| { | |
| "best_metric": 0.8127490039840638, | |
| "best_model_checkpoint": "distilbert-base-multilingual-cased-hyper-matt/run-xcydzmg4/checkpoint-300", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 300, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 3.299391984939575, | |
| "learning_rate": 3.163832728034676e-05, | |
| "loss": 0.6312, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 2.2213969230651855, | |
| "learning_rate": 3.0547350477576185e-05, | |
| "loss": 0.5159, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 1.8511892557144165, | |
| "learning_rate": 2.9456373674805607e-05, | |
| "loss": 0.5112, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 3.693004608154297, | |
| "learning_rate": 2.836539687203503e-05, | |
| "loss": 0.3914, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 6.287681579589844, | |
| "learning_rate": 2.727442006926445e-05, | |
| "loss": 0.3439, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 3.235572576522827, | |
| "learning_rate": 2.6183443266493874e-05, | |
| "loss": 0.4263, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 9.508350372314453, | |
| "learning_rate": 2.5092466463723296e-05, | |
| "loss": 0.3324, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 3.4181947708129883, | |
| "learning_rate": 2.4001489660952715e-05, | |
| "loss": 0.4191, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 3.315913200378418, | |
| "learning_rate": 2.2910512858182137e-05, | |
| "loss": 0.3647, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 3.4039485454559326, | |
| "learning_rate": 2.181953605541156e-05, | |
| "loss": 0.2834, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.87, | |
| "eval_f1": 0.792, | |
| "eval_loss": 0.31944602727890015, | |
| "eval_precision": 0.7795275590551181, | |
| "eval_recall": 0.8048780487804879, | |
| "eval_runtime": 1.5401, | |
| "eval_samples_per_second": 259.717, | |
| "eval_steps_per_second": 16.232, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 8.336113929748535, | |
| "learning_rate": 2.072855925264098e-05, | |
| "loss": 0.2983, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 9.28201675415039, | |
| "learning_rate": 1.9637582449870403e-05, | |
| "loss": 0.3149, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 6.901111125946045, | |
| "learning_rate": 1.8546605647099826e-05, | |
| "loss": 0.3158, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 1.6393898725509644, | |
| "learning_rate": 1.7455628844329248e-05, | |
| "loss": 0.2477, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 3.3284659385681152, | |
| "learning_rate": 1.636465204155867e-05, | |
| "loss": 0.1975, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 12.852226257324219, | |
| "learning_rate": 1.5273675238788092e-05, | |
| "loss": 0.2251, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "grad_norm": 10.670600891113281, | |
| "learning_rate": 1.4182698436017515e-05, | |
| "loss": 0.263, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 2.7927801609039307, | |
| "learning_rate": 1.3091721633246937e-05, | |
| "loss": 0.168, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 1.2662396430969238, | |
| "learning_rate": 1.2000744830476357e-05, | |
| "loss": 0.2702, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 7.965029239654541, | |
| "learning_rate": 1.090976802770578e-05, | |
| "loss": 0.2884, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.87, | |
| "eval_f1": 0.7815126050420168, | |
| "eval_loss": 0.3038490414619446, | |
| "eval_precision": 0.808695652173913, | |
| "eval_recall": 0.7560975609756098, | |
| "eval_runtime": 1.5371, | |
| "eval_samples_per_second": 260.227, | |
| "eval_steps_per_second": 16.264, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "grad_norm": 6.486904144287109, | |
| "learning_rate": 9.818791224935202e-06, | |
| "loss": 0.1707, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 1.786522626876831, | |
| "learning_rate": 8.727814422164624e-06, | |
| "loss": 0.2129, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "grad_norm": 8.180079460144043, | |
| "learning_rate": 7.636837619394046e-06, | |
| "loss": 0.1224, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 9.521204948425293, | |
| "learning_rate": 6.545860816623468e-06, | |
| "loss": 0.1735, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 4.741390705108643, | |
| "learning_rate": 5.45488401385289e-06, | |
| "loss": 0.1549, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "grad_norm": 2.954021453857422, | |
| "learning_rate": 4.363907211082312e-06, | |
| "loss": 0.1712, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "grad_norm": 14.713294982910156, | |
| "learning_rate": 3.272930408311734e-06, | |
| "loss": 0.2493, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 0.9680449366569519, | |
| "learning_rate": 2.181953605541156e-06, | |
| "loss": 0.1804, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "grad_norm": 14.325002670288086, | |
| "learning_rate": 1.090976802770578e-06, | |
| "loss": 0.2148, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.2194037437438965, | |
| "learning_rate": 0.0, | |
| "loss": 0.1031, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8825, | |
| "eval_f1": 0.8127490039840638, | |
| "eval_loss": 0.3291611075401306, | |
| "eval_precision": 0.796875, | |
| "eval_recall": 0.8292682926829268, | |
| "eval_runtime": 1.5817, | |
| "eval_samples_per_second": 252.886, | |
| "eval_steps_per_second": 15.805, | |
| "step": 300 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 300, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 635446111352832.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": { | |
| "_wandb": {}, | |
| "assignments": {}, | |
| "learning_rate": 3.272930408311734e-05, | |
| "metric": "eval/loss", | |
| "num_train_epochs": 3, | |
| "per_device_train_batch_size": 16, | |
| "seed": 40 | |
| } | |
| } | |