| { | |
| "best_metric": 0.7984717269485482, | |
| "best_model_checkpoint": "/home/user/emrecan/models/dbmdz_bert-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": 0.8559, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "eval_accuracy": 0.679775853285787, | |
| "eval_loss": 0.7577011585235596, | |
| "eval_runtime": 35.4894, | |
| "eval_samples_per_second": 276.562, | |
| "eval_steps_per_second": 8.65, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 1.9547101449275363e-05, | |
| "loss": 0.6612, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_accuracy": 0.6957717778909832, | |
| "eval_loss": 0.7263472676277161, | |
| "eval_runtime": 37.7507, | |
| "eval_samples_per_second": 259.995, | |
| "eval_steps_per_second": 8.132, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 1.9320652173913047e-05, | |
| "loss": 0.6115, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_accuracy": 0.7364238410596027, | |
| "eval_loss": 0.6430743932723999, | |
| "eval_runtime": 37.122, | |
| "eval_samples_per_second": 264.399, | |
| "eval_steps_per_second": 8.27, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 1.9094202898550727e-05, | |
| "loss": 0.5916, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "eval_accuracy": 0.7407030056036679, | |
| "eval_loss": 0.6347211599349976, | |
| "eval_runtime": 36.4648, | |
| "eval_samples_per_second": 269.164, | |
| "eval_steps_per_second": 8.419, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 1.8867753623188408e-05, | |
| "loss": 0.5719, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "eval_accuracy": 0.7483443708609272, | |
| "eval_loss": 0.6316782236099243, | |
| "eval_runtime": 36.2383, | |
| "eval_samples_per_second": 270.846, | |
| "eval_steps_per_second": 8.472, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1.864130434782609e-05, | |
| "loss": 0.5575, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_accuracy": 0.7543555781966378, | |
| "eval_loss": 0.6033961772918701, | |
| "eval_runtime": 36.9743, | |
| "eval_samples_per_second": 265.455, | |
| "eval_steps_per_second": 8.303, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 1.841485507246377e-05, | |
| "loss": 0.5521, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_accuracy": 0.7568008150789608, | |
| "eval_loss": 0.6147794723510742, | |
| "eval_runtime": 35.6284, | |
| "eval_samples_per_second": 275.483, | |
| "eval_steps_per_second": 8.617, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 1.818840579710145e-05, | |
| "loss": 0.5393, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "eval_accuracy": 0.7609780947529292, | |
| "eval_loss": 0.593109667301178, | |
| "eval_runtime": 35.1284, | |
| "eval_samples_per_second": 279.404, | |
| "eval_steps_per_second": 8.739, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 1.7961956521739134e-05, | |
| "loss": 0.5382, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_accuracy": 0.7664798777381558, | |
| "eval_loss": 0.5865770578384399, | |
| "eval_runtime": 36.7058, | |
| "eval_samples_per_second": 267.396, | |
| "eval_steps_per_second": 8.364, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 1.7735507246376815e-05, | |
| "loss": 0.5306, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "eval_accuracy": 0.7594498217014773, | |
| "eval_loss": 0.5881215929985046, | |
| "eval_runtime": 36.4456, | |
| "eval_samples_per_second": 269.305, | |
| "eval_steps_per_second": 8.424, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 1.7509057971014495e-05, | |
| "loss": 0.5295, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "eval_accuracy": 0.7632195618950586, | |
| "eval_loss": 0.6120319962501526, | |
| "eval_runtime": 36.0374, | |
| "eval_samples_per_second": 272.356, | |
| "eval_steps_per_second": 8.519, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 1.7282608695652176e-05, | |
| "loss": 0.5225, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "eval_accuracy": 0.7758532857870606, | |
| "eval_loss": 0.5619738698005676, | |
| "eval_runtime": 36.3737, | |
| "eval_samples_per_second": 269.838, | |
| "eval_steps_per_second": 8.44, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 1.7056159420289856e-05, | |
| "loss": 0.5112, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_accuracy": 0.7768721344880285, | |
| "eval_loss": 0.5641229152679443, | |
| "eval_runtime": 36.0192, | |
| "eval_samples_per_second": 272.494, | |
| "eval_steps_per_second": 8.523, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 1.6829710144927537e-05, | |
| "loss": 0.5133, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_accuracy": 0.7798267957208355, | |
| "eval_loss": 0.557083010673523, | |
| "eval_runtime": 37.2598, | |
| "eval_samples_per_second": 263.42, | |
| "eval_steps_per_second": 8.239, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 1.6603260869565218e-05, | |
| "loss": 0.5023, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_accuracy": 0.7721854304635761, | |
| "eval_loss": 0.5719286203384399, | |
| "eval_runtime": 36.5731, | |
| "eval_samples_per_second": 268.366, | |
| "eval_steps_per_second": 8.394, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 1.6376811594202898e-05, | |
| "loss": 0.5017, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "eval_accuracy": 0.7844116148751911, | |
| "eval_loss": 0.5482256412506104, | |
| "eval_runtime": 35.2562, | |
| "eval_samples_per_second": 278.391, | |
| "eval_steps_per_second": 8.708, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 1.615036231884058e-05, | |
| "loss": 0.5111, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_accuracy": 0.780030565461029, | |
| "eval_loss": 0.5503329634666443, | |
| "eval_runtime": 35.8945, | |
| "eval_samples_per_second": 273.44, | |
| "eval_steps_per_second": 8.553, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 1.5923913043478263e-05, | |
| "loss": 0.4929, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "eval_accuracy": 0.7835965359144167, | |
| "eval_loss": 0.5501914620399475, | |
| "eval_runtime": 36.1699, | |
| "eval_samples_per_second": 271.358, | |
| "eval_steps_per_second": 8.488, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 1.5697463768115943e-05, | |
| "loss": 0.4923, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "eval_accuracy": 0.7843097300050943, | |
| "eval_loss": 0.5424306988716125, | |
| "eval_runtime": 35.606, | |
| "eval_samples_per_second": 275.656, | |
| "eval_steps_per_second": 8.622, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 1.5471014492753624e-05, | |
| "loss": 0.4894, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "eval_accuracy": 0.7851248089658686, | |
| "eval_loss": 0.541650116443634, | |
| "eval_runtime": 35.3356, | |
| "eval_samples_per_second": 277.765, | |
| "eval_steps_per_second": 8.688, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 1.5244565217391305e-05, | |
| "loss": 0.4877, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "eval_accuracy": 0.7841059602649006, | |
| "eval_loss": 0.5514414310455322, | |
| "eval_runtime": 36.8151, | |
| "eval_samples_per_second": 266.602, | |
| "eval_steps_per_second": 8.339, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 1.5018115942028985e-05, | |
| "loss": 0.4818, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_accuracy": 0.7848191543555781, | |
| "eval_loss": 0.5493878126144409, | |
| "eval_runtime": 37.2449, | |
| "eval_samples_per_second": 263.526, | |
| "eval_steps_per_second": 8.243, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 1.479166666666667e-05, | |
| "loss": 0.4898, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "eval_accuracy": 0.7859398879266429, | |
| "eval_loss": 0.5450355410575867, | |
| "eval_runtime": 37.2862, | |
| "eval_samples_per_second": 263.234, | |
| "eval_steps_per_second": 8.234, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 1.456521739130435e-05, | |
| "loss": 0.4823, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "eval_accuracy": 0.7877738155883851, | |
| "eval_loss": 0.5417405366897583, | |
| "eval_runtime": 37.3765, | |
| "eval_samples_per_second": 262.598, | |
| "eval_steps_per_second": 8.214, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 1.433876811594203e-05, | |
| "loss": 0.4806, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "eval_accuracy": 0.7874681609780948, | |
| "eval_loss": 0.5354494452476501, | |
| "eval_runtime": 37.6863, | |
| "eval_samples_per_second": 260.44, | |
| "eval_steps_per_second": 8.146, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 1.4112318840579711e-05, | |
| "loss": 0.4779, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "eval_accuracy": 0.7848191543555781, | |
| "eval_loss": 0.5337836146354675, | |
| "eval_runtime": 36.4797, | |
| "eval_samples_per_second": 269.054, | |
| "eval_steps_per_second": 8.416, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 1.3885869565217392e-05, | |
| "loss": 0.4744, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "eval_accuracy": 0.7933774834437086, | |
| "eval_loss": 0.5277203917503357, | |
| "eval_runtime": 36.8856, | |
| "eval_samples_per_second": 266.093, | |
| "eval_steps_per_second": 8.323, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 1.3659420289855074e-05, | |
| "loss": 0.4678, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_accuracy": 0.7870606214977076, | |
| "eval_loss": 0.5506766438484192, | |
| "eval_runtime": 36.0153, | |
| "eval_samples_per_second": 272.523, | |
| "eval_steps_per_second": 8.524, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 1.3432971014492755e-05, | |
| "loss": 0.4727, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_accuracy": 0.7789098318899643, | |
| "eval_loss": 0.5603240132331848, | |
| "eval_runtime": 35.8744, | |
| "eval_samples_per_second": 273.593, | |
| "eval_steps_per_second": 8.558, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 1.3206521739130435e-05, | |
| "loss": 0.4243, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_accuracy": 0.7894039735099337, | |
| "eval_loss": 0.562574565410614, | |
| "eval_runtime": 35.9331, | |
| "eval_samples_per_second": 273.146, | |
| "eval_steps_per_second": 8.544, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 1.2980072463768116e-05, | |
| "loss": 0.3955, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "eval_accuracy": 0.7938869077941926, | |
| "eval_loss": 0.5324372053146362, | |
| "eval_runtime": 37.5987, | |
| "eval_samples_per_second": 261.046, | |
| "eval_steps_per_second": 8.165, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 1.2753623188405797e-05, | |
| "loss": 0.4022, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "eval_accuracy": 0.7924605196128375, | |
| "eval_loss": 0.5322304368019104, | |
| "eval_runtime": 37.3438, | |
| "eval_samples_per_second": 262.828, | |
| "eval_steps_per_second": 8.221, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 1.252717391304348e-05, | |
| "loss": 0.3976, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "eval_accuracy": 0.7919510952623535, | |
| "eval_loss": 0.545001745223999, | |
| "eval_runtime": 36.321, | |
| "eval_samples_per_second": 270.229, | |
| "eval_steps_per_second": 8.452, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 1.2300724637681161e-05, | |
| "loss": 0.3913, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "eval_accuracy": 0.7948038716250637, | |
| "eval_loss": 0.5464295744895935, | |
| "eval_runtime": 36.0101, | |
| "eval_samples_per_second": 272.562, | |
| "eval_steps_per_second": 8.525, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 1.2074275362318842e-05, | |
| "loss": 0.406, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "eval_accuracy": 0.7958227203260316, | |
| "eval_loss": 0.5405891537666321, | |
| "eval_runtime": 35.967, | |
| "eval_samples_per_second": 272.889, | |
| "eval_steps_per_second": 8.536, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 1.1847826086956522e-05, | |
| "loss": 0.3875, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "eval_accuracy": 0.7877738155883851, | |
| "eval_loss": 0.5489205718040466, | |
| "eval_runtime": 36.1107, | |
| "eval_samples_per_second": 271.803, | |
| "eval_steps_per_second": 8.502, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 1.1621376811594205e-05, | |
| "loss": 0.4024, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "eval_accuracy": 0.7924605196128375, | |
| "eval_loss": 0.542732834815979, | |
| "eval_runtime": 36.87, | |
| "eval_samples_per_second": 266.206, | |
| "eval_steps_per_second": 8.327, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 1.1394927536231885e-05, | |
| "loss": 0.3988, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "eval_accuracy": 0.7904228222109017, | |
| "eval_loss": 0.5334596037864685, | |
| "eval_runtime": 36.0057, | |
| "eval_samples_per_second": 272.596, | |
| "eval_steps_per_second": 8.526, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 1.1168478260869566e-05, | |
| "loss": 0.393, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "eval_accuracy": 0.7922567498726439, | |
| "eval_loss": 0.5415398478507996, | |
| "eval_runtime": 35.9966, | |
| "eval_samples_per_second": 272.665, | |
| "eval_steps_per_second": 8.529, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 1.0942028985507247e-05, | |
| "loss": 0.3988, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "eval_accuracy": 0.7962302598064187, | |
| "eval_loss": 0.5384690761566162, | |
| "eval_runtime": 36.005, | |
| "eval_samples_per_second": 272.601, | |
| "eval_steps_per_second": 8.527, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 1.0715579710144927e-05, | |
| "loss": 0.3912, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "eval_accuracy": 0.7950076413652573, | |
| "eval_loss": 0.5382993817329407, | |
| "eval_runtime": 37.5898, | |
| "eval_samples_per_second": 261.108, | |
| "eval_steps_per_second": 8.167, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 1.0489130434782611e-05, | |
| "loss": 0.3949, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "eval_accuracy": 0.7930718288334182, | |
| "eval_loss": 0.5415102243423462, | |
| "eval_runtime": 37.2008, | |
| "eval_samples_per_second": 263.838, | |
| "eval_steps_per_second": 8.253, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 1.0262681159420292e-05, | |
| "loss": 0.3902, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "eval_accuracy": 0.789302088639837, | |
| "eval_loss": 0.5438172817230225, | |
| "eval_runtime": 36.0003, | |
| "eval_samples_per_second": 272.636, | |
| "eval_steps_per_second": 8.528, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 1.0036231884057972e-05, | |
| "loss": 0.3948, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "eval_accuracy": 0.7906265919510953, | |
| "eval_loss": 0.5348154902458191, | |
| "eval_runtime": 36.2261, | |
| "eval_samples_per_second": 270.937, | |
| "eval_steps_per_second": 8.475, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 9.809782608695653e-06, | |
| "loss": 0.3921, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "eval_accuracy": 0.7889964340295467, | |
| "eval_loss": 0.5361019968986511, | |
| "eval_runtime": 36.0927, | |
| "eval_samples_per_second": 271.938, | |
| "eval_steps_per_second": 8.506, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 9.583333333333335e-06, | |
| "loss": 0.3944, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "eval_accuracy": 0.7953132959755477, | |
| "eval_loss": 0.5419210195541382, | |
| "eval_runtime": 36.1865, | |
| "eval_samples_per_second": 271.234, | |
| "eval_steps_per_second": 8.484, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 9.356884057971016e-06, | |
| "loss": 0.3959, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_accuracy": 0.7967396841569027, | |
| "eval_loss": 0.5401512980461121, | |
| "eval_runtime": 36.1559, | |
| "eval_samples_per_second": 271.463, | |
| "eval_steps_per_second": 8.491, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 9.130434782608697e-06, | |
| "loss": 0.3926, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "eval_accuracy": 0.7924605196128375, | |
| "eval_loss": 0.5428866744041443, | |
| "eval_runtime": 36.1454, | |
| "eval_samples_per_second": 271.542, | |
| "eval_steps_per_second": 8.493, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 8.903985507246377e-06, | |
| "loss": 0.3854, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "eval_accuracy": 0.7959246051961284, | |
| "eval_loss": 0.534604012966156, | |
| "eval_runtime": 36.0997, | |
| "eval_samples_per_second": 271.886, | |
| "eval_steps_per_second": 8.504, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 8.677536231884058e-06, | |
| "loss": 0.3864, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "eval_accuracy": 0.7978604177279673, | |
| "eval_loss": 0.5241352915763855, | |
| "eval_runtime": 35.9203, | |
| "eval_samples_per_second": 273.244, | |
| "eval_steps_per_second": 8.547, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 8.45108695652174e-06, | |
| "loss": 0.385, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "eval_accuracy": 0.8002037697401936, | |
| "eval_loss": 0.514944851398468, | |
| "eval_runtime": 37.3533, | |
| "eval_samples_per_second": 262.762, | |
| "eval_steps_per_second": 8.219, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 8.22463768115942e-06, | |
| "loss": 0.3871, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "eval_accuracy": 0.8002037697401936, | |
| "eval_loss": 0.5325358510017395, | |
| "eval_runtime": 36.687, | |
| "eval_samples_per_second": 267.534, | |
| "eval_steps_per_second": 8.368, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 7.998188405797103e-06, | |
| "loss": 0.3819, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "eval_accuracy": 0.8022414671421294, | |
| "eval_loss": 0.5332015752792358, | |
| "eval_runtime": 35.9183, | |
| "eval_samples_per_second": 273.259, | |
| "eval_steps_per_second": 8.547, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 7.771739130434784e-06, | |
| "loss": 0.384, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "eval_accuracy": 0.7872643912379012, | |
| "eval_loss": 0.5419135093688965, | |
| "eval_runtime": 35.939, | |
| "eval_samples_per_second": 273.102, | |
| "eval_steps_per_second": 8.542, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 7.545289855072464e-06, | |
| "loss": 0.3899, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "eval_accuracy": 0.7973509933774835, | |
| "eval_loss": 0.522521436214447, | |
| "eval_runtime": 36.1317, | |
| "eval_samples_per_second": 271.645, | |
| "eval_steps_per_second": 8.497, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 7.318840579710146e-06, | |
| "loss": 0.3894, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "eval_accuracy": 0.7976566479877738, | |
| "eval_loss": 0.5357836484909058, | |
| "eval_runtime": 36.1326, | |
| "eval_samples_per_second": 271.639, | |
| "eval_steps_per_second": 8.496, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 7.092391304347826e-06, | |
| "loss": 0.3838, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "eval_accuracy": 0.7987773815588385, | |
| "eval_loss": 0.5263946652412415, | |
| "eval_runtime": 35.9423, | |
| "eval_samples_per_second": 273.077, | |
| "eval_steps_per_second": 8.541, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 6.865942028985509e-06, | |
| "loss": 0.3881, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "eval_accuracy": 0.795618950585838, | |
| "eval_loss": 0.5279687643051147, | |
| "eval_runtime": 35.9441, | |
| "eval_samples_per_second": 273.063, | |
| "eval_steps_per_second": 8.541, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 6.639492753623189e-06, | |
| "loss": 0.3756, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7969434538970963, | |
| "eval_loss": 0.5600506663322449, | |
| "eval_runtime": 36.2008, | |
| "eval_samples_per_second": 271.126, | |
| "eval_steps_per_second": 8.48, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 6.41304347826087e-06, | |
| "loss": 0.3156, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "eval_accuracy": 0.7924605196128375, | |
| "eval_loss": 0.5936241745948792, | |
| "eval_runtime": 36.6834, | |
| "eval_samples_per_second": 267.56, | |
| "eval_steps_per_second": 8.369, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 6.186594202898551e-06, | |
| "loss": 0.3125, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "eval_accuracy": 0.7937850229240958, | |
| "eval_loss": 0.5897734761238098, | |
| "eval_runtime": 43.1106, | |
| "eval_samples_per_second": 227.67, | |
| "eval_steps_per_second": 7.121, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 5.960144927536232e-06, | |
| "loss": 0.3179, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "eval_accuracy": 0.798064187468161, | |
| "eval_loss": 0.5591443777084351, | |
| "eval_runtime": 36.4374, | |
| "eval_samples_per_second": 269.366, | |
| "eval_steps_per_second": 8.425, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 5.733695652173914e-06, | |
| "loss": 0.315, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "eval_accuracy": 0.7970453387671931, | |
| "eval_loss": 0.585314154624939, | |
| "eval_runtime": 35.8475, | |
| "eval_samples_per_second": 273.799, | |
| "eval_steps_per_second": 8.564, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 5.507246376811595e-06, | |
| "loss": 0.3122, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "eval_accuracy": 0.7978604177279673, | |
| "eval_loss": 0.5801787376403809, | |
| "eval_runtime": 35.9545, | |
| "eval_samples_per_second": 272.984, | |
| "eval_steps_per_second": 8.539, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 5.2807971014492755e-06, | |
| "loss": 0.3105, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "eval_accuracy": 0.7978604177279673, | |
| "eval_loss": 0.5757964849472046, | |
| "eval_runtime": 36.2052, | |
| "eval_samples_per_second": 271.094, | |
| "eval_steps_per_second": 8.479, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 5.054347826086957e-06, | |
| "loss": 0.3076, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "eval_accuracy": 0.7979623025980642, | |
| "eval_loss": 0.5684827566146851, | |
| "eval_runtime": 37.6302, | |
| "eval_samples_per_second": 260.828, | |
| "eval_steps_per_second": 8.158, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 4.8278985507246375e-06, | |
| "loss": 0.3117, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "eval_accuracy": 0.7943963321446765, | |
| "eval_loss": 0.5799030661582947, | |
| "eval_runtime": 35.9518, | |
| "eval_samples_per_second": 273.004, | |
| "eval_steps_per_second": 8.539, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 4.601449275362319e-06, | |
| "loss": 0.3108, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "eval_accuracy": 0.7987773815588385, | |
| "eval_loss": 0.5741643905639648, | |
| "eval_runtime": 36.1228, | |
| "eval_samples_per_second": 271.712, | |
| "eval_steps_per_second": 8.499, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 4.3750000000000005e-06, | |
| "loss": 0.3047, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "eval_accuracy": 0.7920529801324503, | |
| "eval_loss": 0.5906898975372314, | |
| "eval_runtime": 36.4484, | |
| "eval_samples_per_second": 269.284, | |
| "eval_steps_per_second": 8.423, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 4.148550724637682e-06, | |
| "loss": 0.3114, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "eval_accuracy": 0.7936831380539989, | |
| "eval_loss": 0.5723035335540771, | |
| "eval_runtime": 37.6048, | |
| "eval_samples_per_second": 261.004, | |
| "eval_steps_per_second": 8.164, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 3.9221014492753625e-06, | |
| "loss": 0.3035, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "eval_accuracy": 0.7955170657157412, | |
| "eval_loss": 0.5944135189056396, | |
| "eval_runtime": 37.5234, | |
| "eval_samples_per_second": 261.57, | |
| "eval_steps_per_second": 8.182, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 3.6956521739130436e-06, | |
| "loss": 0.3129, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "eval_accuracy": 0.7927661742231279, | |
| "eval_loss": 0.5837991833686829, | |
| "eval_runtime": 37.2958, | |
| "eval_samples_per_second": 263.167, | |
| "eval_steps_per_second": 8.231, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 3.4692028985507246e-06, | |
| "loss": 0.3071, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "eval_accuracy": 0.7949057564951605, | |
| "eval_loss": 0.5928712487220764, | |
| "eval_runtime": 37.6873, | |
| "eval_samples_per_second": 260.433, | |
| "eval_steps_per_second": 8.146, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 3.242753623188406e-06, | |
| "loss": 0.3061, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "eval_accuracy": 0.7967396841569027, | |
| "eval_loss": 0.579406201839447, | |
| "eval_runtime": 37.7053, | |
| "eval_samples_per_second": 260.308, | |
| "eval_steps_per_second": 8.142, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 3.016304347826087e-06, | |
| "loss": 0.3068, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "eval_accuracy": 0.7954151808456444, | |
| "eval_loss": 0.5892155766487122, | |
| "eval_runtime": 36.0876, | |
| "eval_samples_per_second": 271.977, | |
| "eval_steps_per_second": 8.507, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 2.7898550724637686e-06, | |
| "loss": 0.3053, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "eval_accuracy": 0.7962302598064187, | |
| "eval_loss": 0.5795509815216064, | |
| "eval_runtime": 36.0517, | |
| "eval_samples_per_second": 272.248, | |
| "eval_steps_per_second": 8.516, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 2.563405797101449e-06, | |
| "loss": 0.3117, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "eval_accuracy": 0.798064187468161, | |
| "eval_loss": 0.5763091444969177, | |
| "eval_runtime": 36.0281, | |
| "eval_samples_per_second": 272.426, | |
| "eval_steps_per_second": 8.521, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 2.3369565217391307e-06, | |
| "loss": 0.3062, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "eval_accuracy": 0.7964340295466124, | |
| "eval_loss": 0.5851861238479614, | |
| "eval_runtime": 37.7429, | |
| "eval_samples_per_second": 260.049, | |
| "eval_steps_per_second": 8.134, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 2.1105072463768117e-06, | |
| "loss": 0.3004, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "eval_accuracy": 0.7966377992868059, | |
| "eval_loss": 0.5793375968933105, | |
| "eval_runtime": 37.5317, | |
| "eval_samples_per_second": 261.512, | |
| "eval_steps_per_second": 8.18, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 1.884057971014493e-06, | |
| "loss": 0.3146, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "eval_accuracy": 0.7984717269485482, | |
| "eval_loss": 0.569298505783081, | |
| "eval_runtime": 36.0904, | |
| "eval_samples_per_second": 271.956, | |
| "eval_steps_per_second": 8.506, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 1.657608695652174e-06, | |
| "loss": 0.3146, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "eval_accuracy": 0.7981660723382578, | |
| "eval_loss": 0.5787694454193115, | |
| "eval_runtime": 36.6231, | |
| "eval_samples_per_second": 268.0, | |
| "eval_steps_per_second": 8.383, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 1.4311594202898552e-06, | |
| "loss": 0.3079, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "eval_accuracy": 0.7977585328578706, | |
| "eval_loss": 0.5726253986358643, | |
| "eval_runtime": 37.51, | |
| "eval_samples_per_second": 261.663, | |
| "eval_steps_per_second": 8.184, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "learning_rate": 1.2047101449275363e-06, | |
| "loss": 0.3058, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "eval_accuracy": 0.7987773815588385, | |
| "eval_loss": 0.5677123665809631, | |
| "eval_runtime": 37.4144, | |
| "eval_samples_per_second": 262.332, | |
| "eval_steps_per_second": 8.205, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 9.782608695652175e-07, | |
| "loss": 0.3055, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "eval_accuracy": 0.7981660723382578, | |
| "eval_loss": 0.5700839161872864, | |
| "eval_runtime": 36.0967, | |
| "eval_samples_per_second": 271.908, | |
| "eval_steps_per_second": 8.505, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 7.518115942028987e-07, | |
| "loss": 0.3049, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "eval_accuracy": 0.7970453387671931, | |
| "eval_loss": 0.5808990001678467, | |
| "eval_runtime": 37.6784, | |
| "eval_samples_per_second": 260.494, | |
| "eval_steps_per_second": 8.148, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 5.253623188405797e-07, | |
| "loss": 0.3044, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "eval_accuracy": 0.798573611818645, | |
| "eval_loss": 0.5740913152694702, | |
| "eval_runtime": 37.5561, | |
| "eval_samples_per_second": 261.342, | |
| "eval_steps_per_second": 8.174, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 2.989130434782609e-07, | |
| "loss": 0.3057, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "eval_accuracy": 0.7979623025980642, | |
| "eval_loss": 0.5742570757865906, | |
| "eval_runtime": 36.0024, | |
| "eval_samples_per_second": 272.621, | |
| "eval_steps_per_second": 8.527, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 7.246376811594204e-08, | |
| "loss": 0.3081, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "eval_accuracy": 0.7977585328578706, | |
| "eval_loss": 0.5770518183708191, | |
| "eval_runtime": 35.9949, | |
| "eval_samples_per_second": 272.677, | |
| "eval_steps_per_second": 8.529, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 88320, | |
| "total_flos": 1.0090192573609696e+17, | |
| "train_loss": 0.41023221776105356, | |
| "train_runtime": 36982.4162, | |
| "train_samples_per_second": 76.42, | |
| "train_steps_per_second": 2.388 | |
| } | |
| ], | |
| "max_steps": 88320, | |
| "num_train_epochs": 3, | |
| "total_flos": 1.0090192573609696e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |