| { | |
| "best_metric": 0.7855386128873586, | |
| "best_model_checkpoint": "wav2vec2-xls-r-300m-ic-finetuning/checkpoint-2790", | |
| "epoch": 35.0, | |
| "global_step": 3150, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 1.5555555555555555e-05, | |
| "loss": 4.0884, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.06443679291687161, | |
| "eval_loss": 3.94179630279541, | |
| "eval_runtime": 17.2679, | |
| "eval_samples_per_second": 117.733, | |
| "eval_steps_per_second": 0.927, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 3.142857142857143e-05, | |
| "loss": 4.0116, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 4.73015873015873e-05, | |
| "loss": 3.8285, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.06443679291687161, | |
| "eval_loss": 3.742227792739868, | |
| "eval_runtime": 17.4046, | |
| "eval_samples_per_second": 116.808, | |
| "eval_steps_per_second": 0.919, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 6.317460317460318e-05, | |
| "loss": 3.7787, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 7.904761904761905e-05, | |
| "loss": 3.7603, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.06443679291687161, | |
| "eval_loss": 3.730922222137451, | |
| "eval_runtime": 17.2612, | |
| "eval_samples_per_second": 117.779, | |
| "eval_steps_per_second": 0.927, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 9.46031746031746e-05, | |
| "loss": 3.7493, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 9.883597883597884e-05, | |
| "loss": 3.7475, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.08460403344810624, | |
| "eval_loss": 3.719400405883789, | |
| "eval_runtime": 17.1221, | |
| "eval_samples_per_second": 118.736, | |
| "eval_steps_per_second": 0.934, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "learning_rate": 9.707231040564374e-05, | |
| "loss": 3.7259, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 9.530864197530865e-05, | |
| "loss": 3.6843, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.102803738317757, | |
| "eval_loss": 3.6270625591278076, | |
| "eval_runtime": 16.8572, | |
| "eval_samples_per_second": 120.601, | |
| "eval_steps_per_second": 0.949, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 5.56, | |
| "learning_rate": 9.354497354497354e-05, | |
| "loss": 3.6367, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.12690605017215936, | |
| "eval_loss": 3.4777231216430664, | |
| "eval_runtime": 16.8395, | |
| "eval_samples_per_second": 120.728, | |
| "eval_steps_per_second": 0.95, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 6.11, | |
| "learning_rate": 9.178130511463845e-05, | |
| "loss": 3.5853, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "learning_rate": 9.001763668430336e-05, | |
| "loss": 3.4016, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.21446138711264143, | |
| "eval_loss": 3.1104345321655273, | |
| "eval_runtime": 16.8901, | |
| "eval_samples_per_second": 120.366, | |
| "eval_steps_per_second": 0.947, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 7.22, | |
| "learning_rate": 8.825396825396825e-05, | |
| "loss": 3.213, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 7.78, | |
| "learning_rate": 8.649029982363316e-05, | |
| "loss": 2.946, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.31234628627643873, | |
| "eval_loss": 2.6535961627960205, | |
| "eval_runtime": 16.8793, | |
| "eval_samples_per_second": 120.444, | |
| "eval_steps_per_second": 0.948, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "learning_rate": 8.472663139329807e-05, | |
| "loss": 2.7358, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 8.89, | |
| "learning_rate": 8.296296296296296e-05, | |
| "loss": 2.514, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.4176094441711756, | |
| "eval_loss": 2.3149795532226562, | |
| "eval_runtime": 16.8771, | |
| "eval_samples_per_second": 120.459, | |
| "eval_steps_per_second": 0.948, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 9.44, | |
| "learning_rate": 8.119929453262787e-05, | |
| "loss": 2.3267, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 7.947089947089948e-05, | |
| "loss": 2.163, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.48106246925725527, | |
| "eval_loss": 2.0448246002197266, | |
| "eval_runtime": 16.8844, | |
| "eval_samples_per_second": 120.407, | |
| "eval_steps_per_second": 0.948, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 10.56, | |
| "learning_rate": 7.770723104056437e-05, | |
| "loss": 1.947, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.5700934579439252, | |
| "eval_loss": 1.7913175821304321, | |
| "eval_runtime": 16.8279, | |
| "eval_samples_per_second": 120.811, | |
| "eval_steps_per_second": 0.951, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 11.11, | |
| "learning_rate": 7.594356261022928e-05, | |
| "loss": 1.8568, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 11.67, | |
| "learning_rate": 7.417989417989419e-05, | |
| "loss": 1.7119, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.6242006886374816, | |
| "eval_loss": 1.616714596748352, | |
| "eval_runtime": 16.8545, | |
| "eval_samples_per_second": 120.62, | |
| "eval_steps_per_second": 0.949, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 12.22, | |
| "learning_rate": 7.241622574955908e-05, | |
| "loss": 1.5846, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 12.78, | |
| "learning_rate": 7.065255731922399e-05, | |
| "loss": 1.4849, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.6620757501229709, | |
| "eval_loss": 1.5006158351898193, | |
| "eval_runtime": 16.8739, | |
| "eval_samples_per_second": 120.482, | |
| "eval_steps_per_second": 0.948, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 13.33, | |
| "learning_rate": 6.88888888888889e-05, | |
| "loss": 1.3323, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 13.89, | |
| "learning_rate": 6.712522045855379e-05, | |
| "loss": 1.3281, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.6699458927693065, | |
| "eval_loss": 1.4369958639144897, | |
| "eval_runtime": 16.7964, | |
| "eval_samples_per_second": 121.038, | |
| "eval_steps_per_second": 0.953, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 14.44, | |
| "learning_rate": 6.53615520282187e-05, | |
| "loss": 1.1692, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 6.35978835978836e-05, | |
| "loss": 1.1518, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.6984751598622725, | |
| "eval_loss": 1.3894743919372559, | |
| "eval_runtime": 16.9538, | |
| "eval_samples_per_second": 119.914, | |
| "eval_steps_per_second": 0.944, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 15.56, | |
| "learning_rate": 6.183421516754851e-05, | |
| "loss": 0.9798, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.7329070339399901, | |
| "eval_loss": 1.3057465553283691, | |
| "eval_runtime": 16.8273, | |
| "eval_samples_per_second": 120.816, | |
| "eval_steps_per_second": 0.951, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 16.11, | |
| "learning_rate": 6.0070546737213404e-05, | |
| "loss": 1.0168, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 16.67, | |
| "learning_rate": 5.830687830687831e-05, | |
| "loss": 0.9296, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.7353664535169699, | |
| "eval_loss": 1.2697882652282715, | |
| "eval_runtime": 16.8455, | |
| "eval_samples_per_second": 120.685, | |
| "eval_steps_per_second": 0.95, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 17.22, | |
| "learning_rate": 5.654320987654321e-05, | |
| "loss": 0.8497, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 17.78, | |
| "learning_rate": 5.477954144620812e-05, | |
| "loss": 0.7886, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.750122970978849, | |
| "eval_loss": 1.2826693058013916, | |
| "eval_runtime": 16.966, | |
| "eval_samples_per_second": 119.828, | |
| "eval_steps_per_second": 0.943, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 18.33, | |
| "learning_rate": 5.301587301587302e-05, | |
| "loss": 0.7637, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 18.89, | |
| "learning_rate": 5.125220458553792e-05, | |
| "loss": 0.7191, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.749139203148057, | |
| "eval_loss": 1.3038619756698608, | |
| "eval_runtime": 16.7983, | |
| "eval_samples_per_second": 121.024, | |
| "eval_steps_per_second": 0.952, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 19.44, | |
| "learning_rate": 4.948853615520283e-05, | |
| "loss": 0.6952, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 4.772486772486773e-05, | |
| "loss": 0.609, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.7594687653713723, | |
| "eval_loss": 1.2997246980667114, | |
| "eval_runtime": 16.7476, | |
| "eval_samples_per_second": 121.39, | |
| "eval_steps_per_second": 0.955, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 20.56, | |
| "learning_rate": 4.596119929453263e-05, | |
| "loss": 0.5703, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.76586325627152, | |
| "eval_loss": 1.2955206632614136, | |
| "eval_runtime": 16.7956, | |
| "eval_samples_per_second": 121.044, | |
| "eval_steps_per_second": 0.953, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 21.11, | |
| "learning_rate": 4.4197530864197535e-05, | |
| "loss": 0.5608, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 21.67, | |
| "learning_rate": 4.2433862433862435e-05, | |
| "loss": 0.5251, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.7614363010329562, | |
| "eval_loss": 1.3140268325805664, | |
| "eval_runtime": 16.7643, | |
| "eval_samples_per_second": 121.27, | |
| "eval_steps_per_second": 0.954, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 22.22, | |
| "learning_rate": 4.067019400352734e-05, | |
| "loss": 0.5351, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 22.78, | |
| "learning_rate": 3.890652557319224e-05, | |
| "loss": 0.5041, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 0.7668470241023119, | |
| "eval_loss": 1.319765567779541, | |
| "eval_runtime": 16.8549, | |
| "eval_samples_per_second": 120.618, | |
| "eval_steps_per_second": 0.949, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 23.33, | |
| "learning_rate": 3.717813051146385e-05, | |
| "loss": 0.435, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 23.89, | |
| "learning_rate": 3.5414462081128755e-05, | |
| "loss": 0.4455, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.7816035415641909, | |
| "eval_loss": 1.3251405954360962, | |
| "eval_runtime": 16.7377, | |
| "eval_samples_per_second": 121.462, | |
| "eval_steps_per_second": 0.956, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 24.44, | |
| "learning_rate": 3.3650793650793656e-05, | |
| "loss": 0.4152, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 3.1887125220458556e-05, | |
| "loss": 0.4046, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.7825873093949828, | |
| "eval_loss": 1.3233106136322021, | |
| "eval_runtime": 16.7729, | |
| "eval_samples_per_second": 121.207, | |
| "eval_steps_per_second": 0.954, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 25.56, | |
| "learning_rate": 3.012345679012346e-05, | |
| "loss": 0.3861, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.7737333989178554, | |
| "eval_loss": 1.3713818788528442, | |
| "eval_runtime": 16.8666, | |
| "eval_samples_per_second": 120.534, | |
| "eval_steps_per_second": 0.949, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 26.11, | |
| "learning_rate": 2.8359788359788357e-05, | |
| "loss": 0.3571, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 26.67, | |
| "learning_rate": 2.659611992945326e-05, | |
| "loss": 0.3798, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 0.7747171667486473, | |
| "eval_loss": 1.3577079772949219, | |
| "eval_runtime": 16.7848, | |
| "eval_samples_per_second": 121.122, | |
| "eval_steps_per_second": 0.953, | |
| "step": 2430 | |
| }, | |
| { | |
| "epoch": 27.22, | |
| "learning_rate": 2.4832451499118168e-05, | |
| "loss": 0.3686, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 27.78, | |
| "learning_rate": 2.3068783068783072e-05, | |
| "loss": 0.3402, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.7757009345794392, | |
| "eval_loss": 1.3933676481246948, | |
| "eval_runtime": 16.7209, | |
| "eval_samples_per_second": 121.585, | |
| "eval_steps_per_second": 0.957, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 28.33, | |
| "learning_rate": 2.1305114638447972e-05, | |
| "loss": 0.3556, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 28.89, | |
| "learning_rate": 1.9541446208112876e-05, | |
| "loss": 0.2951, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 0.779636005902607, | |
| "eval_loss": 1.3655041456222534, | |
| "eval_runtime": 16.7839, | |
| "eval_samples_per_second": 121.128, | |
| "eval_steps_per_second": 0.953, | |
| "step": 2610 | |
| }, | |
| { | |
| "epoch": 29.44, | |
| "learning_rate": 1.777777777777778e-05, | |
| "loss": 0.3126, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "learning_rate": 1.6014109347442684e-05, | |
| "loss": 0.2953, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.7776684702410231, | |
| "eval_loss": 1.3812841176986694, | |
| "eval_runtime": 16.7955, | |
| "eval_samples_per_second": 121.044, | |
| "eval_steps_per_second": 0.953, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 30.56, | |
| "learning_rate": 1.4250440917107586e-05, | |
| "loss": 0.3125, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_accuracy": 0.7855386128873586, | |
| "eval_loss": 1.3787145614624023, | |
| "eval_runtime": 16.8143, | |
| "eval_samples_per_second": 120.909, | |
| "eval_steps_per_second": 0.952, | |
| "step": 2790 | |
| }, | |
| { | |
| "epoch": 31.11, | |
| "learning_rate": 1.2486772486772486e-05, | |
| "loss": 0.2608, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 31.67, | |
| "learning_rate": 1.072310405643739e-05, | |
| "loss": 0.2564, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.778652238071815, | |
| "eval_loss": 1.3935987949371338, | |
| "eval_runtime": 16.7896, | |
| "eval_samples_per_second": 121.087, | |
| "eval_steps_per_second": 0.953, | |
| "step": 2880 | |
| }, | |
| { | |
| "epoch": 32.22, | |
| "learning_rate": 8.959435626102292e-06, | |
| "loss": 0.2534, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 32.78, | |
| "learning_rate": 7.195767195767196e-06, | |
| "loss": 0.2427, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_accuracy": 0.7855386128873586, | |
| "eval_loss": 1.388238787651062, | |
| "eval_runtime": 16.8064, | |
| "eval_samples_per_second": 120.966, | |
| "eval_steps_per_second": 0.952, | |
| "step": 2970 | |
| }, | |
| { | |
| "epoch": 33.33, | |
| "learning_rate": 5.432098765432099e-06, | |
| "loss": 0.2691, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 33.89, | |
| "learning_rate": 3.668430335097002e-06, | |
| "loss": 0.279, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.7855386128873586, | |
| "eval_loss": 1.4113564491271973, | |
| "eval_runtime": 16.7799, | |
| "eval_samples_per_second": 121.157, | |
| "eval_steps_per_second": 0.954, | |
| "step": 3060 | |
| }, | |
| { | |
| "epoch": 34.44, | |
| "learning_rate": 1.9047619047619051e-06, | |
| "loss": 0.2582, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "learning_rate": 1.4109347442680778e-07, | |
| "loss": 0.2308, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_accuracy": 0.7855386128873586, | |
| "eval_loss": 1.4130984544754028, | |
| "eval_runtime": 16.8643, | |
| "eval_samples_per_second": 120.551, | |
| "eval_steps_per_second": 0.949, | |
| "step": 3150 | |
| } | |
| ], | |
| "max_steps": 3150, | |
| "num_train_epochs": 35, | |
| "total_flos": 1.221413841164159e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |