| { | |
| "best_metric": 1.3231624364852905, | |
| "best_model_checkpoint": "hBERTv2_data_aug_rte/checkpoint-568", | |
| "epoch": 6.0, | |
| "global_step": 3408, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.900704225352113e-05, | |
| "loss": 0.6262, | |
| "step": 568 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.49097472924187724, | |
| "eval_loss": 1.3231624364852905, | |
| "eval_runtime": 0.3367, | |
| "eval_samples_per_second": 822.619, | |
| "eval_steps_per_second": 5.939, | |
| "step": 568 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8007042253521126e-05, | |
| "loss": 0.0855, | |
| "step": 1136 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.49458483754512633, | |
| "eval_loss": 2.34574818611145, | |
| "eval_runtime": 0.3344, | |
| "eval_samples_per_second": 828.433, | |
| "eval_steps_per_second": 5.981, | |
| "step": 1136 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.700704225352113e-05, | |
| "loss": 0.022, | |
| "step": 1704 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5018050541516246, | |
| "eval_loss": 2.9796979427337646, | |
| "eval_runtime": 0.3341, | |
| "eval_samples_per_second": 829.097, | |
| "eval_steps_per_second": 5.986, | |
| "step": 1704 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.6007042253521125e-05, | |
| "loss": 0.0128, | |
| "step": 2272 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5270758122743683, | |
| "eval_loss": 2.639472246170044, | |
| "eval_runtime": 0.3358, | |
| "eval_samples_per_second": 824.985, | |
| "eval_steps_per_second": 5.957, | |
| "step": 2272 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.500704225352113e-05, | |
| "loss": 0.0085, | |
| "step": 2840 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5379061371841155, | |
| "eval_loss": 3.163405418395996, | |
| "eval_runtime": 0.3351, | |
| "eval_samples_per_second": 826.541, | |
| "eval_steps_per_second": 5.968, | |
| "step": 2840 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.400704225352113e-05, | |
| "loss": 0.0059, | |
| "step": 3408 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.51985559566787, | |
| "eval_loss": 3.5948262214660645, | |
| "eval_runtime": 0.3496, | |
| "eval_samples_per_second": 792.375, | |
| "eval_steps_per_second": 5.721, | |
| "step": 3408 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "step": 3408, | |
| "total_flos": 1.086589187873833e+17, | |
| "train_loss": 0.1268328929171316, | |
| "train_runtime": 3113.9793, | |
| "train_samples_per_second": 2334.601, | |
| "train_steps_per_second": 9.12 | |
| } | |
| ], | |
| "max_steps": 28400, | |
| "num_train_epochs": 50, | |
| "total_flos": 1.086589187873833e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |