| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.932735426008969, | |
| "eval_steps": 500, | |
| "global_step": 5500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 1.820986547085202e-05, | |
| "loss": 1.626, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 1.641614349775785e-05, | |
| "loss": 1.4201, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_bleu": 33.3044, | |
| "eval_gen_len": 28.8563, | |
| "eval_loss": 1.3592603206634521, | |
| "eval_runtime": 172.5107, | |
| "eval_samples_per_second": 9.762, | |
| "eval_steps_per_second": 0.614, | |
| "step": 1115 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 1.462242152466368e-05, | |
| "loss": 1.2694, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 1.2828699551569507e-05, | |
| "loss": 1.2085, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_bleu": 34.5144, | |
| "eval_gen_len": 29.0564, | |
| "eval_loss": 1.3197557926177979, | |
| "eval_runtime": 172.4795, | |
| "eval_samples_per_second": 9.763, | |
| "eval_steps_per_second": 0.615, | |
| "step": 2230 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 1.1034977578475336e-05, | |
| "loss": 1.1369, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 9.244843049327356e-06, | |
| "loss": 1.0786, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_bleu": 34.6844, | |
| "eval_gen_len": 29.1496, | |
| "eval_loss": 1.3080627918243408, | |
| "eval_runtime": 175.6578, | |
| "eval_samples_per_second": 9.587, | |
| "eval_steps_per_second": 0.603, | |
| "step": 3345 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 7.451121076233184e-06, | |
| "loss": 1.0529, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 5.657399103139014e-06, | |
| "loss": 0.9958, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_bleu": 34.862, | |
| "eval_gen_len": 29.0529, | |
| "eval_loss": 1.309422254562378, | |
| "eval_runtime": 174.4734, | |
| "eval_samples_per_second": 9.652, | |
| "eval_steps_per_second": 0.608, | |
| "step": 4460 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 3.863677130044843e-06, | |
| "loss": 0.9933, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "learning_rate": 2.069955156950673e-06, | |
| "loss": 0.9477, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 4.93, | |
| "learning_rate": 2.762331838565023e-07, | |
| "loss": 0.9504, | |
| "step": 5500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 5575, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 1101235524599808.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |