| { | |
| "best_metric": 0.9411764705882353, | |
| "best_model_checkpoint": "vit-base-patch16-224-in21k-finetuned-papsmear/checkpoint-765", | |
| "epoch": 29.80392156862745, | |
| "eval_steps": 500, | |
| "global_step": 1140, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.26143790849673204, | |
| "grad_norm": 1.524962306022644, | |
| "learning_rate": 4.3859649122807014e-06, | |
| "loss": 1.8303, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.5228758169934641, | |
| "grad_norm": 1.3496487140655518, | |
| "learning_rate": 8.771929824561403e-06, | |
| "loss": 1.7834, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.7843137254901961, | |
| "grad_norm": 1.857043981552124, | |
| "learning_rate": 1.3157894736842106e-05, | |
| "loss": 1.6954, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.9934640522875817, | |
| "eval_accuracy": 0.34558823529411764, | |
| "eval_loss": 1.610603928565979, | |
| "eval_runtime": 33.3958, | |
| "eval_samples_per_second": 4.072, | |
| "eval_steps_per_second": 0.509, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.0457516339869282, | |
| "grad_norm": 1.5567628145217896, | |
| "learning_rate": 1.7543859649122806e-05, | |
| "loss": 1.5872, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.3071895424836601, | |
| "grad_norm": 1.5087599754333496, | |
| "learning_rate": 2.1929824561403507e-05, | |
| "loss": 1.4855, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.5686274509803921, | |
| "grad_norm": 1.6773402690887451, | |
| "learning_rate": 2.6315789473684212e-05, | |
| "loss": 1.3823, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.8300653594771243, | |
| "grad_norm": 1.4024033546447754, | |
| "learning_rate": 3.0701754385964913e-05, | |
| "loss": 1.2818, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.9869281045751634, | |
| "eval_accuracy": 0.5735294117647058, | |
| "eval_loss": 1.2412389516830444, | |
| "eval_runtime": 32.7103, | |
| "eval_samples_per_second": 4.158, | |
| "eval_steps_per_second": 0.52, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 2.0915032679738563, | |
| "grad_norm": 1.6061407327651978, | |
| "learning_rate": 3.508771929824561e-05, | |
| "loss": 1.2088, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.3529411764705883, | |
| "grad_norm": 3.8964905738830566, | |
| "learning_rate": 3.9473684210526316e-05, | |
| "loss": 1.0983, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.6143790849673203, | |
| "grad_norm": 3.24979305267334, | |
| "learning_rate": 4.3859649122807014e-05, | |
| "loss": 1.0632, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.8758169934640523, | |
| "grad_norm": 2.1930079460144043, | |
| "learning_rate": 4.824561403508772e-05, | |
| "loss": 1.0023, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.980392156862745, | |
| "eval_accuracy": 0.7132352941176471, | |
| "eval_loss": 0.9875355958938599, | |
| "eval_runtime": 32.8626, | |
| "eval_samples_per_second": 4.138, | |
| "eval_steps_per_second": 0.517, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 3.1372549019607843, | |
| "grad_norm": 2.084902048110962, | |
| "learning_rate": 4.970760233918128e-05, | |
| "loss": 0.823, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.3986928104575163, | |
| "grad_norm": 1.8761796951293945, | |
| "learning_rate": 4.9220272904483435e-05, | |
| "loss": 0.785, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.6601307189542482, | |
| "grad_norm": 3.031520128250122, | |
| "learning_rate": 4.8732943469785574e-05, | |
| "loss": 0.7775, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.9215686274509802, | |
| "grad_norm": 2.1056406497955322, | |
| "learning_rate": 4.824561403508772e-05, | |
| "loss": 0.7163, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6911764705882353, | |
| "eval_loss": 0.8399065136909485, | |
| "eval_runtime": 32.9831, | |
| "eval_samples_per_second": 4.123, | |
| "eval_steps_per_second": 0.515, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 4.183006535947713, | |
| "grad_norm": 3.3159000873565674, | |
| "learning_rate": 4.7758284600389865e-05, | |
| "loss": 0.6677, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 4.444444444444445, | |
| "grad_norm": 2.1595349311828613, | |
| "learning_rate": 4.727095516569201e-05, | |
| "loss": 0.6366, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 4.705882352941177, | |
| "grad_norm": 2.2182297706604004, | |
| "learning_rate": 4.678362573099415e-05, | |
| "loss": 0.5876, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 4.967320261437909, | |
| "grad_norm": 5.272632122039795, | |
| "learning_rate": 4.62962962962963e-05, | |
| "loss": 0.5173, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 4.993464052287582, | |
| "eval_accuracy": 0.8161764705882353, | |
| "eval_loss": 0.6545882821083069, | |
| "eval_runtime": 31.9706, | |
| "eval_samples_per_second": 4.254, | |
| "eval_steps_per_second": 0.532, | |
| "step": 191 | |
| }, | |
| { | |
| "epoch": 5.228758169934641, | |
| "grad_norm": 5.355340003967285, | |
| "learning_rate": 4.580896686159844e-05, | |
| "loss": 0.5057, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 5.490196078431373, | |
| "grad_norm": 2.3517274856567383, | |
| "learning_rate": 4.5321637426900585e-05, | |
| "loss": 0.4577, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 5.751633986928105, | |
| "grad_norm": 7.760730266571045, | |
| "learning_rate": 4.483430799220273e-05, | |
| "loss": 0.5057, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 5.9869281045751634, | |
| "eval_accuracy": 0.8308823529411765, | |
| "eval_loss": 0.6250559091567993, | |
| "eval_runtime": 32.3916, | |
| "eval_samples_per_second": 4.199, | |
| "eval_steps_per_second": 0.525, | |
| "step": 229 | |
| }, | |
| { | |
| "epoch": 6.0130718954248366, | |
| "grad_norm": 3.0986075401306152, | |
| "learning_rate": 4.4346978557504876e-05, | |
| "loss": 0.4856, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 6.2745098039215685, | |
| "grad_norm": 7.31836462020874, | |
| "learning_rate": 4.3859649122807014e-05, | |
| "loss": 0.5036, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 6.5359477124183005, | |
| "grad_norm": 2.5667953491210938, | |
| "learning_rate": 4.3372319688109166e-05, | |
| "loss": 0.4063, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 6.7973856209150325, | |
| "grad_norm": 2.488330602645874, | |
| "learning_rate": 4.2884990253411305e-05, | |
| "loss": 0.4313, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 6.980392156862745, | |
| "eval_accuracy": 0.8308823529411765, | |
| "eval_loss": 0.5696418881416321, | |
| "eval_runtime": 32.7903, | |
| "eval_samples_per_second": 4.148, | |
| "eval_steps_per_second": 0.518, | |
| "step": 267 | |
| }, | |
| { | |
| "epoch": 7.0588235294117645, | |
| "grad_norm": 4.5028228759765625, | |
| "learning_rate": 4.239766081871345e-05, | |
| "loss": 0.3409, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 7.3202614379084965, | |
| "grad_norm": 3.8160030841827393, | |
| "learning_rate": 4.1910331384015596e-05, | |
| "loss": 0.4005, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 7.5816993464052285, | |
| "grad_norm": 7.039416790008545, | |
| "learning_rate": 4.142300194931774e-05, | |
| "loss": 0.4047, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 7.8431372549019605, | |
| "grad_norm": 2.3010506629943848, | |
| "learning_rate": 4.093567251461988e-05, | |
| "loss": 0.325, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8308823529411765, | |
| "eval_loss": 0.5506752133369446, | |
| "eval_runtime": 33.2532, | |
| "eval_samples_per_second": 4.09, | |
| "eval_steps_per_second": 0.511, | |
| "step": 306 | |
| }, | |
| { | |
| "epoch": 8.104575163398692, | |
| "grad_norm": 2.194329023361206, | |
| "learning_rate": 4.044834307992203e-05, | |
| "loss": 0.3016, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 8.366013071895425, | |
| "grad_norm": 4.172213554382324, | |
| "learning_rate": 3.996101364522417e-05, | |
| "loss": 0.3194, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 8.627450980392156, | |
| "grad_norm": 3.775691270828247, | |
| "learning_rate": 3.9473684210526316e-05, | |
| "loss": 0.3115, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 8.88888888888889, | |
| "grad_norm": 3.285763740539551, | |
| "learning_rate": 3.898635477582846e-05, | |
| "loss": 0.3811, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 8.993464052287582, | |
| "eval_accuracy": 0.8676470588235294, | |
| "eval_loss": 0.4429008662700653, | |
| "eval_runtime": 32.2721, | |
| "eval_samples_per_second": 4.214, | |
| "eval_steps_per_second": 0.527, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 9.15032679738562, | |
| "grad_norm": 2.8171193599700928, | |
| "learning_rate": 3.849902534113061e-05, | |
| "loss": 0.3617, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 9.411764705882353, | |
| "grad_norm": 2.766669273376465, | |
| "learning_rate": 3.8011695906432746e-05, | |
| "loss": 0.2679, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 9.673202614379084, | |
| "grad_norm": 5.8598833084106445, | |
| "learning_rate": 3.75243664717349e-05, | |
| "loss": 0.2909, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 9.934640522875817, | |
| "grad_norm": 4.5529608726501465, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.2341, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 9.986928104575163, | |
| "eval_accuracy": 0.875, | |
| "eval_loss": 0.42220985889434814, | |
| "eval_runtime": 32.0969, | |
| "eval_samples_per_second": 4.237, | |
| "eval_steps_per_second": 0.53, | |
| "step": 382 | |
| }, | |
| { | |
| "epoch": 10.196078431372548, | |
| "grad_norm": 3.98348069190979, | |
| "learning_rate": 3.654970760233918e-05, | |
| "loss": 0.2612, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 10.457516339869281, | |
| "grad_norm": 5.887331485748291, | |
| "learning_rate": 3.606237816764133e-05, | |
| "loss": 0.2518, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 10.718954248366012, | |
| "grad_norm": 3.759489059448242, | |
| "learning_rate": 3.557504873294347e-05, | |
| "loss": 0.2361, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 10.980392156862745, | |
| "grad_norm": 6.444194793701172, | |
| "learning_rate": 3.508771929824561e-05, | |
| "loss": 0.3082, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 10.980392156862745, | |
| "eval_accuracy": 0.7720588235294118, | |
| "eval_loss": 0.6572562456130981, | |
| "eval_runtime": 32.1021, | |
| "eval_samples_per_second": 4.236, | |
| "eval_steps_per_second": 0.53, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 11.241830065359476, | |
| "grad_norm": 2.5144402980804443, | |
| "learning_rate": 3.4600389863547764e-05, | |
| "loss": 0.2284, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 11.50326797385621, | |
| "grad_norm": 5.127187728881836, | |
| "learning_rate": 3.41130604288499e-05, | |
| "loss": 0.2353, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 11.764705882352942, | |
| "grad_norm": 3.2781078815460205, | |
| "learning_rate": 3.362573099415205e-05, | |
| "loss": 0.2571, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8897058823529411, | |
| "eval_loss": 0.4229148328304291, | |
| "eval_runtime": 31.8733, | |
| "eval_samples_per_second": 4.267, | |
| "eval_steps_per_second": 0.533, | |
| "step": 459 | |
| }, | |
| { | |
| "epoch": 12.026143790849673, | |
| "grad_norm": 1.4445286989212036, | |
| "learning_rate": 3.313840155945419e-05, | |
| "loss": 0.2537, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 12.287581699346406, | |
| "grad_norm": 3.5303471088409424, | |
| "learning_rate": 3.265107212475634e-05, | |
| "loss": 0.23, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 12.549019607843137, | |
| "grad_norm": 4.927082538604736, | |
| "learning_rate": 3.216374269005848e-05, | |
| "loss": 0.1693, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 12.81045751633987, | |
| "grad_norm": 1.5047080516815186, | |
| "learning_rate": 3.167641325536063e-05, | |
| "loss": 0.2374, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 12.993464052287582, | |
| "eval_accuracy": 0.875, | |
| "eval_loss": 0.42333418130874634, | |
| "eval_runtime": 32.071, | |
| "eval_samples_per_second": 4.241, | |
| "eval_steps_per_second": 0.53, | |
| "step": 497 | |
| }, | |
| { | |
| "epoch": 13.071895424836601, | |
| "grad_norm": 1.6010866165161133, | |
| "learning_rate": 3.118908382066277e-05, | |
| "loss": 0.2091, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 13.333333333333334, | |
| "grad_norm": 1.0056742429733276, | |
| "learning_rate": 3.0701754385964913e-05, | |
| "loss": 0.2203, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 13.594771241830065, | |
| "grad_norm": 0.8992928266525269, | |
| "learning_rate": 3.0214424951267055e-05, | |
| "loss": 0.1784, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 13.856209150326798, | |
| "grad_norm": 1.4664764404296875, | |
| "learning_rate": 2.9727095516569204e-05, | |
| "loss": 0.128, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 13.986928104575163, | |
| "eval_accuracy": 0.8970588235294118, | |
| "eval_loss": 0.367097944021225, | |
| "eval_runtime": 31.5031, | |
| "eval_samples_per_second": 4.317, | |
| "eval_steps_per_second": 0.54, | |
| "step": 535 | |
| }, | |
| { | |
| "epoch": 14.117647058823529, | |
| "grad_norm": 2.8378169536590576, | |
| "learning_rate": 2.9239766081871346e-05, | |
| "loss": 0.1436, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 14.379084967320262, | |
| "grad_norm": 1.6675957441329956, | |
| "learning_rate": 2.875243664717349e-05, | |
| "loss": 0.1986, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 14.640522875816993, | |
| "grad_norm": 8.45456314086914, | |
| "learning_rate": 2.8265107212475634e-05, | |
| "loss": 0.1711, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 14.901960784313726, | |
| "grad_norm": 6.542934417724609, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.1718, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 14.980392156862745, | |
| "eval_accuracy": 0.8970588235294118, | |
| "eval_loss": 0.34296926856040955, | |
| "eval_runtime": 31.8312, | |
| "eval_samples_per_second": 4.273, | |
| "eval_steps_per_second": 0.534, | |
| "step": 573 | |
| }, | |
| { | |
| "epoch": 15.163398692810457, | |
| "grad_norm": 4.025688171386719, | |
| "learning_rate": 2.729044834307992e-05, | |
| "loss": 0.1681, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 15.42483660130719, | |
| "grad_norm": 0.5761296153068542, | |
| "learning_rate": 2.680311890838207e-05, | |
| "loss": 0.1558, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 15.686274509803921, | |
| "grad_norm": 3.5217578411102295, | |
| "learning_rate": 2.6315789473684212e-05, | |
| "loss": 0.1413, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 15.947712418300654, | |
| "grad_norm": 5.222166538238525, | |
| "learning_rate": 2.5828460038986357e-05, | |
| "loss": 0.16, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.875, | |
| "eval_loss": 0.41041284799575806, | |
| "eval_runtime": 32.0124, | |
| "eval_samples_per_second": 4.248, | |
| "eval_steps_per_second": 0.531, | |
| "step": 612 | |
| }, | |
| { | |
| "epoch": 16.209150326797385, | |
| "grad_norm": 4.111217021942139, | |
| "learning_rate": 2.53411306042885e-05, | |
| "loss": 0.136, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 16.470588235294116, | |
| "grad_norm": 4.035433769226074, | |
| "learning_rate": 2.485380116959064e-05, | |
| "loss": 0.1369, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 16.73202614379085, | |
| "grad_norm": 1.081305980682373, | |
| "learning_rate": 2.4366471734892787e-05, | |
| "loss": 0.1331, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 16.99346405228758, | |
| "grad_norm": 0.21553057432174683, | |
| "learning_rate": 2.3879142300194932e-05, | |
| "loss": 0.1096, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 16.99346405228758, | |
| "eval_accuracy": 0.9117647058823529, | |
| "eval_loss": 0.2919999659061432, | |
| "eval_runtime": 32.5572, | |
| "eval_samples_per_second": 4.177, | |
| "eval_steps_per_second": 0.522, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 17.254901960784313, | |
| "grad_norm": 5.312492847442627, | |
| "learning_rate": 2.3391812865497074e-05, | |
| "loss": 0.0943, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 17.516339869281047, | |
| "grad_norm": 0.38019096851348877, | |
| "learning_rate": 2.290448343079922e-05, | |
| "loss": 0.1058, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 17.77777777777778, | |
| "grad_norm": 9.7896089553833, | |
| "learning_rate": 2.2417153996101365e-05, | |
| "loss": 0.1408, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 17.986928104575163, | |
| "eval_accuracy": 0.9044117647058824, | |
| "eval_loss": 0.26298093795776367, | |
| "eval_runtime": 31.7137, | |
| "eval_samples_per_second": 4.288, | |
| "eval_steps_per_second": 0.536, | |
| "step": 688 | |
| }, | |
| { | |
| "epoch": 18.03921568627451, | |
| "grad_norm": 1.3320598602294922, | |
| "learning_rate": 2.1929824561403507e-05, | |
| "loss": 0.1352, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 18.30065359477124, | |
| "grad_norm": 1.0650367736816406, | |
| "learning_rate": 2.1442495126705653e-05, | |
| "loss": 0.1149, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 18.562091503267975, | |
| "grad_norm": 0.5184645056724548, | |
| "learning_rate": 2.0955165692007798e-05, | |
| "loss": 0.1414, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 18.823529411764707, | |
| "grad_norm": 0.20009098947048187, | |
| "learning_rate": 2.046783625730994e-05, | |
| "loss": 0.113, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 18.980392156862745, | |
| "eval_accuracy": 0.8823529411764706, | |
| "eval_loss": 0.30839478969573975, | |
| "eval_runtime": 31.8589, | |
| "eval_samples_per_second": 4.269, | |
| "eval_steps_per_second": 0.534, | |
| "step": 726 | |
| }, | |
| { | |
| "epoch": 19.084967320261438, | |
| "grad_norm": 0.23899152874946594, | |
| "learning_rate": 1.9980506822612085e-05, | |
| "loss": 0.082, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 19.34640522875817, | |
| "grad_norm": 2.6471519470214844, | |
| "learning_rate": 1.949317738791423e-05, | |
| "loss": 0.0818, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 19.607843137254903, | |
| "grad_norm": 3.0529627799987793, | |
| "learning_rate": 1.9005847953216373e-05, | |
| "loss": 0.1271, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 19.869281045751634, | |
| "grad_norm": 0.2476678043603897, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.1272, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9411764705882353, | |
| "eval_loss": 0.2523283064365387, | |
| "eval_runtime": 31.804, | |
| "eval_samples_per_second": 4.276, | |
| "eval_steps_per_second": 0.535, | |
| "step": 765 | |
| }, | |
| { | |
| "epoch": 20.130718954248366, | |
| "grad_norm": 11.856123924255371, | |
| "learning_rate": 1.8031189083820664e-05, | |
| "loss": 0.1052, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 20.392156862745097, | |
| "grad_norm": 3.1671411991119385, | |
| "learning_rate": 1.7543859649122806e-05, | |
| "loss": 0.1264, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 20.65359477124183, | |
| "grad_norm": 0.7181591391563416, | |
| "learning_rate": 1.705653021442495e-05, | |
| "loss": 0.1141, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 20.915032679738562, | |
| "grad_norm": 3.510709762573242, | |
| "learning_rate": 1.6569200779727097e-05, | |
| "loss": 0.119, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 20.99346405228758, | |
| "eval_accuracy": 0.8823529411764706, | |
| "eval_loss": 0.42536023259162903, | |
| "eval_runtime": 32.0193, | |
| "eval_samples_per_second": 4.247, | |
| "eval_steps_per_second": 0.531, | |
| "step": 803 | |
| }, | |
| { | |
| "epoch": 21.176470588235293, | |
| "grad_norm": 2.389382839202881, | |
| "learning_rate": 1.608187134502924e-05, | |
| "loss": 0.1316, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 21.437908496732025, | |
| "grad_norm": 6.068481922149658, | |
| "learning_rate": 1.5594541910331384e-05, | |
| "loss": 0.1126, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 21.69934640522876, | |
| "grad_norm": 0.7834749817848206, | |
| "learning_rate": 1.5107212475633528e-05, | |
| "loss": 0.0782, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 21.96078431372549, | |
| "grad_norm": 2.6613547801971436, | |
| "learning_rate": 1.4619883040935673e-05, | |
| "loss": 0.1068, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 21.986928104575163, | |
| "eval_accuracy": 0.8970588235294118, | |
| "eval_loss": 0.35191574692726135, | |
| "eval_runtime": 31.8243, | |
| "eval_samples_per_second": 4.273, | |
| "eval_steps_per_second": 0.534, | |
| "step": 841 | |
| }, | |
| { | |
| "epoch": 22.22222222222222, | |
| "grad_norm": 2.7536206245422363, | |
| "learning_rate": 1.4132553606237817e-05, | |
| "loss": 0.1058, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 22.483660130718953, | |
| "grad_norm": 5.085578441619873, | |
| "learning_rate": 1.364522417153996e-05, | |
| "loss": 0.0847, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 22.745098039215687, | |
| "grad_norm": 1.596222996711731, | |
| "learning_rate": 1.3157894736842106e-05, | |
| "loss": 0.0723, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 22.980392156862745, | |
| "eval_accuracy": 0.9191176470588235, | |
| "eval_loss": 0.32932135462760925, | |
| "eval_runtime": 33.5379, | |
| "eval_samples_per_second": 4.055, | |
| "eval_steps_per_second": 0.507, | |
| "step": 879 | |
| }, | |
| { | |
| "epoch": 23.00653594771242, | |
| "grad_norm": 1.2901531457901, | |
| "learning_rate": 1.267056530214425e-05, | |
| "loss": 0.0937, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 23.26797385620915, | |
| "grad_norm": 0.6107073426246643, | |
| "learning_rate": 1.2183235867446393e-05, | |
| "loss": 0.076, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 23.529411764705884, | |
| "grad_norm": 0.1317284107208252, | |
| "learning_rate": 1.1695906432748537e-05, | |
| "loss": 0.0742, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 23.790849673202615, | |
| "grad_norm": 1.982415795326233, | |
| "learning_rate": 1.1208576998050683e-05, | |
| "loss": 0.0769, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.9264705882352942, | |
| "eval_loss": 0.2613476812839508, | |
| "eval_runtime": 32.3164, | |
| "eval_samples_per_second": 4.208, | |
| "eval_steps_per_second": 0.526, | |
| "step": 918 | |
| }, | |
| { | |
| "epoch": 24.052287581699346, | |
| "grad_norm": 0.13533662259578705, | |
| "learning_rate": 1.0721247563352826e-05, | |
| "loss": 0.0872, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 24.313725490196077, | |
| "grad_norm": 0.18768663704395294, | |
| "learning_rate": 1.023391812865497e-05, | |
| "loss": 0.0836, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 24.575163398692812, | |
| "grad_norm": 1.9041831493377686, | |
| "learning_rate": 9.746588693957115e-06, | |
| "loss": 0.0968, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 24.836601307189543, | |
| "grad_norm": 1.782492756843567, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 0.095, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 24.99346405228758, | |
| "eval_accuracy": 0.9411764705882353, | |
| "eval_loss": 0.26091232895851135, | |
| "eval_runtime": 32.1423, | |
| "eval_samples_per_second": 4.231, | |
| "eval_steps_per_second": 0.529, | |
| "step": 956 | |
| }, | |
| { | |
| "epoch": 25.098039215686274, | |
| "grad_norm": 0.38502180576324463, | |
| "learning_rate": 8.771929824561403e-06, | |
| "loss": 0.0871, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 25.359477124183005, | |
| "grad_norm": 7.287825584411621, | |
| "learning_rate": 8.284600389863548e-06, | |
| "loss": 0.0662, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 25.62091503267974, | |
| "grad_norm": 3.129642963409424, | |
| "learning_rate": 7.797270955165692e-06, | |
| "loss": 0.0506, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 25.88235294117647, | |
| "grad_norm": 0.24554868042469025, | |
| "learning_rate": 7.3099415204678366e-06, | |
| "loss": 0.0863, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 25.986928104575163, | |
| "eval_accuracy": 0.9264705882352942, | |
| "eval_loss": 0.264960914850235, | |
| "eval_runtime": 32.1045, | |
| "eval_samples_per_second": 4.236, | |
| "eval_steps_per_second": 0.53, | |
| "step": 994 | |
| }, | |
| { | |
| "epoch": 26.143790849673202, | |
| "grad_norm": 1.0143533945083618, | |
| "learning_rate": 6.82261208576998e-06, | |
| "loss": 0.0604, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 26.405228758169933, | |
| "grad_norm": 0.31393057107925415, | |
| "learning_rate": 6.335282651072125e-06, | |
| "loss": 0.0589, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 26.666666666666668, | |
| "grad_norm": 5.615325450897217, | |
| "learning_rate": 5.8479532163742686e-06, | |
| "loss": 0.0654, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 26.9281045751634, | |
| "grad_norm": 0.4632560610771179, | |
| "learning_rate": 5.360623781676413e-06, | |
| "loss": 0.0795, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 26.980392156862745, | |
| "eval_accuracy": 0.9117647058823529, | |
| "eval_loss": 0.297785222530365, | |
| "eval_runtime": 31.5613, | |
| "eval_samples_per_second": 4.309, | |
| "eval_steps_per_second": 0.539, | |
| "step": 1032 | |
| }, | |
| { | |
| "epoch": 27.18954248366013, | |
| "grad_norm": 0.2585015594959259, | |
| "learning_rate": 4.873294346978558e-06, | |
| "loss": 0.0719, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 27.45098039215686, | |
| "grad_norm": 0.364827960729599, | |
| "learning_rate": 4.3859649122807014e-06, | |
| "loss": 0.0837, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 27.712418300653596, | |
| "grad_norm": 0.14460325241088867, | |
| "learning_rate": 3.898635477582846e-06, | |
| "loss": 0.0483, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 27.973856209150327, | |
| "grad_norm": 0.43661826848983765, | |
| "learning_rate": 3.41130604288499e-06, | |
| "loss": 0.0564, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.9191176470588235, | |
| "eval_loss": 0.2736634314060211, | |
| "eval_runtime": 31.9037, | |
| "eval_samples_per_second": 4.263, | |
| "eval_steps_per_second": 0.533, | |
| "step": 1071 | |
| }, | |
| { | |
| "epoch": 28.235294117647058, | |
| "grad_norm": 0.2871550917625427, | |
| "learning_rate": 2.9239766081871343e-06, | |
| "loss": 0.0676, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 28.49673202614379, | |
| "grad_norm": 0.19494622945785522, | |
| "learning_rate": 2.436647173489279e-06, | |
| "loss": 0.055, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 28.758169934640524, | |
| "grad_norm": 0.46231183409690857, | |
| "learning_rate": 1.949317738791423e-06, | |
| "loss": 0.0562, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 28.99346405228758, | |
| "eval_accuracy": 0.9191176470588235, | |
| "eval_loss": 0.2940830886363983, | |
| "eval_runtime": 32.3077, | |
| "eval_samples_per_second": 4.21, | |
| "eval_steps_per_second": 0.526, | |
| "step": 1109 | |
| }, | |
| { | |
| "epoch": 29.019607843137255, | |
| "grad_norm": 0.13441592454910278, | |
| "learning_rate": 1.4619883040935671e-06, | |
| "loss": 0.049, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 29.281045751633986, | |
| "grad_norm": 5.47345495223999, | |
| "learning_rate": 9.746588693957115e-07, | |
| "loss": 0.0712, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 29.54248366013072, | |
| "grad_norm": 0.6587017178535461, | |
| "learning_rate": 4.873294346978557e-07, | |
| "loss": 0.0679, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 29.80392156862745, | |
| "grad_norm": 0.12627951800823212, | |
| "learning_rate": 0.0, | |
| "loss": 0.0751, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 29.80392156862745, | |
| "eval_accuracy": 0.9191176470588235, | |
| "eval_loss": 0.3110995590686798, | |
| "eval_runtime": 33.2625, | |
| "eval_samples_per_second": 4.089, | |
| "eval_steps_per_second": 0.511, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 29.80392156862745, | |
| "step": 1140, | |
| "total_flos": 2.827009127861453e+18, | |
| "train_loss": 0.3311853929046999, | |
| "train_runtime": 10131.822, | |
| "train_samples_per_second": 3.624, | |
| "train_steps_per_second": 0.113 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1140, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 30, | |
| "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": 2.827009127861453e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |