metadata
library_name: transformers
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: babylm-base2.5m-roberta
results: []
babylm-base2.5m-roberta
This model is a fine-tuned version of on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 5.5461
- Accuracy: 0.1659
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: 5e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 195
- training_steps: 19500
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 6.8743 | 0.1122 | 200 | 6.5233 | 0.1022 |
| 6.1033 | 0.2245 | 400 | 6.2006 | 0.1237 |
| 5.9405 | 0.3367 | 600 | 6.0823 | 0.1329 |
| 5.8236 | 0.4489 | 800 | 6.0260 | 0.1374 |
| 5.777 | 0.5612 | 1000 | 5.9901 | 0.1381 |
| 5.7245 | 0.6734 | 1200 | 5.9425 | 0.1430 |
| 5.7086 | 0.7856 | 1400 | 5.9186 | 0.1472 |
| 5.6794 | 0.8979 | 1600 | 5.8993 | 0.1473 |
| 5.6582 | 1.0101 | 1800 | 5.8833 | 0.1476 |
| 5.5798 | 1.1223 | 2000 | 5.8732 | 0.1492 |
| 5.4905 | 2.2447 | 4000 | 5.7809 | 0.1542 |
| 5.3537 | 3.3670 | 6000 | 5.7190 | 0.1586 |
| 5.3199 | 4.4893 | 8000 | 5.6712 | 0.1601 |
| 5.2578 | 5.6117 | 10000 | 5.6395 | 0.1610 |
| 5.2367 | 6.7340 | 12000 | 5.6065 | 0.1616 |
| 5.2088 | 7.8563 | 14000 | 5.5854 | 0.1632 |
| 5.1523 | 8.9787 | 16000 | 5.5658 | 0.1644 |
| 5.1263 | 10.1010 | 18000 | 5.5506 | 0.1649 |
Framework versions
- Transformers 4.50.3
- Pytorch 2.7.1+cu126
- Datasets 3.6.0
- Tokenizers 0.21.4