| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 50.0, | |
| "eval_steps": 500, | |
| "global_step": 7400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.218454865161452, | |
| "eval_loss": 6.010812759399414, | |
| "eval_runtime": 12.8832, | |
| "eval_samples_per_second": 90.738, | |
| "eval_steps_per_second": 2.872, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.2297482956165591, | |
| "eval_loss": 5.830421447753906, | |
| "eval_runtime": 12.8689, | |
| "eval_samples_per_second": 90.839, | |
| "eval_steps_per_second": 2.875, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.23940862863018553, | |
| "eval_loss": 5.686750888824463, | |
| "eval_runtime": 12.8635, | |
| "eval_samples_per_second": 90.877, | |
| "eval_steps_per_second": 2.876, | |
| "step": 444 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "learning_rate": 4.6621621621621625e-06, | |
| "loss": 5.9726, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.24915536891584797, | |
| "eval_loss": 5.57318639755249, | |
| "eval_runtime": 12.8537, | |
| "eval_samples_per_second": 90.947, | |
| "eval_steps_per_second": 2.879, | |
| "step": 592 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.2586860910214204, | |
| "eval_loss": 5.474750518798828, | |
| "eval_runtime": 12.849, | |
| "eval_samples_per_second": 90.98, | |
| "eval_steps_per_second": 2.88, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.26733545895222544, | |
| "eval_loss": 5.387089729309082, | |
| "eval_runtime": 12.8597, | |
| "eval_samples_per_second": 90.904, | |
| "eval_steps_per_second": 2.877, | |
| "step": 888 | |
| }, | |
| { | |
| "epoch": 6.76, | |
| "learning_rate": 4.324324324324325e-06, | |
| "loss": 5.5397, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.27559599415886843, | |
| "eval_loss": 5.308462142944336, | |
| "eval_runtime": 12.8451, | |
| "eval_samples_per_second": 91.008, | |
| "eval_steps_per_second": 2.88, | |
| "step": 1036 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.2826554682842108, | |
| "eval_loss": 5.240093231201172, | |
| "eval_runtime": 12.8308, | |
| "eval_samples_per_second": 91.109, | |
| "eval_steps_per_second": 2.884, | |
| "step": 1184 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.2887731031443606, | |
| "eval_loss": 5.181127071380615, | |
| "eval_runtime": 12.8466, | |
| "eval_samples_per_second": 90.997, | |
| "eval_steps_per_second": 2.88, | |
| "step": 1332 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.2933094849262514, | |
| "eval_loss": 5.127747535705566, | |
| "eval_runtime": 12.8511, | |
| "eval_samples_per_second": 90.965, | |
| "eval_steps_per_second": 2.879, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 10.14, | |
| "learning_rate": 3.986486486486487e-06, | |
| "loss": 5.2883, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.29827790306832225, | |
| "eval_loss": 5.07957124710083, | |
| "eval_runtime": 12.8494, | |
| "eval_samples_per_second": 90.977, | |
| "eval_steps_per_second": 2.88, | |
| "step": 1628 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.3029957401214886, | |
| "eval_loss": 5.035754680633545, | |
| "eval_runtime": 12.8272, | |
| "eval_samples_per_second": 91.135, | |
| "eval_steps_per_second": 2.885, | |
| "step": 1776 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.30672853427344443, | |
| "eval_loss": 4.995058059692383, | |
| "eval_runtime": 12.8564, | |
| "eval_samples_per_second": 90.927, | |
| "eval_steps_per_second": 2.878, | |
| "step": 1924 | |
| }, | |
| { | |
| "epoch": 13.51, | |
| "learning_rate": 3.648648648648649e-06, | |
| "loss": 5.1076, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.31034035824454986, | |
| "eval_loss": 4.957174777984619, | |
| "eval_runtime": 12.8689, | |
| "eval_samples_per_second": 90.839, | |
| "eval_steps_per_second": 2.875, | |
| "step": 2072 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.3138916971252301, | |
| "eval_loss": 4.919981479644775, | |
| "eval_runtime": 12.8448, | |
| "eval_samples_per_second": 91.009, | |
| "eval_steps_per_second": 2.881, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.3172183770986166, | |
| "eval_loss": 4.887693405151367, | |
| "eval_runtime": 12.8684, | |
| "eval_samples_per_second": 90.842, | |
| "eval_steps_per_second": 2.875, | |
| "step": 2368 | |
| }, | |
| { | |
| "epoch": 16.89, | |
| "learning_rate": 3.310810810810811e-06, | |
| "loss": 4.9674, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.32033767961911674, | |
| "eval_loss": 4.855071544647217, | |
| "eval_runtime": 12.87, | |
| "eval_samples_per_second": 90.832, | |
| "eval_steps_per_second": 2.875, | |
| "step": 2516 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.32318911959630525, | |
| "eval_loss": 4.825829982757568, | |
| "eval_runtime": 12.8411, | |
| "eval_samples_per_second": 91.036, | |
| "eval_steps_per_second": 2.881, | |
| "step": 2664 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.3264725959336738, | |
| "eval_loss": 4.800779342651367, | |
| "eval_runtime": 12.8589, | |
| "eval_samples_per_second": 90.91, | |
| "eval_steps_per_second": 2.877, | |
| "step": 2812 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.3288919995506822, | |
| "eval_loss": 4.774311542510986, | |
| "eval_runtime": 12.8368, | |
| "eval_samples_per_second": 91.066, | |
| "eval_steps_per_second": 2.882, | |
| "step": 2960 | |
| }, | |
| { | |
| "epoch": 20.27, | |
| "learning_rate": 2.9729729729729736e-06, | |
| "loss": 4.858, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.3316915951646491, | |
| "eval_loss": 4.749689102172852, | |
| "eval_runtime": 12.8577, | |
| "eval_samples_per_second": 90.919, | |
| "eval_steps_per_second": 2.878, | |
| "step": 3108 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.33381721405673503, | |
| "eval_loss": 4.727055549621582, | |
| "eval_runtime": 12.849, | |
| "eval_samples_per_second": 90.98, | |
| "eval_steps_per_second": 2.88, | |
| "step": 3256 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 0.33634030640018664, | |
| "eval_loss": 4.705799102783203, | |
| "eval_runtime": 12.8414, | |
| "eval_samples_per_second": 91.034, | |
| "eval_steps_per_second": 2.881, | |
| "step": 3404 | |
| }, | |
| { | |
| "epoch": 23.65, | |
| "learning_rate": 2.6351351351351353e-06, | |
| "loss": 4.76, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.3384486438378654, | |
| "eval_loss": 4.686633586883545, | |
| "eval_runtime": 12.852, | |
| "eval_samples_per_second": 90.959, | |
| "eval_steps_per_second": 2.879, | |
| "step": 3552 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.3403841667314721, | |
| "eval_loss": 4.668373107910156, | |
| "eval_runtime": 12.8631, | |
| "eval_samples_per_second": 90.88, | |
| "eval_steps_per_second": 2.876, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.34248386344194726, | |
| "eval_loss": 4.648622512817383, | |
| "eval_runtime": 12.8633, | |
| "eval_samples_per_second": 90.879, | |
| "eval_steps_per_second": 2.876, | |
| "step": 3848 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 0.3443416197907216, | |
| "eval_loss": 4.632272243499756, | |
| "eval_runtime": 12.8435, | |
| "eval_samples_per_second": 91.019, | |
| "eval_steps_per_second": 2.881, | |
| "step": 3996 | |
| }, | |
| { | |
| "epoch": 27.03, | |
| "learning_rate": 2.297297297297298e-06, | |
| "loss": 4.6863, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.34594015432338787, | |
| "eval_loss": 4.615506172180176, | |
| "eval_runtime": 12.8661, | |
| "eval_samples_per_second": 90.859, | |
| "eval_steps_per_second": 2.876, | |
| "step": 4144 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 0.34756461103766495, | |
| "eval_loss": 4.601576805114746, | |
| "eval_runtime": 12.8739, | |
| "eval_samples_per_second": 90.804, | |
| "eval_steps_per_second": 2.874, | |
| "step": 4292 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.34898169029905557, | |
| "eval_loss": 4.5874128341674805, | |
| "eval_runtime": 12.8754, | |
| "eval_samples_per_second": 90.793, | |
| "eval_steps_per_second": 2.874, | |
| "step": 4440 | |
| }, | |
| { | |
| "epoch": 30.41, | |
| "learning_rate": 1.9594594594594595e-06, | |
| "loss": 4.6168, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_accuracy": 0.3504592546508714, | |
| "eval_loss": 4.574199676513672, | |
| "eval_runtime": 12.8467, | |
| "eval_samples_per_second": 90.996, | |
| "eval_steps_per_second": 2.88, | |
| "step": 4588 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.35180720809463323, | |
| "eval_loss": 4.562849998474121, | |
| "eval_runtime": 12.8628, | |
| "eval_samples_per_second": 90.882, | |
| "eval_steps_per_second": 2.877, | |
| "step": 4736 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_accuracy": 0.3534230240817067, | |
| "eval_loss": 4.550704479217529, | |
| "eval_runtime": 12.8788, | |
| "eval_samples_per_second": 90.769, | |
| "eval_steps_per_second": 2.873, | |
| "step": 4884 | |
| }, | |
| { | |
| "epoch": 33.78, | |
| "learning_rate": 1.6216216216216219e-06, | |
| "loss": 4.5684, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.35426117462045603, | |
| "eval_loss": 4.541166305541992, | |
| "eval_runtime": 12.9077, | |
| "eval_samples_per_second": 90.566, | |
| "eval_steps_per_second": 2.866, | |
| "step": 5032 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_accuracy": 0.3557905833354935, | |
| "eval_loss": 4.531555652618408, | |
| "eval_runtime": 12.8613, | |
| "eval_samples_per_second": 90.893, | |
| "eval_steps_per_second": 2.877, | |
| "step": 5180 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.3569570815079797, | |
| "eval_loss": 4.520727157592773, | |
| "eval_runtime": 12.9563, | |
| "eval_samples_per_second": 90.226, | |
| "eval_steps_per_second": 2.856, | |
| "step": 5328 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_accuracy": 0.3579594058635975, | |
| "eval_loss": 4.513218879699707, | |
| "eval_runtime": 12.9069, | |
| "eval_samples_per_second": 90.572, | |
| "eval_steps_per_second": 2.867, | |
| "step": 5476 | |
| }, | |
| { | |
| "epoch": 37.16, | |
| "learning_rate": 1.2837837837837838e-06, | |
| "loss": 4.5277, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_accuracy": 0.3587629934935324, | |
| "eval_loss": 4.505405426025391, | |
| "eval_runtime": 12.8784, | |
| "eval_samples_per_second": 90.772, | |
| "eval_steps_per_second": 2.873, | |
| "step": 5624 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_accuracy": 0.35967891057711415, | |
| "eval_loss": 4.499256610870361, | |
| "eval_runtime": 12.8813, | |
| "eval_samples_per_second": 90.752, | |
| "eval_steps_per_second": 2.872, | |
| "step": 5772 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.3604306538438275, | |
| "eval_loss": 4.493128776550293, | |
| "eval_runtime": 12.8448, | |
| "eval_samples_per_second": 91.009, | |
| "eval_steps_per_second": 2.881, | |
| "step": 5920 | |
| }, | |
| { | |
| "epoch": 40.54, | |
| "learning_rate": 9.459459459459461e-07, | |
| "loss": 4.4886, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_accuracy": 0.3610959898385048, | |
| "eval_loss": 4.487875461578369, | |
| "eval_runtime": 12.8279, | |
| "eval_samples_per_second": 91.13, | |
| "eval_steps_per_second": 2.884, | |
| "step": 6068 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_accuracy": 0.3616749185611461, | |
| "eval_loss": 4.482149124145508, | |
| "eval_runtime": 12.9187, | |
| "eval_samples_per_second": 90.489, | |
| "eval_steps_per_second": 2.864, | |
| "step": 6216 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_accuracy": 0.3622365658293802, | |
| "eval_loss": 4.477830410003662, | |
| "eval_runtime": 12.8586, | |
| "eval_samples_per_second": 90.912, | |
| "eval_steps_per_second": 2.877, | |
| "step": 6364 | |
| }, | |
| { | |
| "epoch": 43.92, | |
| "learning_rate": 6.081081081081082e-07, | |
| "loss": 4.4727, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.36262539855354226, | |
| "eval_loss": 4.474149703979492, | |
| "eval_runtime": 12.8383, | |
| "eval_samples_per_second": 91.056, | |
| "eval_steps_per_second": 2.882, | |
| "step": 6512 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_accuracy": 0.3629969498232971, | |
| "eval_loss": 4.471028804779053, | |
| "eval_runtime": 12.8363, | |
| "eval_samples_per_second": 91.07, | |
| "eval_steps_per_second": 2.882, | |
| "step": 6660 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_accuracy": 0.36328209382101595, | |
| "eval_loss": 4.469077110290527, | |
| "eval_runtime": 12.8487, | |
| "eval_samples_per_second": 90.982, | |
| "eval_steps_per_second": 2.88, | |
| "step": 6808 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_accuracy": 0.3634462676378844, | |
| "eval_loss": 4.4663591384887695, | |
| "eval_runtime": 12.8695, | |
| "eval_samples_per_second": 90.835, | |
| "eval_steps_per_second": 2.875, | |
| "step": 6956 | |
| }, | |
| { | |
| "epoch": 47.3, | |
| "learning_rate": 2.702702702702703e-07, | |
| "loss": 4.4542, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.36362772290916, | |
| "eval_loss": 4.465246200561523, | |
| "eval_runtime": 12.8565, | |
| "eval_samples_per_second": 90.926, | |
| "eval_steps_per_second": 2.878, | |
| "step": 7104 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "eval_accuracy": 0.36365364509077086, | |
| "eval_loss": 4.464395046234131, | |
| "eval_runtime": 12.8452, | |
| "eval_samples_per_second": 91.007, | |
| "eval_steps_per_second": 2.88, | |
| "step": 7252 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_accuracy": 0.36367956727238165, | |
| "eval_loss": 4.464205741882324, | |
| "eval_runtime": 12.8483, | |
| "eval_samples_per_second": 90.984, | |
| "eval_steps_per_second": 2.88, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "step": 7400, | |
| "total_flos": 1.00265577216e+17, | |
| "train_loss": 4.855868909681165, | |
| "train_runtime": 6356.4736, | |
| "train_samples_per_second": 37.112, | |
| "train_steps_per_second": 1.164 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 7400, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "total_flos": 1.00265577216e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |