| # Data validation only. No model is trained by this entry point. | |
| token_dir = "artifacts/token-data/corpus-v1-16k" | |
| index_dir = "artifacts/training-data/corpus-v1-c128" | |
| context_length = 128 | |
| batch_size = 128 | |
| num_workers = 4 | |
| seed = 42 | |
| check_batches = 8 | |
| # Data validation only. No model is trained by this entry point. | |
| token_dir = "artifacts/token-data/corpus-v1-16k" | |
| index_dir = "artifacts/training-data/corpus-v1-c128" | |
| context_length = 128 | |
| batch_size = 128 | |
| num_workers = 4 | |
| seed = 42 | |
| check_batches = 8 | |