| { | |
| "best_metric": 0.7371370351502802, | |
| "best_model_checkpoint": "/home/user/emrecan/models/dbmdz_distilbert-base-turkish-cased_allnli_tr/checkpoint-80000", | |
| "epoch": 3.0, | |
| "global_step": 88320, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 1.9773550724637682e-05, | |
| "loss": 0.94, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "eval_accuracy": 0.5812531839021905, | |
| "eval_loss": 0.9074209928512573, | |
| "eval_runtime": 19.0819, | |
| "eval_samples_per_second": 514.361, | |
| "eval_steps_per_second": 16.089, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 1.9547101449275363e-05, | |
| "loss": 0.8102, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_accuracy": 0.5949057564951604, | |
| "eval_loss": 0.8801841139793396, | |
| "eval_runtime": 18.8995, | |
| "eval_samples_per_second": 519.327, | |
| "eval_steps_per_second": 16.244, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 1.9320652173913047e-05, | |
| "loss": 0.7737, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_accuracy": 0.6154865002547122, | |
| "eval_loss": 0.8491194248199463, | |
| "eval_runtime": 18.862, | |
| "eval_samples_per_second": 520.36, | |
| "eval_steps_per_second": 16.276, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 1.9094202898550727e-05, | |
| "loss": 0.7576, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "eval_accuracy": 0.6260825267447784, | |
| "eval_loss": 0.8283059000968933, | |
| "eval_runtime": 19.0535, | |
| "eval_samples_per_second": 515.129, | |
| "eval_steps_per_second": 16.113, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 1.8867753623188408e-05, | |
| "loss": 0.7286, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "eval_accuracy": 0.6361691288843607, | |
| "eval_loss": 0.8149797916412354, | |
| "eval_runtime": 18.9255, | |
| "eval_samples_per_second": 518.613, | |
| "eval_steps_per_second": 16.222, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1.864130434782609e-05, | |
| "loss": 0.7162, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_accuracy": 0.6400407539480387, | |
| "eval_loss": 0.7998170852661133, | |
| "eval_runtime": 18.8499, | |
| "eval_samples_per_second": 520.693, | |
| "eval_steps_per_second": 16.287, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 1.841485507246377e-05, | |
| "loss": 0.7092, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_accuracy": 0.6565461029037188, | |
| "eval_loss": 0.7830348014831543, | |
| "eval_runtime": 19.051, | |
| "eval_samples_per_second": 515.195, | |
| "eval_steps_per_second": 16.115, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 1.818840579710145e-05, | |
| "loss": 0.6962, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "eval_accuracy": 0.6628629648497198, | |
| "eval_loss": 0.7653073072433472, | |
| "eval_runtime": 19.1088, | |
| "eval_samples_per_second": 513.638, | |
| "eval_steps_per_second": 16.066, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 1.7961956521739134e-05, | |
| "loss": 0.6876, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_accuracy": 0.6686704024452369, | |
| "eval_loss": 0.7630106210708618, | |
| "eval_runtime": 19.1562, | |
| "eval_samples_per_second": 512.367, | |
| "eval_steps_per_second": 16.026, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 1.7735507246376815e-05, | |
| "loss": 0.6778, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "eval_accuracy": 0.6738665308201732, | |
| "eval_loss": 0.7475225329399109, | |
| "eval_runtime": 18.8719, | |
| "eval_samples_per_second": 520.085, | |
| "eval_steps_per_second": 16.268, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 1.7509057971014495e-05, | |
| "loss": 0.6737, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "eval_accuracy": 0.6781456953642384, | |
| "eval_loss": 0.7495400309562683, | |
| "eval_runtime": 24.231, | |
| "eval_samples_per_second": 405.059, | |
| "eval_steps_per_second": 12.67, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 1.7282608695652176e-05, | |
| "loss": 0.6712, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "eval_accuracy": 0.6826286296484972, | |
| "eval_loss": 0.7350295782089233, | |
| "eval_runtime": 25.183, | |
| "eval_samples_per_second": 389.747, | |
| "eval_steps_per_second": 12.191, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 1.7056159420289856e-05, | |
| "loss": 0.6559, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_accuracy": 0.6896586856851757, | |
| "eval_loss": 0.7274259328842163, | |
| "eval_runtime": 20.5021, | |
| "eval_samples_per_second": 478.731, | |
| "eval_steps_per_second": 14.974, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 1.6829710144927537e-05, | |
| "loss": 0.6493, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_accuracy": 0.6901681100356597, | |
| "eval_loss": 0.7247650623321533, | |
| "eval_runtime": 19.0556, | |
| "eval_samples_per_second": 515.071, | |
| "eval_steps_per_second": 16.111, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 1.6603260869565218e-05, | |
| "loss": 0.6483, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_accuracy": 0.6857870606214977, | |
| "eval_loss": 0.726345419883728, | |
| "eval_runtime": 19.1114, | |
| "eval_samples_per_second": 513.568, | |
| "eval_steps_per_second": 16.064, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 1.6376811594202898e-05, | |
| "loss": 0.6445, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "eval_accuracy": 0.697809475292919, | |
| "eval_loss": 0.7070016860961914, | |
| "eval_runtime": 19.0198, | |
| "eval_samples_per_second": 516.041, | |
| "eval_steps_per_second": 16.141, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 1.615036231884058e-05, | |
| "loss": 0.6467, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_accuracy": 0.6981151299032093, | |
| "eval_loss": 0.7083250284194946, | |
| "eval_runtime": 19.0519, | |
| "eval_samples_per_second": 515.172, | |
| "eval_steps_per_second": 16.114, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 1.5923913043478263e-05, | |
| "loss": 0.6332, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "eval_accuracy": 0.7003565970453388, | |
| "eval_loss": 0.6996482014656067, | |
| "eval_runtime": 18.8872, | |
| "eval_samples_per_second": 519.665, | |
| "eval_steps_per_second": 16.254, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 1.5697463768115943e-05, | |
| "loss": 0.6288, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "eval_accuracy": 0.697809475292919, | |
| "eval_loss": 0.6978936195373535, | |
| "eval_runtime": 19.1365, | |
| "eval_samples_per_second": 512.895, | |
| "eval_steps_per_second": 16.043, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 1.5471014492753624e-05, | |
| "loss": 0.6308, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "eval_accuracy": 0.7040244523688233, | |
| "eval_loss": 0.6911584734916687, | |
| "eval_runtime": 19.2976, | |
| "eval_samples_per_second": 508.612, | |
| "eval_steps_per_second": 15.909, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 1.5244565217391305e-05, | |
| "loss": 0.622, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "eval_accuracy": 0.7092205807437596, | |
| "eval_loss": 0.6903654932975769, | |
| "eval_runtime": 20.766, | |
| "eval_samples_per_second": 472.648, | |
| "eval_steps_per_second": 14.784, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 1.5018115942028985e-05, | |
| "loss": 0.615, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_accuracy": 0.7094243504839531, | |
| "eval_loss": 0.6871742010116577, | |
| "eval_runtime": 20.0444, | |
| "eval_samples_per_second": 489.662, | |
| "eval_steps_per_second": 15.316, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 1.479166666666667e-05, | |
| "loss": 0.6186, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "eval_accuracy": 0.7074885379521141, | |
| "eval_loss": 0.6876837611198425, | |
| "eval_runtime": 18.8534, | |
| "eval_samples_per_second": 520.594, | |
| "eval_steps_per_second": 16.283, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 1.456521739130435e-05, | |
| "loss": 0.6183, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "eval_accuracy": 0.7110545084055018, | |
| "eval_loss": 0.6818488836288452, | |
| "eval_runtime": 18.9963, | |
| "eval_samples_per_second": 516.679, | |
| "eval_steps_per_second": 16.161, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 1.433876811594203e-05, | |
| "loss": 0.6115, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "eval_accuracy": 0.7121752419765665, | |
| "eval_loss": 0.6855539083480835, | |
| "eval_runtime": 18.8559, | |
| "eval_samples_per_second": 520.528, | |
| "eval_steps_per_second": 16.281, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 1.4112318840579711e-05, | |
| "loss": 0.608, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "eval_accuracy": 0.7178807947019867, | |
| "eval_loss": 0.6696537137031555, | |
| "eval_runtime": 18.8735, | |
| "eval_samples_per_second": 520.042, | |
| "eval_steps_per_second": 16.266, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 1.3885869565217392e-05, | |
| "loss": 0.6071, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "eval_accuracy": 0.7180845644421804, | |
| "eval_loss": 0.6726976633071899, | |
| "eval_runtime": 18.9354, | |
| "eval_samples_per_second": 518.342, | |
| "eval_steps_per_second": 16.213, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 1.3659420289855074e-05, | |
| "loss": 0.601, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_accuracy": 0.7117677024961793, | |
| "eval_loss": 0.6798397898674011, | |
| "eval_runtime": 18.8349, | |
| "eval_samples_per_second": 521.108, | |
| "eval_steps_per_second": 16.3, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 1.3432971014492755e-05, | |
| "loss": 0.6018, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_accuracy": 0.7070809984717269, | |
| "eval_loss": 0.685390293598175, | |
| "eval_runtime": 19.049, | |
| "eval_samples_per_second": 515.251, | |
| "eval_steps_per_second": 16.116, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 1.3206521739130435e-05, | |
| "loss": 0.5762, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_accuracy": 0.7214467651553744, | |
| "eval_loss": 0.6697466373443604, | |
| "eval_runtime": 18.9286, | |
| "eval_samples_per_second": 518.528, | |
| "eval_steps_per_second": 16.219, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 1.2980072463768116e-05, | |
| "loss": 0.5507, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "eval_accuracy": 0.7184921039225675, | |
| "eval_loss": 0.6709696054458618, | |
| "eval_runtime": 20.3622, | |
| "eval_samples_per_second": 482.02, | |
| "eval_steps_per_second": 15.077, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 1.2753623188405797e-05, | |
| "loss": 0.5575, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "eval_accuracy": 0.7225674987264391, | |
| "eval_loss": 0.6709049344062805, | |
| "eval_runtime": 20.0855, | |
| "eval_samples_per_second": 488.662, | |
| "eval_steps_per_second": 15.285, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 1.252717391304348e-05, | |
| "loss": 0.5493, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "eval_accuracy": 0.7191034131431483, | |
| "eval_loss": 0.6659159064292908, | |
| "eval_runtime": 20.41, | |
| "eval_samples_per_second": 480.892, | |
| "eval_steps_per_second": 15.042, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 1.2300724637681161e-05, | |
| "loss": 0.5464, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "eval_accuracy": 0.7231788079470198, | |
| "eval_loss": 0.6709375381469727, | |
| "eval_runtime": 18.9369, | |
| "eval_samples_per_second": 518.301, | |
| "eval_steps_per_second": 16.212, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 1.2074275362318842e-05, | |
| "loss": 0.5595, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "eval_accuracy": 0.7219561895058584, | |
| "eval_loss": 0.6641885638237, | |
| "eval_runtime": 18.8852, | |
| "eval_samples_per_second": 519.718, | |
| "eval_steps_per_second": 16.256, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 1.1847826086956522e-05, | |
| "loss": 0.5446, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "eval_accuracy": 0.720224146714213, | |
| "eval_loss": 0.6708667874336243, | |
| "eval_runtime": 18.8795, | |
| "eval_samples_per_second": 519.875, | |
| "eval_steps_per_second": 16.261, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 1.1621376811594205e-05, | |
| "loss": 0.5524, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "eval_accuracy": 0.714824248599083, | |
| "eval_loss": 0.6751348972320557, | |
| "eval_runtime": 18.8954, | |
| "eval_samples_per_second": 519.438, | |
| "eval_steps_per_second": 16.247, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 1.1394927536231885e-05, | |
| "loss": 0.5473, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "eval_accuracy": 0.7209373408048905, | |
| "eval_loss": 0.6641535758972168, | |
| "eval_runtime": 18.8349, | |
| "eval_samples_per_second": 521.107, | |
| "eval_steps_per_second": 16.3, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 1.1168478260869566e-05, | |
| "loss": 0.5477, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "eval_accuracy": 0.7222618441161488, | |
| "eval_loss": 0.6661530137062073, | |
| "eval_runtime": 18.9172, | |
| "eval_samples_per_second": 518.841, | |
| "eval_steps_per_second": 16.229, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 1.0942028985507247e-05, | |
| "loss": 0.5522, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "eval_accuracy": 0.7226693835965359, | |
| "eval_loss": 0.6586304306983948, | |
| "eval_runtime": 19.03, | |
| "eval_samples_per_second": 515.763, | |
| "eval_steps_per_second": 16.132, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 1.0715579710144927e-05, | |
| "loss": 0.5406, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "eval_accuracy": 0.7258278145695364, | |
| "eval_loss": 0.6601831912994385, | |
| "eval_runtime": 20.4522, | |
| "eval_samples_per_second": 479.9, | |
| "eval_steps_per_second": 15.011, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 1.0489130434782611e-05, | |
| "loss": 0.54, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "eval_accuracy": 0.7272542027508915, | |
| "eval_loss": 0.6564027070999146, | |
| "eval_runtime": 20.4433, | |
| "eval_samples_per_second": 480.108, | |
| "eval_steps_per_second": 15.017, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 1.0262681159420292e-05, | |
| "loss": 0.5458, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "eval_accuracy": 0.7213448802852777, | |
| "eval_loss": 0.6779993176460266, | |
| "eval_runtime": 20.545, | |
| "eval_samples_per_second": 477.731, | |
| "eval_steps_per_second": 14.943, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 1.0036231884057972e-05, | |
| "loss": 0.5448, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "eval_accuracy": 0.7234844625573102, | |
| "eval_loss": 0.6560543775558472, | |
| "eval_runtime": 18.9249, | |
| "eval_samples_per_second": 518.629, | |
| "eval_steps_per_second": 16.222, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 9.809782608695653e-06, | |
| "loss": 0.5418, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "eval_accuracy": 0.7253183902190524, | |
| "eval_loss": 0.6599519848823547, | |
| "eval_runtime": 19.0375, | |
| "eval_samples_per_second": 515.561, | |
| "eval_steps_per_second": 16.126, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 9.583333333333335e-06, | |
| "loss": 0.5408, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "eval_accuracy": 0.7273560876209882, | |
| "eval_loss": 0.6616316437721252, | |
| "eval_runtime": 19.3854, | |
| "eval_samples_per_second": 506.308, | |
| "eval_steps_per_second": 15.837, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 9.356884057971016e-06, | |
| "loss": 0.5451, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_accuracy": 0.7282730514518594, | |
| "eval_loss": 0.655745267868042, | |
| "eval_runtime": 18.8581, | |
| "eval_samples_per_second": 520.467, | |
| "eval_steps_per_second": 16.28, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 9.130434782608697e-06, | |
| "loss": 0.5385, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "eval_accuracy": 0.7294956698930208, | |
| "eval_loss": 0.6583314538002014, | |
| "eval_runtime": 19.1341, | |
| "eval_samples_per_second": 512.958, | |
| "eval_steps_per_second": 16.045, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 8.903985507246377e-06, | |
| "loss": 0.5261, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "eval_accuracy": 0.7324503311258278, | |
| "eval_loss": 0.6467986106872559, | |
| "eval_runtime": 19.0986, | |
| "eval_samples_per_second": 513.912, | |
| "eval_steps_per_second": 16.074, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 8.677536231884058e-06, | |
| "loss": 0.5364, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "eval_accuracy": 0.732857870606215, | |
| "eval_loss": 0.644736111164093, | |
| "eval_runtime": 18.8951, | |
| "eval_samples_per_second": 519.447, | |
| "eval_steps_per_second": 16.248, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 8.45108695652174e-06, | |
| "loss": 0.5294, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "eval_accuracy": 0.7320427916454406, | |
| "eval_loss": 0.6429356932640076, | |
| "eval_runtime": 20.711, | |
| "eval_samples_per_second": 473.903, | |
| "eval_steps_per_second": 14.823, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 8.22463768115942e-06, | |
| "loss": 0.5332, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "eval_accuracy": 0.7271523178807947, | |
| "eval_loss": 0.6508257985115051, | |
| "eval_runtime": 20.8391, | |
| "eval_samples_per_second": 470.989, | |
| "eval_steps_per_second": 14.732, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 7.998188405797103e-06, | |
| "loss": 0.5274, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "eval_accuracy": 0.7325522159959246, | |
| "eval_loss": 0.6491544842720032, | |
| "eval_runtime": 19.9398, | |
| "eval_samples_per_second": 492.231, | |
| "eval_steps_per_second": 15.396, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 7.771739130434784e-06, | |
| "loss": 0.5286, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "eval_accuracy": 0.7318390219052471, | |
| "eval_loss": 0.6470006704330444, | |
| "eval_runtime": 18.918, | |
| "eval_samples_per_second": 518.818, | |
| "eval_steps_per_second": 16.228, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 7.545289855072464e-06, | |
| "loss": 0.5359, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "eval_accuracy": 0.7354049923586348, | |
| "eval_loss": 0.6392927169799805, | |
| "eval_runtime": 19.0427, | |
| "eval_samples_per_second": 515.422, | |
| "eval_steps_per_second": 16.122, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 7.318840579710146e-06, | |
| "loss": 0.5366, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "eval_accuracy": 0.736729495669893, | |
| "eval_loss": 0.6445118188858032, | |
| "eval_runtime": 19.0599, | |
| "eval_samples_per_second": 514.955, | |
| "eval_steps_per_second": 16.107, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 7.092391304347826e-06, | |
| "loss": 0.5296, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "eval_accuracy": 0.7313295975547631, | |
| "eval_loss": 0.6413289904594421, | |
| "eval_runtime": 18.9234, | |
| "eval_samples_per_second": 518.671, | |
| "eval_steps_per_second": 16.223, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 6.865942028985509e-06, | |
| "loss": 0.5346, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "eval_accuracy": 0.7315333672949567, | |
| "eval_loss": 0.6392691135406494, | |
| "eval_runtime": 18.8526, | |
| "eval_samples_per_second": 520.617, | |
| "eval_steps_per_second": 16.284, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 6.639492753623189e-06, | |
| "loss": 0.5264, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7357106469689251, | |
| "eval_loss": 0.6448105573654175, | |
| "eval_runtime": 18.9027, | |
| "eval_samples_per_second": 519.238, | |
| "eval_steps_per_second": 16.241, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 6.41304347826087e-06, | |
| "loss": 0.4857, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "eval_accuracy": 0.7334691798267957, | |
| "eval_loss": 0.6640365123748779, | |
| "eval_runtime": 19.0337, | |
| "eval_samples_per_second": 515.664, | |
| "eval_steps_per_second": 16.129, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 6.186594202898551e-06, | |
| "loss": 0.4888, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "eval_accuracy": 0.7318390219052471, | |
| "eval_loss": 0.6611655950546265, | |
| "eval_runtime": 20.389, | |
| "eval_samples_per_second": 481.387, | |
| "eval_steps_per_second": 15.057, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 5.960144927536232e-06, | |
| "loss": 0.4964, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "eval_accuracy": 0.7336729495669893, | |
| "eval_loss": 0.6516362428665161, | |
| "eval_runtime": 20.5019, | |
| "eval_samples_per_second": 478.737, | |
| "eval_steps_per_second": 14.974, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 5.733695652173914e-06, | |
| "loss": 0.493, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "eval_accuracy": 0.7356087620988283, | |
| "eval_loss": 0.6503166556358337, | |
| "eval_runtime": 20.2046, | |
| "eval_samples_per_second": 485.78, | |
| "eval_steps_per_second": 15.195, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 5.507246376811595e-06, | |
| "loss": 0.4961, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "eval_accuracy": 0.734793683138054, | |
| "eval_loss": 0.6519193053245544, | |
| "eval_runtime": 18.9078, | |
| "eval_samples_per_second": 519.097, | |
| "eval_steps_per_second": 16.237, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 5.2807971014492755e-06, | |
| "loss": 0.4847, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "eval_accuracy": 0.7326541008660213, | |
| "eval_loss": 0.6516821980476379, | |
| "eval_runtime": 19.1631, | |
| "eval_samples_per_second": 512.183, | |
| "eval_steps_per_second": 16.02, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 5.054347826086957e-06, | |
| "loss": 0.483, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "eval_accuracy": 0.7310239429444727, | |
| "eval_loss": 0.655524492263794, | |
| "eval_runtime": 19.0256, | |
| "eval_samples_per_second": 515.884, | |
| "eval_steps_per_second": 16.136, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 4.8278985507246375e-06, | |
| "loss": 0.4857, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "eval_accuracy": 0.7312277126846664, | |
| "eval_loss": 0.6524935364723206, | |
| "eval_runtime": 19.0792, | |
| "eval_samples_per_second": 514.435, | |
| "eval_steps_per_second": 16.091, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 4.601449275362319e-06, | |
| "loss": 0.484, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "eval_accuracy": 0.7341823739174732, | |
| "eval_loss": 0.6443938612937927, | |
| "eval_runtime": 19.0647, | |
| "eval_samples_per_second": 514.826, | |
| "eval_steps_per_second": 16.103, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 4.3750000000000005e-06, | |
| "loss": 0.4792, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "eval_accuracy": 0.7329597554763118, | |
| "eval_loss": 0.6508365273475647, | |
| "eval_runtime": 19.1608, | |
| "eval_samples_per_second": 512.243, | |
| "eval_steps_per_second": 16.022, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 4.148550724637682e-06, | |
| "loss": 0.488, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "eval_accuracy": 0.7343861436576669, | |
| "eval_loss": 0.6512596607208252, | |
| "eval_runtime": 20.0389, | |
| "eval_samples_per_second": 489.797, | |
| "eval_steps_per_second": 15.32, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 3.9221014492753625e-06, | |
| "loss": 0.472, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "eval_accuracy": 0.7345899133978604, | |
| "eval_loss": 0.6546801328659058, | |
| "eval_runtime": 21.0064, | |
| "eval_samples_per_second": 467.238, | |
| "eval_steps_per_second": 14.615, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 3.6956521739130436e-06, | |
| "loss": 0.4872, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "eval_accuracy": 0.7341823739174732, | |
| "eval_loss": 0.650049090385437, | |
| "eval_runtime": 20.3969, | |
| "eval_samples_per_second": 481.201, | |
| "eval_steps_per_second": 15.051, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 3.4692028985507246e-06, | |
| "loss": 0.4782, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "eval_accuracy": 0.735812531839022, | |
| "eval_loss": 0.6584815979003906, | |
| "eval_runtime": 20.5297, | |
| "eval_samples_per_second": 478.088, | |
| "eval_steps_per_second": 14.954, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 3.242753623188406e-06, | |
| "loss": 0.481, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "eval_accuracy": 0.7356087620988283, | |
| "eval_loss": 0.647666335105896, | |
| "eval_runtime": 18.9629, | |
| "eval_samples_per_second": 517.589, | |
| "eval_steps_per_second": 16.189, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 3.016304347826087e-06, | |
| "loss": 0.4822, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "eval_accuracy": 0.7345899133978604, | |
| "eval_loss": 0.6586557626724243, | |
| "eval_runtime": 19.0494, | |
| "eval_samples_per_second": 515.239, | |
| "eval_steps_per_second": 16.116, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 2.7898550724637686e-06, | |
| "loss": 0.4728, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "eval_accuracy": 0.7339786041772797, | |
| "eval_loss": 0.6572148203849792, | |
| "eval_runtime": 18.9091, | |
| "eval_samples_per_second": 519.063, | |
| "eval_steps_per_second": 16.236, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 2.563405797101449e-06, | |
| "loss": 0.4841, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "eval_accuracy": 0.7374426897605706, | |
| "eval_loss": 0.6442714929580688, | |
| "eval_runtime": 19.0799, | |
| "eval_samples_per_second": 514.415, | |
| "eval_steps_per_second": 16.09, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 2.3369565217391307e-06, | |
| "loss": 0.4885, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "eval_accuracy": 0.736220071319409, | |
| "eval_loss": 0.6493647694587708, | |
| "eval_runtime": 18.9053, | |
| "eval_samples_per_second": 519.165, | |
| "eval_steps_per_second": 16.239, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 2.1105072463768117e-06, | |
| "loss": 0.4752, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "eval_accuracy": 0.7381558838512481, | |
| "eval_loss": 0.6508816480636597, | |
| "eval_runtime": 19.0533, | |
| "eval_samples_per_second": 515.135, | |
| "eval_steps_per_second": 16.113, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 1.884057971014493e-06, | |
| "loss": 0.4883, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "eval_accuracy": 0.7371370351502802, | |
| "eval_loss": 0.6456698179244995, | |
| "eval_runtime": 18.855, | |
| "eval_samples_per_second": 520.551, | |
| "eval_steps_per_second": 16.282, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 1.657608695652174e-06, | |
| "loss": 0.4888, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "eval_accuracy": 0.7364238410596027, | |
| "eval_loss": 0.6496989130973816, | |
| "eval_runtime": 20.8847, | |
| "eval_samples_per_second": 469.962, | |
| "eval_steps_per_second": 14.7, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 1.4311594202898552e-06, | |
| "loss": 0.4844, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "eval_accuracy": 0.7376464595007641, | |
| "eval_loss": 0.6481407880783081, | |
| "eval_runtime": 20.1004, | |
| "eval_samples_per_second": 488.299, | |
| "eval_steps_per_second": 15.273, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "learning_rate": 1.2047101449275363e-06, | |
| "loss": 0.4833, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "eval_accuracy": 0.7388690779419256, | |
| "eval_loss": 0.6450786590576172, | |
| "eval_runtime": 20.1432, | |
| "eval_samples_per_second": 487.26, | |
| "eval_steps_per_second": 15.241, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 9.782608695652175e-07, | |
| "loss": 0.48, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "eval_accuracy": 0.7373408048904737, | |
| "eval_loss": 0.6423162221908569, | |
| "eval_runtime": 18.9117, | |
| "eval_samples_per_second": 518.992, | |
| "eval_steps_per_second": 16.233, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 7.518115942028987e-07, | |
| "loss": 0.4832, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "eval_accuracy": 0.7357106469689251, | |
| "eval_loss": 0.6477082371711731, | |
| "eval_runtime": 18.9455, | |
| "eval_samples_per_second": 518.066, | |
| "eval_steps_per_second": 16.204, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 5.253623188405797e-07, | |
| "loss": 0.4805, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "eval_accuracy": 0.7378502292409577, | |
| "eval_loss": 0.6464207768440247, | |
| "eval_runtime": 18.9094, | |
| "eval_samples_per_second": 519.053, | |
| "eval_steps_per_second": 16.235, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 2.989130434782609e-07, | |
| "loss": 0.4775, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "eval_accuracy": 0.7379521141110545, | |
| "eval_loss": 0.6477315425872803, | |
| "eval_runtime": 19.0715, | |
| "eval_samples_per_second": 514.642, | |
| "eval_steps_per_second": 16.097, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 7.246376811594204e-08, | |
| "loss": 0.4843, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "eval_accuracy": 0.7380539989811513, | |
| "eval_loss": 0.6481098532676697, | |
| "eval_runtime": 18.8988, | |
| "eval_samples_per_second": 519.345, | |
| "eval_steps_per_second": 16.244, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 88320, | |
| "total_flos": 5.0801111188255944e+16, | |
| "train_loss": 0.565458259202432, | |
| "train_runtime": 14832.9522, | |
| "train_samples_per_second": 190.536, | |
| "train_steps_per_second": 5.954 | |
| } | |
| ], | |
| "max_steps": 88320, | |
| "num_train_epochs": 3, | |
| "total_flos": 5.0801111188255944e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |