| { | |
| "best_metric": 0.6929702162742615, | |
| "best_model_checkpoint": "hBERTv2_qnli/checkpoint-3280", | |
| "epoch": 13.0, | |
| "global_step": 5330, | |
| "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": 0.6968, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5053999633900788, | |
| "eval_loss": 0.6952407360076904, | |
| "eval_runtime": 6.3931, | |
| "eval_samples_per_second": 854.51, | |
| "eval_steps_per_second": 3.441, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.6943, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.4946000366099213, | |
| "eval_loss": 0.6931500434875488, | |
| "eval_runtime": 6.373, | |
| "eval_samples_per_second": 857.211, | |
| "eval_steps_per_second": 3.452, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.6937, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5053999633900788, | |
| "eval_loss": 0.6933012008666992, | |
| "eval_runtime": 6.3501, | |
| "eval_samples_per_second": 860.304, | |
| "eval_steps_per_second": 3.465, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.6934, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5053999633900788, | |
| "eval_loss": 0.6930809617042542, | |
| "eval_runtime": 6.372, | |
| "eval_samples_per_second": 857.351, | |
| "eval_steps_per_second": 3.453, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.6934, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5053999633900788, | |
| "eval_loss": 0.6931115388870239, | |
| "eval_runtime": 6.375, | |
| "eval_samples_per_second": 856.943, | |
| "eval_steps_per_second": 3.451, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.6933, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5053999633900788, | |
| "eval_loss": 0.6930229663848877, | |
| "eval_runtime": 6.349, | |
| "eval_samples_per_second": 860.452, | |
| "eval_steps_per_second": 3.465, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.6933, | |
| "step": 2870 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5053999633900788, | |
| "eval_loss": 0.6931431889533997, | |
| "eval_runtime": 6.3713, | |
| "eval_samples_per_second": 857.434, | |
| "eval_steps_per_second": 3.453, | |
| "step": 2870 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.6932, | |
| "step": 3280 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.5053999633900788, | |
| "eval_loss": 0.6929702162742615, | |
| "eval_runtime": 6.3689, | |
| "eval_samples_per_second": 857.766, | |
| "eval_steps_per_second": 3.454, | |
| "step": 3280 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.6932, | |
| "step": 3690 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.4946000366099213, | |
| "eval_loss": 0.6933699250221252, | |
| "eval_runtime": 6.3817, | |
| "eval_samples_per_second": 856.035, | |
| "eval_steps_per_second": 3.447, | |
| "step": 3690 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 4e-05, | |
| "loss": 0.6932, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.5053999633900788, | |
| "eval_loss": 0.6930176615715027, | |
| "eval_runtime": 6.3795, | |
| "eval_samples_per_second": 856.338, | |
| "eval_steps_per_second": 3.449, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 0.6932, | |
| "step": 4510 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.4946000366099213, | |
| "eval_loss": 0.6931337118148804, | |
| "eval_runtime": 6.3738, | |
| "eval_samples_per_second": 857.109, | |
| "eval_steps_per_second": 3.452, | |
| "step": 4510 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 3.8e-05, | |
| "loss": 0.6933, | |
| "step": 4920 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.4946000366099213, | |
| "eval_loss": 0.6934068202972412, | |
| "eval_runtime": 6.3622, | |
| "eval_samples_per_second": 858.665, | |
| "eval_steps_per_second": 3.458, | |
| "step": 4920 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 3.7e-05, | |
| "loss": 0.6932, | |
| "step": 5330 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.4946000366099213, | |
| "eval_loss": 0.6931495070457458, | |
| "eval_runtime": 6.3927, | |
| "eval_samples_per_second": 854.572, | |
| "eval_steps_per_second": 3.441, | |
| "step": 5330 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "step": 5330, | |
| "total_flos": 1.6959930780719514e+17, | |
| "train_loss": 0.6936567925601694, | |
| "train_runtime": 4977.7257, | |
| "train_samples_per_second": 1052.117, | |
| "train_steps_per_second": 4.118 | |
| } | |
| ], | |
| "max_steps": 20500, | |
| "num_train_epochs": 50, | |
| "total_flos": 1.6959930780719514e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |