| { | |
| "best_metric": 0.9746632456703015, | |
| "best_model_checkpoint": "/data3/akoudounas/speech_unlearning/models/fsc/facebook/wav2vec2-base/4/_retain/checkpoint-2000", | |
| "epoch": 2.869440459110473, | |
| "eval_steps": 400, | |
| "global_step": 2000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.14347202295552366, | |
| "grad_norm": 10.184388160705566, | |
| "learning_rate": 5e-05, | |
| "loss": 13.3354, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.28694404591104733, | |
| "grad_norm": 22.545684814453125, | |
| "learning_rate": 0.0001, | |
| "loss": 10.7429, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.430416068866571, | |
| "grad_norm": 56.65978240966797, | |
| "learning_rate": 9.444444444444444e-05, | |
| "loss": 6.6371, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5738880918220947, | |
| "grad_norm": 58.855796813964844, | |
| "learning_rate": 8.888888888888889e-05, | |
| "loss": 4.8056, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.5738880918220947, | |
| "eval_accuracy": 0.6023091725465042, | |
| "eval_f1_macro": 0.5295300525459568, | |
| "eval_loss": 1.0795540809631348, | |
| "eval_runtime": 13.9266, | |
| "eval_samples_per_second": 223.887, | |
| "eval_steps_per_second": 28.004, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.7173601147776184, | |
| "grad_norm": 19.51607322692871, | |
| "learning_rate": 8.333333333333334e-05, | |
| "loss": 3.8297, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.860832137733142, | |
| "grad_norm": 290.1480712890625, | |
| "learning_rate": 7.777777777777778e-05, | |
| "loss": 2.8656, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.0043041606886658, | |
| "grad_norm": 20.89418601989746, | |
| "learning_rate": 7.222222222222222e-05, | |
| "loss": 2.0125, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.1477761836441893, | |
| "grad_norm": 12.759949684143066, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 1.3945, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.1477761836441893, | |
| "eval_accuracy": 0.9207825529185375, | |
| "eval_f1_macro": 0.9143966950672161, | |
| "eval_loss": 0.3709124028682709, | |
| "eval_runtime": 19.3466, | |
| "eval_samples_per_second": 161.165, | |
| "eval_steps_per_second": 20.159, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.291248206599713, | |
| "grad_norm": 4.4212541580200195, | |
| "learning_rate": 6.111111111111112e-05, | |
| "loss": 1.0552, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.4347202295552366, | |
| "grad_norm": 3.2309963703155518, | |
| "learning_rate": 5.555555555555556e-05, | |
| "loss": 0.9127, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.5781922525107603, | |
| "grad_norm": 50.245872497558594, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8042, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.721664275466284, | |
| "grad_norm": 190.7835235595703, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.6361, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.721664275466284, | |
| "eval_accuracy": 0.9528543938422065, | |
| "eval_f1_macro": 0.9475574056240644, | |
| "eval_loss": 0.2340688854455948, | |
| "eval_runtime": 13.9095, | |
| "eval_samples_per_second": 224.163, | |
| "eval_steps_per_second": 28.038, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.8651362984218078, | |
| "grad_norm": 0.5275015234947205, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.6868, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.0086083213773316, | |
| "grad_norm": 0.28355035185813904, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.6716, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.152080344332855, | |
| "grad_norm": 0.17906032502651215, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.4126, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.2955523672883786, | |
| "grad_norm": 0.20960959792137146, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.3714, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.2955523672883786, | |
| "eval_accuracy": 0.967928159076331, | |
| "eval_f1_macro": 0.9641675805500564, | |
| "eval_loss": 0.16309456527233124, | |
| "eval_runtime": 13.748, | |
| "eval_samples_per_second": 226.797, | |
| "eval_steps_per_second": 28.368, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.4390243902439024, | |
| "grad_norm": 5.835017204284668, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.4885, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.582496413199426, | |
| "grad_norm": 0.15750645101070404, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.3515, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.72596843615495, | |
| "grad_norm": 29.3380126953125, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.3854, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.869440459110473, | |
| "grad_norm": 12.397686004638672, | |
| "learning_rate": 0.0, | |
| "loss": 0.3996, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.869440459110473, | |
| "eval_accuracy": 0.9746632456703015, | |
| "eval_f1_macro": 0.9700002035002927, | |
| "eval_loss": 0.13398364186286926, | |
| "eval_runtime": 13.9617, | |
| "eval_samples_per_second": 223.325, | |
| "eval_steps_per_second": 27.933, | |
| "step": 2000 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 2000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 2000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.324313120768e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |