| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.922136886324097, | |
| "global_step": 116000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 4.999957567785463e-05, | |
| "loss": 2.0156, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 4.915135570925447e-05, | |
| "loss": 1.305, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 4.8302711418508934e-05, | |
| "loss": 1.1809, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 4.74540671277634e-05, | |
| "loss": 1.1207, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 4.6605422837017866e-05, | |
| "loss": 1.0869, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 4.575677854627233e-05, | |
| "loss": 1.0604, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 4.49081342555268e-05, | |
| "loss": 1.0425, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 4.4059489964781264e-05, | |
| "loss": 1.0303, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 4.321084567403573e-05, | |
| "loss": 1.0101, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 4.2362201383290196e-05, | |
| "loss": 0.9935, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 4.151355709254466e-05, | |
| "loss": 0.9833, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 4.066491280179913e-05, | |
| "loss": 0.9833, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.1714409589767456, | |
| "eval_runtime": 285.104, | |
| "eval_samples_per_second": 50.936, | |
| "step": 23567 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 3.9816268511053594e-05, | |
| "loss": 0.9653, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 3.896762422030806e-05, | |
| "loss": 0.9378, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 3.8118979929562526e-05, | |
| "loss": 0.9398, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 3.727033563881699e-05, | |
| "loss": 0.9269, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 3.642169134807146e-05, | |
| "loss": 0.9311, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 3.5573047057325924e-05, | |
| "loss": 0.9179, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 3.472440276658039e-05, | |
| "loss": 0.91, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 3.3875758475834856e-05, | |
| "loss": 0.9146, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 3.302711418508932e-05, | |
| "loss": 0.9101, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 3.217846989434379e-05, | |
| "loss": 0.9024, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 3.132982560359825e-05, | |
| "loss": 0.9078, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 3.048118131285272e-05, | |
| "loss": 0.8919, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.1437541246414185, | |
| "eval_runtime": 285.2156, | |
| "eval_samples_per_second": 50.916, | |
| "step": 47134 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 2.9632537022107186e-05, | |
| "loss": 0.8866, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 2.878389273136165e-05, | |
| "loss": 0.876, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 2.793524844061612e-05, | |
| "loss": 0.8733, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 2.708660414987058e-05, | |
| "loss": 0.8685, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 2.623795985912505e-05, | |
| "loss": 0.8666, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 2.5389315568379513e-05, | |
| "loss": 0.8661, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 2.4540671277633982e-05, | |
| "loss": 0.8542, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 2.3692026986888445e-05, | |
| "loss": 0.8638, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 2.284338269614291e-05, | |
| "loss": 0.854, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 2.1994738405397377e-05, | |
| "loss": 0.8596, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 2.1146094114651846e-05, | |
| "loss": 0.8566, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 2.0297449823906312e-05, | |
| "loss": 0.8531, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.1323479413986206, | |
| "eval_runtime": 285.2208, | |
| "eval_samples_per_second": 50.915, | |
| "step": 70701 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 1.944880553316078e-05, | |
| "loss": 0.8331, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 1.8600161242415244e-05, | |
| "loss": 0.8402, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 1.7751516951669707e-05, | |
| "loss": 0.8381, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 1.6902872660924173e-05, | |
| "loss": 0.8362, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 1.605422837017864e-05, | |
| "loss": 0.8318, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "learning_rate": 1.5205584079433105e-05, | |
| "loss": 0.835, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 1.4356939788687573e-05, | |
| "loss": 0.8344, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "learning_rate": 1.3508295497942039e-05, | |
| "loss": 0.8354, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "learning_rate": 1.2659651207196505e-05, | |
| "loss": 0.8283, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "learning_rate": 1.181100691645097e-05, | |
| "loss": 0.8341, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "learning_rate": 1.0962362625705435e-05, | |
| "loss": 0.8225, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "learning_rate": 1.0113718334959903e-05, | |
| "loss": 0.8161, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.1247462034225464, | |
| "eval_runtime": 285.2246, | |
| "eval_samples_per_second": 50.914, | |
| "step": 94268 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 9.265074044214369e-06, | |
| "loss": 0.8112, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "learning_rate": 8.416429753468833e-06, | |
| "loss": 0.8188, | |
| "step": 98000 | |
| }, | |
| { | |
| "epoch": 4.24, | |
| "learning_rate": 7.567785462723299e-06, | |
| "loss": 0.8161, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 4.33, | |
| "learning_rate": 6.719141171977767e-06, | |
| "loss": 0.8157, | |
| "step": 102000 | |
| }, | |
| { | |
| "epoch": 4.41, | |
| "learning_rate": 5.870496881232232e-06, | |
| "loss": 0.8058, | |
| "step": 104000 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "learning_rate": 5.021852590486698e-06, | |
| "loss": 0.8122, | |
| "step": 106000 | |
| }, | |
| { | |
| "epoch": 4.58, | |
| "learning_rate": 4.173208299741164e-06, | |
| "loss": 0.8135, | |
| "step": 108000 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "learning_rate": 3.3245640089956294e-06, | |
| "loss": 0.816, | |
| "step": 110000 | |
| }, | |
| { | |
| "epoch": 4.75, | |
| "learning_rate": 2.475919718250096e-06, | |
| "loss": 0.8163, | |
| "step": 112000 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "learning_rate": 1.6272754275045616e-06, | |
| "loss": 0.813, | |
| "step": 114000 | |
| }, | |
| { | |
| "epoch": 4.92, | |
| "learning_rate": 7.786311367590275e-07, | |
| "loss": 0.8102, | |
| "step": 116000 | |
| } | |
| ], | |
| "max_steps": 117835, | |
| "num_train_epochs": 5, | |
| "total_flos": 443431511673274368, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |