| { | |
| "best_metric": 0.6899781823158264, | |
| "best_model_checkpoint": "mobilebert_add_GLUE_Experiment_wnli_256/checkpoint-5", | |
| "epoch": 6.0, | |
| "global_step": 30, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.6944, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5633802816901409, | |
| "eval_loss": 0.6899781823158264, | |
| "eval_runtime": 0.1477, | |
| "eval_samples_per_second": 480.678, | |
| "eval_steps_per_second": 6.77, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.6936, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5633802816901409, | |
| "eval_loss": 0.6921276450157166, | |
| "eval_runtime": 0.1426, | |
| "eval_samples_per_second": 497.746, | |
| "eval_steps_per_second": 7.011, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.6933, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5633802816901409, | |
| "eval_loss": 0.6930466890335083, | |
| "eval_runtime": 0.1484, | |
| "eval_samples_per_second": 478.358, | |
| "eval_steps_per_second": 6.737, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.693, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5633802816901409, | |
| "eval_loss": 0.6919631361961365, | |
| "eval_runtime": 0.1441, | |
| "eval_samples_per_second": 492.739, | |
| "eval_steps_per_second": 6.94, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.693, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5633802816901409, | |
| "eval_loss": 0.690990686416626, | |
| "eval_runtime": 0.1446, | |
| "eval_samples_per_second": 491.001, | |
| "eval_steps_per_second": 6.916, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.6931, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5633802816901409, | |
| "eval_loss": 0.6907631754875183, | |
| "eval_runtime": 0.1426, | |
| "eval_samples_per_second": 498.028, | |
| "eval_steps_per_second": 7.014, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "step": 30, | |
| "total_flos": 98203563196416.0, | |
| "train_loss": 0.6933950026830037, | |
| "train_runtime": 42.2893, | |
| "train_samples_per_second": 750.78, | |
| "train_steps_per_second": 5.912 | |
| } | |
| ], | |
| "max_steps": 250, | |
| "num_train_epochs": 50, | |
| "total_flos": 98203563196416.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |