| { | |
| "best_metric": 0.4670983850955963, | |
| "best_model_checkpoint": "mobilebert_add_GLUE_Experiment_sst2/checkpoint-4743", | |
| "epoch": 14.0, | |
| "global_step": 7378, | |
| "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": 0.6635, | |
| "step": 527 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5389908256880734, | |
| "eval_loss": 0.6994264721870422, | |
| "eval_runtime": 1.8725, | |
| "eval_samples_per_second": 465.675, | |
| "eval_steps_per_second": 3.738, | |
| "step": 527 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.5959, | |
| "step": 1054 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5665137614678899, | |
| "eval_loss": 0.6921247839927673, | |
| "eval_runtime": 1.8745, | |
| "eval_samples_per_second": 465.202, | |
| "eval_steps_per_second": 3.734, | |
| "step": 1054 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.5684, | |
| "step": 1581 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.551605504587156, | |
| "eval_loss": 0.7081714868545532, | |
| "eval_runtime": 1.8708, | |
| "eval_samples_per_second": 466.112, | |
| "eval_steps_per_second": 3.742, | |
| "step": 1581 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.5544, | |
| "step": 2108 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5619266055045872, | |
| "eval_loss": 0.688310980796814, | |
| "eval_runtime": 1.8754, | |
| "eval_samples_per_second": 464.97, | |
| "eval_steps_per_second": 3.733, | |
| "step": 2108 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.5471, | |
| "step": 2635 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5940366972477065, | |
| "eval_loss": 0.6938039064407349, | |
| "eval_runtime": 1.8755, | |
| "eval_samples_per_second": 464.943, | |
| "eval_steps_per_second": 3.732, | |
| "step": 2635 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.5414, | |
| "step": 3162 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5802752293577982, | |
| "eval_loss": 0.7045485973358154, | |
| "eval_runtime": 1.8702, | |
| "eval_samples_per_second": 466.254, | |
| "eval_steps_per_second": 3.743, | |
| "step": 3162 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.5381, | |
| "step": 3689 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5653669724770642, | |
| "eval_loss": 0.7353540658950806, | |
| "eval_runtime": 1.8751, | |
| "eval_samples_per_second": 465.044, | |
| "eval_steps_per_second": 3.733, | |
| "step": 3689 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.5338, | |
| "step": 4216 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.5825688073394495, | |
| "eval_loss": 0.7316306829452515, | |
| "eval_runtime": 1.8737, | |
| "eval_samples_per_second": 465.401, | |
| "eval_steps_per_second": 3.736, | |
| "step": 4216 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.3529, | |
| "step": 4743 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.7970183486238532, | |
| "eval_loss": 0.4670983850955963, | |
| "eval_runtime": 1.8694, | |
| "eval_samples_per_second": 466.456, | |
| "eval_steps_per_second": 3.744, | |
| "step": 4743 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 4e-05, | |
| "loss": 0.2415, | |
| "step": 5270 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.7981651376146789, | |
| "eval_loss": 0.4721962809562683, | |
| "eval_runtime": 1.8749, | |
| "eval_samples_per_second": 465.103, | |
| "eval_steps_per_second": 3.734, | |
| "step": 5270 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 0.2075, | |
| "step": 5797 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8061926605504587, | |
| "eval_loss": 0.4797086715698242, | |
| "eval_runtime": 1.8721, | |
| "eval_samples_per_second": 465.778, | |
| "eval_steps_per_second": 3.739, | |
| "step": 5797 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 3.8e-05, | |
| "loss": 0.1862, | |
| "step": 6324 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.7993119266055045, | |
| "eval_loss": 0.5133755803108215, | |
| "eval_runtime": 1.8749, | |
| "eval_samples_per_second": 465.095, | |
| "eval_steps_per_second": 3.734, | |
| "step": 6324 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 3.7e-05, | |
| "loss": 0.1724, | |
| "step": 6851 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.7993119266055045, | |
| "eval_loss": 0.5256267786026001, | |
| "eval_runtime": 1.8734, | |
| "eval_samples_per_second": 465.457, | |
| "eval_steps_per_second": 3.736, | |
| "step": 6851 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.1662, | |
| "step": 7378 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8027522935779816, | |
| "eval_loss": 0.5705782771110535, | |
| "eval_runtime": 1.8724, | |
| "eval_samples_per_second": 465.701, | |
| "eval_steps_per_second": 3.738, | |
| "step": 7378 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "step": 7378, | |
| "total_flos": 2.789091149296435e+16, | |
| "train_loss": 0.4192447450368441, | |
| "train_runtime": 6419.7515, | |
| "train_samples_per_second": 524.545, | |
| "train_steps_per_second": 4.105 | |
| } | |
| ], | |
| "max_steps": 26350, | |
| "num_train_epochs": 50, | |
| "total_flos": 2.789091149296435e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |