| # Path to the training dataset CSV file | |
| file_path: data/grpo/train.csv | |
| # Split of the dataset | |
| split: train | |
| # Maximum number of training samples to use | |
| max_rows: 10000 | |
| # Random seed for dataset shuffling | |
| seed: 42 | |
| # Path to the training dataset CSV file | |
| file_path: data/grpo/train.csv | |
| # Split of the dataset | |
| split: train | |
| # Maximum number of training samples to use | |
| max_rows: 10000 | |
| # Random seed for dataset shuffling | |
| seed: 42 | |