| { | |
| "best_metric": 0.7951807228915663, | |
| "best_model_checkpoint": "distilbert-base-multilingual-cased-hyper-matt/run-jvfupcpb/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": 2.15091609954834, | |
| "learning_rate": 4.3626180253789615e-05, | |
| "loss": 0.5952, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 2.9856648445129395, | |
| "learning_rate": 4.288675346982709e-05, | |
| "loss": 0.5841, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 1.2374624013900757, | |
| "learning_rate": 4.2147326685864546e-05, | |
| "loss": 0.4749, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 2.3952836990356445, | |
| "learning_rate": 4.140789990190201e-05, | |
| "loss": 0.5499, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 6.6773223876953125, | |
| "learning_rate": 4.0668473117939477e-05, | |
| "loss": 0.4286, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 3.308351993560791, | |
| "learning_rate": 3.992904633397694e-05, | |
| "loss": 0.5564, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 5.943718910217285, | |
| "learning_rate": 3.918961955001441e-05, | |
| "loss": 0.6481, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 2.2899556159973145, | |
| "learning_rate": 3.845019276605187e-05, | |
| "loss": 0.3176, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 8.657326698303223, | |
| "learning_rate": 3.771076598208933e-05, | |
| "loss": 0.332, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 3.0114917755126953, | |
| "learning_rate": 3.6971339198126796e-05, | |
| "loss": 0.5015, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 1.8786035776138306, | |
| "learning_rate": 3.623191241416426e-05, | |
| "loss": 0.4642, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 1.8745856285095215, | |
| "learning_rate": 3.549248563020173e-05, | |
| "loss": 0.4042, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 7.431813716888428, | |
| "learning_rate": 3.475305884623919e-05, | |
| "loss": 0.278, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 2.6287291049957275, | |
| "learning_rate": 3.401363206227666e-05, | |
| "loss": 0.3648, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 8.826187133789062, | |
| "learning_rate": 3.3274205278314116e-05, | |
| "loss": 0.4816, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 3.6470208168029785, | |
| "learning_rate": 3.253477849435158e-05, | |
| "loss": 0.4655, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 3.792525053024292, | |
| "learning_rate": 3.1795351710389047e-05, | |
| "loss": 0.391, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 2.4055352210998535, | |
| "learning_rate": 3.1055924926426505e-05, | |
| "loss": 0.3063, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 2.7378880977630615, | |
| "learning_rate": 3.0316498142463974e-05, | |
| "loss": 0.4052, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 5.600724697113037, | |
| "learning_rate": 2.9577071358501436e-05, | |
| "loss": 0.2775, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8675, | |
| "eval_f1": 0.7854251012145749, | |
| "eval_loss": 0.3443775475025177, | |
| "eval_precision": 0.782258064516129, | |
| "eval_recall": 0.7886178861788617, | |
| "eval_runtime": 1.5164, | |
| "eval_samples_per_second": 263.79, | |
| "eval_steps_per_second": 16.487, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 0.5725231766700745, | |
| "learning_rate": 2.8837644574538904e-05, | |
| "loss": 0.2948, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 9.40677547454834, | |
| "learning_rate": 2.8098217790576366e-05, | |
| "loss": 0.2623, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 0.5530999898910522, | |
| "learning_rate": 2.735879100661383e-05, | |
| "loss": 0.2702, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 15.44442081451416, | |
| "learning_rate": 2.6619364222651293e-05, | |
| "loss": 0.5376, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 0.7881362438201904, | |
| "learning_rate": 2.587993743868876e-05, | |
| "loss": 0.3554, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 4.4421067237854, | |
| "learning_rate": 2.514051065472622e-05, | |
| "loss": 0.1891, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "grad_norm": 4.921718120574951, | |
| "learning_rate": 2.440108387076369e-05, | |
| "loss": 0.2338, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 3.178603172302246, | |
| "learning_rate": 2.366165708680115e-05, | |
| "loss": 0.3782, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "grad_norm": 0.419980525970459, | |
| "learning_rate": 2.2922230302838617e-05, | |
| "loss": 0.1908, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 8.025728225708008, | |
| "learning_rate": 2.218280351887608e-05, | |
| "loss": 0.2183, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "grad_norm": 0.9326677918434143, | |
| "learning_rate": 2.1443376734913544e-05, | |
| "loss": 0.3055, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 5.646274566650391, | |
| "learning_rate": 2.0703949950951006e-05, | |
| "loss": 0.2639, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "grad_norm": 14.51612663269043, | |
| "learning_rate": 1.996452316698847e-05, | |
| "loss": 0.2841, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "grad_norm": 2.912120819091797, | |
| "learning_rate": 1.9225096383025936e-05, | |
| "loss": 0.1919, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 11.444270133972168, | |
| "learning_rate": 1.8485669599063398e-05, | |
| "loss": 0.1804, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 3.0050899982452393, | |
| "learning_rate": 1.7746242815100863e-05, | |
| "loss": 0.166, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "grad_norm": 24.114974975585938, | |
| "learning_rate": 1.700681603113833e-05, | |
| "loss": 0.3688, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 0.28243792057037354, | |
| "learning_rate": 1.626738924717579e-05, | |
| "loss": 0.3356, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "grad_norm": 0.6977348327636719, | |
| "learning_rate": 1.5527962463213253e-05, | |
| "loss": 0.4389, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 21.475341796875, | |
| "learning_rate": 1.4788535679250718e-05, | |
| "loss": 0.2895, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.87, | |
| "eval_f1": 0.792, | |
| "eval_loss": 0.33494508266448975, | |
| "eval_precision": 0.7795275590551181, | |
| "eval_recall": 0.8048780487804879, | |
| "eval_runtime": 1.5155, | |
| "eval_samples_per_second": 263.946, | |
| "eval_steps_per_second": 16.497, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "grad_norm": 4.863823413848877, | |
| "learning_rate": 1.4049108895288183e-05, | |
| "loss": 0.1339, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "grad_norm": 21.860219955444336, | |
| "learning_rate": 1.3309682111325647e-05, | |
| "loss": 0.2416, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "grad_norm": 0.2086387276649475, | |
| "learning_rate": 1.257025532736311e-05, | |
| "loss": 0.1266, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 0.49468812346458435, | |
| "learning_rate": 1.1830828543400576e-05, | |
| "loss": 0.2281, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "grad_norm": 4.723079204559326, | |
| "learning_rate": 1.109140175943804e-05, | |
| "loss": 0.0827, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "grad_norm": 14.231110572814941, | |
| "learning_rate": 1.0351974975475503e-05, | |
| "loss": 0.1488, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "grad_norm": 0.563101053237915, | |
| "learning_rate": 9.612548191512968e-06, | |
| "loss": 0.31, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 20.05416488647461, | |
| "learning_rate": 8.873121407550432e-06, | |
| "loss": 0.1491, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "grad_norm": 7.252449989318848, | |
| "learning_rate": 8.133694623587895e-06, | |
| "loss": 0.2171, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 1.7218003273010254, | |
| "learning_rate": 7.394267839625359e-06, | |
| "loss": 0.0633, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "grad_norm": 0.06682740151882172, | |
| "learning_rate": 6.654841055662823e-06, | |
| "loss": 0.2153, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "grad_norm": 0.25739797949790955, | |
| "learning_rate": 5.915414271700288e-06, | |
| "loss": 0.0267, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "grad_norm": 0.05432148650288582, | |
| "learning_rate": 5.175987487737751e-06, | |
| "loss": 0.1115, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "grad_norm": 3.8702685832977295, | |
| "learning_rate": 4.436560703775216e-06, | |
| "loss": 0.249, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "grad_norm": 28.443191528320312, | |
| "learning_rate": 3.6971339198126795e-06, | |
| "loss": 0.3686, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 0.1164081022143364, | |
| "learning_rate": 2.957707135850144e-06, | |
| "loss": 0.0713, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "grad_norm": 0.35371047258377075, | |
| "learning_rate": 2.218280351887608e-06, | |
| "loss": 0.1125, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "grad_norm": 17.213769912719727, | |
| "learning_rate": 1.478853567925072e-06, | |
| "loss": 0.2296, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "grad_norm": 0.3524082899093628, | |
| "learning_rate": 7.39426783962536e-07, | |
| "loss": 0.1454, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.2137354463338852, | |
| "learning_rate": 0.0, | |
| "loss": 0.0681, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8725, | |
| "eval_f1": 0.7951807228915663, | |
| "eval_loss": 0.4910232424736023, | |
| "eval_precision": 0.7857142857142857, | |
| "eval_recall": 0.8048780487804879, | |
| "eval_runtime": 1.5682, | |
| "eval_samples_per_second": 255.073, | |
| "eval_steps_per_second": 15.942, | |
| "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": 4.436560703775216e-05, | |
| "metric": "eval/loss", | |
| "num_train_epochs": 3, | |
| "per_device_train_batch_size": 8, | |
| "seed": 40 | |
| } | |
| } | |