| { | |
| "best_metric": 0.9448421901784814, | |
| "best_model_checkpoint": "rubert_toxicity_classifier/checkpoint-14777", | |
| "epoch": 2.9997969817960346, | |
| "eval_steps": 500, | |
| "global_step": 22164, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06767273465520741, | |
| "grad_norm": 19.48541831970215, | |
| "learning_rate": 2.2552999548940012e-06, | |
| "loss": 0.2951, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.13534546931041483, | |
| "grad_norm": 11.972173690795898, | |
| "learning_rate": 4.5105999097880024e-06, | |
| "loss": 0.2026, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.20301820396562226, | |
| "grad_norm": 14.64987564086914, | |
| "learning_rate": 6.765899864682003e-06, | |
| "loss": 0.1768, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.27069093862082966, | |
| "grad_norm": 17.7277774810791, | |
| "learning_rate": 9.021199819576005e-06, | |
| "loss": 0.1687, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.33836367327603706, | |
| "grad_norm": 17.497051239013672, | |
| "learning_rate": 9.858124028675992e-06, | |
| "loss": 0.169, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.4060364079312445, | |
| "grad_norm": 0.5624993443489075, | |
| "learning_rate": 9.607459768386225e-06, | |
| "loss": 0.1655, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.4737091425864519, | |
| "grad_norm": 46.436702728271484, | |
| "learning_rate": 9.356795508096457e-06, | |
| "loss": 0.1631, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.5413818772416593, | |
| "grad_norm": 0.23696815967559814, | |
| "learning_rate": 9.106131247806689e-06, | |
| "loss": 0.1504, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.6090546118968667, | |
| "grad_norm": 73.74691772460938, | |
| "learning_rate": 8.855466987516921e-06, | |
| "loss": 0.1449, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.6767273465520741, | |
| "grad_norm": 11.027336120605469, | |
| "learning_rate": 8.604802727227154e-06, | |
| "loss": 0.1405, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.7444000812072816, | |
| "grad_norm": 91.75090026855469, | |
| "learning_rate": 8.354138466937384e-06, | |
| "loss": 0.1555, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.812072815862489, | |
| "grad_norm": 37.738136291503906, | |
| "learning_rate": 8.103474206647616e-06, | |
| "loss": 0.1542, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.8797455505176964, | |
| "grad_norm": 0.06403321027755737, | |
| "learning_rate": 7.852809946357848e-06, | |
| "loss": 0.1544, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.9474182851729038, | |
| "grad_norm": 57.167232513427734, | |
| "learning_rate": 7.602145686068081e-06, | |
| "loss": 0.1562, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.9999323272653448, | |
| "eval_accuracy": 0.9771611282402649, | |
| "eval_f1": 0.9400239904038384, | |
| "eval_loss": 0.13267828524112701, | |
| "eval_precision": 0.9255905511811023, | |
| "eval_recall": 0.954914703493095, | |
| "eval_runtime": 41.5418, | |
| "eval_samples_per_second": 632.4, | |
| "eval_steps_per_second": 79.053, | |
| "step": 7388 | |
| }, | |
| { | |
| "epoch": 1.0150910198281113, | |
| "grad_norm": 46.099124908447266, | |
| "learning_rate": 7.351481425778313e-06, | |
| "loss": 0.1436, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.0827637544833186, | |
| "grad_norm": 0.1531323343515396, | |
| "learning_rate": 7.100817165488545e-06, | |
| "loss": 0.1, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.1504364891385261, | |
| "grad_norm": 0.17008908092975616, | |
| "learning_rate": 6.850152905198777e-06, | |
| "loss": 0.1309, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.2181092237937334, | |
| "grad_norm": 12.858524322509766, | |
| "learning_rate": 6.599488644909009e-06, | |
| "loss": 0.1145, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.285781958448941, | |
| "grad_norm": 0.05379267409443855, | |
| "learning_rate": 6.348824384619241e-06, | |
| "loss": 0.1286, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.3534546931041485, | |
| "grad_norm": 0.09806462377309799, | |
| "learning_rate": 6.098160124329474e-06, | |
| "loss": 0.1252, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.4211274277593557, | |
| "grad_norm": 15.07020378112793, | |
| "learning_rate": 5.847495864039706e-06, | |
| "loss": 0.1055, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.4888001624145633, | |
| "grad_norm": 4.208822250366211, | |
| "learning_rate": 5.596831603749938e-06, | |
| "loss": 0.1069, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.5564728970697705, | |
| "grad_norm": 53.62128829956055, | |
| "learning_rate": 5.3461673434601705e-06, | |
| "loss": 0.1181, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.624145631724978, | |
| "grad_norm": 2.466780424118042, | |
| "learning_rate": 5.095503083170402e-06, | |
| "loss": 0.1126, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.6918183663801853, | |
| "grad_norm": 7.860719203948975, | |
| "learning_rate": 4.844838822880634e-06, | |
| "loss": 0.1149, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.7594911010353929, | |
| "grad_norm": 1.2510418891906738, | |
| "learning_rate": 4.594174562590866e-06, | |
| "loss": 0.1162, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.8271638356906004, | |
| "grad_norm": 14.688665390014648, | |
| "learning_rate": 4.3435103023010985e-06, | |
| "loss": 0.1258, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.8948365703458077, | |
| "grad_norm": 0.057993557304143906, | |
| "learning_rate": 4.09284604201133e-06, | |
| "loss": 0.1097, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.962509305001015, | |
| "grad_norm": 1.119680404663086, | |
| "learning_rate": 3.842181781721562e-06, | |
| "loss": 0.1108, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9791785619123748, | |
| "eval_f1": 0.9448421901784814, | |
| "eval_loss": 0.1426609307527542, | |
| "eval_precision": 0.9383136390947326, | |
| "eval_recall": 0.9514622258326564, | |
| "eval_runtime": 41.7833, | |
| "eval_samples_per_second": 628.744, | |
| "eval_steps_per_second": 78.596, | |
| "step": 14777 | |
| }, | |
| { | |
| "epoch": 2.0301820396562227, | |
| "grad_norm": 0.01971724070608616, | |
| "learning_rate": 3.5915175214317943e-06, | |
| "loss": 0.0907, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.09785477431143, | |
| "grad_norm": 3.985325336456299, | |
| "learning_rate": 3.340853261142027e-06, | |
| "loss": 0.073, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 2.1655275089666373, | |
| "grad_norm": 32.28806686401367, | |
| "learning_rate": 3.090189000852259e-06, | |
| "loss": 0.0692, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 2.2332002436218445, | |
| "grad_norm": 0.15957437455654144, | |
| "learning_rate": 2.839524740562491e-06, | |
| "loss": 0.0819, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 2.3008729782770523, | |
| "grad_norm": 132.57357788085938, | |
| "learning_rate": 2.588860480272723e-06, | |
| "loss": 0.0693, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 2.3685457129322596, | |
| "grad_norm": 8.654540061950684, | |
| "learning_rate": 2.338196219982955e-06, | |
| "loss": 0.0715, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 2.436218447587467, | |
| "grad_norm": 9.029056549072266, | |
| "learning_rate": 2.087531959693187e-06, | |
| "loss": 0.0652, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 2.5038911822426746, | |
| "grad_norm": 9.801325798034668, | |
| "learning_rate": 1.8368676994034193e-06, | |
| "loss": 0.0646, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 2.571563916897882, | |
| "grad_norm": 0.0120752714574337, | |
| "learning_rate": 1.5862034391136513e-06, | |
| "loss": 0.0772, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 2.639236651553089, | |
| "grad_norm": 409.3838806152344, | |
| "learning_rate": 1.3355391788238833e-06, | |
| "loss": 0.0588, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 2.706909386208297, | |
| "grad_norm": 0.07232743501663208, | |
| "learning_rate": 1.0848749185341155e-06, | |
| "loss": 0.0667, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 2.774582120863504, | |
| "grad_norm": 0.0916425883769989, | |
| "learning_rate": 8.342106582443475e-07, | |
| "loss": 0.0617, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 2.8422548555187115, | |
| "grad_norm": 0.2864920198917389, | |
| "learning_rate": 5.835463979545797e-07, | |
| "loss": 0.076, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 2.9099275901739188, | |
| "grad_norm": 0.04937116429209709, | |
| "learning_rate": 3.328821376648118e-07, | |
| "loss": 0.0644, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 2.9776003248291265, | |
| "grad_norm": 0.08246259391307831, | |
| "learning_rate": 8.221787737504388e-08, | |
| "loss": 0.0695, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 2.9997969817960346, | |
| "eval_accuracy": 0.9790643675535762, | |
| "eval_f1": 0.9445452712240371, | |
| "eval_loss": 0.17905421555042267, | |
| "eval_precision": 0.9379255106127353, | |
| "eval_recall": 0.9512591389114541, | |
| "eval_runtime": 42.1003, | |
| "eval_samples_per_second": 624.01, | |
| "eval_steps_per_second": 78.004, | |
| "step": 22164 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 22164, | |
| "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": 4.665261589246464e+16, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |