| { | |
| "best_metric": 0.8776137828826904, | |
| "best_model_checkpoint": "mobilebert_sa_GLUE_Experiment_mnli_128/checkpoint-27612", | |
| "epoch": 14.0, | |
| "global_step": 42952, | |
| "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": 1.0129, | |
| "step": 3068 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5437595517065715, | |
| "eval_loss": 0.9528754353523254, | |
| "eval_runtime": 21.6888, | |
| "eval_samples_per_second": 452.537, | |
| "eval_steps_per_second": 3.55, | |
| "step": 3068 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.9284, | |
| "step": 6136 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5593479368313805, | |
| "eval_loss": 0.9265554547309875, | |
| "eval_runtime": 21.6883, | |
| "eval_samples_per_second": 452.548, | |
| "eval_steps_per_second": 3.55, | |
| "step": 6136 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.8999, | |
| "step": 9204 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5774834437086093, | |
| "eval_loss": 0.9054868817329407, | |
| "eval_runtime": 21.6795, | |
| "eval_samples_per_second": 452.731, | |
| "eval_steps_per_second": 3.552, | |
| "step": 9204 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.8803, | |
| "step": 12272 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5854304635761589, | |
| "eval_loss": 0.8951208591461182, | |
| "eval_runtime": 21.7169, | |
| "eval_samples_per_second": 451.952, | |
| "eval_steps_per_second": 3.546, | |
| "step": 12272 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.8637, | |
| "step": 15340 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5885888945491594, | |
| "eval_loss": 0.8991015553474426, | |
| "eval_runtime": 21.7239, | |
| "eval_samples_per_second": 451.806, | |
| "eval_steps_per_second": 3.544, | |
| "step": 15340 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.8472, | |
| "step": 18408 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5913397860417728, | |
| "eval_loss": 0.8906782269477844, | |
| "eval_runtime": 21.7654, | |
| "eval_samples_per_second": 450.945, | |
| "eval_steps_per_second": 3.538, | |
| "step": 18408 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.8309, | |
| "step": 21476 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5927661742231278, | |
| "eval_loss": 0.8940486907958984, | |
| "eval_runtime": 21.665, | |
| "eval_samples_per_second": 453.036, | |
| "eval_steps_per_second": 3.554, | |
| "step": 21476 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.814, | |
| "step": 24544 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.5987773815588385, | |
| "eval_loss": 0.8879620432853699, | |
| "eval_runtime": 21.7009, | |
| "eval_samples_per_second": 452.286, | |
| "eval_steps_per_second": 3.548, | |
| "step": 24544 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.7988, | |
| "step": 27612 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.6022414671421293, | |
| "eval_loss": 0.8776137828826904, | |
| "eval_runtime": 21.7212, | |
| "eval_samples_per_second": 451.863, | |
| "eval_steps_per_second": 3.545, | |
| "step": 27612 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 4e-05, | |
| "loss": 0.7825, | |
| "step": 30680 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.6022414671421293, | |
| "eval_loss": 0.8958155512809753, | |
| "eval_runtime": 21.7108, | |
| "eval_samples_per_second": 452.08, | |
| "eval_steps_per_second": 3.547, | |
| "step": 30680 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 0.7662, | |
| "step": 33748 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.6061130922058074, | |
| "eval_loss": 0.8834749460220337, | |
| "eval_runtime": 21.6874, | |
| "eval_samples_per_second": 452.567, | |
| "eval_steps_per_second": 3.55, | |
| "step": 33748 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 3.8e-05, | |
| "loss": 0.7504, | |
| "step": 36816 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.6040753948038716, | |
| "eval_loss": 0.9004417657852173, | |
| "eval_runtime": 21.7036, | |
| "eval_samples_per_second": 452.229, | |
| "eval_steps_per_second": 3.548, | |
| "step": 36816 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 3.7e-05, | |
| "loss": 0.7359, | |
| "step": 39884 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.6, | |
| "eval_loss": 0.9251638650894165, | |
| "eval_runtime": 21.7106, | |
| "eval_samples_per_second": 452.083, | |
| "eval_steps_per_second": 3.547, | |
| "step": 39884 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.7204, | |
| "step": 42952 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.6007131940906776, | |
| "eval_loss": 0.9130809903144836, | |
| "eval_runtime": 21.7023, | |
| "eval_samples_per_second": 452.257, | |
| "eval_steps_per_second": 3.548, | |
| "step": 42952 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "step": 42952, | |
| "total_flos": 1.3103958963991347e+17, | |
| "train_loss": 0.8308279570709018, | |
| "train_runtime": 35893.6313, | |
| "train_samples_per_second": 547.036, | |
| "train_steps_per_second": 4.274 | |
| } | |
| ], | |
| "max_steps": 153400, | |
| "num_train_epochs": 50, | |
| "total_flos": 1.3103958963991347e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |