| { | |
| "best_metric": 0.7712177121771218, | |
| "best_model_checkpoint": "/srv/scratch2/dbielik/.cache/huggingface/checkpoints/final/allenai/scibert_scivocab_cased-zo_up/checkpoint-432", | |
| "epoch": 16.0, | |
| "eval_steps": 500, | |
| "global_step": 432, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 9.934002876281738, | |
| "learning_rate": 9.5e-06, | |
| "loss": 2.8387, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.1033210332103321, | |
| "eval_f1": 0.07709920253414887, | |
| "eval_loss": 2.758859395980835, | |
| "eval_precision": 0.10450163475993733, | |
| "eval_recall": 0.1033210332103321, | |
| "eval_runtime": 1.3397, | |
| "eval_samples_per_second": 202.279, | |
| "eval_steps_per_second": 8.957, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 8.466582298278809, | |
| "learning_rate": 9e-06, | |
| "loss": 2.6687, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.25830258302583026, | |
| "eval_f1": 0.2284463868863592, | |
| "eval_loss": 2.6112310886383057, | |
| "eval_precision": 0.28493199473562614, | |
| "eval_recall": 0.25830258302583026, | |
| "eval_runtime": 1.3145, | |
| "eval_samples_per_second": 206.155, | |
| "eval_steps_per_second": 9.129, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 11.356733322143555, | |
| "learning_rate": 8.5e-06, | |
| "loss": 2.4398, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.4354243542435424, | |
| "eval_f1": 0.381253286711997, | |
| "eval_loss": 2.360955238342285, | |
| "eval_precision": 0.40620812098283454, | |
| "eval_recall": 0.4354243542435424, | |
| "eval_runtime": 1.2981, | |
| "eval_samples_per_second": 208.77, | |
| "eval_steps_per_second": 9.244, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 13.148231506347656, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 2.1223, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5756457564575646, | |
| "eval_f1": 0.5459700567948665, | |
| "eval_loss": 2.0634419918060303, | |
| "eval_precision": 0.5859418937087851, | |
| "eval_recall": 0.5756457564575646, | |
| "eval_runtime": 1.2527, | |
| "eval_samples_per_second": 216.329, | |
| "eval_steps_per_second": 9.579, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 10.06527042388916, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 1.8247, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6531365313653137, | |
| "eval_f1": 0.6254590334029755, | |
| "eval_loss": 1.8168519735336304, | |
| "eval_precision": 0.6682549359951895, | |
| "eval_recall": 0.6531365313653137, | |
| "eval_runtime": 1.3328, | |
| "eval_samples_per_second": 203.332, | |
| "eval_steps_per_second": 9.004, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 14.93287467956543, | |
| "learning_rate": 7e-06, | |
| "loss": 1.5665, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7084870848708487, | |
| "eval_f1": 0.6909028725740709, | |
| "eval_loss": 1.590735912322998, | |
| "eval_precision": 0.6983839902665424, | |
| "eval_recall": 0.7084870848708487, | |
| "eval_runtime": 1.5285, | |
| "eval_samples_per_second": 177.303, | |
| "eval_steps_per_second": 7.851, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 13.733495712280273, | |
| "learning_rate": 6.5000000000000004e-06, | |
| "loss": 1.3405, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7269372693726938, | |
| "eval_f1": 0.7118111768214795, | |
| "eval_loss": 1.422202229499817, | |
| "eval_precision": 0.7273552449363285, | |
| "eval_recall": 0.7269372693726938, | |
| "eval_runtime": 1.3212, | |
| "eval_samples_per_second": 205.115, | |
| "eval_steps_per_second": 9.083, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 12.636699676513672, | |
| "learning_rate": 6e-06, | |
| "loss": 1.1401, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7527675276752768, | |
| "eval_f1": 0.7446176669003964, | |
| "eval_loss": 1.2776741981506348, | |
| "eval_precision": 0.7541910880891607, | |
| "eval_recall": 0.7527675276752768, | |
| "eval_runtime": 1.5275, | |
| "eval_samples_per_second": 177.414, | |
| "eval_steps_per_second": 7.856, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 12.018023490905762, | |
| "learning_rate": 5.500000000000001e-06, | |
| "loss": 0.9726, | |
| "step": 243 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.7380073800738007, | |
| "eval_f1": 0.7283869268501074, | |
| "eval_loss": 1.191291332244873, | |
| "eval_precision": 0.7393310690625704, | |
| "eval_recall": 0.7380073800738007, | |
| "eval_runtime": 1.3288, | |
| "eval_samples_per_second": 203.948, | |
| "eval_steps_per_second": 9.031, | |
| "step": 243 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 12.431900024414062, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8393, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.7564575645756457, | |
| "eval_f1": 0.7476978666822599, | |
| "eval_loss": 1.1135777235031128, | |
| "eval_precision": 0.7561729602587051, | |
| "eval_recall": 0.7564575645756457, | |
| "eval_runtime": 1.3261, | |
| "eval_samples_per_second": 204.36, | |
| "eval_steps_per_second": 9.049, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 14.116780281066895, | |
| "learning_rate": 4.5e-06, | |
| "loss": 0.7209, | |
| "step": 297 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.7527675276752768, | |
| "eval_f1": 0.7446247192959354, | |
| "eval_loss": 1.053971529006958, | |
| "eval_precision": 0.7499033596795298, | |
| "eval_recall": 0.7527675276752768, | |
| "eval_runtime": 1.3614, | |
| "eval_samples_per_second": 199.067, | |
| "eval_steps_per_second": 8.815, | |
| "step": 297 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 4.1638922691345215, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.6125, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.7601476014760148, | |
| "eval_f1": 0.752596751208132, | |
| "eval_loss": 1.0184677839279175, | |
| "eval_precision": 0.7563512209074453, | |
| "eval_recall": 0.7601476014760148, | |
| "eval_runtime": 1.329, | |
| "eval_samples_per_second": 203.918, | |
| "eval_steps_per_second": 9.03, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 14.60648250579834, | |
| "learning_rate": 3.5e-06, | |
| "loss": 0.5704, | |
| "step": 351 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.7564575645756457, | |
| "eval_f1": 0.7496637461027665, | |
| "eval_loss": 0.995012104511261, | |
| "eval_precision": 0.7548258823457891, | |
| "eval_recall": 0.7564575645756457, | |
| "eval_runtime": 1.3367, | |
| "eval_samples_per_second": 202.734, | |
| "eval_steps_per_second": 8.977, | |
| "step": 351 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 3.8977322578430176, | |
| "learning_rate": 3e-06, | |
| "loss": 0.4939, | |
| "step": 378 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.7675276752767528, | |
| "eval_f1": 0.7618750835491949, | |
| "eval_loss": 0.9737688302993774, | |
| "eval_precision": 0.7664142901616198, | |
| "eval_recall": 0.7675276752767528, | |
| "eval_runtime": 1.3436, | |
| "eval_samples_per_second": 201.691, | |
| "eval_steps_per_second": 8.931, | |
| "step": 378 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 15.754039764404297, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.467, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.7601476014760148, | |
| "eval_f1": 0.7550647299408412, | |
| "eval_loss": 0.9670635461807251, | |
| "eval_precision": 0.7608612393303595, | |
| "eval_recall": 0.7601476014760148, | |
| "eval_runtime": 1.3334, | |
| "eval_samples_per_second": 203.247, | |
| "eval_steps_per_second": 9.0, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 4.923061847686768, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 0.4176, | |
| "step": 432 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.7712177121771218, | |
| "eval_f1": 0.767099786720493, | |
| "eval_loss": 0.9529668688774109, | |
| "eval_precision": 0.7719573695224307, | |
| "eval_recall": 0.7712177121771218, | |
| "eval_runtime": 1.5294, | |
| "eval_samples_per_second": 177.199, | |
| "eval_steps_per_second": 7.846, | |
| "step": 432 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 540, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2652516628070400.0, | |
| "train_batch_size": 24, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |