.2176 | 5.0 | 93005 | 3.4699 | 0.3940 |
| 3.1687 | 6.0 | 111606 | 3.4478 | 0.3967 |
| 3.1232 | 7.0 | 130207 | 3.4306 | 0.3992 |
| 3.0903 | 8.0 | 148808 | 3.4360 | 0.4007 |
| 3.0619 | 9.0 | 167409 | 3.4178 | 0.4020 |
| 3.0326 | 10.0 | 186010 | 3.4113 | 0.4025 |
| 3.0057 | 11.0 | 204611 | 3.4085 | 0.4036 |
| 2.9825 | 12.0 | 223212 | 3.4075 | 0.4043 |
| 2.9605 | 13.0 | 241813 | 3.4155 | 0.4047 |
| 2.9399 | 14.0 | 260414 | 3.4355 | 0.4047 |
| 2.9261 | 15.0 | 279015 | 3.4277 | 0.4055 |
| 2.9036 | 16.0 | 297616 | 3.4246 | 0.4057 |
| 2.8909 | 17.0 | 316217 | 3.4307 | 0.4058 |
| 2.8715 | 18.0 | 334818 | 3.4373 | 0.4058 |
| 2.8532 | 19.0 | 353419 | 3.4386 | 0.4062 |
| 2.8408 | 20.0 | 372020 | 3.4453 | 0.4059 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.1.0+cu121
- Datasets 2.12.0
- Tokenizers 0.14.1