| { | |
| "best_metric": 0.5091743119266054, | |
| "best_model_checkpoint": "outputs/soft_prompt/roberta-base/sst2/checkpoint-200", | |
| "epoch": 3.0, | |
| "global_step": 6315, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "eval_accuracy": 0.5091743119266054, | |
| "eval_loss": 1.2113580703735352, | |
| "eval_runtime": 1.5734, | |
| "eval_samples_per_second": 554.211, | |
| "eval_steps_per_second": 17.796, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "eval_accuracy": 0.5091743119266054, | |
| "eval_loss": 1.1833394765853882, | |
| "eval_runtime": 1.3274, | |
| "eval_samples_per_second": 656.934, | |
| "eval_steps_per_second": 21.094, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.027624703087885987, | |
| "loss": 1.7048, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_accuracy": 0.5091743119266054, | |
| "eval_loss": 1.4068197011947632, | |
| "eval_runtime": 1.6024, | |
| "eval_samples_per_second": 544.198, | |
| "eval_steps_per_second": 17.474, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "eval_accuracy": 0.5091743119266054, | |
| "eval_loss": 1.2358722686767578, | |
| "eval_runtime": 1.3312, | |
| "eval_samples_per_second": 655.031, | |
| "eval_steps_per_second": 21.033, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.02524940617577197, | |
| "loss": 1.4721, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_accuracy": 0.4908256880733945, | |
| "eval_loss": 1.5192540884017944, | |
| "eval_runtime": 1.4576, | |
| "eval_samples_per_second": 598.233, | |
| "eval_steps_per_second": 19.209, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "eval_accuracy": 0.4908256880733945, | |
| "eval_loss": 1.0336543321609497, | |
| "eval_runtime": 1.7394, | |
| "eval_samples_per_second": 501.315, | |
| "eval_steps_per_second": 16.097, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "eval_accuracy": 0.5091743119266054, | |
| "eval_loss": 2.6292548179626465, | |
| "eval_runtime": 1.5911, | |
| "eval_samples_per_second": 548.047, | |
| "eval_steps_per_second": 17.598, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.022874109263657957, | |
| "loss": 1.821, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "eval_accuracy": 0.4908256880733945, | |
| "eval_loss": 1.309380054473877, | |
| "eval_runtime": 1.4381, | |
| "eval_samples_per_second": 606.369, | |
| "eval_steps_per_second": 19.471, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "eval_accuracy": 0.4908256880733945, | |
| "eval_loss": 1.0032501220703125, | |
| "eval_runtime": 1.3929, | |
| "eval_samples_per_second": 626.051, | |
| "eval_steps_per_second": 20.103, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 0.02049881235154394, | |
| "loss": 1.6669, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_accuracy": 0.5091743119266054, | |
| "eval_loss": 2.3615148067474365, | |
| "eval_runtime": 1.4549, | |
| "eval_samples_per_second": 599.336, | |
| "eval_steps_per_second": 19.245, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "eval_accuracy": 0.4908256880733945, | |
| "eval_loss": 1.1986268758773804, | |
| "eval_runtime": 1.518, | |
| "eval_samples_per_second": 574.444, | |
| "eval_steps_per_second": 18.445, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "eval_accuracy": 0.4908256880733945, | |
| "eval_loss": 0.9300673007965088, | |
| "eval_runtime": 1.4024, | |
| "eval_samples_per_second": 621.772, | |
| "eval_steps_per_second": 19.965, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 0.01812351543942993, | |
| "loss": 1.5726, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "eval_accuracy": 0.4908256880733945, | |
| "eval_loss": 1.1378525495529175, | |
| "eval_runtime": 1.3734, | |
| "eval_samples_per_second": 634.927, | |
| "eval_steps_per_second": 20.388, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "eval_accuracy": 0.4908256880733945, | |
| "eval_loss": 0.8935280442237854, | |
| "eval_runtime": 1.4758, | |
| "eval_samples_per_second": 590.85, | |
| "eval_steps_per_second": 18.972, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 0.015748218527315914, | |
| "loss": 2.2264, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "eval_accuracy": 0.4908256880733945, | |
| "eval_loss": 0.8003481030464172, | |
| "eval_runtime": 1.4567, | |
| "eval_samples_per_second": 598.601, | |
| "eval_steps_per_second": 19.221, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "eval_accuracy": 0.5091743119266054, | |
| "eval_loss": 1.1951367855072021, | |
| "eval_runtime": 1.2323, | |
| "eval_samples_per_second": 707.612, | |
| "eval_steps_per_second": 22.721, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "eval_accuracy": 0.5091743119266054, | |
| "eval_loss": 1.1763222217559814, | |
| "eval_runtime": 1.3939, | |
| "eval_samples_per_second": 625.563, | |
| "eval_steps_per_second": 20.087, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 0.0133729216152019, | |
| "loss": 1.5653, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "eval_accuracy": 0.5091743119266054, | |
| "eval_loss": 1.1043964624404907, | |
| "eval_runtime": 1.3672, | |
| "eval_samples_per_second": 637.781, | |
| "eval_steps_per_second": 20.479, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "eval_accuracy": 0.4908256880733945, | |
| "eval_loss": 2.281111717224121, | |
| "eval_runtime": 1.6251, | |
| "eval_samples_per_second": 536.597, | |
| "eval_steps_per_second": 17.23, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 0.010997624703087886, | |
| "loss": 1.3918, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "eval_accuracy": 0.4908256880733945, | |
| "eval_loss": 1.7353538274765015, | |
| "eval_runtime": 1.3011, | |
| "eval_samples_per_second": 670.193, | |
| "eval_steps_per_second": 21.52, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5091743119266054, | |
| "eval_loss": 2.1966099739074707, | |
| "eval_runtime": 1.3231, | |
| "eval_samples_per_second": 659.044, | |
| "eval_steps_per_second": 21.162, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "eval_accuracy": 0.5091743119266054, | |
| "eval_loss": 0.9623039364814758, | |
| "eval_runtime": 1.428, | |
| "eval_samples_per_second": 610.649, | |
| "eval_steps_per_second": 19.608, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 0.008622327790973872, | |
| "loss": 1.3465, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "eval_accuracy": 0.5091743119266054, | |
| "eval_loss": 0.7837328314781189, | |
| "eval_runtime": 1.6722, | |
| "eval_samples_per_second": 521.453, | |
| "eval_steps_per_second": 16.744, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "eval_accuracy": 0.4908256880733945, | |
| "eval_loss": 0.6955246329307556, | |
| "eval_runtime": 1.5518, | |
| "eval_samples_per_second": 561.939, | |
| "eval_steps_per_second": 18.044, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 0.006247030878859857, | |
| "loss": 1.0154, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "eval_accuracy": 0.5091743119266054, | |
| "eval_loss": 1.5205951929092407, | |
| "eval_runtime": 1.4264, | |
| "eval_samples_per_second": 611.325, | |
| "eval_steps_per_second": 19.63, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "eval_accuracy": 0.4908256880733945, | |
| "eval_loss": 1.8486016988754272, | |
| "eval_runtime": 1.5808, | |
| "eval_samples_per_second": 551.633, | |
| "eval_steps_per_second": 17.713, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "eval_accuracy": 0.5091743119266054, | |
| "eval_loss": 0.806239128112793, | |
| "eval_runtime": 1.0228, | |
| "eval_samples_per_second": 852.573, | |
| "eval_steps_per_second": 27.376, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 0.003871733966745843, | |
| "loss": 0.892, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "eval_accuracy": 0.4908256880733945, | |
| "eval_loss": 0.7133888602256775, | |
| "eval_runtime": 1.4851, | |
| "eval_samples_per_second": 587.161, | |
| "eval_steps_per_second": 18.854, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "eval_accuracy": 0.5091743119266054, | |
| "eval_loss": 0.8787677884101868, | |
| "eval_runtime": 1.0808, | |
| "eval_samples_per_second": 806.785, | |
| "eval_steps_per_second": 25.906, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 0.001496437054631829, | |
| "loss": 0.784, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "eval_accuracy": 0.4908256880733945, | |
| "eval_loss": 0.8426281809806824, | |
| "eval_runtime": 1.6584, | |
| "eval_samples_per_second": 525.823, | |
| "eval_steps_per_second": 16.884, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "eval_accuracy": 0.5091743119266054, | |
| "eval_loss": 0.8049508333206177, | |
| "eval_runtime": 1.4457, | |
| "eval_samples_per_second": 603.185, | |
| "eval_steps_per_second": 19.368, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 6315, | |
| "total_flos": 1.330211706755328e+16, | |
| "train_loss": 1.4194242838348365, | |
| "train_runtime": 1125.0881, | |
| "train_samples_per_second": 179.583, | |
| "train_steps_per_second": 5.613 | |
| } | |
| ], | |
| "max_steps": 6315, | |
| "num_train_epochs": 3, | |
| "total_flos": 1.330211706755328e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |