| { | |
| "best_metric": 0.6930651664733887, | |
| "best_model_checkpoint": "hBERTv1_no_pretrain_qnli/checkpoint-9009", | |
| "epoch": 16.0, | |
| "global_step": 13104, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.00049002442002442, | |
| "loss": 0.715, | |
| "step": 819 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.4946000366099213, | |
| "eval_loss": 0.6931389570236206, | |
| "eval_runtime": 9.5675, | |
| "eval_samples_per_second": 570.995, | |
| "eval_steps_per_second": 4.494, | |
| "step": 819 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.00048002442002442004, | |
| "loss": 0.6932, | |
| "step": 1638 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.4946000366099213, | |
| "eval_loss": 0.6931495070457458, | |
| "eval_runtime": 9.5694, | |
| "eval_samples_per_second": 570.88, | |
| "eval_steps_per_second": 4.493, | |
| "step": 1638 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 0.00047002442002442, | |
| "loss": 0.6936, | |
| "step": 2457 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5053999633900788, | |
| "eval_loss": 0.6931167840957642, | |
| "eval_runtime": 9.5644, | |
| "eval_samples_per_second": 571.18, | |
| "eval_steps_per_second": 4.496, | |
| "step": 2457 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 0.00046002442002442004, | |
| "loss": 0.6932, | |
| "step": 3276 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.4946000366099213, | |
| "eval_loss": 0.6931969523429871, | |
| "eval_runtime": 9.5713, | |
| "eval_samples_per_second": 570.77, | |
| "eval_steps_per_second": 4.493, | |
| "step": 3276 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.00045002442002442, | |
| "loss": 0.6932, | |
| "step": 4095 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5053999633900788, | |
| "eval_loss": 0.6933488249778748, | |
| "eval_runtime": 9.573, | |
| "eval_samples_per_second": 570.669, | |
| "eval_steps_per_second": 4.492, | |
| "step": 4095 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 0.00044002442002442004, | |
| "loss": 0.6932, | |
| "step": 4914 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5053999633900788, | |
| "eval_loss": 0.6930915117263794, | |
| "eval_runtime": 9.5567, | |
| "eval_samples_per_second": 571.639, | |
| "eval_steps_per_second": 4.499, | |
| "step": 4914 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 0.00043002442002442007, | |
| "loss": 0.6932, | |
| "step": 5733 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5053999633900788, | |
| "eval_loss": 0.6931073069572449, | |
| "eval_runtime": 9.5808, | |
| "eval_samples_per_second": 570.202, | |
| "eval_steps_per_second": 4.488, | |
| "step": 5733 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 0.00042002442002442005, | |
| "loss": 0.6932, | |
| "step": 6552 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.5053999633900788, | |
| "eval_loss": 0.6930862069129944, | |
| "eval_runtime": 9.5904, | |
| "eval_samples_per_second": 569.63, | |
| "eval_steps_per_second": 4.484, | |
| "step": 6552 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 0.00041003663003663003, | |
| "loss": 0.6935, | |
| "step": 7371 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.5053999633900788, | |
| "eval_loss": 0.693478524684906, | |
| "eval_runtime": 9.5344, | |
| "eval_samples_per_second": 572.981, | |
| "eval_steps_per_second": 4.51, | |
| "step": 7371 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 0.00040003663003663006, | |
| "loss": 0.6932, | |
| "step": 8190 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.5053999633900788, | |
| "eval_loss": 0.6930915117263794, | |
| "eval_runtime": 9.5321, | |
| "eval_samples_per_second": 573.116, | |
| "eval_steps_per_second": 4.511, | |
| "step": 8190 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 0.00039003663003663004, | |
| "loss": 0.6932, | |
| "step": 9009 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.5053999633900788, | |
| "eval_loss": 0.6930651664733887, | |
| "eval_runtime": 9.5448, | |
| "eval_samples_per_second": 572.352, | |
| "eval_steps_per_second": 4.505, | |
| "step": 9009 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 0.00038004884004884, | |
| "loss": 0.6932, | |
| "step": 9828 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.5053999633900788, | |
| "eval_loss": 0.6931073069572449, | |
| "eval_runtime": 9.5697, | |
| "eval_samples_per_second": 570.866, | |
| "eval_steps_per_second": 4.493, | |
| "step": 9828 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 0.00037004884004884005, | |
| "loss": 0.6932, | |
| "step": 10647 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.5053999633900788, | |
| "eval_loss": 0.6931073069572449, | |
| "eval_runtime": 9.5909, | |
| "eval_samples_per_second": 569.603, | |
| "eval_steps_per_second": 4.483, | |
| "step": 10647 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 0.00036004884004884, | |
| "loss": 0.6932, | |
| "step": 11466 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.4946000366099213, | |
| "eval_loss": 0.6931231617927551, | |
| "eval_runtime": 9.5446, | |
| "eval_samples_per_second": 572.363, | |
| "eval_steps_per_second": 4.505, | |
| "step": 11466 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 0.00035004884004884005, | |
| "loss": 0.6932, | |
| "step": 12285 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.4946000366099213, | |
| "eval_loss": 0.693396270275116, | |
| "eval_runtime": 9.5573, | |
| "eval_samples_per_second": 571.603, | |
| "eval_steps_per_second": 4.499, | |
| "step": 12285 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 0.0003400488400488401, | |
| "loss": 0.6932, | |
| "step": 13104 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.4946000366099213, | |
| "eval_loss": 0.6931284070014954, | |
| "eval_runtime": 9.5555, | |
| "eval_samples_per_second": 571.71, | |
| "eval_steps_per_second": 4.5, | |
| "step": 13104 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "step": 13104, | |
| "total_flos": 2.481936015514665e+17, | |
| "train_loss": 0.694585192334521, | |
| "train_runtime": 8903.1517, | |
| "train_samples_per_second": 588.236, | |
| "train_steps_per_second": 4.599 | |
| } | |
| ], | |
| "max_steps": 40950, | |
| "num_train_epochs": 50, | |
| "total_flos": 2.481936015514665e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |