| { | |
| "best_metric": 0.7874015748031497, | |
| "best_model_checkpoint": "distilbert-base-multilingual-cased-hyper-matt/run-y728az4t/checkpoint-800", | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 800, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.025, | |
| "grad_norm": 4.511281490325928, | |
| "learning_rate": 5.742539561589771e-05, | |
| "loss": 0.5537, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 21.07503890991211, | |
| "learning_rate": 5.713682578868214e-05, | |
| "loss": 0.635, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.075, | |
| "grad_norm": 6.964919567108154, | |
| "learning_rate": 5.6848255961466576e-05, | |
| "loss": 0.6774, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 1.1289658546447754, | |
| "learning_rate": 5.6559686134251004e-05, | |
| "loss": 0.5411, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.125, | |
| "grad_norm": 7.270464897155762, | |
| "learning_rate": 5.627111630703544e-05, | |
| "loss": 0.6009, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 16.581676483154297, | |
| "learning_rate": 5.598254647981987e-05, | |
| "loss": 0.5943, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.175, | |
| "grad_norm": 23.3575382232666, | |
| "learning_rate": 5.569397665260431e-05, | |
| "loss": 0.647, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 10.81038761138916, | |
| "learning_rate": 5.540540682538874e-05, | |
| "loss": 0.4295, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.225, | |
| "grad_norm": 19.09532928466797, | |
| "learning_rate": 5.5116836998173175e-05, | |
| "loss": 0.6079, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 19.97694206237793, | |
| "learning_rate": 5.482826717095761e-05, | |
| "loss": 0.6182, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.275, | |
| "grad_norm": 19.958236694335938, | |
| "learning_rate": 5.453969734374204e-05, | |
| "loss": 0.5946, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 6.271239757537842, | |
| "learning_rate": 5.425112751652647e-05, | |
| "loss": 0.26, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.325, | |
| "grad_norm": 3.8671107292175293, | |
| "learning_rate": 5.396255768931091e-05, | |
| "loss": 0.346, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 4.9892425537109375, | |
| "learning_rate": 5.367398786209535e-05, | |
| "loss": 0.3437, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.375, | |
| "grad_norm": 0.6231233477592468, | |
| "learning_rate": 5.338541803487978e-05, | |
| "loss": 0.2564, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 7.22336483001709, | |
| "learning_rate": 5.3096848207664216e-05, | |
| "loss": 0.8952, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.425, | |
| "grad_norm": 10.588624000549316, | |
| "learning_rate": 5.280827838044865e-05, | |
| "loss": 0.4772, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 1.097969651222229, | |
| "learning_rate": 5.251970855323308e-05, | |
| "loss": 0.3356, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.475, | |
| "grad_norm": 19.84050941467285, | |
| "learning_rate": 5.223113872601751e-05, | |
| "loss": 0.647, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 8.249490737915039, | |
| "learning_rate": 5.1942568898801946e-05, | |
| "loss": 0.4706, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.525, | |
| "grad_norm": 2.187814712524414, | |
| "learning_rate": 5.165399907158638e-05, | |
| "loss": 0.4385, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 7.703945159912109, | |
| "learning_rate": 5.1365429244370815e-05, | |
| "loss": 0.3925, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.575, | |
| "grad_norm": 9.645771980285645, | |
| "learning_rate": 5.107685941715525e-05, | |
| "loss": 0.4914, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 5.81449556350708, | |
| "learning_rate": 5.0788289589939684e-05, | |
| "loss": 0.5944, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.625, | |
| "grad_norm": 12.817668914794922, | |
| "learning_rate": 5.049971976272411e-05, | |
| "loss": 0.4712, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 4.410495758056641, | |
| "learning_rate": 5.0211149935508546e-05, | |
| "loss": 0.451, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.675, | |
| "grad_norm": 2.6431968212127686, | |
| "learning_rate": 4.992258010829298e-05, | |
| "loss": 0.1548, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 0.6477805972099304, | |
| "learning_rate": 4.9634010281077414e-05, | |
| "loss": 0.5608, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.725, | |
| "grad_norm": 5.920861721038818, | |
| "learning_rate": 4.934544045386185e-05, | |
| "loss": 0.4278, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 5.376490116119385, | |
| "learning_rate": 4.905687062664628e-05, | |
| "loss": 0.2859, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.775, | |
| "grad_norm": 17.03923988342285, | |
| "learning_rate": 4.876830079943072e-05, | |
| "loss": 0.2231, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 15.971346855163574, | |
| "learning_rate": 4.8479730972215145e-05, | |
| "loss": 0.5975, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.825, | |
| "grad_norm": 72.75434112548828, | |
| "learning_rate": 4.819116114499958e-05, | |
| "loss": 0.652, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 41.155269622802734, | |
| "learning_rate": 4.7902591317784014e-05, | |
| "loss": 0.6102, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.875, | |
| "grad_norm": 17.774869918823242, | |
| "learning_rate": 4.761402149056845e-05, | |
| "loss": 0.4959, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 8.649765014648438, | |
| "learning_rate": 4.732545166335288e-05, | |
| "loss": 0.7816, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.925, | |
| "grad_norm": 6.404572486877441, | |
| "learning_rate": 4.703688183613732e-05, | |
| "loss": 0.4796, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 13.620793342590332, | |
| "learning_rate": 4.674831200892175e-05, | |
| "loss": 0.3894, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.975, | |
| "grad_norm": 10.787906646728516, | |
| "learning_rate": 4.6459742181706185e-05, | |
| "loss": 0.2826, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.09124770760536194, | |
| "learning_rate": 4.617117235449062e-05, | |
| "loss": 0.3121, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8175, | |
| "eval_f1": 0.7345454545454545, | |
| "eval_loss": 0.5034022331237793, | |
| "eval_precision": 0.6644736842105263, | |
| "eval_recall": 0.8211382113821138, | |
| "eval_runtime": 1.5181, | |
| "eval_samples_per_second": 263.484, | |
| "eval_steps_per_second": 16.468, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.025, | |
| "grad_norm": 8.29990291595459, | |
| "learning_rate": 4.5882602527275054e-05, | |
| "loss": 0.4983, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 1.2915505170822144, | |
| "learning_rate": 4.559403270005949e-05, | |
| "loss": 0.3114, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.075, | |
| "grad_norm": 36.152198791503906, | |
| "learning_rate": 4.530546287284392e-05, | |
| "loss": 0.5468, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 17.699785232543945, | |
| "learning_rate": 4.501689304562836e-05, | |
| "loss": 0.3794, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.125, | |
| "grad_norm": 0.3559480607509613, | |
| "learning_rate": 4.4728323218412785e-05, | |
| "loss": 0.4284, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 6.532162666320801, | |
| "learning_rate": 4.443975339119722e-05, | |
| "loss": 0.4065, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.175, | |
| "grad_norm": 0.3651299774646759, | |
| "learning_rate": 4.4151183563981653e-05, | |
| "loss": 0.4627, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 0.09214647114276886, | |
| "learning_rate": 4.386261373676609e-05, | |
| "loss": 0.2161, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.225, | |
| "grad_norm": 35.60824203491211, | |
| "learning_rate": 4.357404390955052e-05, | |
| "loss": 0.4197, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 0.518726110458374, | |
| "learning_rate": 4.3285474082334956e-05, | |
| "loss": 0.3998, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.275, | |
| "grad_norm": 0.9324079155921936, | |
| "learning_rate": 4.299690425511939e-05, | |
| "loss": 0.5359, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 0.24253502488136292, | |
| "learning_rate": 4.270833442790382e-05, | |
| "loss": 0.2544, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.325, | |
| "grad_norm": 58.45193099975586, | |
| "learning_rate": 4.241976460068825e-05, | |
| "loss": 0.3068, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "grad_norm": 0.18596920371055603, | |
| "learning_rate": 4.213119477347269e-05, | |
| "loss": 0.1352, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.375, | |
| "grad_norm": 0.09809605032205582, | |
| "learning_rate": 4.184262494625712e-05, | |
| "loss": 0.2002, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 99.6656723022461, | |
| "learning_rate": 4.1554055119041556e-05, | |
| "loss": 0.5991, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.425, | |
| "grad_norm": 0.20418468117713928, | |
| "learning_rate": 4.126548529182599e-05, | |
| "loss": 0.1033, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "grad_norm": 6.0724196434021, | |
| "learning_rate": 4.0976915464610424e-05, | |
| "loss": 0.2963, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.475, | |
| "grad_norm": 32.76648712158203, | |
| "learning_rate": 4.068834563739485e-05, | |
| "loss": 0.5506, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 6.148083686828613, | |
| "learning_rate": 4.0399775810179286e-05, | |
| "loss": 0.4567, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.525, | |
| "grad_norm": 0.12815533578395844, | |
| "learning_rate": 4.011120598296372e-05, | |
| "loss": 0.0237, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "grad_norm": 0.6783960461616516, | |
| "learning_rate": 3.9822636155748155e-05, | |
| "loss": 0.5228, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.575, | |
| "grad_norm": 0.062089353799819946, | |
| "learning_rate": 3.9534066328532596e-05, | |
| "loss": 0.2315, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 4.188582420349121, | |
| "learning_rate": 3.924549650131703e-05, | |
| "loss": 0.0173, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.625, | |
| "grad_norm": 0.3549104928970337, | |
| "learning_rate": 3.8956926674101465e-05, | |
| "loss": 0.3312, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "grad_norm": 0.1310347616672516, | |
| "learning_rate": 3.866835684688589e-05, | |
| "loss": 0.1958, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.675, | |
| "grad_norm": 19.055049896240234, | |
| "learning_rate": 3.837978701967033e-05, | |
| "loss": 0.5, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "grad_norm": 0.1129627674818039, | |
| "learning_rate": 3.809121719245476e-05, | |
| "loss": 0.2486, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.725, | |
| "grad_norm": 0.21243779361248016, | |
| "learning_rate": 3.7802647365239195e-05, | |
| "loss": 0.0712, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 0.06997116655111313, | |
| "learning_rate": 3.751407753802363e-05, | |
| "loss": 0.48, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.775, | |
| "grad_norm": 0.12350662797689438, | |
| "learning_rate": 3.7225507710808064e-05, | |
| "loss": 0.1861, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 9.771814346313477, | |
| "learning_rate": 3.69369378835925e-05, | |
| "loss": 0.5302, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.825, | |
| "grad_norm": 1.4791642427444458, | |
| "learning_rate": 3.6648368056376926e-05, | |
| "loss": 0.4248, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "grad_norm": 9.070107460021973, | |
| "learning_rate": 3.635979822916136e-05, | |
| "loss": 0.4693, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 1.875, | |
| "grad_norm": 25.12293815612793, | |
| "learning_rate": 3.6071228401945795e-05, | |
| "loss": 0.3694, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 0.19772209227085114, | |
| "learning_rate": 3.578265857473023e-05, | |
| "loss": 0.3738, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 1.925, | |
| "grad_norm": 18.277198791503906, | |
| "learning_rate": 3.5494088747514663e-05, | |
| "loss": 0.3531, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "grad_norm": 0.32979539036750793, | |
| "learning_rate": 3.52055189202991e-05, | |
| "loss": 0.0431, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 1.975, | |
| "grad_norm": 0.03908811882138252, | |
| "learning_rate": 3.491694909308353e-05, | |
| "loss": 0.4406, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.19753161072731018, | |
| "learning_rate": 3.462837926586796e-05, | |
| "loss": 0.2476, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.865, | |
| "eval_f1": 0.7874015748031497, | |
| "eval_loss": 0.5722174644470215, | |
| "eval_precision": 0.7633587786259542, | |
| "eval_recall": 0.8130081300813008, | |
| "eval_runtime": 1.5153, | |
| "eval_samples_per_second": 263.971, | |
| "eval_steps_per_second": 16.498, | |
| "step": 800 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 2000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "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": 423630740901888.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": { | |
| "_wandb": {}, | |
| "assignments": {}, | |
| "learning_rate": 5.771396544311327e-05, | |
| "metric": "eval/loss", | |
| "num_train_epochs": 5, | |
| "per_device_train_batch_size": 4, | |
| "seed": 4 | |
| } | |
| } | |