| { | |
| "best_metric": 0.9087967049132097, | |
| "best_model_checkpoint": "trillsson3-ft-keyword-spotting-11/checkpoint-7980", | |
| "epoch": 19.999373825923605, | |
| "global_step": 7980, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 3.571428571428571e-05, | |
| "loss": 7.4329, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 7.330827067669172e-05, | |
| "loss": 4.5207, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.00011090225563909774, | |
| "loss": 2.9219, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6216534274786703, | |
| "eval_loss": 1.202303409576416, | |
| "eval_runtime": 139.1467, | |
| "eval_samples_per_second": 48.855, | |
| "eval_steps_per_second": 1.531, | |
| "step": 399 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.00014849624060150375, | |
| "loss": 2.0315, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 0.00018609022556390976, | |
| "loss": 1.4541, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 0.00022368421052631576, | |
| "loss": 1.1541, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 0.00026127819548872177, | |
| "loss": 0.9604, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8117093262724331, | |
| "eval_loss": 0.5436997413635254, | |
| "eval_runtime": 140.1586, | |
| "eval_samples_per_second": 48.502, | |
| "eval_steps_per_second": 1.52, | |
| "step": 798 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 0.0002988721804511278, | |
| "loss": 0.8683, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 0.00029594820384294066, | |
| "loss": 0.8309, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 0.0002917710944026733, | |
| "loss": 0.7717, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 0.000287593984962406, | |
| "loss": 0.7608, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8887908208296558, | |
| "eval_loss": 0.42219480872154236, | |
| "eval_runtime": 140.2584, | |
| "eval_samples_per_second": 48.468, | |
| "eval_steps_per_second": 1.519, | |
| "step": 1197 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 0.00028341687552213866, | |
| "loss": 0.7384, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "learning_rate": 0.0002792397660818713, | |
| "loss": 0.7127, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "learning_rate": 0.000275062656641604, | |
| "loss": 0.7025, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "learning_rate": 0.00027088554720133666, | |
| "loss": 0.7045, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8932038834951457, | |
| "eval_loss": 0.3880767524242401, | |
| "eval_runtime": 139.5385, | |
| "eval_samples_per_second": 48.718, | |
| "eval_steps_per_second": 1.526, | |
| "step": 1596 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 0.0002667084377610693, | |
| "loss": 0.7082, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 4.26, | |
| "learning_rate": 0.000262531328320802, | |
| "loss": 0.6833, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 4.51, | |
| "learning_rate": 0.00025835421888053466, | |
| "loss": 0.6881, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "learning_rate": 0.0002541771094402673, | |
| "loss": 0.659, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8846719623418653, | |
| "eval_loss": 0.37056076526641846, | |
| "eval_runtime": 141.2433, | |
| "eval_samples_per_second": 48.13, | |
| "eval_steps_per_second": 1.508, | |
| "step": 1995 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 0.00025, | |
| "loss": 0.6621, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.26, | |
| "learning_rate": 0.00024582289055973265, | |
| "loss": 0.6643, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 5.51, | |
| "learning_rate": 0.00024164578111946533, | |
| "loss": 0.6421, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "learning_rate": 0.00023746867167919798, | |
| "loss": 0.6541, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.891732862606649, | |
| "eval_loss": 0.3553307056427002, | |
| "eval_runtime": 140.013, | |
| "eval_samples_per_second": 48.553, | |
| "eval_steps_per_second": 1.521, | |
| "step": 2394 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "learning_rate": 0.00023329156223893065, | |
| "loss": 0.6477, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 6.27, | |
| "learning_rate": 0.00022911445279866333, | |
| "loss": 0.651, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 6.52, | |
| "learning_rate": 0.00022493734335839597, | |
| "loss": 0.6509, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 6.77, | |
| "learning_rate": 0.00022076023391812865, | |
| "loss": 0.6448, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8952633127390409, | |
| "eval_loss": 0.3482230305671692, | |
| "eval_runtime": 140.3109, | |
| "eval_samples_per_second": 48.45, | |
| "eval_steps_per_second": 1.518, | |
| "step": 2793 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "learning_rate": 0.00021658312447786132, | |
| "loss": 0.6421, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 7.27, | |
| "learning_rate": 0.00021240601503759397, | |
| "loss": 0.6418, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 7.52, | |
| "learning_rate": 0.00020822890559732665, | |
| "loss": 0.6453, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 7.77, | |
| "learning_rate": 0.00020405179615705932, | |
| "loss": 0.6288, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8989408649602825, | |
| "eval_loss": 0.34094032645225525, | |
| "eval_runtime": 141.0541, | |
| "eval_samples_per_second": 48.194, | |
| "eval_steps_per_second": 1.51, | |
| "step": 3192 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 0.00019987468671679197, | |
| "loss": 0.6336, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 8.27, | |
| "learning_rate": 0.00019569757727652464, | |
| "loss": 0.6441, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 8.52, | |
| "learning_rate": 0.00019152046783625732, | |
| "loss": 0.6557, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 8.77, | |
| "learning_rate": 0.00018734335839598997, | |
| "loss": 0.641, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9051191526919682, | |
| "eval_loss": 0.3297290503978729, | |
| "eval_runtime": 140.3281, | |
| "eval_samples_per_second": 48.444, | |
| "eval_steps_per_second": 1.518, | |
| "step": 3591 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "learning_rate": 0.00018316624895572264, | |
| "loss": 0.6411, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 9.27, | |
| "learning_rate": 0.00017898913951545532, | |
| "loss": 0.6455, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 9.52, | |
| "learning_rate": 0.00017481203007518797, | |
| "loss": 0.6464, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 9.77, | |
| "learning_rate": 0.00017063492063492064, | |
| "loss": 0.6369, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9042365401588702, | |
| "eval_loss": 0.33252373337745667, | |
| "eval_runtime": 140.3318, | |
| "eval_samples_per_second": 48.442, | |
| "eval_steps_per_second": 1.518, | |
| "step": 3990 | |
| }, | |
| { | |
| "epoch": 10.03, | |
| "learning_rate": 0.0001664578111946533, | |
| "loss": 0.6225, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 10.28, | |
| "learning_rate": 0.00016228070175438596, | |
| "loss": 0.6295, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 10.53, | |
| "learning_rate": 0.0001581453634085213, | |
| "loss": 0.6255, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 10.78, | |
| "learning_rate": 0.00015396825396825397, | |
| "loss": 0.6218, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.9064430714916152, | |
| "eval_loss": 0.3250460922718048, | |
| "eval_runtime": 140.1683, | |
| "eval_samples_per_second": 48.499, | |
| "eval_steps_per_second": 1.52, | |
| "step": 4389 | |
| }, | |
| { | |
| "epoch": 11.03, | |
| "learning_rate": 0.00014979114452798661, | |
| "loss": 0.6507, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 11.28, | |
| "learning_rate": 0.0001456140350877193, | |
| "loss": 0.623, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 11.53, | |
| "learning_rate": 0.00014143692564745194, | |
| "loss": 0.6278, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 11.78, | |
| "learning_rate": 0.0001372598162071846, | |
| "loss": 0.6247, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8958517210944396, | |
| "eval_loss": 0.3311932682991028, | |
| "eval_runtime": 140.5537, | |
| "eval_samples_per_second": 48.366, | |
| "eval_steps_per_second": 1.515, | |
| "step": 4788 | |
| }, | |
| { | |
| "epoch": 12.03, | |
| "learning_rate": 0.0001330827067669173, | |
| "loss": 0.6149, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 12.28, | |
| "learning_rate": 0.00012890559732664993, | |
| "loss": 0.6244, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 12.53, | |
| "learning_rate": 0.0001247284878863826, | |
| "loss": 0.6294, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 12.78, | |
| "learning_rate": 0.00012055137844611527, | |
| "loss": 0.6284, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.9068843777581642, | |
| "eval_loss": 0.321657657623291, | |
| "eval_runtime": 139.9572, | |
| "eval_samples_per_second": 48.572, | |
| "eval_steps_per_second": 1.522, | |
| "step": 5187 | |
| }, | |
| { | |
| "epoch": 13.03, | |
| "learning_rate": 0.00011637426900584793, | |
| "loss": 0.6317, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 13.28, | |
| "learning_rate": 0.00011219715956558061, | |
| "loss": 0.6397, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 13.53, | |
| "learning_rate": 0.00010802005012531327, | |
| "loss": 0.6041, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 13.78, | |
| "learning_rate": 0.00010384294068504593, | |
| "loss": 0.6213, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8977640482494852, | |
| "eval_loss": 0.33012473583221436, | |
| "eval_runtime": 139.8104, | |
| "eval_samples_per_second": 48.623, | |
| "eval_steps_per_second": 1.523, | |
| "step": 5586 | |
| }, | |
| { | |
| "epoch": 14.04, | |
| "learning_rate": 9.96658312447786e-05, | |
| "loss": 0.6206, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 14.29, | |
| "learning_rate": 9.548872180451127e-05, | |
| "loss": 0.6379, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 14.54, | |
| "learning_rate": 9.131161236424393e-05, | |
| "loss": 0.6269, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 14.79, | |
| "learning_rate": 8.713450292397659e-05, | |
| "loss": 0.6274, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.9080611944689615, | |
| "eval_loss": 0.31795457005500793, | |
| "eval_runtime": 139.5988, | |
| "eval_samples_per_second": 48.697, | |
| "eval_steps_per_second": 1.526, | |
| "step": 5985 | |
| }, | |
| { | |
| "epoch": 15.04, | |
| "learning_rate": 8.295739348370926e-05, | |
| "loss": 0.6263, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 15.29, | |
| "learning_rate": 7.878028404344193e-05, | |
| "loss": 0.6171, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 15.54, | |
| "learning_rate": 7.460317460317459e-05, | |
| "loss": 0.626, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 15.79, | |
| "learning_rate": 7.042606516290726e-05, | |
| "loss": 0.627, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.9020300088261254, | |
| "eval_loss": 0.32565683126449585, | |
| "eval_runtime": 139.2707, | |
| "eval_samples_per_second": 48.811, | |
| "eval_steps_per_second": 1.529, | |
| "step": 6384 | |
| }, | |
| { | |
| "epoch": 16.04, | |
| "learning_rate": 6.624895572263992e-05, | |
| "loss": 0.6166, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 16.29, | |
| "learning_rate": 6.207184628237259e-05, | |
| "loss": 0.6146, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 16.54, | |
| "learning_rate": 5.7894736842105253e-05, | |
| "loss": 0.6126, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 16.79, | |
| "learning_rate": 5.3759398496240595e-05, | |
| "loss": 0.6227, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.9055604589585172, | |
| "eval_loss": 0.3193184435367584, | |
| "eval_runtime": 140.199, | |
| "eval_samples_per_second": 48.488, | |
| "eval_steps_per_second": 1.519, | |
| "step": 6783 | |
| }, | |
| { | |
| "epoch": 17.04, | |
| "learning_rate": 4.9582289055973256e-05, | |
| "loss": 0.629, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 17.29, | |
| "learning_rate": 4.5405179615705925e-05, | |
| "loss": 0.6055, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 17.54, | |
| "learning_rate": 4.122807017543859e-05, | |
| "loss": 0.6088, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 17.79, | |
| "learning_rate": 3.7050960735171254e-05, | |
| "loss": 0.6192, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.9065901735804649, | |
| "eval_loss": 0.31994640827178955, | |
| "eval_runtime": 139.8576, | |
| "eval_samples_per_second": 48.607, | |
| "eval_steps_per_second": 1.523, | |
| "step": 7182 | |
| }, | |
| { | |
| "epoch": 18.05, | |
| "learning_rate": 3.287385129490392e-05, | |
| "loss": 0.624, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 18.3, | |
| "learning_rate": 2.8696741854636587e-05, | |
| "loss": 0.6102, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 18.55, | |
| "learning_rate": 2.4519632414369252e-05, | |
| "loss": 0.6261, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 18.8, | |
| "learning_rate": 2.0342522974101917e-05, | |
| "loss": 0.6075, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.9073256840247131, | |
| "eval_loss": 0.3183441460132599, | |
| "eval_runtime": 145.1619, | |
| "eval_samples_per_second": 46.83, | |
| "eval_steps_per_second": 1.467, | |
| "step": 7581 | |
| }, | |
| { | |
| "epoch": 19.05, | |
| "learning_rate": 1.6165413533834585e-05, | |
| "loss": 0.6066, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 19.3, | |
| "learning_rate": 1.198830409356725e-05, | |
| "loss": 0.6057, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 19.55, | |
| "learning_rate": 7.811194653299915e-06, | |
| "loss": 0.6276, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 19.8, | |
| "learning_rate": 3.634085213032581e-06, | |
| "loss": 0.6196, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9087967049132097, | |
| "eval_loss": 0.3166075348854065, | |
| "eval_runtime": 146.0633, | |
| "eval_samples_per_second": 46.541, | |
| "eval_steps_per_second": 1.458, | |
| "step": 7980 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 7980, | |
| "total_flos": 0.0, | |
| "train_loss": 0.8469889265552798, | |
| "train_runtime": 24285.4703, | |
| "train_samples_per_second": 42.078, | |
| "train_steps_per_second": 0.329 | |
| } | |
| ], | |
| "max_steps": 7980, | |
| "num_train_epochs": 20, | |
| "total_flos": 0.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |