| { | |
| "best_metric": 1.4430099725723267, | |
| "best_model_checkpoint": "./zhko_mbartLarge_19p_run3/checkpoint-6250", | |
| "epoch": 3.589375448671931, | |
| "eval_steps": 1250, | |
| "global_step": 10000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 1.1666666666666665e-05, | |
| "loss": 2.5039, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 2.333333333333333e-05, | |
| "loss": 1.957, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "eval_bleu": 12.1283, | |
| "eval_gen_len": 15.4356, | |
| "eval_loss": 1.7572811841964722, | |
| "eval_runtime": 683.8515, | |
| "eval_samples_per_second": 16.293, | |
| "eval_steps_per_second": 1.019, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 3.5e-05, | |
| "loss": 1.815, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 3.4336115326251896e-05, | |
| "loss": 1.7411, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 3.367223065250379e-05, | |
| "loss": 1.6568, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "eval_bleu": 14.3528, | |
| "eval_gen_len": 14.9632, | |
| "eval_loss": 1.5792840719223022, | |
| "eval_runtime": 663.5666, | |
| "eval_samples_per_second": 16.791, | |
| "eval_steps_per_second": 1.05, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 3.300834597875569e-05, | |
| "loss": 1.5955, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 3.234446130500758e-05, | |
| "loss": 1.4964, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "eval_bleu": 16.0215, | |
| "eval_gen_len": 14.9365, | |
| "eval_loss": 1.4805573225021362, | |
| "eval_runtime": 662.6976, | |
| "eval_samples_per_second": 16.813, | |
| "eval_steps_per_second": 1.052, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 3.168057663125948e-05, | |
| "loss": 1.3799, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 3.101669195751138e-05, | |
| "loss": 1.2679, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 3.0352807283763275e-05, | |
| "loss": 1.2204, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "eval_bleu": 16.5619, | |
| "eval_gen_len": 14.8621, | |
| "eval_loss": 1.4681929349899292, | |
| "eval_runtime": 647.0579, | |
| "eval_samples_per_second": 17.219, | |
| "eval_steps_per_second": 1.077, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 2.968892261001517e-05, | |
| "loss": 1.2044, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 2.9025037936267067e-05, | |
| "loss": 1.1678, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "eval_bleu": 16.9926, | |
| "eval_gen_len": 14.7721, | |
| "eval_loss": 1.4430099725723267, | |
| "eval_runtime": 652.3055, | |
| "eval_samples_per_second": 17.081, | |
| "eval_steps_per_second": 1.069, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 2.8361153262518967e-05, | |
| "loss": 1.0948, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 2.7697268588770863e-05, | |
| "loss": 1.004, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 2.7033383915022762e-05, | |
| "loss": 0.9448, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "eval_bleu": 17.0327, | |
| "eval_gen_len": 14.5641, | |
| "eval_loss": 1.4615132808685303, | |
| "eval_runtime": 639.6142, | |
| "eval_samples_per_second": 17.42, | |
| "eval_steps_per_second": 1.09, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "learning_rate": 2.6369499241274658e-05, | |
| "loss": 0.9289, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 2.570561456752655e-05, | |
| "loss": 0.9083, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "eval_bleu": 17.863, | |
| "eval_gen_len": 14.6946, | |
| "eval_loss": 1.4744235277175903, | |
| "eval_runtime": 639.363, | |
| "eval_samples_per_second": 17.427, | |
| "eval_steps_per_second": 1.09, | |
| "step": 8750 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 2.504172989377845e-05, | |
| "loss": 0.8756, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.41, | |
| "learning_rate": 2.4377845220030346e-05, | |
| "loss": 0.8075, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 2.3713960546282245e-05, | |
| "loss": 0.7389, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "eval_bleu": 17.3256, | |
| "eval_gen_len": 14.7824, | |
| "eval_loss": 1.512853741645813, | |
| "eval_runtime": 646.652, | |
| "eval_samples_per_second": 17.23, | |
| "eval_steps_per_second": 1.078, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "step": 10000, | |
| "total_flos": 6.9348152573952e+17, | |
| "train_loss": 1.3154443878173827, | |
| "train_runtime": 22326.8302, | |
| "train_samples_per_second": 39.928, | |
| "train_steps_per_second": 1.248 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 27860, | |
| "num_train_epochs": 10, | |
| "save_steps": 1250, | |
| "total_flos": 6.9348152573952e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |