| { | |
| "best_metric": 0.9234042553191489, | |
| "best_model_checkpoint": "vit-msn-base-finetuned-lf-invalidation/checkpoint-62", | |
| "epoch": 76.8, | |
| "eval_steps": 500, | |
| "global_step": 480, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.96, | |
| "eval_accuracy": 0.6957446808510638, | |
| "eval_loss": 0.6511951088905334, | |
| "eval_runtime": 4.1128, | |
| "eval_samples_per_second": 114.277, | |
| "eval_steps_per_second": 3.647, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 100.99421691894531, | |
| "learning_rate": 1.0416666666666668e-05, | |
| "loss": 0.7053, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "eval_accuracy": 0.6808510638297872, | |
| "eval_loss": 0.6310930848121643, | |
| "eval_runtime": 4.1662, | |
| "eval_samples_per_second": 112.813, | |
| "eval_steps_per_second": 3.6, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "eval_accuracy": 0.7276595744680852, | |
| "eval_loss": 0.5360996127128601, | |
| "eval_runtime": 4.0581, | |
| "eval_samples_per_second": 115.819, | |
| "eval_steps_per_second": 3.696, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 14.486075401306152, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 0.5163, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8680851063829788, | |
| "eval_loss": 0.3341110050678253, | |
| "eval_runtime": 3.919, | |
| "eval_samples_per_second": 119.928, | |
| "eval_steps_per_second": 3.827, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 19.806690216064453, | |
| "learning_rate": 3.125e-05, | |
| "loss": 0.3242, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "eval_accuracy": 0.8808510638297873, | |
| "eval_loss": 0.3167176842689514, | |
| "eval_runtime": 3.8854, | |
| "eval_samples_per_second": 120.965, | |
| "eval_steps_per_second": 3.861, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 5.92, | |
| "eval_accuracy": 0.8191489361702128, | |
| "eval_loss": 0.39598795771598816, | |
| "eval_runtime": 4.17, | |
| "eval_samples_per_second": 112.71, | |
| "eval_steps_per_second": 3.597, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "grad_norm": 18.975563049316406, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.2779, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 6.88, | |
| "eval_accuracy": 0.825531914893617, | |
| "eval_loss": 0.3817645013332367, | |
| "eval_runtime": 3.9864, | |
| "eval_samples_per_second": 117.9, | |
| "eval_steps_per_second": 3.763, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 50.12718963623047, | |
| "learning_rate": 4.976851851851852e-05, | |
| "loss": 0.2348, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7361702127659574, | |
| "eval_loss": 0.5018748641014099, | |
| "eval_runtime": 4.0244, | |
| "eval_samples_per_second": 116.789, | |
| "eval_steps_per_second": 3.727, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 8.96, | |
| "eval_accuracy": 0.8851063829787233, | |
| "eval_loss": 0.29437732696533203, | |
| "eval_runtime": 3.863, | |
| "eval_samples_per_second": 121.668, | |
| "eval_steps_per_second": 3.883, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "grad_norm": 87.47682189941406, | |
| "learning_rate": 4.8611111111111115e-05, | |
| "loss": 0.26, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 9.92, | |
| "eval_accuracy": 0.9234042553191489, | |
| "eval_loss": 0.24138842523097992, | |
| "eval_runtime": 3.9931, | |
| "eval_samples_per_second": 117.702, | |
| "eval_steps_per_second": 3.756, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 10.88, | |
| "eval_accuracy": 0.8297872340425532, | |
| "eval_loss": 0.36644989252090454, | |
| "eval_runtime": 3.8761, | |
| "eval_samples_per_second": 121.257, | |
| "eval_steps_per_second": 3.87, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 11.2, | |
| "grad_norm": 37.262699127197266, | |
| "learning_rate": 4.745370370370371e-05, | |
| "loss": 0.2778, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.9042553191489362, | |
| "eval_loss": 0.2505495548248291, | |
| "eval_runtime": 3.8837, | |
| "eval_samples_per_second": 121.018, | |
| "eval_steps_per_second": 3.862, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 12.8, | |
| "grad_norm": 39.793601989746094, | |
| "learning_rate": 4.62962962962963e-05, | |
| "loss": 0.2271, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 12.96, | |
| "eval_accuracy": 0.6297872340425532, | |
| "eval_loss": 0.6277480721473694, | |
| "eval_runtime": 3.9373, | |
| "eval_samples_per_second": 119.372, | |
| "eval_steps_per_second": 3.81, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 13.92, | |
| "eval_accuracy": 0.874468085106383, | |
| "eval_loss": 0.275258332490921, | |
| "eval_runtime": 3.9907, | |
| "eval_samples_per_second": 117.773, | |
| "eval_steps_per_second": 3.759, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 14.4, | |
| "grad_norm": 45.56210708618164, | |
| "learning_rate": 4.5138888888888894e-05, | |
| "loss": 0.2488, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 14.88, | |
| "eval_accuracy": 0.6957446808510638, | |
| "eval_loss": 0.6249393820762634, | |
| "eval_runtime": 4.032, | |
| "eval_samples_per_second": 116.567, | |
| "eval_steps_per_second": 3.72, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 39.393192291259766, | |
| "learning_rate": 4.3981481481481486e-05, | |
| "loss": 0.2729, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.7148936170212766, | |
| "eval_loss": 0.519493043422699, | |
| "eval_runtime": 4.1537, | |
| "eval_samples_per_second": 113.151, | |
| "eval_steps_per_second": 3.611, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 16.96, | |
| "eval_accuracy": 0.574468085106383, | |
| "eval_loss": 0.7983953952789307, | |
| "eval_runtime": 4.0505, | |
| "eval_samples_per_second": 116.036, | |
| "eval_steps_per_second": 3.703, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 17.6, | |
| "grad_norm": 52.11545944213867, | |
| "learning_rate": 4.282407407407408e-05, | |
| "loss": 0.3261, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 17.92, | |
| "eval_accuracy": 0.7723404255319148, | |
| "eval_loss": 0.4630971848964691, | |
| "eval_runtime": 4.0296, | |
| "eval_samples_per_second": 116.636, | |
| "eval_steps_per_second": 3.722, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 18.88, | |
| "eval_accuracy": 0.5148936170212766, | |
| "eval_loss": 1.100958228111267, | |
| "eval_runtime": 4.0952, | |
| "eval_samples_per_second": 114.768, | |
| "eval_steps_per_second": 3.663, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 19.2, | |
| "grad_norm": 80.0373764038086, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.2212, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9170212765957447, | |
| "eval_loss": 0.23374585807323456, | |
| "eval_runtime": 4.0961, | |
| "eval_samples_per_second": 114.744, | |
| "eval_steps_per_second": 3.662, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 20.8, | |
| "grad_norm": 2.179731607437134, | |
| "learning_rate": 4.0509259259259265e-05, | |
| "loss": 0.2802, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 20.96, | |
| "eval_accuracy": 0.7574468085106383, | |
| "eval_loss": 0.46376925706863403, | |
| "eval_runtime": 4.1103, | |
| "eval_samples_per_second": 114.347, | |
| "eval_steps_per_second": 3.649, | |
| "step": 131 | |
| }, | |
| { | |
| "epoch": 21.92, | |
| "eval_accuracy": 0.8361702127659575, | |
| "eval_loss": 0.38592880964279175, | |
| "eval_runtime": 4.1405, | |
| "eval_samples_per_second": 113.513, | |
| "eval_steps_per_second": 3.623, | |
| "step": 137 | |
| }, | |
| { | |
| "epoch": 22.4, | |
| "grad_norm": 4.62510871887207, | |
| "learning_rate": 3.935185185185186e-05, | |
| "loss": 0.2112, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 22.88, | |
| "eval_accuracy": 0.6893617021276596, | |
| "eval_loss": 0.6708246469497681, | |
| "eval_runtime": 4.1195, | |
| "eval_samples_per_second": 114.091, | |
| "eval_steps_per_second": 3.641, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "grad_norm": 5.395915508270264, | |
| "learning_rate": 3.8194444444444444e-05, | |
| "loss": 0.2231, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.8680851063829788, | |
| "eval_loss": 0.3386794626712799, | |
| "eval_runtime": 4.0237, | |
| "eval_samples_per_second": 116.808, | |
| "eval_steps_per_second": 3.728, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 24.96, | |
| "eval_accuracy": 0.6553191489361702, | |
| "eval_loss": 0.7044735550880432, | |
| "eval_runtime": 4.1207, | |
| "eval_samples_per_second": 114.059, | |
| "eval_steps_per_second": 3.64, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 25.6, | |
| "grad_norm": 12.411273002624512, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.2037, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 25.92, | |
| "eval_accuracy": 0.8276595744680851, | |
| "eval_loss": 0.3957701325416565, | |
| "eval_runtime": 4.0677, | |
| "eval_samples_per_second": 115.543, | |
| "eval_steps_per_second": 3.688, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 26.88, | |
| "eval_accuracy": 0.7702127659574468, | |
| "eval_loss": 0.5082454681396484, | |
| "eval_runtime": 4.0429, | |
| "eval_samples_per_second": 116.254, | |
| "eval_steps_per_second": 3.71, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 27.2, | |
| "grad_norm": 13.522443771362305, | |
| "learning_rate": 3.587962962962963e-05, | |
| "loss": 0.1845, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.723404255319149, | |
| "eval_loss": 0.5990515351295471, | |
| "eval_runtime": 4.0619, | |
| "eval_samples_per_second": 115.71, | |
| "eval_steps_per_second": 3.693, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 28.8, | |
| "grad_norm": 40.79707336425781, | |
| "learning_rate": 3.472222222222222e-05, | |
| "loss": 0.1898, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 28.96, | |
| "eval_accuracy": 0.7617021276595745, | |
| "eval_loss": 0.510837197303772, | |
| "eval_runtime": 4.1139, | |
| "eval_samples_per_second": 114.248, | |
| "eval_steps_per_second": 3.646, | |
| "step": 181 | |
| }, | |
| { | |
| "epoch": 29.92, | |
| "eval_accuracy": 0.9085106382978724, | |
| "eval_loss": 0.27203500270843506, | |
| "eval_runtime": 4.1071, | |
| "eval_samples_per_second": 114.435, | |
| "eval_steps_per_second": 3.652, | |
| "step": 187 | |
| }, | |
| { | |
| "epoch": 30.4, | |
| "grad_norm": 34.6284065246582, | |
| "learning_rate": 3.3564814814814815e-05, | |
| "loss": 0.2118, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 30.88, | |
| "eval_accuracy": 0.7851063829787234, | |
| "eval_loss": 0.4935612976551056, | |
| "eval_runtime": 4.1447, | |
| "eval_samples_per_second": 113.398, | |
| "eval_steps_per_second": 3.619, | |
| "step": 193 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "grad_norm": 5.604860782623291, | |
| "learning_rate": 3.240740740740741e-05, | |
| "loss": 0.2097, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.8404255319148937, | |
| "eval_loss": 0.37482374906539917, | |
| "eval_runtime": 4.0858, | |
| "eval_samples_per_second": 115.032, | |
| "eval_steps_per_second": 3.671, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 32.96, | |
| "eval_accuracy": 0.776595744680851, | |
| "eval_loss": 0.5048179626464844, | |
| "eval_runtime": 4.0089, | |
| "eval_samples_per_second": 117.24, | |
| "eval_steps_per_second": 3.742, | |
| "step": 206 | |
| }, | |
| { | |
| "epoch": 33.6, | |
| "grad_norm": 7.735608100891113, | |
| "learning_rate": 3.125e-05, | |
| "loss": 0.1704, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 33.92, | |
| "eval_accuracy": 0.7957446808510639, | |
| "eval_loss": 0.43682861328125, | |
| "eval_runtime": 4.0913, | |
| "eval_samples_per_second": 114.879, | |
| "eval_steps_per_second": 3.666, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 34.88, | |
| "eval_accuracy": 0.6829787234042554, | |
| "eval_loss": 0.6958675384521484, | |
| "eval_runtime": 4.1104, | |
| "eval_samples_per_second": 114.345, | |
| "eval_steps_per_second": 3.649, | |
| "step": 218 | |
| }, | |
| { | |
| "epoch": 35.2, | |
| "grad_norm": 25.477895736694336, | |
| "learning_rate": 3.0092592592592593e-05, | |
| "loss": 0.1962, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.5957446808510638, | |
| "eval_loss": 1.009740948677063, | |
| "eval_runtime": 4.0288, | |
| "eval_samples_per_second": 116.66, | |
| "eval_steps_per_second": 3.723, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 36.8, | |
| "grad_norm": 7.080097198486328, | |
| "learning_rate": 2.8935185185185186e-05, | |
| "loss": 0.1686, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 36.96, | |
| "eval_accuracy": 0.7914893617021277, | |
| "eval_loss": 0.4992178976535797, | |
| "eval_runtime": 4.0814, | |
| "eval_samples_per_second": 115.157, | |
| "eval_steps_per_second": 3.675, | |
| "step": 231 | |
| }, | |
| { | |
| "epoch": 37.92, | |
| "eval_accuracy": 0.7574468085106383, | |
| "eval_loss": 0.5373654365539551, | |
| "eval_runtime": 4.2322, | |
| "eval_samples_per_second": 111.052, | |
| "eval_steps_per_second": 3.544, | |
| "step": 237 | |
| }, | |
| { | |
| "epoch": 38.4, | |
| "grad_norm": 39.29030227661133, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.1855, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 38.88, | |
| "eval_accuracy": 0.8340425531914893, | |
| "eval_loss": 0.371025025844574, | |
| "eval_runtime": 4.1514, | |
| "eval_samples_per_second": 113.216, | |
| "eval_steps_per_second": 3.613, | |
| "step": 243 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 21.52515983581543, | |
| "learning_rate": 2.6620370370370372e-05, | |
| "loss": 0.1528, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.8446808510638298, | |
| "eval_loss": 0.3630984425544739, | |
| "eval_runtime": 4.1723, | |
| "eval_samples_per_second": 112.647, | |
| "eval_steps_per_second": 3.595, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 40.96, | |
| "eval_accuracy": 0.7680851063829788, | |
| "eval_loss": 0.5588864088058472, | |
| "eval_runtime": 4.2314, | |
| "eval_samples_per_second": 111.075, | |
| "eval_steps_per_second": 3.545, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 41.6, | |
| "grad_norm": 2.9336180686950684, | |
| "learning_rate": 2.5462962962962965e-05, | |
| "loss": 0.1523, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 41.92, | |
| "eval_accuracy": 0.7808510638297872, | |
| "eval_loss": 0.5147323608398438, | |
| "eval_runtime": 4.1942, | |
| "eval_samples_per_second": 112.059, | |
| "eval_steps_per_second": 3.576, | |
| "step": 262 | |
| }, | |
| { | |
| "epoch": 42.88, | |
| "eval_accuracy": 0.7638297872340426, | |
| "eval_loss": 0.5298714637756348, | |
| "eval_runtime": 4.0956, | |
| "eval_samples_per_second": 114.756, | |
| "eval_steps_per_second": 3.662, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 43.2, | |
| "grad_norm": 20.56193733215332, | |
| "learning_rate": 2.4305555555555558e-05, | |
| "loss": 0.1709, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.7446808510638298, | |
| "eval_loss": 0.5937234163284302, | |
| "eval_runtime": 4.0352, | |
| "eval_samples_per_second": 116.474, | |
| "eval_steps_per_second": 3.717, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 44.8, | |
| "grad_norm": 17.483304977416992, | |
| "learning_rate": 2.314814814814815e-05, | |
| "loss": 0.1527, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 44.96, | |
| "eval_accuracy": 0.7382978723404255, | |
| "eval_loss": 0.5969159603118896, | |
| "eval_runtime": 4.1383, | |
| "eval_samples_per_second": 113.574, | |
| "eval_steps_per_second": 3.625, | |
| "step": 281 | |
| }, | |
| { | |
| "epoch": 45.92, | |
| "eval_accuracy": 0.725531914893617, | |
| "eval_loss": 0.6439131498336792, | |
| "eval_runtime": 4.1256, | |
| "eval_samples_per_second": 113.922, | |
| "eval_steps_per_second": 3.636, | |
| "step": 287 | |
| }, | |
| { | |
| "epoch": 46.4, | |
| "grad_norm": 13.123701095581055, | |
| "learning_rate": 2.1990740740740743e-05, | |
| "loss": 0.1397, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 46.88, | |
| "eval_accuracy": 0.6723404255319149, | |
| "eval_loss": 0.7720506191253662, | |
| "eval_runtime": 4.0907, | |
| "eval_samples_per_second": 114.894, | |
| "eval_steps_per_second": 3.667, | |
| "step": 293 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "grad_norm": 15.003984451293945, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 0.1538, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.7702127659574468, | |
| "eval_loss": 0.5767794251441956, | |
| "eval_runtime": 4.0083, | |
| "eval_samples_per_second": 117.257, | |
| "eval_steps_per_second": 3.742, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 48.96, | |
| "eval_accuracy": 0.7595744680851064, | |
| "eval_loss": 0.5801470875740051, | |
| "eval_runtime": 3.9427, | |
| "eval_samples_per_second": 119.209, | |
| "eval_steps_per_second": 3.805, | |
| "step": 306 | |
| }, | |
| { | |
| "epoch": 49.6, | |
| "grad_norm": 2.649744987487793, | |
| "learning_rate": 1.967592592592593e-05, | |
| "loss": 0.1466, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 49.92, | |
| "eval_accuracy": 0.7574468085106383, | |
| "eval_loss": 0.5672721266746521, | |
| "eval_runtime": 4.0569, | |
| "eval_samples_per_second": 115.852, | |
| "eval_steps_per_second": 3.697, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 50.88, | |
| "eval_accuracy": 0.7085106382978723, | |
| "eval_loss": 0.6468719244003296, | |
| "eval_runtime": 4.0234, | |
| "eval_samples_per_second": 116.818, | |
| "eval_steps_per_second": 3.728, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 51.2, | |
| "grad_norm": 12.5094633102417, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.1302, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.6957446808510638, | |
| "eval_loss": 0.7276235222816467, | |
| "eval_runtime": 4.0008, | |
| "eval_samples_per_second": 117.475, | |
| "eval_steps_per_second": 3.749, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 52.8, | |
| "grad_norm": 6.599560737609863, | |
| "learning_rate": 1.736111111111111e-05, | |
| "loss": 0.1565, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 52.96, | |
| "eval_accuracy": 0.6723404255319149, | |
| "eval_loss": 0.8247136473655701, | |
| "eval_runtime": 4.0526, | |
| "eval_samples_per_second": 115.976, | |
| "eval_steps_per_second": 3.701, | |
| "step": 331 | |
| }, | |
| { | |
| "epoch": 53.92, | |
| "eval_accuracy": 0.7978723404255319, | |
| "eval_loss": 0.4810582399368286, | |
| "eval_runtime": 4.0005, | |
| "eval_samples_per_second": 117.486, | |
| "eval_steps_per_second": 3.75, | |
| "step": 337 | |
| }, | |
| { | |
| "epoch": 54.4, | |
| "grad_norm": 12.574357986450195, | |
| "learning_rate": 1.6203703703703704e-05, | |
| "loss": 0.1267, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 54.88, | |
| "eval_accuracy": 0.7021276595744681, | |
| "eval_loss": 0.6372675895690918, | |
| "eval_runtime": 4.047, | |
| "eval_samples_per_second": 116.135, | |
| "eval_steps_per_second": 3.706, | |
| "step": 343 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "grad_norm": 29.667768478393555, | |
| "learning_rate": 1.5046296296296297e-05, | |
| "loss": 0.1424, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.6723404255319149, | |
| "eval_loss": 0.7251705527305603, | |
| "eval_runtime": 4.0519, | |
| "eval_samples_per_second": 115.995, | |
| "eval_steps_per_second": 3.702, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 56.96, | |
| "eval_accuracy": 0.7489361702127659, | |
| "eval_loss": 0.5696622729301453, | |
| "eval_runtime": 4.0808, | |
| "eval_samples_per_second": 115.174, | |
| "eval_steps_per_second": 3.676, | |
| "step": 356 | |
| }, | |
| { | |
| "epoch": 57.6, | |
| "grad_norm": 3.276287317276001, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.1053, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 57.92, | |
| "eval_accuracy": 0.6957446808510638, | |
| "eval_loss": 0.7066917419433594, | |
| "eval_runtime": 4.0845, | |
| "eval_samples_per_second": 115.07, | |
| "eval_steps_per_second": 3.672, | |
| "step": 362 | |
| }, | |
| { | |
| "epoch": 58.88, | |
| "eval_accuracy": 0.7063829787234043, | |
| "eval_loss": 0.6576955318450928, | |
| "eval_runtime": 4.1852, | |
| "eval_samples_per_second": 112.301, | |
| "eval_steps_per_second": 3.584, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 59.2, | |
| "grad_norm": 7.9189982414245605, | |
| "learning_rate": 1.2731481481481482e-05, | |
| "loss": 0.1301, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_accuracy": 0.774468085106383, | |
| "eval_loss": 0.5325801372528076, | |
| "eval_runtime": 4.0787, | |
| "eval_samples_per_second": 115.233, | |
| "eval_steps_per_second": 3.678, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 60.8, | |
| "grad_norm": 14.686637878417969, | |
| "learning_rate": 1.1574074074074075e-05, | |
| "loss": 0.0906, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 60.96, | |
| "eval_accuracy": 0.7851063829787234, | |
| "eval_loss": 0.546753466129303, | |
| "eval_runtime": 4.0812, | |
| "eval_samples_per_second": 115.163, | |
| "eval_steps_per_second": 3.675, | |
| "step": 381 | |
| }, | |
| { | |
| "epoch": 61.92, | |
| "eval_accuracy": 0.8276595744680851, | |
| "eval_loss": 0.4413163959980011, | |
| "eval_runtime": 4.1408, | |
| "eval_samples_per_second": 113.504, | |
| "eval_steps_per_second": 3.622, | |
| "step": 387 | |
| }, | |
| { | |
| "epoch": 62.4, | |
| "grad_norm": 9.14445686340332, | |
| "learning_rate": 1.0416666666666668e-05, | |
| "loss": 0.0974, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 62.88, | |
| "eval_accuracy": 0.7659574468085106, | |
| "eval_loss": 0.5478885173797607, | |
| "eval_runtime": 4.1286, | |
| "eval_samples_per_second": 113.839, | |
| "eval_steps_per_second": 3.633, | |
| "step": 393 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "grad_norm": 4.1058526039123535, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 0.1133, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_accuracy": 0.7042553191489361, | |
| "eval_loss": 0.7109193801879883, | |
| "eval_runtime": 4.2149, | |
| "eval_samples_per_second": 111.508, | |
| "eval_steps_per_second": 3.559, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 64.96, | |
| "eval_accuracy": 0.7617021276595745, | |
| "eval_loss": 0.5734679102897644, | |
| "eval_runtime": 4.1133, | |
| "eval_samples_per_second": 114.265, | |
| "eval_steps_per_second": 3.647, | |
| "step": 406 | |
| }, | |
| { | |
| "epoch": 65.6, | |
| "grad_norm": 5.876250267028809, | |
| "learning_rate": 8.101851851851852e-06, | |
| "loss": 0.1189, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 65.92, | |
| "eval_accuracy": 0.8297872340425532, | |
| "eval_loss": 0.4084050953388214, | |
| "eval_runtime": 4.1883, | |
| "eval_samples_per_second": 112.218, | |
| "eval_steps_per_second": 3.581, | |
| "step": 412 | |
| }, | |
| { | |
| "epoch": 66.88, | |
| "eval_accuracy": 0.7489361702127659, | |
| "eval_loss": 0.5716192722320557, | |
| "eval_runtime": 4.1124, | |
| "eval_samples_per_second": 114.289, | |
| "eval_steps_per_second": 3.648, | |
| "step": 418 | |
| }, | |
| { | |
| "epoch": 67.2, | |
| "grad_norm": 2.931035280227661, | |
| "learning_rate": 6.944444444444445e-06, | |
| "loss": 0.1064, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "eval_accuracy": 0.7553191489361702, | |
| "eval_loss": 0.5537174940109253, | |
| "eval_runtime": 4.0965, | |
| "eval_samples_per_second": 114.731, | |
| "eval_steps_per_second": 3.662, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 68.8, | |
| "grad_norm": 4.387136936187744, | |
| "learning_rate": 5.787037037037038e-06, | |
| "loss": 0.1084, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 68.96, | |
| "eval_accuracy": 0.8021276595744681, | |
| "eval_loss": 0.456912100315094, | |
| "eval_runtime": 4.1477, | |
| "eval_samples_per_second": 113.315, | |
| "eval_steps_per_second": 3.616, | |
| "step": 431 | |
| }, | |
| { | |
| "epoch": 69.92, | |
| "eval_accuracy": 0.7617021276595745, | |
| "eval_loss": 0.5227068066596985, | |
| "eval_runtime": 4.1656, | |
| "eval_samples_per_second": 112.828, | |
| "eval_steps_per_second": 3.601, | |
| "step": 437 | |
| }, | |
| { | |
| "epoch": 70.4, | |
| "grad_norm": 6.693394184112549, | |
| "learning_rate": 4.6296296296296296e-06, | |
| "loss": 0.1054, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 70.88, | |
| "eval_accuracy": 0.7276595744680852, | |
| "eval_loss": 0.5995042324066162, | |
| "eval_runtime": 4.1654, | |
| "eval_samples_per_second": 112.834, | |
| "eval_steps_per_second": 3.601, | |
| "step": 443 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "grad_norm": 8.600502014160156, | |
| "learning_rate": 3.4722222222222224e-06, | |
| "loss": 0.1005, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_accuracy": 0.7638297872340426, | |
| "eval_loss": 0.5560170412063599, | |
| "eval_runtime": 4.1827, | |
| "eval_samples_per_second": 112.367, | |
| "eval_steps_per_second": 3.586, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 72.96, | |
| "eval_accuracy": 0.8063829787234043, | |
| "eval_loss": 0.45502665638923645, | |
| "eval_runtime": 4.2071, | |
| "eval_samples_per_second": 111.715, | |
| "eval_steps_per_second": 3.565, | |
| "step": 456 | |
| }, | |
| { | |
| "epoch": 73.6, | |
| "grad_norm": 10.198132514953613, | |
| "learning_rate": 2.3148148148148148e-06, | |
| "loss": 0.1028, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 73.92, | |
| "eval_accuracy": 0.823404255319149, | |
| "eval_loss": 0.4404470920562744, | |
| "eval_runtime": 4.1806, | |
| "eval_samples_per_second": 112.425, | |
| "eval_steps_per_second": 3.588, | |
| "step": 462 | |
| }, | |
| { | |
| "epoch": 74.88, | |
| "eval_accuracy": 0.7957446808510639, | |
| "eval_loss": 0.4761447310447693, | |
| "eval_runtime": 4.1871, | |
| "eval_samples_per_second": 112.251, | |
| "eval_steps_per_second": 3.582, | |
| "step": 468 | |
| }, | |
| { | |
| "epoch": 75.2, | |
| "grad_norm": 7.506448268890381, | |
| "learning_rate": 1.1574074074074074e-06, | |
| "loss": 0.0917, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "eval_accuracy": 0.7680851063829788, | |
| "eval_loss": 0.5278272032737732, | |
| "eval_runtime": 4.1615, | |
| "eval_samples_per_second": 112.939, | |
| "eval_steps_per_second": 3.604, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 76.8, | |
| "grad_norm": 5.474030494689941, | |
| "learning_rate": 0.0, | |
| "loss": 0.1009, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 76.8, | |
| "eval_accuracy": 0.7617021276595745, | |
| "eval_loss": 0.5345979332923889, | |
| "eval_runtime": 4.1813, | |
| "eval_samples_per_second": 112.405, | |
| "eval_steps_per_second": 3.587, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 76.8, | |
| "step": 480, | |
| "total_flos": 4.5903154968099717e+18, | |
| "train_loss": 0.19423907659947873, | |
| "train_runtime": 1687.359, | |
| "train_samples_per_second": 36.554, | |
| "train_steps_per_second": 0.284 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 480, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 80, | |
| "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": 4.5903154968099717e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |