| { | |
| "best_metric": 0.7991586011098205, | |
| "best_model_checkpoint": "results/roberta-base-2e-5-16-256/checkpoint-1000", | |
| "epoch": 5.0, | |
| "global_step": 2300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 8.695652173913044e-06, | |
| "loss": 1.3164, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_accuracy": 0.8347826086956521, | |
| "eval_f1": 0.15165876777251183, | |
| "eval_loss": 0.659010648727417, | |
| "eval_runtime": 1.8554, | |
| "eval_samples_per_second": 185.947, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 1.730434782608696e-05, | |
| "loss": 0.7929, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "eval_accuracy": 0.8956521739130435, | |
| "eval_f1": 0.3681933057098357, | |
| "eval_loss": 0.4090113639831543, | |
| "eval_runtime": 1.8693, | |
| "eval_samples_per_second": 184.559, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 1.934299516908213e-05, | |
| "loss": 0.4876, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "eval_accuracy": 0.9101449275362319, | |
| "eval_f1": 0.45359927879197653, | |
| "eval_loss": 0.2804220914840698, | |
| "eval_runtime": 1.8812, | |
| "eval_samples_per_second": 183.389, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 1.83768115942029e-05, | |
| "loss": 0.3733, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_accuracy": 0.9565217391304348, | |
| "eval_f1": 0.6091620621342052, | |
| "eval_loss": 0.20709100365638733, | |
| "eval_runtime": 1.8861, | |
| "eval_samples_per_second": 182.913, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 1.7410628019323673e-05, | |
| "loss": 0.2586, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "eval_accuracy": 0.9710144927536232, | |
| "eval_f1": 0.7547141566653762, | |
| "eval_loss": 0.14703336358070374, | |
| "eval_runtime": 1.8842, | |
| "eval_samples_per_second": 183.1, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 1.6444444444444444e-05, | |
| "loss": 0.1932, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "eval_accuracy": 0.9652173913043478, | |
| "eval_f1": 0.7599251142269877, | |
| "eval_loss": 0.14065894484519958, | |
| "eval_runtime": 1.8891, | |
| "eval_samples_per_second": 182.625, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 1.5478260869565217e-05, | |
| "loss": 0.1952, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "eval_accuracy": 0.9739130434782609, | |
| "eval_f1": 0.7965866610767565, | |
| "eval_loss": 0.15803956985473633, | |
| "eval_runtime": 1.8925, | |
| "eval_samples_per_second": 182.3, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 1.4521739130434785e-05, | |
| "loss": 0.1734, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "eval_accuracy": 0.9623188405797102, | |
| "eval_f1": 0.7857069143446852, | |
| "eval_loss": 0.1854896992444992, | |
| "eval_runtime": 1.8925, | |
| "eval_samples_per_second": 182.299, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 1.3555555555555557e-05, | |
| "loss": 0.1515, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "eval_accuracy": 0.9681159420289855, | |
| "eval_f1": 0.7569010619010618, | |
| "eval_loss": 0.1608872264623642, | |
| "eval_runtime": 1.8927, | |
| "eval_samples_per_second": 182.276, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 1.2589371980676331e-05, | |
| "loss": 0.1214, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "eval_accuracy": 0.9739130434782609, | |
| "eval_f1": 0.7991586011098205, | |
| "eval_loss": 0.14474469423294067, | |
| "eval_runtime": 1.8926, | |
| "eval_samples_per_second": 182.29, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 1.1623188405797103e-05, | |
| "loss": 0.1145, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "eval_accuracy": 0.9710144927536232, | |
| "eval_f1": 0.7922399586487883, | |
| "eval_loss": 0.14988109469413757, | |
| "eval_runtime": 1.8948, | |
| "eval_samples_per_second": 182.077, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 1.0666666666666667e-05, | |
| "loss": 0.0978, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "eval_accuracy": 0.9478260869565217, | |
| "eval_f1": 0.7102037440793926, | |
| "eval_loss": 0.21022763848304749, | |
| "eval_runtime": 1.8939, | |
| "eval_samples_per_second": 182.162, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 9.700483091787441e-06, | |
| "loss": 0.112, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "eval_accuracy": 0.9681159420289855, | |
| "eval_f1": 0.7895157299399082, | |
| "eval_loss": 0.15368108451366425, | |
| "eval_runtime": 1.8983, | |
| "eval_samples_per_second": 181.738, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 8.734299516908213e-06, | |
| "loss": 0.0879, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "eval_accuracy": 0.9681159420289855, | |
| "eval_f1": 0.7895157299399082, | |
| "eval_loss": 0.160787895321846, | |
| "eval_runtime": 1.8947, | |
| "eval_samples_per_second": 182.086, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "learning_rate": 7.768115942028987e-06, | |
| "loss": 0.0721, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "eval_accuracy": 0.9710144927536232, | |
| "eval_f1": 0.7932638888888889, | |
| "eval_loss": 0.1662783920764923, | |
| "eval_runtime": 1.895, | |
| "eval_samples_per_second": 182.054, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "learning_rate": 6.801932367149759e-06, | |
| "loss": 0.0744, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "eval_accuracy": 0.9710144927536232, | |
| "eval_f1": 0.7932638888888889, | |
| "eval_loss": 0.17374691367149353, | |
| "eval_runtime": 1.8929, | |
| "eval_samples_per_second": 182.259, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "learning_rate": 5.835748792270531e-06, | |
| "loss": 0.0614, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "eval_accuracy": 0.9710144927536232, | |
| "eval_f1": 0.7929378560367236, | |
| "eval_loss": 0.15358248353004456, | |
| "eval_runtime": 1.8971, | |
| "eval_samples_per_second": 181.86, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "learning_rate": 4.869565217391305e-06, | |
| "loss": 0.0702, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "eval_accuracy": 0.9710144927536232, | |
| "eval_f1": 0.7929378560367236, | |
| "eval_loss": 0.1646239012479782, | |
| "eval_runtime": 1.8948, | |
| "eval_samples_per_second": 182.073, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "learning_rate": 3.903381642512077e-06, | |
| "loss": 0.0536, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "eval_accuracy": 0.9739130434782609, | |
| "eval_f1": 0.7966849998231528, | |
| "eval_loss": 0.1578085571527481, | |
| "eval_runtime": 1.8964, | |
| "eval_samples_per_second": 181.922, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "learning_rate": 2.9371980676328504e-06, | |
| "loss": 0.0489, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "eval_accuracy": 0.9710144927536232, | |
| "eval_f1": 0.7932638888888889, | |
| "eval_loss": 0.16712208092212677, | |
| "eval_runtime": 1.9014, | |
| "eval_samples_per_second": 181.448, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "learning_rate": 1.9710144927536233e-06, | |
| "loss": 0.0267, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "eval_accuracy": 0.9739130434782609, | |
| "eval_f1": 0.7982255213856485, | |
| "eval_loss": 0.1600024402141571, | |
| "eval_runtime": 1.8949, | |
| "eval_samples_per_second": 182.068, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "learning_rate": 1.0048309178743963e-06, | |
| "loss": 0.0511, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "eval_accuracy": 0.9710144927536232, | |
| "eval_f1": 0.7932638888888889, | |
| "eval_loss": 0.17408017814159393, | |
| "eval_runtime": 1.8946, | |
| "eval_samples_per_second": 182.094, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 3.864734299516908e-08, | |
| "loss": 0.0291, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9710144927536232, | |
| "eval_f1": 0.7932638888888889, | |
| "eval_loss": 0.180899977684021, | |
| "eval_runtime": 1.8959, | |
| "eval_samples_per_second": 181.972, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 2300, | |
| "total_flos": 0, | |
| "train_runtime": 487.2382, | |
| "train_samples_per_second": 4.72 | |
| } | |
| ], | |
| "max_steps": 2300, | |
| "num_train_epochs": 5, | |
| "total_flos": 0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |