| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.5045728266982867, | |
| "global_step": 420000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 4.8208841872978234e-05, | |
| "loss": 0.0755, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 4.6417683745956466e-05, | |
| "loss": 0.0721, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_loss": 0.06494459509849548, | |
| "eval_runtime": 754.7473, | |
| "eval_samples_per_second": 82.19, | |
| "eval_steps_per_second": 20.549, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 4.462652561893469e-05, | |
| "loss": 0.0704, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 4.2835367491912923e-05, | |
| "loss": 0.0701, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "eval_loss": 0.06424280256032944, | |
| "eval_runtime": 676.9024, | |
| "eval_samples_per_second": 91.642, | |
| "eval_steps_per_second": 22.912, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 4.104420936489115e-05, | |
| "loss": 0.0713, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 3.925305123786938e-05, | |
| "loss": 0.0712, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "eval_loss": 0.0624563954770565, | |
| "eval_runtime": 678.8069, | |
| "eval_samples_per_second": 91.385, | |
| "eval_steps_per_second": 22.847, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 3.746189311084761e-05, | |
| "loss": 0.0698, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 3.5670734983825844e-05, | |
| "loss": 0.0667, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 0.06348983943462372, | |
| "eval_runtime": 675.6836, | |
| "eval_samples_per_second": 91.808, | |
| "eval_steps_per_second": 22.953, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 3.3879576856804076e-05, | |
| "loss": 0.0657, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 3.20884187297823e-05, | |
| "loss": 0.0665, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_loss": 0.06186612322926521, | |
| "eval_runtime": 676.1915, | |
| "eval_samples_per_second": 91.739, | |
| "eval_steps_per_second": 22.936, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 3.0297260602760537e-05, | |
| "loss": 0.0657, | |
| "step": 110000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 2.8506102475738766e-05, | |
| "loss": 0.0653, | |
| "step": 120000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "eval_loss": 0.061661478132009506, | |
| "eval_runtime": 677.4648, | |
| "eval_samples_per_second": 91.566, | |
| "eval_steps_per_second": 22.893, | |
| "step": 120000 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 2.6714944348716997e-05, | |
| "loss": 0.0658, | |
| "step": 130000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 2.4923786221695226e-05, | |
| "loss": 0.0656, | |
| "step": 140000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 0.06098250672221184, | |
| "eval_runtime": 677.3485, | |
| "eval_samples_per_second": 91.582, | |
| "eval_steps_per_second": 22.897, | |
| "step": 140000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 2.3132628094673455e-05, | |
| "loss": 0.0649, | |
| "step": 150000 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 2.1341469967651687e-05, | |
| "loss": 0.0648, | |
| "step": 160000 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "eval_loss": 0.06077565252780914, | |
| "eval_runtime": 677.1708, | |
| "eval_samples_per_second": 91.606, | |
| "eval_steps_per_second": 22.903, | |
| "step": 160000 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 1.9550311840629915e-05, | |
| "loss": 0.0642, | |
| "step": 170000 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 1.7759153713608147e-05, | |
| "loss": 0.0649, | |
| "step": 180000 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "eval_loss": 0.05941811203956604, | |
| "eval_runtime": 677.9109, | |
| "eval_samples_per_second": 91.506, | |
| "eval_steps_per_second": 22.878, | |
| "step": 180000 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 1.5967995586586376e-05, | |
| "loss": 0.0652, | |
| "step": 190000 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 1.4176837459564608e-05, | |
| "loss": 0.0638, | |
| "step": 200000 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "eval_loss": 0.06042506918311119, | |
| "eval_runtime": 678.1922, | |
| "eval_samples_per_second": 91.468, | |
| "eval_steps_per_second": 22.868, | |
| "step": 200000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 1.2385679332542836e-05, | |
| "loss": 0.0621, | |
| "step": 210000 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 1.0594521205521066e-05, | |
| "loss": 0.0643, | |
| "step": 220000 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "eval_loss": 0.059569716453552246, | |
| "eval_runtime": 688.4187, | |
| "eval_samples_per_second": 90.109, | |
| "eval_steps_per_second": 22.528, | |
| "step": 220000 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 8.803363078499297e-06, | |
| "loss": 0.0623, | |
| "step": 230000 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 7.012204951477527e-06, | |
| "loss": 0.0646, | |
| "step": 240000 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "eval_loss": 0.059120796620845795, | |
| "eval_runtime": 715.8659, | |
| "eval_samples_per_second": 86.654, | |
| "eval_steps_per_second": 21.665, | |
| "step": 240000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 4.5522104682445576e-05, | |
| "loss": 0.0634, | |
| "step": 250000 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 4.53429888697434e-05, | |
| "loss": 0.0648, | |
| "step": 260000 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "eval_loss": 0.06051425263285637, | |
| "eval_runtime": 694.96, | |
| "eval_samples_per_second": 89.261, | |
| "eval_steps_per_second": 22.316, | |
| "step": 260000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 4.516387305704122e-05, | |
| "loss": 0.0644, | |
| "step": 270000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.4984757244339046e-05, | |
| "loss": 0.0655, | |
| "step": 280000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.06069951504468918, | |
| "eval_runtime": 679.0968, | |
| "eval_samples_per_second": 91.346, | |
| "eval_steps_per_second": 22.838, | |
| "step": 280000 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 4.4805641431636866e-05, | |
| "loss": 0.0652, | |
| "step": 290000 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 4.462652561893469e-05, | |
| "loss": 0.0631, | |
| "step": 300000 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "eval_loss": 0.06102127209305763, | |
| "eval_runtime": 705.9715, | |
| "eval_samples_per_second": 87.869, | |
| "eval_steps_per_second": 21.968, | |
| "step": 300000 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 4.444740980623252e-05, | |
| "loss": 0.064, | |
| "step": 310000 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 4.426829399353034e-05, | |
| "loss": 0.0643, | |
| "step": 320000 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "eval_loss": 0.05971848592162132, | |
| "eval_runtime": 698.4495, | |
| "eval_samples_per_second": 88.815, | |
| "eval_steps_per_second": 22.205, | |
| "step": 320000 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 4.408917818082816e-05, | |
| "loss": 0.0638, | |
| "step": 330000 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 4.391006236812599e-05, | |
| "loss": 0.064, | |
| "step": 340000 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "eval_loss": 0.05945688858628273, | |
| "eval_runtime": 700.9514, | |
| "eval_samples_per_second": 88.498, | |
| "eval_steps_per_second": 22.126, | |
| "step": 340000 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 4.373094655542381e-05, | |
| "loss": 0.0613, | |
| "step": 350000 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 4.355183074272163e-05, | |
| "loss": 0.0635, | |
| "step": 360000 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "eval_loss": 0.06002034246921539, | |
| "eval_runtime": 697.4531, | |
| "eval_samples_per_second": 88.942, | |
| "eval_steps_per_second": 22.237, | |
| "step": 360000 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 4.337271493001945e-05, | |
| "loss": 0.0635, | |
| "step": 370000 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 4.319359911731728e-05, | |
| "loss": 0.0627, | |
| "step": 380000 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "eval_loss": 0.0593414306640625, | |
| "eval_runtime": 720.1404, | |
| "eval_samples_per_second": 86.14, | |
| "eval_steps_per_second": 21.536, | |
| "step": 380000 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 4.30144833046151e-05, | |
| "loss": 0.0629, | |
| "step": 390000 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 4.2835367491912923e-05, | |
| "loss": 0.0642, | |
| "step": 400000 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "eval_loss": 0.059706129133701324, | |
| "eval_runtime": 706.6838, | |
| "eval_samples_per_second": 87.78, | |
| "eval_steps_per_second": 21.946, | |
| "step": 400000 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 4.265625167921074e-05, | |
| "loss": 0.062, | |
| "step": 410000 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 4.247713586650857e-05, | |
| "loss": 0.0616, | |
| "step": 420000 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "eval_loss": 0.058674536645412445, | |
| "eval_runtime": 740.0142, | |
| "eval_samples_per_second": 83.827, | |
| "eval_steps_per_second": 20.958, | |
| "step": 420000 | |
| } | |
| ], | |
| "max_steps": 2791490, | |
| "num_train_epochs": 10, | |
| "total_flos": 2.8075475590415155e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |