| { | |
| "best_metric": 0.7954545454545454, | |
| "best_model_checkpoint": "distilbert-base-multilingual-cased-hyper-matt/run-vdjsiuit/checkpoint-600", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 600, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 3.0975422859191895, | |
| "learning_rate": 5.939561377324845e-05, | |
| "loss": 0.5513, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 8.386995315551758, | |
| "learning_rate": 5.83889084550578e-05, | |
| "loss": 0.5725, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 2.8792715072631836, | |
| "learning_rate": 5.738220313686715e-05, | |
| "loss": 0.5265, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 4.608365535736084, | |
| "learning_rate": 5.63754978186765e-05, | |
| "loss": 0.5106, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 3.161665201187134, | |
| "learning_rate": 5.5368792500485845e-05, | |
| "loss": 0.4356, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 2.6395349502563477, | |
| "learning_rate": 5.4362087182295196e-05, | |
| "loss": 0.5053, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 2.152228832244873, | |
| "learning_rate": 5.335538186410454e-05, | |
| "loss": 0.3511, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 22.35204315185547, | |
| "learning_rate": 5.234867654591389e-05, | |
| "loss": 0.5189, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 3.6376490592956543, | |
| "learning_rate": 5.134197122772324e-05, | |
| "loss": 0.3371, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 3.2198097705841064, | |
| "learning_rate": 5.033526590953259e-05, | |
| "loss": 0.4037, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 7.86613130569458, | |
| "learning_rate": 4.9328560591341935e-05, | |
| "loss": 0.4833, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 3.6074087619781494, | |
| "learning_rate": 4.8321855273151286e-05, | |
| "loss": 0.2857, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 6.2979230880737305, | |
| "learning_rate": 4.731514995496063e-05, | |
| "loss": 0.4716, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 8.780591011047363, | |
| "learning_rate": 4.630844463676999e-05, | |
| "loss": 0.4018, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 21.96924591064453, | |
| "learning_rate": 4.530173931857933e-05, | |
| "loss": 0.412, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 5.481058597564697, | |
| "learning_rate": 4.4295034000388675e-05, | |
| "loss": 0.5499, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 6.030953884124756, | |
| "learning_rate": 4.3288328682198025e-05, | |
| "loss": 0.4422, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 1.1474953889846802, | |
| "learning_rate": 4.228162336400737e-05, | |
| "loss": 0.2512, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 13.46764087677002, | |
| "learning_rate": 4.127491804581672e-05, | |
| "loss": 0.4896, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 6.764571666717529, | |
| "learning_rate": 4.026821272762607e-05, | |
| "loss": 0.3959, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8775, | |
| "eval_f1": 0.776255707762557, | |
| "eval_loss": 0.42286473512649536, | |
| "eval_precision": 0.8854166666666666, | |
| "eval_recall": 0.6910569105691057, | |
| "eval_runtime": 1.5336, | |
| "eval_samples_per_second": 260.823, | |
| "eval_steps_per_second": 16.301, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 13.774749755859375, | |
| "learning_rate": 3.926150740943542e-05, | |
| "loss": 0.2466, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 0.9035859704017639, | |
| "learning_rate": 3.8254802091244765e-05, | |
| "loss": 0.258, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 10.26409912109375, | |
| "learning_rate": 3.7248096773054115e-05, | |
| "loss": 0.2796, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 16.81245994567871, | |
| "learning_rate": 3.624139145486346e-05, | |
| "loss": 0.2748, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 0.9752582311630249, | |
| "learning_rate": 3.5234686136672817e-05, | |
| "loss": 0.3794, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 0.8559211492538452, | |
| "learning_rate": 3.422798081848216e-05, | |
| "loss": 0.3013, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "grad_norm": 0.20962999761104584, | |
| "learning_rate": 3.322127550029151e-05, | |
| "loss": 0.3029, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 9.075166702270508, | |
| "learning_rate": 3.2214570182100855e-05, | |
| "loss": 0.3187, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "grad_norm": 0.8023856282234192, | |
| "learning_rate": 3.1207864863910205e-05, | |
| "loss": 0.4383, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 8.567304611206055, | |
| "learning_rate": 3.0201159545719553e-05, | |
| "loss": 0.1829, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "grad_norm": 8.392529487609863, | |
| "learning_rate": 2.91944542275289e-05, | |
| "loss": 0.3928, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.7387467622756958, | |
| "learning_rate": 2.818774890933825e-05, | |
| "loss": 0.1683, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "grad_norm": 8.625945091247559, | |
| "learning_rate": 2.7181043591147598e-05, | |
| "loss": 0.2039, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "grad_norm": 0.30814623832702637, | |
| "learning_rate": 2.6174338272956945e-05, | |
| "loss": 0.1959, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 7.88859748840332, | |
| "learning_rate": 2.5167632954766296e-05, | |
| "loss": 0.4723, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 0.31379058957099915, | |
| "learning_rate": 2.4160927636575643e-05, | |
| "loss": 0.3442, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "grad_norm": 9.962308883666992, | |
| "learning_rate": 2.3154222318384993e-05, | |
| "loss": 0.4158, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 3.3801684379577637, | |
| "learning_rate": 2.2147517000194337e-05, | |
| "loss": 0.2539, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "grad_norm": 2.923280954360962, | |
| "learning_rate": 2.1140811682003685e-05, | |
| "loss": 0.1824, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 4.60498046875, | |
| "learning_rate": 2.0134106363813035e-05, | |
| "loss": 0.4522, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8725, | |
| "eval_f1": 0.7792207792207793, | |
| "eval_loss": 0.318487286567688, | |
| "eval_precision": 0.8333333333333334, | |
| "eval_recall": 0.7317073170731707, | |
| "eval_runtime": 1.5288, | |
| "eval_samples_per_second": 261.647, | |
| "eval_steps_per_second": 16.353, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "grad_norm": 0.41749486327171326, | |
| "learning_rate": 1.9127401045622382e-05, | |
| "loss": 0.1156, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "grad_norm": 0.3070080876350403, | |
| "learning_rate": 1.812069572743173e-05, | |
| "loss": 0.0349, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "grad_norm": 2.974935293197632, | |
| "learning_rate": 1.711399040924108e-05, | |
| "loss": 0.133, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 0.4423005282878876, | |
| "learning_rate": 1.6107285091050427e-05, | |
| "loss": 0.0806, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "grad_norm": 0.2689347267150879, | |
| "learning_rate": 1.5100579772859776e-05, | |
| "loss": 0.1243, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "grad_norm": 0.1341865360736847, | |
| "learning_rate": 1.4093874454669125e-05, | |
| "loss": 0.1945, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "grad_norm": 0.14216558635234833, | |
| "learning_rate": 1.3087169136478472e-05, | |
| "loss": 0.2593, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 5.250720977783203, | |
| "learning_rate": 1.2080463818287821e-05, | |
| "loss": 0.3948, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "grad_norm": 18.877355575561523, | |
| "learning_rate": 1.1073758500097169e-05, | |
| "loss": 0.2856, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 0.6348363161087036, | |
| "learning_rate": 1.0067053181906518e-05, | |
| "loss": 0.4102, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "grad_norm": 0.16438539326190948, | |
| "learning_rate": 9.060347863715865e-06, | |
| "loss": 0.2346, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "grad_norm": 0.23815393447875977, | |
| "learning_rate": 8.053642545525214e-06, | |
| "loss": 0.1046, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "grad_norm": 25.9287166595459, | |
| "learning_rate": 7.046937227334563e-06, | |
| "loss": 0.0799, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "grad_norm": 0.2302125245332718, | |
| "learning_rate": 6.040231909143911e-06, | |
| "loss": 0.1963, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "grad_norm": 0.14851589500904083, | |
| "learning_rate": 5.033526590953259e-06, | |
| "loss": 0.1668, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 13.211697578430176, | |
| "learning_rate": 4.026821272762607e-06, | |
| "loss": 0.3321, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "grad_norm": 0.20074860751628876, | |
| "learning_rate": 3.0201159545719554e-06, | |
| "loss": 0.1587, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "grad_norm": 1.0804275274276733, | |
| "learning_rate": 2.0134106363813034e-06, | |
| "loss": 0.1044, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "grad_norm": 0.2024952471256256, | |
| "learning_rate": 1.0067053181906517e-06, | |
| "loss": 0.3374, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.07060264050960541, | |
| "learning_rate": 0.0, | |
| "loss": 0.0739, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.865, | |
| "eval_f1": 0.7954545454545454, | |
| "eval_loss": 0.45124685764312744, | |
| "eval_precision": 0.7446808510638298, | |
| "eval_recall": 0.8536585365853658, | |
| "eval_runtime": 1.5828, | |
| "eval_samples_per_second": 252.719, | |
| "eval_steps_per_second": 15.795, | |
| "step": 600 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 600, | |
| "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": 8, | |
| "trial_name": null, | |
| "trial_params": { | |
| "_wandb": {}, | |
| "assignments": {}, | |
| "learning_rate": 6.0402319091439105e-05, | |
| "metric": "eval/loss", | |
| "num_train_epochs": 3, | |
| "per_device_train_batch_size": 8, | |
| "seed": 33 | |
| } | |
| } | |