| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.1212475633528265, | |
| "eval_steps": 500, | |
| "global_step": 1000, | |
| "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 | |
| }, | |
| { | |
| "epoch": 1.5914230019493179, | |
| "grad_norm": 13.9375, | |
| "learning_rate": 1.9348256763960146e-05, | |
| "loss": 0.4519, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.6226120857699806, | |
| "grad_norm": 21.375, | |
| "learning_rate": 1.9297764858882516e-05, | |
| "loss": 0.5471, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.6538011695906434, | |
| "grad_norm": 34.25, | |
| "learning_rate": 1.9245460336123136e-05, | |
| "loss": 0.4767, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.6849902534113061, | |
| "grad_norm": 23.625, | |
| "learning_rate": 1.9191353392552346e-05, | |
| "loss": 0.4921, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.716179337231969, | |
| "grad_norm": 21.125, | |
| "learning_rate": 1.913545457642601e-05, | |
| "loss": 0.4537, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.7473684210526317, | |
| "grad_norm": 28.5, | |
| "learning_rate": 1.907777478532909e-05, | |
| "loss": 0.466, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.7785575048732944, | |
| "grad_norm": 15.375, | |
| "learning_rate": 1.901832526405114e-05, | |
| "loss": 0.4803, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.8097465886939572, | |
| "grad_norm": 12.625, | |
| "learning_rate": 1.895711760239413e-05, | |
| "loss": 0.4508, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.84093567251462, | |
| "grad_norm": 11.25, | |
| "learning_rate": 1.889416373291298e-05, | |
| "loss": 0.4905, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.8721247563352827, | |
| "grad_norm": 11.375, | |
| "learning_rate": 1.8829475928589272e-05, | |
| "loss": 0.4482, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.9033138401559455, | |
| "grad_norm": 43.25, | |
| "learning_rate": 1.8763066800438638e-05, | |
| "loss": 0.4531, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.9345029239766083, | |
| "grad_norm": 24.75, | |
| "learning_rate": 1.869494929505219e-05, | |
| "loss": 0.4268, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.965692007797271, | |
| "grad_norm": 49.5, | |
| "learning_rate": 1.8625136692072577e-05, | |
| "loss": 0.4737, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.9968810916179338, | |
| "grad_norm": 22.5, | |
| "learning_rate": 1.855364260160507e-05, | |
| "loss": 0.487, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.028849902534113, | |
| "grad_norm": 15.1875, | |
| "learning_rate": 1.848048096156426e-05, | |
| "loss": 0.4299, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.0600389863547757, | |
| "grad_norm": 8.5, | |
| "learning_rate": 1.8405666034956842e-05, | |
| "loss": 0.4246, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.0912280701754384, | |
| "grad_norm": 7.78125, | |
| "learning_rate": 1.8329212407100996e-05, | |
| "loss": 0.4017, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.122417153996101, | |
| "grad_norm": 23.75, | |
| "learning_rate": 1.8251134982782952e-05, | |
| "loss": 0.4378, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.153606237816764, | |
| "grad_norm": 17.125, | |
| "learning_rate": 1.8171448983351284e-05, | |
| "loss": 0.4024, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.1847953216374267, | |
| "grad_norm": 19.125, | |
| "learning_rate": 1.8090169943749477e-05, | |
| "loss": 0.395, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.2159844054580895, | |
| "grad_norm": 21.125, | |
| "learning_rate": 1.8007313709487334e-05, | |
| "loss": 0.3597, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.2471734892787523, | |
| "grad_norm": 4.6875, | |
| "learning_rate": 1.792289643355191e-05, | |
| "loss": 0.3673, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.278362573099415, | |
| "grad_norm": 8.625, | |
| "learning_rate": 1.78369345732584e-05, | |
| "loss": 0.3961, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.309551656920078, | |
| "grad_norm": 12.875, | |
| "learning_rate": 1.7749444887041797e-05, | |
| "loss": 0.3944, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.3407407407407406, | |
| "grad_norm": 17.375, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.3777, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.3719298245614033, | |
| "grad_norm": 12.375, | |
| "learning_rate": 1.7569950556517566e-05, | |
| "loss": 0.3799, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.403118908382066, | |
| "grad_norm": 23.625, | |
| "learning_rate": 1.747798090498532e-05, | |
| "loss": 0.374, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.434307992202729, | |
| "grad_norm": 23.125, | |
| "learning_rate": 1.7384553406258842e-05, | |
| "loss": 0.3891, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.4654970760233916, | |
| "grad_norm": 15.1875, | |
| "learning_rate": 1.7289686274214116e-05, | |
| "loss": 0.3792, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.4966861598440544, | |
| "grad_norm": 34.5, | |
| "learning_rate": 1.7193398003386514e-05, | |
| "loss": 0.374, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.5278752436647176, | |
| "grad_norm": 15.875, | |
| "learning_rate": 1.709570736536521e-05, | |
| "loss": 0.3616, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 2.5590643274853804, | |
| "grad_norm": 5.09375, | |
| "learning_rate": 1.6996633405133656e-05, | |
| "loss": 0.3745, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.590253411306043, | |
| "grad_norm": 20.125, | |
| "learning_rate": 1.68961954373567e-05, | |
| "loss": 0.3756, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 2.621442495126706, | |
| "grad_norm": 17.875, | |
| "learning_rate": 1.6794413042615168e-05, | |
| "loss": 0.3736, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 2.6526315789473687, | |
| "grad_norm": 7.84375, | |
| "learning_rate": 1.6691306063588583e-05, | |
| "loss": 0.3827, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.6838206627680314, | |
| "grad_norm": 16.625, | |
| "learning_rate": 1.6586894601186804e-05, | |
| "loss": 0.376, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 2.715009746588694, | |
| "grad_norm": 20.125, | |
| "learning_rate": 1.6481199010631312e-05, | |
| "loss": 0.3706, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 2.746198830409357, | |
| "grad_norm": 13.6875, | |
| "learning_rate": 1.63742398974869e-05, | |
| "loss": 0.3692, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 2.7773879142300197, | |
| "grad_norm": 10.5625, | |
| "learning_rate": 1.6266038113644605e-05, | |
| "loss": 0.3232, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 2.8085769980506825, | |
| "grad_norm": 11.75, | |
| "learning_rate": 1.6156614753256583e-05, | |
| "loss": 0.3514, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.8397660818713453, | |
| "grad_norm": 16.625, | |
| "learning_rate": 1.6045991148623752e-05, | |
| "loss": 0.3483, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 2.870955165692008, | |
| "grad_norm": 6.9375, | |
| "learning_rate": 1.5934188866037017e-05, | |
| "loss": 0.3546, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 2.902144249512671, | |
| "grad_norm": 28.0, | |
| "learning_rate": 1.5821229701572897e-05, | |
| "loss": 0.3382, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 2.9333333333333336, | |
| "grad_norm": 4.9375, | |
| "learning_rate": 1.570713567684432e-05, | |
| "loss": 0.327, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 2.9645224171539963, | |
| "grad_norm": 6.71875, | |
| "learning_rate": 1.5591929034707468e-05, | |
| "loss": 0.3372, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.995711500974659, | |
| "grad_norm": 14.375, | |
| "learning_rate": 1.5475632234925505e-05, | |
| "loss": 0.3392, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 3.027680311890838, | |
| "grad_norm": 21.0, | |
| "learning_rate": 1.5358267949789968e-05, | |
| "loss": 0.3177, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 3.058869395711501, | |
| "grad_norm": 17.0, | |
| "learning_rate": 1.5239859059700794e-05, | |
| "loss": 0.3119, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 3.0900584795321637, | |
| "grad_norm": 8.1875, | |
| "learning_rate": 1.5120428648705716e-05, | |
| "loss": 0.304, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 3.1212475633528265, | |
| "grad_norm": 10.75, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.2962, | |
| "step": 1000 | |
| } | |
| ], | |
| "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": 3.82656621576192e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |