| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 8.0, |
| "eval_steps": 500, |
| "global_step": 56, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.7142857142857143, |
| "grad_norm": 231.44915771484375, |
| "learning_rate": 9.81132075471698e-07, |
| "loss": 19.46338806152344, |
| "step": 5 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 2.776050567626953, |
| "eval_map": 0.8851, |
| "eval_map_50": 1.0, |
| "eval_map_75": 1.0, |
| "eval_map_large": 0.8799, |
| "eval_map_medium": 1.0, |
| "eval_map_small": -1.0, |
| "eval_map_weight_plate": 0.8851, |
| "eval_mar_1": 0.325, |
| "eval_mar_10": 0.925, |
| "eval_mar_100": 0.975, |
| "eval_mar_100_weight_plate": 0.975, |
| "eval_mar_large": 0.9714, |
| "eval_mar_medium": 1.0, |
| "eval_mar_small": -1.0, |
| "eval_runtime": 1.265, |
| "eval_samples_per_second": 2.372, |
| "eval_steps_per_second": 0.791, |
| "step": 7 |
| }, |
| { |
| "epoch": 1.4285714285714286, |
| "grad_norm": 285.1370849609375, |
| "learning_rate": 8.867924528301887e-07, |
| "loss": 21.529798889160155, |
| "step": 10 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 3.012538194656372, |
| "eval_map": 0.8701, |
| "eval_map_50": 1.0, |
| "eval_map_75": 1.0, |
| "eval_map_large": 0.869, |
| "eval_map_medium": 0.9, |
| "eval_map_small": -1.0, |
| "eval_map_weight_plate": 0.8701, |
| "eval_mar_1": 0.325, |
| "eval_mar_10": 0.9125, |
| "eval_mar_100": 0.9375, |
| "eval_mar_100_weight_plate": 0.9375, |
| "eval_mar_large": 0.9429, |
| "eval_mar_medium": 0.9, |
| "eval_mar_small": -1.0, |
| "eval_runtime": 0.1527, |
| "eval_samples_per_second": 19.652, |
| "eval_steps_per_second": 6.551, |
| "step": 14 |
| }, |
| { |
| "epoch": 2.142857142857143, |
| "grad_norm": 249.71810913085938, |
| "learning_rate": 7.924528301886793e-07, |
| "loss": 17.562179565429688, |
| "step": 15 |
| }, |
| { |
| "epoch": 2.857142857142857, |
| "grad_norm": 200.77024841308594, |
| "learning_rate": 6.981132075471697e-07, |
| "loss": 17.92266387939453, |
| "step": 20 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 3.1628379821777344, |
| "eval_map": 0.8969, |
| "eval_map_50": 1.0, |
| "eval_map_75": 1.0, |
| "eval_map_large": 0.8855, |
| "eval_map_medium": 1.0, |
| "eval_map_small": -1.0, |
| "eval_map_weight_plate": 0.8969, |
| "eval_mar_1": 0.3375, |
| "eval_mar_10": 0.9625, |
| "eval_mar_100": 0.975, |
| "eval_mar_100_weight_plate": 0.975, |
| "eval_mar_large": 0.9714, |
| "eval_mar_medium": 1.0, |
| "eval_mar_small": -1.0, |
| "eval_runtime": 0.1745, |
| "eval_samples_per_second": 17.195, |
| "eval_steps_per_second": 5.732, |
| "step": 21 |
| }, |
| { |
| "epoch": 3.571428571428571, |
| "grad_norm": 150.89956665039062, |
| "learning_rate": 6.037735849056604e-07, |
| "loss": 20.374853515625, |
| "step": 25 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 3.289055585861206, |
| "eval_map": 0.8691, |
| "eval_map_50": 0.9743, |
| "eval_map_75": 0.9743, |
| "eval_map_large": 0.8843, |
| "eval_map_medium": 1.0, |
| "eval_map_small": -1.0, |
| "eval_map_weight_plate": 0.8691, |
| "eval_mar_1": 0.35, |
| "eval_mar_10": 0.95, |
| "eval_mar_100": 0.9625, |
| "eval_mar_100_weight_plate": 0.9625, |
| "eval_mar_large": 0.9571, |
| "eval_mar_medium": 1.0, |
| "eval_mar_small": -1.0, |
| "eval_runtime": 0.1762, |
| "eval_samples_per_second": 17.024, |
| "eval_steps_per_second": 5.675, |
| "step": 28 |
| }, |
| { |
| "epoch": 4.285714285714286, |
| "grad_norm": 213.67095947265625, |
| "learning_rate": 5.094339622641509e-07, |
| "loss": 17.484503173828124, |
| "step": 30 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 152.34652709960938, |
| "learning_rate": 4.150943396226415e-07, |
| "loss": 17.022750854492188, |
| "step": 35 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_loss": 3.3509438037872314, |
| "eval_map": 0.8836, |
| "eval_map_50": 1.0, |
| "eval_map_75": 1.0, |
| "eval_map_large": 0.8844, |
| "eval_map_medium": 0.9, |
| "eval_map_small": -1.0, |
| "eval_map_weight_plate": 0.8836, |
| "eval_mar_1": 0.3375, |
| "eval_mar_10": 0.9375, |
| "eval_mar_100": 0.95, |
| "eval_mar_100_weight_plate": 0.95, |
| "eval_mar_large": 0.9571, |
| "eval_mar_medium": 0.9, |
| "eval_mar_small": -1.0, |
| "eval_runtime": 0.1647, |
| "eval_samples_per_second": 18.218, |
| "eval_steps_per_second": 6.073, |
| "step": 35 |
| }, |
| { |
| "epoch": 5.714285714285714, |
| "grad_norm": 268.59552001953125, |
| "learning_rate": 3.2075471698113204e-07, |
| "loss": 19.07172088623047, |
| "step": 40 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_loss": 3.491076707839966, |
| "eval_map": 0.8787, |
| "eval_map_50": 1.0, |
| "eval_map_75": 1.0, |
| "eval_map_large": 0.8791, |
| "eval_map_medium": 0.9, |
| "eval_map_small": -1.0, |
| "eval_map_weight_plate": 0.8787, |
| "eval_mar_1": 0.325, |
| "eval_mar_10": 0.925, |
| "eval_mar_100": 0.925, |
| "eval_mar_100_weight_plate": 0.925, |
| "eval_mar_large": 0.9286, |
| "eval_mar_medium": 0.9, |
| "eval_mar_small": -1.0, |
| "eval_runtime": 0.1529, |
| "eval_samples_per_second": 19.625, |
| "eval_steps_per_second": 6.542, |
| "step": 42 |
| }, |
| { |
| "epoch": 6.428571428571429, |
| "grad_norm": 123.03663635253906, |
| "learning_rate": 2.2641509433962263e-07, |
| "loss": 16.49561462402344, |
| "step": 45 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_loss": 3.59818959236145, |
| "eval_map": 0.8778, |
| "eval_map_50": 1.0, |
| "eval_map_75": 1.0, |
| "eval_map_large": 0.8756, |
| "eval_map_medium": 0.9, |
| "eval_map_small": -1.0, |
| "eval_map_weight_plate": 0.8778, |
| "eval_mar_1": 0.325, |
| "eval_mar_10": 0.925, |
| "eval_mar_100": 0.925, |
| "eval_mar_100_weight_plate": 0.925, |
| "eval_mar_large": 0.9286, |
| "eval_mar_medium": 0.9, |
| "eval_mar_small": -1.0, |
| "eval_runtime": 0.1821, |
| "eval_samples_per_second": 16.473, |
| "eval_steps_per_second": 5.491, |
| "step": 49 |
| }, |
| { |
| "epoch": 7.142857142857143, |
| "grad_norm": 91.42852783203125, |
| "learning_rate": 1.320754716981132e-07, |
| "loss": 17.98704376220703, |
| "step": 50 |
| }, |
| { |
| "epoch": 7.857142857142857, |
| "grad_norm": 290.90447998046875, |
| "learning_rate": 3.7735849056603774e-08, |
| "loss": 18.0838623046875, |
| "step": 55 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_loss": 3.5716047286987305, |
| "eval_map": 0.8835, |
| "eval_map_50": 1.0, |
| "eval_map_75": 1.0, |
| "eval_map_large": 0.8831, |
| "eval_map_medium": 0.9, |
| "eval_map_small": -1.0, |
| "eval_map_weight_plate": 0.8835, |
| "eval_mar_1": 0.3375, |
| "eval_mar_10": 0.9375, |
| "eval_mar_100": 0.95, |
| "eval_mar_100_weight_plate": 0.95, |
| "eval_mar_large": 0.9571, |
| "eval_mar_medium": 0.9, |
| "eval_mar_small": -1.0, |
| "eval_runtime": 0.1542, |
| "eval_samples_per_second": 19.459, |
| "eval_steps_per_second": 6.486, |
| "step": 56 |
| }, |
| { |
| "epoch": 8.0, |
| "step": 56, |
| "total_flos": 1.65946124832768e+16, |
| "train_loss": 18.40784318106515, |
| "train_runtime": 29.1797, |
| "train_samples_per_second": 6.854, |
| "train_steps_per_second": 1.919 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 56, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 8, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.65946124832768e+16, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
