| { | |
| "best_metric": 1.0819282531738281, | |
| "best_model_checkpoint": "./jako_mbartLarge_13p_run1/checkpoint-7500", | |
| "epoch": 3.116073747078681, | |
| "eval_steps": 1500, | |
| "global_step": 12000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 4.97382884061764e-05, | |
| "loss": 1.9323, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 4.9084009421617376e-05, | |
| "loss": 1.487, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 4.8429730437058366e-05, | |
| "loss": 1.3826, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "eval_bleu": 22.0729, | |
| "eval_gen_len": 18.9717, | |
| "eval_loss": 1.2989100217819214, | |
| "eval_runtime": 581.8634, | |
| "eval_samples_per_second": 13.235, | |
| "eval_steps_per_second": 0.828, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 4.777545145249935e-05, | |
| "loss": 1.2846, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 4.712117246794033e-05, | |
| "loss": 1.2292, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 4.646689348338132e-05, | |
| "loss": 1.1964, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "eval_bleu": 25.4863, | |
| "eval_gen_len": 19.1908, | |
| "eval_loss": 1.1629722118377686, | |
| "eval_runtime": 599.5574, | |
| "eval_samples_per_second": 12.844, | |
| "eval_steps_per_second": 0.804, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 4.58126144988223e-05, | |
| "loss": 1.1636, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 4.515833551426328e-05, | |
| "loss": 1.1243, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 4.450405652970427e-05, | |
| "loss": 0.9449, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "eval_bleu": 27.385, | |
| "eval_gen_len": 18.2955, | |
| "eval_loss": 1.112542748451233, | |
| "eval_runtime": 534.7746, | |
| "eval_samples_per_second": 14.4, | |
| "eval_steps_per_second": 0.901, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 4.3849777545145253e-05, | |
| "loss": 0.8708, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 4.3195498560586236e-05, | |
| "loss": 0.8446, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 4.254121957602722e-05, | |
| "loss": 0.8102, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "eval_bleu": 28.0041, | |
| "eval_gen_len": 18.6572, | |
| "eval_loss": 1.0919792652130127, | |
| "eval_runtime": 549.0767, | |
| "eval_samples_per_second": 14.025, | |
| "eval_steps_per_second": 0.878, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 4.18869405914682e-05, | |
| "loss": 0.7903, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 4.123266160690919e-05, | |
| "loss": 0.7814, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 4.057838262235017e-05, | |
| "loss": 0.7692, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "eval_bleu": 29.1055, | |
| "eval_gen_len": 18.2731, | |
| "eval_loss": 1.0819282531738281, | |
| "eval_runtime": 529.4168, | |
| "eval_samples_per_second": 14.546, | |
| "eval_steps_per_second": 0.91, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 3.992410363779115e-05, | |
| "loss": 0.7152, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 3.926982465323214e-05, | |
| "loss": 0.6146, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 3.8615545668673124e-05, | |
| "loss": 0.5741, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "eval_bleu": 28.1574, | |
| "eval_gen_len": 18.3485, | |
| "eval_loss": 1.1368560791015625, | |
| "eval_runtime": 531.4915, | |
| "eval_samples_per_second": 14.489, | |
| "eval_steps_per_second": 0.907, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 3.7961266684114114e-05, | |
| "loss": 0.5501, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 3.730698769955509e-05, | |
| "loss": 0.5317, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 3.665270871499607e-05, | |
| "loss": 0.5198, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "eval_bleu": 28.657, | |
| "eval_gen_len": 18.4527, | |
| "eval_loss": 1.1538134813308716, | |
| "eval_runtime": 540.4496, | |
| "eval_samples_per_second": 14.249, | |
| "eval_steps_per_second": 0.892, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "learning_rate": 3.599842973043706e-05, | |
| "loss": 0.5201, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 3.5344150745878046e-05, | |
| "loss": 0.5022, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 3.468987176131903e-05, | |
| "loss": 0.4532, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "eval_bleu": 28.6914, | |
| "eval_gen_len": 18.4562, | |
| "eval_loss": 1.158207893371582, | |
| "eval_runtime": 536.18, | |
| "eval_samples_per_second": 14.363, | |
| "eval_steps_per_second": 0.899, | |
| "step": 12000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 38510, | |
| "num_train_epochs": 10, | |
| "save_steps": 1500, | |
| "total_flos": 4.161929374041375e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |