| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 12.0, | |
| "eval_steps": 500, | |
| "global_step": 3936, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_gen_len": 15.577319587628866, | |
| "eval_loss": 8.096635818481445, | |
| "eval_rouge1": 16.5418, | |
| "eval_rouge2": 10.3523, | |
| "eval_rougeL": 13.972, | |
| "eval_rougeLsum": 14.1918, | |
| "eval_runtime": 157.1024, | |
| "eval_samples_per_second": 1.852, | |
| "eval_steps_per_second": 0.236, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 1.8394004282655248e-05, | |
| "loss": 18.3143, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_gen_len": 8.896907216494846, | |
| "eval_loss": 0.9259825944900513, | |
| "eval_rouge1": 31.5806, | |
| "eval_rouge2": 27.0287, | |
| "eval_rougeL": 29.6916, | |
| "eval_rougeLsum": 30.0327, | |
| "eval_runtime": 155.4292, | |
| "eval_samples_per_second": 1.872, | |
| "eval_steps_per_second": 0.238, | |
| "step": 656 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_gen_len": 13.807560137457045, | |
| "eval_loss": 0.7708158493041992, | |
| "eval_rouge1": 22.6847, | |
| "eval_rouge2": 15.805, | |
| "eval_rougeL": 19.6336, | |
| "eval_rougeLsum": 19.8945, | |
| "eval_runtime": 155.3127, | |
| "eval_samples_per_second": 1.874, | |
| "eval_steps_per_second": 0.238, | |
| "step": 984 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 1.5717344753747325e-05, | |
| "loss": 1.0739, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_gen_len": 9.618556701030927, | |
| "eval_loss": 0.7307576537132263, | |
| "eval_rouge1": 35.1675, | |
| "eval_rouge2": 27.3998, | |
| "eval_rougeL": 31.8527, | |
| "eval_rougeLsum": 32.0356, | |
| "eval_runtime": 133.4505, | |
| "eval_samples_per_second": 2.181, | |
| "eval_steps_per_second": 0.277, | |
| "step": 1312 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "learning_rate": 1.3040685224839403e-05, | |
| "loss": 0.8085, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_gen_len": 10.116838487972508, | |
| "eval_loss": 0.708371102809906, | |
| "eval_rouge1": 34.4346, | |
| "eval_rouge2": 26.202, | |
| "eval_rougeL": 30.8999, | |
| "eval_rougeLsum": 31.212, | |
| "eval_runtime": 139.0571, | |
| "eval_samples_per_second": 2.093, | |
| "eval_steps_per_second": 0.266, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_gen_len": 10.268041237113403, | |
| "eval_loss": 0.6923775672912598, | |
| "eval_rouge1": 34.3345, | |
| "eval_rouge2": 26.0144, | |
| "eval_rougeL": 30.692, | |
| "eval_rougeLsum": 31.0384, | |
| "eval_runtime": 141.1422, | |
| "eval_samples_per_second": 2.062, | |
| "eval_steps_per_second": 0.262, | |
| "step": 1968 | |
| }, | |
| { | |
| "epoch": 6.1, | |
| "learning_rate": 1.036402569593148e-05, | |
| "loss": 0.7597, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_gen_len": 10.31958762886598, | |
| "eval_loss": 0.6812536716461182, | |
| "eval_rouge1": 34.3854, | |
| "eval_rouge2": 26.0495, | |
| "eval_rougeL": 30.8335, | |
| "eval_rougeLsum": 31.1696, | |
| "eval_runtime": 143.6068, | |
| "eval_samples_per_second": 2.026, | |
| "eval_steps_per_second": 0.258, | |
| "step": 2296 | |
| }, | |
| { | |
| "epoch": 7.62, | |
| "learning_rate": 7.687366167023556e-06, | |
| "loss": 0.7442, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_gen_len": 10.360824742268042, | |
| "eval_loss": 0.6728662252426147, | |
| "eval_rouge1": 34.3758, | |
| "eval_rouge2": 26.0079, | |
| "eval_rougeL": 30.7863, | |
| "eval_rougeLsum": 31.1239, | |
| "eval_runtime": 142.7676, | |
| "eval_samples_per_second": 2.038, | |
| "eval_steps_per_second": 0.259, | |
| "step": 2624 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_gen_len": 10.49828178694158, | |
| "eval_loss": 0.6669635772705078, | |
| "eval_rouge1": 34.2115, | |
| "eval_rouge2": 25.7443, | |
| "eval_rougeL": 30.5369, | |
| "eval_rougeLsum": 30.9282, | |
| "eval_runtime": 145.2485, | |
| "eval_samples_per_second": 2.003, | |
| "eval_steps_per_second": 0.255, | |
| "step": 2952 | |
| }, | |
| { | |
| "epoch": 9.15, | |
| "learning_rate": 5.010706638115633e-06, | |
| "loss": 0.7252, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_gen_len": 10.529209621993127, | |
| "eval_loss": 0.6624875664710999, | |
| "eval_rouge1": 34.2518, | |
| "eval_rouge2": 25.7147, | |
| "eval_rougeL": 30.5433, | |
| "eval_rougeLsum": 30.9116, | |
| "eval_runtime": 146.9477, | |
| "eval_samples_per_second": 1.98, | |
| "eval_steps_per_second": 0.252, | |
| "step": 3280 | |
| }, | |
| { | |
| "epoch": 10.67, | |
| "learning_rate": 2.334047109207709e-06, | |
| "loss": 0.7168, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_gen_len": 10.618556701030927, | |
| "eval_loss": 0.6601226925849915, | |
| "eval_rouge1": 34.0539, | |
| "eval_rouge2": 25.5073, | |
| "eval_rougeL": 30.329, | |
| "eval_rougeLsum": 30.6828, | |
| "eval_runtime": 146.8209, | |
| "eval_samples_per_second": 1.982, | |
| "eval_steps_per_second": 0.252, | |
| "step": 3608 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_gen_len": 10.532646048109966, | |
| "eval_loss": 0.6593531370162964, | |
| "eval_rouge1": 34.2696, | |
| "eval_rouge2": 25.7973, | |
| "eval_rougeL": 30.5609, | |
| "eval_rougeLsum": 30.9651, | |
| "eval_runtime": 145.5529, | |
| "eval_samples_per_second": 1.999, | |
| "eval_steps_per_second": 0.254, | |
| "step": 3936 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 3936, | |
| "num_train_epochs": 12, | |
| "save_steps": 500, | |
| "total_flos": 2.949770616687821e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |