| { | |
| "best_metric": 0.3247086703777313, | |
| "best_model_checkpoint": "outputs/electra-nli-efl/checkpoint-1050", | |
| "epoch": 17.995555555555555, | |
| "global_step": 3780, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 9.499999999999999e-07, | |
| "loss": 0.4175, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7316666666666667, | |
| "eval_f1": 0.7305281480923569, | |
| "eval_loss": 0.4019607603549957, | |
| "eval_runtime": 11.5986, | |
| "eval_samples_per_second": 258.651, | |
| "eval_steps_per_second": 32.331, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 9e-07, | |
| "loss": 0.3061, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.768, | |
| "eval_f1": 0.7674936528439713, | |
| "eval_loss": 0.3519645631313324, | |
| "eval_runtime": 11.6033, | |
| "eval_samples_per_second": 258.548, | |
| "eval_steps_per_second": 32.319, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 8.499999999999999e-07, | |
| "loss": 0.2588, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.79, | |
| "eval_f1": 0.7887799932409598, | |
| "eval_loss": 0.32531991600990295, | |
| "eval_runtime": 11.5951, | |
| "eval_samples_per_second": 258.729, | |
| "eval_steps_per_second": 32.341, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 8e-07, | |
| "loss": 0.234, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.788, | |
| "eval_f1": 0.7877041067016072, | |
| "eval_loss": 0.33730262517929077, | |
| "eval_runtime": 11.6935, | |
| "eval_samples_per_second": 256.553, | |
| "eval_steps_per_second": 32.069, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 7.5e-07, | |
| "loss": 0.2116, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.804, | |
| "eval_f1": 0.8032759681298146, | |
| "eval_loss": 0.3247086703777313, | |
| "eval_runtime": 11.5938, | |
| "eval_samples_per_second": 258.758, | |
| "eval_steps_per_second": 32.345, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 7e-07, | |
| "loss": 0.1974, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.793, | |
| "eval_f1": 0.7928414316247181, | |
| "eval_loss": 0.34000951051712036, | |
| "eval_runtime": 11.6038, | |
| "eval_samples_per_second": 258.537, | |
| "eval_steps_per_second": 32.317, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 6.5e-07, | |
| "loss": 0.1807, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7973333333333333, | |
| "eval_f1": 0.7969281824938732, | |
| "eval_loss": 0.35112178325653076, | |
| "eval_runtime": 11.589, | |
| "eval_samples_per_second": 258.866, | |
| "eval_steps_per_second": 32.358, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 6e-07, | |
| "loss": 0.1715, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7993333333333333, | |
| "eval_f1": 0.79885693716362, | |
| "eval_loss": 0.34963640570640564, | |
| "eval_runtime": 11.6019, | |
| "eval_samples_per_second": 258.579, | |
| "eval_steps_per_second": 32.322, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 5.5e-07, | |
| "loss": 0.1577, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8043333333333333, | |
| "eval_f1": 0.803166306363764, | |
| "eval_loss": 0.3507426679134369, | |
| "eval_runtime": 11.6051, | |
| "eval_samples_per_second": 258.507, | |
| "eval_steps_per_second": 32.313, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 5e-07, | |
| "loss": 0.1469, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.798, | |
| "eval_f1": 0.7970242731280681, | |
| "eval_loss": 0.3604212701320648, | |
| "eval_runtime": 11.7325, | |
| "eval_samples_per_second": 255.699, | |
| "eval_steps_per_second": 31.962, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 4.5e-07, | |
| "loss": 0.1394, | |
| "step": 2310 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.7966666666666666, | |
| "eval_f1": 0.7957033854293423, | |
| "eval_loss": 0.37343233823776245, | |
| "eval_runtime": 11.6069, | |
| "eval_samples_per_second": 258.468, | |
| "eval_steps_per_second": 32.308, | |
| "step": 2310 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 4e-07, | |
| "loss": 0.1322, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.7913333333333333, | |
| "eval_f1": 0.7905794192426068, | |
| "eval_loss": 0.39285749197006226, | |
| "eval_runtime": 11.6585, | |
| "eval_samples_per_second": 257.322, | |
| "eval_steps_per_second": 32.165, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 3.5e-07, | |
| "loss": 0.1231, | |
| "step": 2730 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.795, | |
| "eval_f1": 0.7941480244179493, | |
| "eval_loss": 0.39544808864593506, | |
| "eval_runtime": 11.5711, | |
| "eval_samples_per_second": 259.267, | |
| "eval_steps_per_second": 32.408, | |
| "step": 2730 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 3e-07, | |
| "loss": 0.1189, | |
| "step": 2940 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.7976666666666666, | |
| "eval_f1": 0.7963303903578043, | |
| "eval_loss": 0.39935171604156494, | |
| "eval_runtime": 11.7059, | |
| "eval_samples_per_second": 256.282, | |
| "eval_steps_per_second": 32.035, | |
| "step": 2940 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 2.5e-07, | |
| "loss": 0.1143, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.7993333333333333, | |
| "eval_f1": 0.798040794417606, | |
| "eval_loss": 0.39949560165405273, | |
| "eval_runtime": 11.6007, | |
| "eval_samples_per_second": 258.605, | |
| "eval_steps_per_second": 32.326, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 2e-07, | |
| "loss": 0.1083, | |
| "step": 3360 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.7926666666666666, | |
| "eval_f1": 0.7918316804069658, | |
| "eval_loss": 0.41252416372299194, | |
| "eval_runtime": 11.6447, | |
| "eval_samples_per_second": 257.627, | |
| "eval_steps_per_second": 32.203, | |
| "step": 3360 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 1.5e-07, | |
| "loss": 0.1079, | |
| "step": 3570 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.7993333333333333, | |
| "eval_f1": 0.79793006993007, | |
| "eval_loss": 0.4035763442516327, | |
| "eval_runtime": 11.5923, | |
| "eval_samples_per_second": 258.793, | |
| "eval_steps_per_second": 32.349, | |
| "step": 3570 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 1e-07, | |
| "loss": 0.1055, | |
| "step": 3780 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.7966666666666666, | |
| "eval_f1": 0.7955672728897627, | |
| "eval_loss": 0.41211840510368347, | |
| "eval_runtime": 11.6637, | |
| "eval_samples_per_second": 257.208, | |
| "eval_steps_per_second": 32.151, | |
| "step": 3780 | |
| } | |
| ], | |
| "max_steps": 4200, | |
| "num_train_epochs": 20, | |
| "total_flos": 7.487790926232038e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |