| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.560233918128655, | |
| "eval_steps": 500, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.031189083820662766, | |
| "grad_norm": 126.5, | |
| "learning_rate": 8.000000000000001e-07, | |
| "loss": 0.5428, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06237816764132553, | |
| "grad_norm": 326.0, | |
| "learning_rate": 1.6000000000000001e-06, | |
| "loss": 0.4758, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0935672514619883, | |
| "grad_norm": 524.0, | |
| "learning_rate": 2.4000000000000003e-06, | |
| "loss": 0.5854, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.12475633528265107, | |
| "grad_norm": 446.0, | |
| "learning_rate": 3.2000000000000003e-06, | |
| "loss": 0.8299, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.15594541910331383, | |
| "grad_norm": 268.0, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.0183, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.1871345029239766, | |
| "grad_norm": 1592.0, | |
| "learning_rate": 4.800000000000001e-06, | |
| "loss": 1.316, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.21832358674463936, | |
| "grad_norm": 139.0, | |
| "learning_rate": 5.600000000000001e-06, | |
| "loss": 1.5705, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.24951267056530213, | |
| "grad_norm": 374.0, | |
| "learning_rate": 6.4000000000000006e-06, | |
| "loss": 5.897, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.2807017543859649, | |
| "grad_norm": 115.0, | |
| "learning_rate": 7.2000000000000005e-06, | |
| "loss": 1.1907, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.31189083820662766, | |
| "grad_norm": 824.0, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 1.602, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.34307992202729043, | |
| "grad_norm": 74.5, | |
| "learning_rate": 8.8e-06, | |
| "loss": 1.3299, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.3742690058479532, | |
| "grad_norm": 129.0, | |
| "learning_rate": 9.600000000000001e-06, | |
| "loss": 1.106, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.40545808966861596, | |
| "grad_norm": 168.0, | |
| "learning_rate": 1.04e-05, | |
| "loss": 1.5673, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.43664717348927873, | |
| "grad_norm": 52.75, | |
| "learning_rate": 1.1200000000000001e-05, | |
| "loss": 0.9792, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.4678362573099415, | |
| "grad_norm": 95.5, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.9398, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.49902534113060426, | |
| "grad_norm": 66.0, | |
| "learning_rate": 1.2800000000000001e-05, | |
| "loss": 0.8719, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.530214424951267, | |
| "grad_norm": 46.75, | |
| "learning_rate": 1.3600000000000002e-05, | |
| "loss": 1.0627, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.5614035087719298, | |
| "grad_norm": 41.75, | |
| "learning_rate": 1.4400000000000001e-05, | |
| "loss": 0.8263, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.5925925925925926, | |
| "grad_norm": 45.5, | |
| "learning_rate": 1.5200000000000002e-05, | |
| "loss": 0.8172, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.6237816764132553, | |
| "grad_norm": 41.5, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.8474, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.6549707602339181, | |
| "grad_norm": 41.75, | |
| "learning_rate": 1.6800000000000002e-05, | |
| "loss": 0.8815, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.6861598440545809, | |
| "grad_norm": 45.75, | |
| "learning_rate": 1.76e-05, | |
| "loss": 0.8097, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.7173489278752436, | |
| "grad_norm": 412.0, | |
| "learning_rate": 1.8400000000000003e-05, | |
| "loss": 1.0962, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.7485380116959064, | |
| "grad_norm": 52.25, | |
| "learning_rate": 1.9200000000000003e-05, | |
| "loss": 10.0622, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.7797270955165692, | |
| "grad_norm": 38.0, | |
| "learning_rate": 2e-05, | |
| "loss": 1.358, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.8109161793372319, | |
| "grad_norm": 47.25, | |
| "learning_rate": 1.9999025240093045e-05, | |
| "loss": 1.0034, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.8421052631578947, | |
| "grad_norm": 22.5, | |
| "learning_rate": 1.9996101150403543e-05, | |
| "loss": 0.7018, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.8732943469785575, | |
| "grad_norm": 42.0, | |
| "learning_rate": 1.9991228300988586e-05, | |
| "loss": 0.6776, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.9044834307992202, | |
| "grad_norm": 29.0, | |
| "learning_rate": 1.9984407641819812e-05, | |
| "loss": 0.7348, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.935672514619883, | |
| "grad_norm": 42.5, | |
| "learning_rate": 1.9975640502598243e-05, | |
| "loss": 0.6959, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.9668615984405458, | |
| "grad_norm": 45.75, | |
| "learning_rate": 1.9964928592495046e-05, | |
| "loss": 0.7085, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.9980506822612085, | |
| "grad_norm": 38.0, | |
| "learning_rate": 1.9952273999818312e-05, | |
| "loss": 0.6367, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.0300194931773878, | |
| "grad_norm": 44.0, | |
| "learning_rate": 1.9937679191605964e-05, | |
| "loss": 0.6069, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.0612085769980506, | |
| "grad_norm": 17.625, | |
| "learning_rate": 1.9921147013144782e-05, | |
| "loss": 0.6435, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.0923976608187134, | |
| "grad_norm": 26.875, | |
| "learning_rate": 1.9902680687415704e-05, | |
| "loss": 0.5493, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.1235867446393761, | |
| "grad_norm": 33.5, | |
| "learning_rate": 1.988228381446553e-05, | |
| "loss": 0.5624, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.154775828460039, | |
| "grad_norm": 40.75, | |
| "learning_rate": 1.985996037070505e-05, | |
| "loss": 0.5981, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.1859649122807017, | |
| "grad_norm": 17.125, | |
| "learning_rate": 1.983571470813386e-05, | |
| "loss": 0.656, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.2171539961013644, | |
| "grad_norm": 21.5, | |
| "learning_rate": 1.9809551553491918e-05, | |
| "loss": 0.4947, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.2483430799220272, | |
| "grad_norm": 29.0, | |
| "learning_rate": 1.9781476007338058e-05, | |
| "loss": 0.562, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.2795321637426902, | |
| "grad_norm": 8.1875, | |
| "learning_rate": 1.9751493543055634e-05, | |
| "loss": 0.4717, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.310721247563353, | |
| "grad_norm": 41.5, | |
| "learning_rate": 1.9719610005785466e-05, | |
| "loss": 0.5865, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.3419103313840157, | |
| "grad_norm": 23.375, | |
| "learning_rate": 1.9685831611286312e-05, | |
| "loss": 0.4994, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.3730994152046785, | |
| "grad_norm": 18.375, | |
| "learning_rate": 1.9650164944723116e-05, | |
| "loss": 0.4635, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.4042884990253413, | |
| "grad_norm": 25.0, | |
| "learning_rate": 1.961261695938319e-05, | |
| "loss": 0.5149, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.435477582846004, | |
| "grad_norm": 37.5, | |
| "learning_rate": 1.9573194975320672e-05, | |
| "loss": 0.5077, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.4666666666666668, | |
| "grad_norm": 21.0, | |
| "learning_rate": 1.9531906677929472e-05, | |
| "loss": 0.5361, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.4978557504873296, | |
| "grad_norm": 11.3125, | |
| "learning_rate": 1.9488760116444966e-05, | |
| "loss": 0.5036, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.5290448343079923, | |
| "grad_norm": 39.75, | |
| "learning_rate": 1.944376370237481e-05, | |
| "loss": 0.4732, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.560233918128655, | |
| "grad_norm": 17.5, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.4622, | |
| "step": 500 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 2500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 8, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.91328310788096e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |