| { | |
| "best_metric": 1.0987577438354492, | |
| "best_model_checkpoint": "hBERTv2_data_aug_mnli/checkpoint-314400", | |
| "epoch": 15.0, | |
| "global_step": 471600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.900038167938932e-05, | |
| "loss": 1.0988, | |
| "step": 31440 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.31818644931227713, | |
| "eval_loss": 1.09883713722229, | |
| "eval_runtime": 11.411, | |
| "eval_samples_per_second": 860.133, | |
| "eval_steps_per_second": 3.418, | |
| "step": 31440 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8000731552162856e-05, | |
| "loss": 1.0985, | |
| "step": 62880 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.31818644931227713, | |
| "eval_loss": 1.0991657972335815, | |
| "eval_runtime": 11.4006, | |
| "eval_samples_per_second": 860.917, | |
| "eval_steps_per_second": 3.421, | |
| "step": 62880 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7001176844783717e-05, | |
| "loss": 1.0985, | |
| "step": 94320 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.31818644931227713, | |
| "eval_loss": 1.099130392074585, | |
| "eval_runtime": 11.4603, | |
| "eval_samples_per_second": 856.436, | |
| "eval_steps_per_second": 3.403, | |
| "step": 94320 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.6001685750636135e-05, | |
| "loss": 1.0985, | |
| "step": 125760 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.31818644931227713, | |
| "eval_loss": 1.0990595817565918, | |
| "eval_runtime": 11.4198, | |
| "eval_samples_per_second": 859.471, | |
| "eval_steps_per_second": 3.415, | |
| "step": 125760 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5002131043257e-05, | |
| "loss": 1.0985, | |
| "step": 157200 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.31818644931227713, | |
| "eval_loss": 1.0987752676010132, | |
| "eval_runtime": 11.4346, | |
| "eval_samples_per_second": 858.363, | |
| "eval_steps_per_second": 3.411, | |
| "step": 157200 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4002639949109414e-05, | |
| "loss": 1.0985, | |
| "step": 188640 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.31818644931227713, | |
| "eval_loss": 1.0987842082977295, | |
| "eval_runtime": 11.6007, | |
| "eval_samples_per_second": 846.071, | |
| "eval_steps_per_second": 3.362, | |
| "step": 188640 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 4.300311704834606e-05, | |
| "loss": 1.0985, | |
| "step": 220080 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.31818644931227713, | |
| "eval_loss": 1.0988376140594482, | |
| "eval_runtime": 11.7431, | |
| "eval_samples_per_second": 835.809, | |
| "eval_steps_per_second": 3.321, | |
| "step": 220080 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 4.200356234096692e-05, | |
| "loss": 1.0985, | |
| "step": 251520 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.31818644931227713, | |
| "eval_loss": 1.0988106727600098, | |
| "eval_runtime": 11.6275, | |
| "eval_samples_per_second": 844.121, | |
| "eval_steps_per_second": 3.354, | |
| "step": 251520 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 4.100400763358779e-05, | |
| "loss": 1.0985, | |
| "step": 282960 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.31818644931227713, | |
| "eval_loss": 1.0987931489944458, | |
| "eval_runtime": 11.4012, | |
| "eval_samples_per_second": 860.872, | |
| "eval_steps_per_second": 3.421, | |
| "step": 282960 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 4.000445292620865e-05, | |
| "loss": 1.0985, | |
| "step": 314400 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.31818644931227713, | |
| "eval_loss": 1.0987577438354492, | |
| "eval_runtime": 11.4144, | |
| "eval_samples_per_second": 859.88, | |
| "eval_steps_per_second": 3.417, | |
| "step": 314400 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 3.9004930025445295e-05, | |
| "loss": 1.0985, | |
| "step": 345840 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.31818644931227713, | |
| "eval_loss": 1.0988106727600098, | |
| "eval_runtime": 11.4047, | |
| "eval_samples_per_second": 860.606, | |
| "eval_steps_per_second": 3.42, | |
| "step": 345840 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 3.800537531806616e-05, | |
| "loss": 1.0985, | |
| "step": 377280 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.31818644931227713, | |
| "eval_loss": 1.0988460779190063, | |
| "eval_runtime": 11.4535, | |
| "eval_samples_per_second": 856.946, | |
| "eval_steps_per_second": 3.405, | |
| "step": 377280 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 3.70058524173028e-05, | |
| "loss": 1.0985, | |
| "step": 408720 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.31818644931227713, | |
| "eval_loss": 1.0987931489944458, | |
| "eval_runtime": 11.41, | |
| "eval_samples_per_second": 860.209, | |
| "eval_steps_per_second": 3.418, | |
| "step": 408720 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 3.600636132315522e-05, | |
| "loss": 1.0985, | |
| "step": 440160 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.31818644931227713, | |
| "eval_loss": 1.0988106727600098, | |
| "eval_runtime": 11.4231, | |
| "eval_samples_per_second": 859.227, | |
| "eval_steps_per_second": 3.414, | |
| "step": 440160 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 3.500680661577608e-05, | |
| "loss": 1.0985, | |
| "step": 471600 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.31818644931227713, | |
| "eval_loss": 1.0987842082977295, | |
| "eval_runtime": 11.3973, | |
| "eval_samples_per_second": 861.165, | |
| "eval_steps_per_second": 3.422, | |
| "step": 471600 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "step": 471600, | |
| "total_flos": 1.5037282485912207e+19, | |
| "train_loss": 1.0985465619036259, | |
| "train_runtime": 409336.4634, | |
| "train_samples_per_second": 983.123, | |
| "train_steps_per_second": 3.84 | |
| } | |
| ], | |
| "max_steps": 1572000, | |
| "num_train_epochs": 50, | |
| "total_flos": 1.5037282485912207e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |