| { | |
| "best_metric": 0.8267716535433071, | |
| "best_model_checkpoint": "videomae-base-finetuned-isl-numbers_2/checkpoint-3572", | |
| "epoch": 49.02, | |
| "eval_steps": 500, | |
| "global_step": 3800, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.002631578947368421, | |
| "grad_norm": 8.72721004486084, | |
| "learning_rate": 1.3157894736842106e-06, | |
| "loss": 3.642, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.005263157894736842, | |
| "grad_norm": 11.387261390686035, | |
| "learning_rate": 2.631578947368421e-06, | |
| "loss": 3.6937, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.007894736842105263, | |
| "grad_norm": 9.664594650268555, | |
| "learning_rate": 3.9473684210526315e-06, | |
| "loss": 3.6305, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.010526315789473684, | |
| "grad_norm": 8.870984077453613, | |
| "learning_rate": 5.263157894736842e-06, | |
| "loss": 3.6148, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.013157894736842105, | |
| "grad_norm": 9.365774154663086, | |
| "learning_rate": 6.578947368421053e-06, | |
| "loss": 3.5805, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.015789473684210527, | |
| "grad_norm": 9.285212516784668, | |
| "learning_rate": 7.894736842105263e-06, | |
| "loss": 3.564, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.018421052631578946, | |
| "grad_norm": 8.437307357788086, | |
| "learning_rate": 9.210526315789474e-06, | |
| "loss": 3.5502, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "eval_accuracy": 0.09448818897637795, | |
| "eval_loss": 3.425055980682373, | |
| "eval_runtime": 196.4771, | |
| "eval_samples_per_second": 0.646, | |
| "eval_steps_per_second": 0.081, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 1.0010526315789474, | |
| "grad_norm": 9.444938659667969, | |
| "learning_rate": 1.0526315789473684e-05, | |
| "loss": 3.4468, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.0036842105263157, | |
| "grad_norm": 9.038703918457031, | |
| "learning_rate": 1.1842105263157895e-05, | |
| "loss": 3.3166, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.0063157894736843, | |
| "grad_norm": 9.809856414794922, | |
| "learning_rate": 1.3157894736842106e-05, | |
| "loss": 3.3522, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0089473684210526, | |
| "grad_norm": 9.944860458374023, | |
| "learning_rate": 1.4473684210526317e-05, | |
| "loss": 3.1998, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.0115789473684211, | |
| "grad_norm": 9.153931617736816, | |
| "learning_rate": 1.5789473684210526e-05, | |
| "loss": 3.175, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.0142105263157895, | |
| "grad_norm": 10.077743530273438, | |
| "learning_rate": 1.7105263157894737e-05, | |
| "loss": 3.1365, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.016842105263158, | |
| "grad_norm": 10.884031295776367, | |
| "learning_rate": 1.8421052631578947e-05, | |
| "loss": 3.1498, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.0194736842105263, | |
| "grad_norm": 10.6647310256958, | |
| "learning_rate": 1.9736842105263158e-05, | |
| "loss": 3.1152, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_accuracy": 0.2755905511811024, | |
| "eval_loss": 3.0364396572113037, | |
| "eval_runtime": 218.5017, | |
| "eval_samples_per_second": 0.581, | |
| "eval_steps_per_second": 0.073, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 2.002105263157895, | |
| "grad_norm": 9.977360725402832, | |
| "learning_rate": 2.105263157894737e-05, | |
| "loss": 2.8494, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.004736842105263, | |
| "grad_norm": 10.521862983703613, | |
| "learning_rate": 2.236842105263158e-05, | |
| "loss": 2.7613, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.0073684210526315, | |
| "grad_norm": 11.251272201538086, | |
| "learning_rate": 2.368421052631579e-05, | |
| "loss": 2.8185, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "grad_norm": 11.353875160217285, | |
| "learning_rate": 2.5e-05, | |
| "loss": 2.9295, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.0126315789473685, | |
| "grad_norm": 13.587601661682129, | |
| "learning_rate": 2.6315789473684212e-05, | |
| "loss": 2.6847, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.015263157894737, | |
| "grad_norm": 9.29124641418457, | |
| "learning_rate": 2.7631578947368426e-05, | |
| "loss": 2.6595, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.017894736842105, | |
| "grad_norm": 10.696941375732422, | |
| "learning_rate": 2.8947368421052634e-05, | |
| "loss": 2.6365, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "eval_accuracy": 0.3779527559055118, | |
| "eval_loss": 2.619691848754883, | |
| "eval_runtime": 215.3407, | |
| "eval_samples_per_second": 0.59, | |
| "eval_steps_per_second": 0.074, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 3.0005263157894735, | |
| "grad_norm": 10.942057609558105, | |
| "learning_rate": 3.0263157894736844e-05, | |
| "loss": 2.6872, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 3.0031578947368422, | |
| "grad_norm": 12.153741836547852, | |
| "learning_rate": 3.157894736842105e-05, | |
| "loss": 2.4375, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.0057894736842106, | |
| "grad_norm": 9.289186477661133, | |
| "learning_rate": 3.289473684210527e-05, | |
| "loss": 2.2951, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.008421052631579, | |
| "grad_norm": 11.481297492980957, | |
| "learning_rate": 3.421052631578947e-05, | |
| "loss": 2.3551, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 3.011052631578947, | |
| "grad_norm": 14.849848747253418, | |
| "learning_rate": 3.5526315789473684e-05, | |
| "loss": 2.4047, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 3.013684210526316, | |
| "grad_norm": 12.359841346740723, | |
| "learning_rate": 3.6842105263157895e-05, | |
| "loss": 2.3038, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 3.0163157894736843, | |
| "grad_norm": 12.507572174072266, | |
| "learning_rate": 3.815789473684211e-05, | |
| "loss": 2.3412, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 3.0189473684210526, | |
| "grad_norm": 11.021321296691895, | |
| "learning_rate": 3.9473684210526316e-05, | |
| "loss": 2.3879, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "eval_accuracy": 0.4645669291338583, | |
| "eval_loss": 2.1518962383270264, | |
| "eval_runtime": 218.0756, | |
| "eval_samples_per_second": 0.582, | |
| "eval_steps_per_second": 0.073, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 4.001578947368421, | |
| "grad_norm": 10.985835075378418, | |
| "learning_rate": 4.078947368421053e-05, | |
| "loss": 2.081, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 4.00421052631579, | |
| "grad_norm": 11.158766746520996, | |
| "learning_rate": 4.210526315789474e-05, | |
| "loss": 2.0336, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 4.006842105263158, | |
| "grad_norm": 18.02586555480957, | |
| "learning_rate": 4.342105263157895e-05, | |
| "loss": 1.9907, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 4.009473684210526, | |
| "grad_norm": 9.385698318481445, | |
| "learning_rate": 4.473684210526316e-05, | |
| "loss": 2.1798, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 4.012105263157895, | |
| "grad_norm": 13.736913681030273, | |
| "learning_rate": 4.605263157894737e-05, | |
| "loss": 1.8809, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 4.014736842105263, | |
| "grad_norm": 13.594823837280273, | |
| "learning_rate": 4.736842105263158e-05, | |
| "loss": 1.9144, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 4.017368421052631, | |
| "grad_norm": 17.99551773071289, | |
| "learning_rate": 4.868421052631579e-05, | |
| "loss": 2.0676, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "grad_norm": 14.397540092468262, | |
| "learning_rate": 5e-05, | |
| "loss": 1.9396, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "eval_accuracy": 0.41732283464566927, | |
| "eval_loss": 2.080385684967041, | |
| "eval_runtime": 207.8742, | |
| "eval_samples_per_second": 0.611, | |
| "eval_steps_per_second": 0.077, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 5.002631578947368, | |
| "grad_norm": 15.16203498840332, | |
| "learning_rate": 4.985380116959065e-05, | |
| "loss": 1.7725, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 5.005263157894737, | |
| "grad_norm": 17.93940544128418, | |
| "learning_rate": 4.970760233918128e-05, | |
| "loss": 1.7235, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 5.007894736842105, | |
| "grad_norm": 16.616912841796875, | |
| "learning_rate": 4.956140350877193e-05, | |
| "loss": 1.877, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 5.010526315789473, | |
| "grad_norm": 12.440061569213867, | |
| "learning_rate": 4.941520467836258e-05, | |
| "loss": 1.6128, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 5.0131578947368425, | |
| "grad_norm": 18.26654815673828, | |
| "learning_rate": 4.926900584795322e-05, | |
| "loss": 1.8575, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 5.015789473684211, | |
| "grad_norm": 17.84028434753418, | |
| "learning_rate": 4.912280701754386e-05, | |
| "loss": 1.8118, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 5.018421052631579, | |
| "grad_norm": 15.778058052062988, | |
| "learning_rate": 4.8976608187134504e-05, | |
| "loss": 1.9285, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "eval_accuracy": 0.44881889763779526, | |
| "eval_loss": 1.9335325956344604, | |
| "eval_runtime": 213.1562, | |
| "eval_samples_per_second": 0.596, | |
| "eval_steps_per_second": 0.075, | |
| "step": 456 | |
| }, | |
| { | |
| "epoch": 6.001052631578947, | |
| "grad_norm": 12.914003372192383, | |
| "learning_rate": 4.883040935672515e-05, | |
| "loss": 1.621, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 6.003684210526316, | |
| "grad_norm": 12.375211715698242, | |
| "learning_rate": 4.868421052631579e-05, | |
| "loss": 1.5505, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 6.0063157894736845, | |
| "grad_norm": 17.16912841796875, | |
| "learning_rate": 4.853801169590643e-05, | |
| "loss": 1.5443, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 6.008947368421053, | |
| "grad_norm": 15.248703002929688, | |
| "learning_rate": 4.839181286549708e-05, | |
| "loss": 1.5193, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 6.011578947368421, | |
| "grad_norm": 10.187657356262207, | |
| "learning_rate": 4.824561403508772e-05, | |
| "loss": 1.4412, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 6.0142105263157895, | |
| "grad_norm": 8.52863883972168, | |
| "learning_rate": 4.8099415204678366e-05, | |
| "loss": 1.5114, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 6.016842105263158, | |
| "grad_norm": 10.42663288116455, | |
| "learning_rate": 4.7953216374269006e-05, | |
| "loss": 1.4669, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 6.019473684210526, | |
| "grad_norm": 13.066956520080566, | |
| "learning_rate": 4.780701754385965e-05, | |
| "loss": 1.5843, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "eval_accuracy": 0.48031496062992124, | |
| "eval_loss": 1.790671706199646, | |
| "eval_runtime": 218.3253, | |
| "eval_samples_per_second": 0.582, | |
| "eval_steps_per_second": 0.073, | |
| "step": 532 | |
| }, | |
| { | |
| "epoch": 7.002105263157895, | |
| "grad_norm": 14.293023109436035, | |
| "learning_rate": 4.7660818713450294e-05, | |
| "loss": 1.4318, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 7.004736842105263, | |
| "grad_norm": 17.056535720825195, | |
| "learning_rate": 4.751461988304094e-05, | |
| "loss": 1.3673, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 7.0073684210526315, | |
| "grad_norm": 15.318123817443848, | |
| "learning_rate": 4.736842105263158e-05, | |
| "loss": 1.3918, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "grad_norm": 10.944533348083496, | |
| "learning_rate": 4.722222222222222e-05, | |
| "loss": 1.4558, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 7.012631578947368, | |
| "grad_norm": 15.36630916595459, | |
| "learning_rate": 4.707602339181287e-05, | |
| "loss": 1.2652, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 7.015263157894736, | |
| "grad_norm": 19.23952865600586, | |
| "learning_rate": 4.6929824561403515e-05, | |
| "loss": 1.3143, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 7.017894736842106, | |
| "grad_norm": 16.251291275024414, | |
| "learning_rate": 4.678362573099415e-05, | |
| "loss": 1.2387, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "eval_accuracy": 0.3858267716535433, | |
| "eval_loss": 1.8962125778198242, | |
| "eval_runtime": 214.1972, | |
| "eval_samples_per_second": 0.593, | |
| "eval_steps_per_second": 0.075, | |
| "step": 608 | |
| }, | |
| { | |
| "epoch": 8.000526315789473, | |
| "grad_norm": 15.009963035583496, | |
| "learning_rate": 4.6637426900584796e-05, | |
| "loss": 1.198, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 8.003157894736843, | |
| "grad_norm": 10.402515411376953, | |
| "learning_rate": 4.649122807017544e-05, | |
| "loss": 1.2153, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 8.00578947368421, | |
| "grad_norm": 15.393932342529297, | |
| "learning_rate": 4.634502923976608e-05, | |
| "loss": 1.1167, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 8.00842105263158, | |
| "grad_norm": 6.643280506134033, | |
| "learning_rate": 4.619883040935672e-05, | |
| "loss": 1.2265, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 8.011052631578947, | |
| "grad_norm": 9.686060905456543, | |
| "learning_rate": 4.605263157894737e-05, | |
| "loss": 1.1322, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 8.013684210526316, | |
| "grad_norm": 15.071520805358887, | |
| "learning_rate": 4.590643274853802e-05, | |
| "loss": 0.9386, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 8.016315789473683, | |
| "grad_norm": 13.912795066833496, | |
| "learning_rate": 4.576023391812866e-05, | |
| "loss": 1.0938, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 8.018947368421053, | |
| "grad_norm": 19.317672729492188, | |
| "learning_rate": 4.56140350877193e-05, | |
| "loss": 1.2578, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "eval_accuracy": 0.44881889763779526, | |
| "eval_loss": 1.7191396951675415, | |
| "eval_runtime": 221.1734, | |
| "eval_samples_per_second": 0.574, | |
| "eval_steps_per_second": 0.072, | |
| "step": 684 | |
| }, | |
| { | |
| "epoch": 9.00157894736842, | |
| "grad_norm": 12.921488761901855, | |
| "learning_rate": 4.5467836257309945e-05, | |
| "loss": 1.0928, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 9.00421052631579, | |
| "grad_norm": 12.777690887451172, | |
| "learning_rate": 4.5321637426900585e-05, | |
| "loss": 1.1085, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 9.006842105263157, | |
| "grad_norm": 13.301621437072754, | |
| "learning_rate": 4.517543859649123e-05, | |
| "loss": 1.2055, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 9.009473684210526, | |
| "grad_norm": 11.155688285827637, | |
| "learning_rate": 4.502923976608187e-05, | |
| "loss": 1.0145, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 9.012105263157896, | |
| "grad_norm": 19.05325698852539, | |
| "learning_rate": 4.488304093567251e-05, | |
| "loss": 1.0998, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 9.014736842105263, | |
| "grad_norm": 14.210393905639648, | |
| "learning_rate": 4.473684210526316e-05, | |
| "loss": 0.8998, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 9.017368421052632, | |
| "grad_norm": 20.1743106842041, | |
| "learning_rate": 4.4590643274853806e-05, | |
| "loss": 1.1693, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "grad_norm": 17.429399490356445, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.9611, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "eval_accuracy": 0.4881889763779528, | |
| "eval_loss": 1.7361595630645752, | |
| "eval_runtime": 221.2054, | |
| "eval_samples_per_second": 0.574, | |
| "eval_steps_per_second": 0.072, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 10.00263157894737, | |
| "grad_norm": 6.412835597991943, | |
| "learning_rate": 4.429824561403509e-05, | |
| "loss": 0.8146, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 10.005263157894737, | |
| "grad_norm": 7.542128562927246, | |
| "learning_rate": 4.4152046783625734e-05, | |
| "loss": 0.8144, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 10.007894736842106, | |
| "grad_norm": 9.76062297821045, | |
| "learning_rate": 4.400584795321638e-05, | |
| "loss": 0.8581, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 10.010526315789473, | |
| "grad_norm": 15.923550605773926, | |
| "learning_rate": 4.3859649122807014e-05, | |
| "loss": 0.9352, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 10.013157894736842, | |
| "grad_norm": 15.5842924118042, | |
| "learning_rate": 4.371345029239766e-05, | |
| "loss": 0.8613, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 10.01578947368421, | |
| "grad_norm": 14.166058540344238, | |
| "learning_rate": 4.356725146198831e-05, | |
| "loss": 0.9478, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 10.01842105263158, | |
| "grad_norm": 15.743328094482422, | |
| "learning_rate": 4.342105263157895e-05, | |
| "loss": 0.9247, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 10.02, | |
| "eval_accuracy": 0.5905511811023622, | |
| "eval_loss": 1.3898028135299683, | |
| "eval_runtime": 219.3829, | |
| "eval_samples_per_second": 0.579, | |
| "eval_steps_per_second": 0.073, | |
| "step": 836 | |
| }, | |
| { | |
| "epoch": 11.001052631578947, | |
| "grad_norm": 15.066838264465332, | |
| "learning_rate": 4.327485380116959e-05, | |
| "loss": 1.0099, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 11.003684210526316, | |
| "grad_norm": 22.22589683532715, | |
| "learning_rate": 4.3128654970760236e-05, | |
| "loss": 0.9153, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 11.006315789473684, | |
| "grad_norm": 12.97102165222168, | |
| "learning_rate": 4.298245614035088e-05, | |
| "loss": 0.7456, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 11.008947368421053, | |
| "grad_norm": 15.3944730758667, | |
| "learning_rate": 4.283625730994152e-05, | |
| "loss": 0.6651, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 11.01157894736842, | |
| "grad_norm": 18.06490707397461, | |
| "learning_rate": 4.269005847953216e-05, | |
| "loss": 0.9046, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 11.01421052631579, | |
| "grad_norm": 21.835529327392578, | |
| "learning_rate": 4.254385964912281e-05, | |
| "loss": 0.9137, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 11.016842105263159, | |
| "grad_norm": 8.073750495910645, | |
| "learning_rate": 4.239766081871345e-05, | |
| "loss": 0.9701, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 11.019473684210526, | |
| "grad_norm": 13.15185260772705, | |
| "learning_rate": 4.22514619883041e-05, | |
| "loss": 0.8107, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 11.02, | |
| "eval_accuracy": 0.4094488188976378, | |
| "eval_loss": 1.9588265419006348, | |
| "eval_runtime": 217.239, | |
| "eval_samples_per_second": 0.585, | |
| "eval_steps_per_second": 0.074, | |
| "step": 912 | |
| }, | |
| { | |
| "epoch": 12.002105263157894, | |
| "grad_norm": 13.381047248840332, | |
| "learning_rate": 4.210526315789474e-05, | |
| "loss": 0.5541, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 12.004736842105263, | |
| "grad_norm": 14.908097267150879, | |
| "learning_rate": 4.195906432748538e-05, | |
| "loss": 0.9128, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 12.007368421052632, | |
| "grad_norm": 13.663619995117188, | |
| "learning_rate": 4.1812865497076025e-05, | |
| "loss": 0.8758, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 12.01, | |
| "grad_norm": 17.63921546936035, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.6901, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 12.012631578947369, | |
| "grad_norm": 15.174348831176758, | |
| "learning_rate": 4.152046783625731e-05, | |
| "loss": 0.6394, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 12.015263157894736, | |
| "grad_norm": 11.022780418395996, | |
| "learning_rate": 4.137426900584795e-05, | |
| "loss": 0.7149, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 12.017894736842106, | |
| "grad_norm": 13.170819282531738, | |
| "learning_rate": 4.12280701754386e-05, | |
| "loss": 0.7618, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 12.02, | |
| "eval_accuracy": 0.6614173228346457, | |
| "eval_loss": 1.141591191291809, | |
| "eval_runtime": 213.5376, | |
| "eval_samples_per_second": 0.595, | |
| "eval_steps_per_second": 0.075, | |
| "step": 988 | |
| }, | |
| { | |
| "epoch": 13.000526315789473, | |
| "grad_norm": 12.721129417419434, | |
| "learning_rate": 4.1081871345029247e-05, | |
| "loss": 0.6057, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 13.003157894736843, | |
| "grad_norm": 13.658514976501465, | |
| "learning_rate": 4.093567251461988e-05, | |
| "loss": 0.7183, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 13.00578947368421, | |
| "grad_norm": 14.460805892944336, | |
| "learning_rate": 4.078947368421053e-05, | |
| "loss": 0.6447, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 13.00842105263158, | |
| "grad_norm": 12.585667610168457, | |
| "learning_rate": 4.0643274853801174e-05, | |
| "loss": 0.7182, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 13.011052631578947, | |
| "grad_norm": 12.871528625488281, | |
| "learning_rate": 4.0497076023391814e-05, | |
| "loss": 0.7526, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 13.013684210526316, | |
| "grad_norm": 26.28831672668457, | |
| "learning_rate": 4.0350877192982455e-05, | |
| "loss": 0.8322, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 13.016315789473683, | |
| "grad_norm": 9.93556022644043, | |
| "learning_rate": 4.02046783625731e-05, | |
| "loss": 0.712, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 13.018947368421053, | |
| "grad_norm": 17.637866973876953, | |
| "learning_rate": 4.005847953216375e-05, | |
| "loss": 0.7083, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 13.02, | |
| "eval_accuracy": 0.6614173228346457, | |
| "eval_loss": 1.281165599822998, | |
| "eval_runtime": 215.2472, | |
| "eval_samples_per_second": 0.59, | |
| "eval_steps_per_second": 0.074, | |
| "step": 1064 | |
| }, | |
| { | |
| "epoch": 14.00157894736842, | |
| "grad_norm": 5.20871114730835, | |
| "learning_rate": 3.991228070175439e-05, | |
| "loss": 0.6391, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 14.00421052631579, | |
| "grad_norm": 9.707756996154785, | |
| "learning_rate": 3.976608187134503e-05, | |
| "loss": 0.6656, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 14.006842105263157, | |
| "grad_norm": 9.207053184509277, | |
| "learning_rate": 3.9619883040935676e-05, | |
| "loss": 0.522, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 14.009473684210526, | |
| "grad_norm": 10.508357048034668, | |
| "learning_rate": 3.9473684210526316e-05, | |
| "loss": 0.5712, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 14.012105263157896, | |
| "grad_norm": 9.813148498535156, | |
| "learning_rate": 3.932748538011696e-05, | |
| "loss": 0.5209, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 14.014736842105263, | |
| "grad_norm": 23.12822723388672, | |
| "learning_rate": 3.9181286549707604e-05, | |
| "loss": 0.6701, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 14.017368421052632, | |
| "grad_norm": 16.006336212158203, | |
| "learning_rate": 3.9035087719298244e-05, | |
| "loss": 0.6569, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 14.02, | |
| "grad_norm": 15.42993450164795, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.7098, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 14.02, | |
| "eval_accuracy": 0.5196850393700787, | |
| "eval_loss": 1.4601401090621948, | |
| "eval_runtime": 222.0416, | |
| "eval_samples_per_second": 0.572, | |
| "eval_steps_per_second": 0.072, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 15.00263157894737, | |
| "grad_norm": 12.818732261657715, | |
| "learning_rate": 3.874269005847954e-05, | |
| "loss": 0.4586, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 15.005263157894737, | |
| "grad_norm": 18.07966423034668, | |
| "learning_rate": 3.859649122807018e-05, | |
| "loss": 0.6118, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 15.007894736842106, | |
| "grad_norm": 10.6485013961792, | |
| "learning_rate": 3.845029239766082e-05, | |
| "loss": 0.5496, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 15.010526315789473, | |
| "grad_norm": 10.82335376739502, | |
| "learning_rate": 3.8304093567251465e-05, | |
| "loss": 0.5104, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 15.013157894736842, | |
| "grad_norm": 16.787870407104492, | |
| "learning_rate": 3.815789473684211e-05, | |
| "loss": 0.5459, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 15.01578947368421, | |
| "grad_norm": 25.27979850769043, | |
| "learning_rate": 3.8011695906432746e-05, | |
| "loss": 0.7882, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 15.01842105263158, | |
| "grad_norm": 6.820335865020752, | |
| "learning_rate": 3.786549707602339e-05, | |
| "loss": 0.4601, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 15.02, | |
| "eval_accuracy": 0.6692913385826772, | |
| "eval_loss": 1.1275523900985718, | |
| "eval_runtime": 223.6984, | |
| "eval_samples_per_second": 0.568, | |
| "eval_steps_per_second": 0.072, | |
| "step": 1216 | |
| }, | |
| { | |
| "epoch": 16.001052631578947, | |
| "grad_norm": 20.168317794799805, | |
| "learning_rate": 3.771929824561404e-05, | |
| "loss": 0.576, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 16.003684210526316, | |
| "grad_norm": 14.202433586120605, | |
| "learning_rate": 3.757309941520468e-05, | |
| "loss": 0.4841, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 16.006315789473685, | |
| "grad_norm": 17.228273391723633, | |
| "learning_rate": 3.742690058479532e-05, | |
| "loss": 0.4666, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 16.00894736842105, | |
| "grad_norm": 14.974773406982422, | |
| "learning_rate": 3.728070175438597e-05, | |
| "loss": 0.6221, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 16.01157894736842, | |
| "grad_norm": 18.37944793701172, | |
| "learning_rate": 3.713450292397661e-05, | |
| "loss": 0.4969, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 16.01421052631579, | |
| "grad_norm": 11.213744163513184, | |
| "learning_rate": 3.6988304093567254e-05, | |
| "loss": 0.3893, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 16.01684210526316, | |
| "grad_norm": 15.736601829528809, | |
| "learning_rate": 3.6842105263157895e-05, | |
| "loss": 0.5969, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 16.019473684210528, | |
| "grad_norm": 21.78984260559082, | |
| "learning_rate": 3.669590643274854e-05, | |
| "loss": 0.5684, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 16.02, | |
| "eval_accuracy": 0.5590551181102362, | |
| "eval_loss": 1.479176640510559, | |
| "eval_runtime": 210.4765, | |
| "eval_samples_per_second": 0.603, | |
| "eval_steps_per_second": 0.076, | |
| "step": 1292 | |
| }, | |
| { | |
| "epoch": 17.002105263157894, | |
| "grad_norm": 18.97657585144043, | |
| "learning_rate": 3.654970760233918e-05, | |
| "loss": 0.4405, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 17.004736842105263, | |
| "grad_norm": 16.370620727539062, | |
| "learning_rate": 3.640350877192983e-05, | |
| "loss": 0.4933, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 17.007368421052632, | |
| "grad_norm": 18.727680206298828, | |
| "learning_rate": 3.625730994152047e-05, | |
| "loss": 0.7038, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 17.01, | |
| "grad_norm": 11.238962173461914, | |
| "learning_rate": 3.611111111111111e-05, | |
| "loss": 0.4704, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 17.012631578947367, | |
| "grad_norm": 8.99775218963623, | |
| "learning_rate": 3.5964912280701756e-05, | |
| "loss": 0.4999, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 17.015263157894736, | |
| "grad_norm": 4.336584091186523, | |
| "learning_rate": 3.5818713450292403e-05, | |
| "loss": 0.4195, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 17.017894736842106, | |
| "grad_norm": 31.79267120361328, | |
| "learning_rate": 3.5672514619883044e-05, | |
| "loss": 0.5044, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 17.02, | |
| "eval_accuracy": 0.6614173228346457, | |
| "eval_loss": 1.1235504150390625, | |
| "eval_runtime": 226.3183, | |
| "eval_samples_per_second": 0.561, | |
| "eval_steps_per_second": 0.071, | |
| "step": 1368 | |
| }, | |
| { | |
| "epoch": 18.000526315789475, | |
| "grad_norm": 5.108359336853027, | |
| "learning_rate": 3.5526315789473684e-05, | |
| "loss": 0.4842, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 18.00315789473684, | |
| "grad_norm": 6.3470354080200195, | |
| "learning_rate": 3.538011695906433e-05, | |
| "loss": 0.3885, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 18.00578947368421, | |
| "grad_norm": 14.960538864135742, | |
| "learning_rate": 3.523391812865498e-05, | |
| "loss": 0.4218, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 18.00842105263158, | |
| "grad_norm": 3.5540506839752197, | |
| "learning_rate": 3.508771929824561e-05, | |
| "loss": 0.5639, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 18.01105263157895, | |
| "grad_norm": 8.745479583740234, | |
| "learning_rate": 3.494152046783626e-05, | |
| "loss": 0.2593, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 18.013684210526314, | |
| "grad_norm": 16.45661735534668, | |
| "learning_rate": 3.4795321637426905e-05, | |
| "loss": 0.5067, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 18.016315789473683, | |
| "grad_norm": 29.834877014160156, | |
| "learning_rate": 3.4649122807017546e-05, | |
| "loss": 0.4747, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 18.018947368421053, | |
| "grad_norm": 3.380971670150757, | |
| "learning_rate": 3.4502923976608186e-05, | |
| "loss": 0.4551, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 18.02, | |
| "eval_accuracy": 0.6062992125984252, | |
| "eval_loss": 1.3894457817077637, | |
| "eval_runtime": 208.6422, | |
| "eval_samples_per_second": 0.609, | |
| "eval_steps_per_second": 0.077, | |
| "step": 1444 | |
| }, | |
| { | |
| "epoch": 19.001578947368422, | |
| "grad_norm": 17.57008171081543, | |
| "learning_rate": 3.435672514619883e-05, | |
| "loss": 0.4354, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 19.004210526315788, | |
| "grad_norm": 16.11405372619629, | |
| "learning_rate": 3.421052631578947e-05, | |
| "loss": 0.4899, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 19.006842105263157, | |
| "grad_norm": 11.66887092590332, | |
| "learning_rate": 3.406432748538012e-05, | |
| "loss": 0.4279, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 19.009473684210526, | |
| "grad_norm": 13.749747276306152, | |
| "learning_rate": 3.391812865497076e-05, | |
| "loss": 0.39, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 19.012105263157896, | |
| "grad_norm": 22.861486434936523, | |
| "learning_rate": 3.377192982456141e-05, | |
| "loss": 0.3361, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 19.014736842105265, | |
| "grad_norm": 19.228759765625, | |
| "learning_rate": 3.362573099415205e-05, | |
| "loss": 0.3906, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 19.01736842105263, | |
| "grad_norm": 21.020246505737305, | |
| "learning_rate": 3.3479532163742695e-05, | |
| "loss": 0.5774, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 19.02, | |
| "grad_norm": 9.788912773132324, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.3488, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 19.02, | |
| "eval_accuracy": 0.6614173228346457, | |
| "eval_loss": 1.2918061017990112, | |
| "eval_runtime": 222.2729, | |
| "eval_samples_per_second": 0.571, | |
| "eval_steps_per_second": 0.072, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 20.00263157894737, | |
| "grad_norm": 15.891140937805176, | |
| "learning_rate": 3.3187134502923975e-05, | |
| "loss": 0.2902, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 20.00526315789474, | |
| "grad_norm": 5.71783971786499, | |
| "learning_rate": 3.304093567251462e-05, | |
| "loss": 0.2562, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 20.007894736842104, | |
| "grad_norm": 5.522655487060547, | |
| "learning_rate": 3.289473684210527e-05, | |
| "loss": 0.3025, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 20.010526315789473, | |
| "grad_norm": 4.9579691886901855, | |
| "learning_rate": 3.274853801169591e-05, | |
| "loss": 0.1642, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 20.013157894736842, | |
| "grad_norm": 15.316770553588867, | |
| "learning_rate": 3.260233918128655e-05, | |
| "loss": 0.4431, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 20.01578947368421, | |
| "grad_norm": 2.228877067565918, | |
| "learning_rate": 3.24561403508772e-05, | |
| "loss": 0.3193, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 20.018421052631577, | |
| "grad_norm": 9.935259819030762, | |
| "learning_rate": 3.230994152046784e-05, | |
| "loss": 0.4711, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 20.02, | |
| "eval_accuracy": 0.6299212598425197, | |
| "eval_loss": 1.2510179281234741, | |
| "eval_runtime": 212.9147, | |
| "eval_samples_per_second": 0.596, | |
| "eval_steps_per_second": 0.075, | |
| "step": 1596 | |
| }, | |
| { | |
| "epoch": 21.001052631578947, | |
| "grad_norm": 15.269537925720215, | |
| "learning_rate": 3.216374269005848e-05, | |
| "loss": 0.2989, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 21.003684210526316, | |
| "grad_norm": 11.242606163024902, | |
| "learning_rate": 3.2017543859649124e-05, | |
| "loss": 0.3384, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 21.006315789473685, | |
| "grad_norm": 14.554858207702637, | |
| "learning_rate": 3.187134502923977e-05, | |
| "loss": 0.305, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 21.00894736842105, | |
| "grad_norm": 3.1583755016326904, | |
| "learning_rate": 3.172514619883041e-05, | |
| "loss": 0.1487, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 21.01157894736842, | |
| "grad_norm": 30.419754028320312, | |
| "learning_rate": 3.157894736842105e-05, | |
| "loss": 0.4566, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 21.01421052631579, | |
| "grad_norm": 7.953301429748535, | |
| "learning_rate": 3.14327485380117e-05, | |
| "loss": 0.4296, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 21.01684210526316, | |
| "grad_norm": 19.097740173339844, | |
| "learning_rate": 3.128654970760234e-05, | |
| "loss": 0.3578, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 21.019473684210528, | |
| "grad_norm": 23.819265365600586, | |
| "learning_rate": 3.1140350877192986e-05, | |
| "loss": 0.3451, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 21.02, | |
| "eval_accuracy": 0.6692913385826772, | |
| "eval_loss": 1.126503586769104, | |
| "eval_runtime": 207.0446, | |
| "eval_samples_per_second": 0.613, | |
| "eval_steps_per_second": 0.077, | |
| "step": 1672 | |
| }, | |
| { | |
| "epoch": 22.002105263157894, | |
| "grad_norm": 1.5578737258911133, | |
| "learning_rate": 3.0994152046783626e-05, | |
| "loss": 0.3176, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 22.004736842105263, | |
| "grad_norm": 1.8389073610305786, | |
| "learning_rate": 3.084795321637427e-05, | |
| "loss": 0.2775, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 22.007368421052632, | |
| "grad_norm": 16.95042610168457, | |
| "learning_rate": 3.0701754385964913e-05, | |
| "loss": 0.2394, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 22.01, | |
| "grad_norm": 5.184008598327637, | |
| "learning_rate": 3.055555555555556e-05, | |
| "loss": 0.4281, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 22.012631578947367, | |
| "grad_norm": 13.091333389282227, | |
| "learning_rate": 3.0409356725146197e-05, | |
| "loss": 0.3197, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 22.015263157894736, | |
| "grad_norm": 18.41486358642578, | |
| "learning_rate": 3.0263157894736844e-05, | |
| "loss": 0.3218, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 22.017894736842106, | |
| "grad_norm": 23.512739181518555, | |
| "learning_rate": 3.0116959064327488e-05, | |
| "loss": 0.394, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 22.02, | |
| "eval_accuracy": 0.6377952755905512, | |
| "eval_loss": 1.167567253112793, | |
| "eval_runtime": 217.2249, | |
| "eval_samples_per_second": 0.585, | |
| "eval_steps_per_second": 0.074, | |
| "step": 1748 | |
| }, | |
| { | |
| "epoch": 23.000526315789475, | |
| "grad_norm": 3.652381181716919, | |
| "learning_rate": 2.997076023391813e-05, | |
| "loss": 0.3088, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 23.00315789473684, | |
| "grad_norm": 9.2783203125, | |
| "learning_rate": 2.9824561403508772e-05, | |
| "loss": 0.2488, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 23.00578947368421, | |
| "grad_norm": 4.100778579711914, | |
| "learning_rate": 2.9678362573099415e-05, | |
| "loss": 0.1505, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 23.00842105263158, | |
| "grad_norm": 11.44543170928955, | |
| "learning_rate": 2.9532163742690062e-05, | |
| "loss": 0.392, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 23.01105263157895, | |
| "grad_norm": 1.7498180866241455, | |
| "learning_rate": 2.9385964912280706e-05, | |
| "loss": 0.264, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 23.013684210526314, | |
| "grad_norm": 16.737688064575195, | |
| "learning_rate": 2.9239766081871346e-05, | |
| "loss": 0.1904, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 23.016315789473683, | |
| "grad_norm": 3.4222452640533447, | |
| "learning_rate": 2.909356725146199e-05, | |
| "loss": 0.3417, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 23.018947368421053, | |
| "grad_norm": 21.721633911132812, | |
| "learning_rate": 2.8947368421052634e-05, | |
| "loss": 0.234, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 23.02, | |
| "eval_accuracy": 0.7086614173228346, | |
| "eval_loss": 1.0714032649993896, | |
| "eval_runtime": 218.1648, | |
| "eval_samples_per_second": 0.582, | |
| "eval_steps_per_second": 0.073, | |
| "step": 1824 | |
| }, | |
| { | |
| "epoch": 24.001578947368422, | |
| "grad_norm": 27.658559799194336, | |
| "learning_rate": 2.8801169590643277e-05, | |
| "loss": 0.3998, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 24.004210526315788, | |
| "grad_norm": 9.68593692779541, | |
| "learning_rate": 2.8654970760233917e-05, | |
| "loss": 0.156, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 24.006842105263157, | |
| "grad_norm": 26.603404998779297, | |
| "learning_rate": 2.850877192982456e-05, | |
| "loss": 0.4107, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 24.009473684210526, | |
| "grad_norm": 15.769572257995605, | |
| "learning_rate": 2.8362573099415208e-05, | |
| "loss": 0.2049, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 24.012105263157896, | |
| "grad_norm": 13.503707885742188, | |
| "learning_rate": 2.821637426900585e-05, | |
| "loss": 0.278, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 24.014736842105265, | |
| "grad_norm": 12.523197174072266, | |
| "learning_rate": 2.8070175438596492e-05, | |
| "loss": 0.2822, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 24.01736842105263, | |
| "grad_norm": 17.988203048706055, | |
| "learning_rate": 2.7923976608187135e-05, | |
| "loss": 0.2812, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 24.02, | |
| "grad_norm": 26.06406593322754, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.2318, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 24.02, | |
| "eval_accuracy": 0.6377952755905512, | |
| "eval_loss": 1.2646870613098145, | |
| "eval_runtime": 209.6259, | |
| "eval_samples_per_second": 0.606, | |
| "eval_steps_per_second": 0.076, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 25.00263157894737, | |
| "grad_norm": 1.2127106189727783, | |
| "learning_rate": 2.7631578947368426e-05, | |
| "loss": 0.2181, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 25.00526315789474, | |
| "grad_norm": 11.234424591064453, | |
| "learning_rate": 2.7485380116959063e-05, | |
| "loss": 0.3224, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 25.007894736842104, | |
| "grad_norm": 5.068539619445801, | |
| "learning_rate": 2.733918128654971e-05, | |
| "loss": 0.2047, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 25.010526315789473, | |
| "grad_norm": 22.778417587280273, | |
| "learning_rate": 2.7192982456140354e-05, | |
| "loss": 0.2739, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 25.013157894736842, | |
| "grad_norm": 9.123680114746094, | |
| "learning_rate": 2.7046783625730997e-05, | |
| "loss": 0.2422, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 25.01578947368421, | |
| "grad_norm": 10.594013214111328, | |
| "learning_rate": 2.6900584795321637e-05, | |
| "loss": 0.3974, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 25.018421052631577, | |
| "grad_norm": 17.55132293701172, | |
| "learning_rate": 2.675438596491228e-05, | |
| "loss": 0.4294, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 25.02, | |
| "eval_accuracy": 0.7480314960629921, | |
| "eval_loss": 1.0250136852264404, | |
| "eval_runtime": 226.4979, | |
| "eval_samples_per_second": 0.561, | |
| "eval_steps_per_second": 0.071, | |
| "step": 1976 | |
| }, | |
| { | |
| "epoch": 26.001052631578947, | |
| "grad_norm": 8.826836585998535, | |
| "learning_rate": 2.6608187134502928e-05, | |
| "loss": 0.1744, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 26.003684210526316, | |
| "grad_norm": 2.1957454681396484, | |
| "learning_rate": 2.6461988304093572e-05, | |
| "loss": 0.3863, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 26.006315789473685, | |
| "grad_norm": 38.29682159423828, | |
| "learning_rate": 2.6315789473684212e-05, | |
| "loss": 0.3005, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 26.00894736842105, | |
| "grad_norm": 0.6198650002479553, | |
| "learning_rate": 2.6169590643274856e-05, | |
| "loss": 0.2257, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 26.01157894736842, | |
| "grad_norm": 17.415178298950195, | |
| "learning_rate": 2.60233918128655e-05, | |
| "loss": 0.4258, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 26.01421052631579, | |
| "grad_norm": 3.995495319366455, | |
| "learning_rate": 2.5877192982456143e-05, | |
| "loss": 0.2217, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 26.01684210526316, | |
| "grad_norm": 19.10511589050293, | |
| "learning_rate": 2.5730994152046783e-05, | |
| "loss": 0.2082, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 26.019473684210528, | |
| "grad_norm": 22.767173767089844, | |
| "learning_rate": 2.5584795321637427e-05, | |
| "loss": 0.2084, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 26.02, | |
| "eval_accuracy": 0.6850393700787402, | |
| "eval_loss": 1.1360818147659302, | |
| "eval_runtime": 212.9674, | |
| "eval_samples_per_second": 0.596, | |
| "eval_steps_per_second": 0.075, | |
| "step": 2052 | |
| }, | |
| { | |
| "epoch": 27.002105263157894, | |
| "grad_norm": 0.7093040943145752, | |
| "learning_rate": 2.5438596491228074e-05, | |
| "loss": 0.3909, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 27.004736842105263, | |
| "grad_norm": 4.107806205749512, | |
| "learning_rate": 2.5292397660818717e-05, | |
| "loss": 0.1481, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 27.007368421052632, | |
| "grad_norm": 8.628134727478027, | |
| "learning_rate": 2.5146198830409358e-05, | |
| "loss": 0.27, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 27.01, | |
| "grad_norm": 5.203458786010742, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2593, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 27.012631578947367, | |
| "grad_norm": 12.856894493103027, | |
| "learning_rate": 2.485380116959064e-05, | |
| "loss": 0.1478, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 27.015263157894736, | |
| "grad_norm": 0.4042441248893738, | |
| "learning_rate": 2.470760233918129e-05, | |
| "loss": 0.3425, | |
| "step": 2110 | |
| }, | |
| { | |
| "epoch": 27.017894736842106, | |
| "grad_norm": 22.888172149658203, | |
| "learning_rate": 2.456140350877193e-05, | |
| "loss": 0.1724, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 27.02, | |
| "eval_accuracy": 0.7401574803149606, | |
| "eval_loss": 0.8791013956069946, | |
| "eval_runtime": 226.6646, | |
| "eval_samples_per_second": 0.56, | |
| "eval_steps_per_second": 0.071, | |
| "step": 2128 | |
| }, | |
| { | |
| "epoch": 28.000526315789475, | |
| "grad_norm": 8.835603713989258, | |
| "learning_rate": 2.4415204678362576e-05, | |
| "loss": 0.2062, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 28.00315789473684, | |
| "grad_norm": 30.87272071838379, | |
| "learning_rate": 2.4269005847953216e-05, | |
| "loss": 0.2398, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 28.00578947368421, | |
| "grad_norm": 3.532879590988159, | |
| "learning_rate": 2.412280701754386e-05, | |
| "loss": 0.152, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 28.00842105263158, | |
| "grad_norm": 7.7977752685546875, | |
| "learning_rate": 2.3976608187134503e-05, | |
| "loss": 0.2509, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 28.01105263157895, | |
| "grad_norm": 22.783979415893555, | |
| "learning_rate": 2.3830409356725147e-05, | |
| "loss": 0.2393, | |
| "step": 2170 | |
| }, | |
| { | |
| "epoch": 28.013684210526314, | |
| "grad_norm": 16.93852424621582, | |
| "learning_rate": 2.368421052631579e-05, | |
| "loss": 0.2569, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 28.016315789473683, | |
| "grad_norm": 15.875041961669922, | |
| "learning_rate": 2.3538011695906434e-05, | |
| "loss": 0.2386, | |
| "step": 2190 | |
| }, | |
| { | |
| "epoch": 28.018947368421053, | |
| "grad_norm": 0.38901668787002563, | |
| "learning_rate": 2.3391812865497074e-05, | |
| "loss": 0.1715, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 28.02, | |
| "eval_accuracy": 0.7559055118110236, | |
| "eval_loss": 0.7549403309822083, | |
| "eval_runtime": 220.4782, | |
| "eval_samples_per_second": 0.576, | |
| "eval_steps_per_second": 0.073, | |
| "step": 2204 | |
| }, | |
| { | |
| "epoch": 29.001578947368422, | |
| "grad_norm": 14.774117469787598, | |
| "learning_rate": 2.324561403508772e-05, | |
| "loss": 0.2363, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 29.004210526315788, | |
| "grad_norm": 25.879791259765625, | |
| "learning_rate": 2.309941520467836e-05, | |
| "loss": 0.1621, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 29.006842105263157, | |
| "grad_norm": 29.524221420288086, | |
| "learning_rate": 2.295321637426901e-05, | |
| "loss": 0.2586, | |
| "step": 2230 | |
| }, | |
| { | |
| "epoch": 29.009473684210526, | |
| "grad_norm": 0.5666794776916504, | |
| "learning_rate": 2.280701754385965e-05, | |
| "loss": 0.1223, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 29.012105263157896, | |
| "grad_norm": 1.050584077835083, | |
| "learning_rate": 2.2660818713450292e-05, | |
| "loss": 0.1105, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 29.014736842105265, | |
| "grad_norm": 23.264785766601562, | |
| "learning_rate": 2.2514619883040936e-05, | |
| "loss": 0.1934, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 29.01736842105263, | |
| "grad_norm": 21.131118774414062, | |
| "learning_rate": 2.236842105263158e-05, | |
| "loss": 0.3142, | |
| "step": 2270 | |
| }, | |
| { | |
| "epoch": 29.02, | |
| "grad_norm": 6.656404495239258, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.2719, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 29.02, | |
| "eval_accuracy": 0.7716535433070866, | |
| "eval_loss": 0.770797073841095, | |
| "eval_runtime": 215.6644, | |
| "eval_samples_per_second": 0.589, | |
| "eval_steps_per_second": 0.074, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 30.00263157894737, | |
| "grad_norm": 5.025905132293701, | |
| "learning_rate": 2.2076023391812867e-05, | |
| "loss": 0.0985, | |
| "step": 2290 | |
| }, | |
| { | |
| "epoch": 30.00526315789474, | |
| "grad_norm": 3.686487913131714, | |
| "learning_rate": 2.1929824561403507e-05, | |
| "loss": 0.1349, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 30.007894736842104, | |
| "grad_norm": 3.4678690433502197, | |
| "learning_rate": 2.1783625730994154e-05, | |
| "loss": 0.0814, | |
| "step": 2310 | |
| }, | |
| { | |
| "epoch": 30.010526315789473, | |
| "grad_norm": 3.107570171356201, | |
| "learning_rate": 2.1637426900584794e-05, | |
| "loss": 0.1936, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 30.013157894736842, | |
| "grad_norm": 2.725902795791626, | |
| "learning_rate": 2.149122807017544e-05, | |
| "loss": 0.1985, | |
| "step": 2330 | |
| }, | |
| { | |
| "epoch": 30.01578947368421, | |
| "grad_norm": 3.2765145301818848, | |
| "learning_rate": 2.134502923976608e-05, | |
| "loss": 0.1352, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 30.018421052631577, | |
| "grad_norm": 16.57495880126953, | |
| "learning_rate": 2.1198830409356725e-05, | |
| "loss": 0.2021, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 30.02, | |
| "eval_accuracy": 0.7165354330708661, | |
| "eval_loss": 1.139369010925293, | |
| "eval_runtime": 219.7556, | |
| "eval_samples_per_second": 0.578, | |
| "eval_steps_per_second": 0.073, | |
| "step": 2356 | |
| }, | |
| { | |
| "epoch": 31.001052631578947, | |
| "grad_norm": 0.4346730709075928, | |
| "learning_rate": 2.105263157894737e-05, | |
| "loss": 0.1554, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 31.003684210526316, | |
| "grad_norm": 15.532111167907715, | |
| "learning_rate": 2.0906432748538013e-05, | |
| "loss": 0.1262, | |
| "step": 2370 | |
| }, | |
| { | |
| "epoch": 31.006315789473685, | |
| "grad_norm": 3.0062544345855713, | |
| "learning_rate": 2.0760233918128656e-05, | |
| "loss": 0.1561, | |
| "step": 2380 | |
| }, | |
| { | |
| "epoch": 31.00894736842105, | |
| "grad_norm": 15.58200740814209, | |
| "learning_rate": 2.06140350877193e-05, | |
| "loss": 0.1591, | |
| "step": 2390 | |
| }, | |
| { | |
| "epoch": 31.01157894736842, | |
| "grad_norm": 9.851778984069824, | |
| "learning_rate": 2.046783625730994e-05, | |
| "loss": 0.1295, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 31.01421052631579, | |
| "grad_norm": 1.939591646194458, | |
| "learning_rate": 2.0321637426900587e-05, | |
| "loss": 0.1785, | |
| "step": 2410 | |
| }, | |
| { | |
| "epoch": 31.01684210526316, | |
| "grad_norm": 13.707688331604004, | |
| "learning_rate": 2.0175438596491227e-05, | |
| "loss": 0.1499, | |
| "step": 2420 | |
| }, | |
| { | |
| "epoch": 31.019473684210528, | |
| "grad_norm": 1.3323267698287964, | |
| "learning_rate": 2.0029239766081874e-05, | |
| "loss": 0.0999, | |
| "step": 2430 | |
| }, | |
| { | |
| "epoch": 31.02, | |
| "eval_accuracy": 0.7716535433070866, | |
| "eval_loss": 0.7838273644447327, | |
| "eval_runtime": 221.1151, | |
| "eval_samples_per_second": 0.574, | |
| "eval_steps_per_second": 0.072, | |
| "step": 2432 | |
| }, | |
| { | |
| "epoch": 32.002105263157894, | |
| "grad_norm": 1.2039517164230347, | |
| "learning_rate": 1.9883040935672515e-05, | |
| "loss": 0.1384, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 32.00473684210527, | |
| "grad_norm": 4.4641499519348145, | |
| "learning_rate": 1.9736842105263158e-05, | |
| "loss": 0.2209, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 32.00736842105263, | |
| "grad_norm": 5.985694885253906, | |
| "learning_rate": 1.9590643274853802e-05, | |
| "loss": 0.1737, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 32.01, | |
| "grad_norm": 9.976003646850586, | |
| "learning_rate": 1.9444444444444445e-05, | |
| "loss": 0.1998, | |
| "step": 2470 | |
| }, | |
| { | |
| "epoch": 32.01263157894737, | |
| "grad_norm": 9.69500732421875, | |
| "learning_rate": 1.929824561403509e-05, | |
| "loss": 0.2171, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 32.015263157894736, | |
| "grad_norm": 21.551578521728516, | |
| "learning_rate": 1.9152046783625733e-05, | |
| "loss": 0.1829, | |
| "step": 2490 | |
| }, | |
| { | |
| "epoch": 32.0178947368421, | |
| "grad_norm": 9.649386405944824, | |
| "learning_rate": 1.9005847953216373e-05, | |
| "loss": 0.1473, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 32.02, | |
| "eval_accuracy": 0.6456692913385826, | |
| "eval_loss": 1.3808568716049194, | |
| "eval_runtime": 212.0868, | |
| "eval_samples_per_second": 0.599, | |
| "eval_steps_per_second": 0.075, | |
| "step": 2508 | |
| }, | |
| { | |
| "epoch": 33.00052631578947, | |
| "grad_norm": 0.5508492588996887, | |
| "learning_rate": 1.885964912280702e-05, | |
| "loss": 0.1658, | |
| "step": 2510 | |
| }, | |
| { | |
| "epoch": 33.003157894736844, | |
| "grad_norm": 1.3304760456085205, | |
| "learning_rate": 1.871345029239766e-05, | |
| "loss": 0.1968, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 33.00578947368421, | |
| "grad_norm": 35.01847457885742, | |
| "learning_rate": 1.8567251461988304e-05, | |
| "loss": 0.1615, | |
| "step": 2530 | |
| }, | |
| { | |
| "epoch": 33.008421052631576, | |
| "grad_norm": 5.112392902374268, | |
| "learning_rate": 1.8421052631578947e-05, | |
| "loss": 0.1648, | |
| "step": 2540 | |
| }, | |
| { | |
| "epoch": 33.01105263157895, | |
| "grad_norm": 2.015471935272217, | |
| "learning_rate": 1.827485380116959e-05, | |
| "loss": 0.2019, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 33.013684210526314, | |
| "grad_norm": 0.9456290602684021, | |
| "learning_rate": 1.8128654970760235e-05, | |
| "loss": 0.1382, | |
| "step": 2560 | |
| }, | |
| { | |
| "epoch": 33.01631578947369, | |
| "grad_norm": 5.0300164222717285, | |
| "learning_rate": 1.7982456140350878e-05, | |
| "loss": 0.1313, | |
| "step": 2570 | |
| }, | |
| { | |
| "epoch": 33.01894736842105, | |
| "grad_norm": 0.5818091034889221, | |
| "learning_rate": 1.7836257309941522e-05, | |
| "loss": 0.0939, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 33.02, | |
| "eval_accuracy": 0.7874015748031497, | |
| "eval_loss": 0.783892035484314, | |
| "eval_runtime": 217.5033, | |
| "eval_samples_per_second": 0.584, | |
| "eval_steps_per_second": 0.074, | |
| "step": 2584 | |
| }, | |
| { | |
| "epoch": 34.00157894736842, | |
| "grad_norm": 1.293542504310608, | |
| "learning_rate": 1.7690058479532165e-05, | |
| "loss": 0.0838, | |
| "step": 2590 | |
| }, | |
| { | |
| "epoch": 34.00421052631579, | |
| "grad_norm": 15.003107070922852, | |
| "learning_rate": 1.7543859649122806e-05, | |
| "loss": 0.2212, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 34.00684210526316, | |
| "grad_norm": 1.133008599281311, | |
| "learning_rate": 1.7397660818713453e-05, | |
| "loss": 0.1399, | |
| "step": 2610 | |
| }, | |
| { | |
| "epoch": 34.009473684210526, | |
| "grad_norm": 14.774809837341309, | |
| "learning_rate": 1.7251461988304093e-05, | |
| "loss": 0.1712, | |
| "step": 2620 | |
| }, | |
| { | |
| "epoch": 34.01210526315789, | |
| "grad_norm": 0.46233054995536804, | |
| "learning_rate": 1.7105263157894737e-05, | |
| "loss": 0.0984, | |
| "step": 2630 | |
| }, | |
| { | |
| "epoch": 34.014736842105265, | |
| "grad_norm": 1.898353934288025, | |
| "learning_rate": 1.695906432748538e-05, | |
| "loss": 0.0641, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 34.01736842105263, | |
| "grad_norm": 0.8380947709083557, | |
| "learning_rate": 1.6812865497076024e-05, | |
| "loss": 0.2255, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 34.02, | |
| "grad_norm": 16.234495162963867, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.0952, | |
| "step": 2660 | |
| }, | |
| { | |
| "epoch": 34.02, | |
| "eval_accuracy": 0.7007874015748031, | |
| "eval_loss": 1.0635521411895752, | |
| "eval_runtime": 209.1412, | |
| "eval_samples_per_second": 0.607, | |
| "eval_steps_per_second": 0.077, | |
| "step": 2660 | |
| }, | |
| { | |
| "epoch": 35.002631578947366, | |
| "grad_norm": 0.17983902990818024, | |
| "learning_rate": 1.652046783625731e-05, | |
| "loss": 0.1539, | |
| "step": 2670 | |
| }, | |
| { | |
| "epoch": 35.00526315789474, | |
| "grad_norm": 8.456697463989258, | |
| "learning_rate": 1.6374269005847955e-05, | |
| "loss": 0.0903, | |
| "step": 2680 | |
| }, | |
| { | |
| "epoch": 35.007894736842104, | |
| "grad_norm": 7.233275413513184, | |
| "learning_rate": 1.62280701754386e-05, | |
| "loss": 0.206, | |
| "step": 2690 | |
| }, | |
| { | |
| "epoch": 35.01052631578948, | |
| "grad_norm": 0.25833821296691895, | |
| "learning_rate": 1.608187134502924e-05, | |
| "loss": 0.1998, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 35.01315789473684, | |
| "grad_norm": 0.7079264521598816, | |
| "learning_rate": 1.5935672514619886e-05, | |
| "loss": 0.1462, | |
| "step": 2710 | |
| }, | |
| { | |
| "epoch": 35.01578947368421, | |
| "grad_norm": 52.23944091796875, | |
| "learning_rate": 1.5789473684210526e-05, | |
| "loss": 0.2127, | |
| "step": 2720 | |
| }, | |
| { | |
| "epoch": 35.01842105263158, | |
| "grad_norm": 17.99128532409668, | |
| "learning_rate": 1.564327485380117e-05, | |
| "loss": 0.2684, | |
| "step": 2730 | |
| }, | |
| { | |
| "epoch": 35.02, | |
| "eval_accuracy": 0.7322834645669292, | |
| "eval_loss": 0.9193896651268005, | |
| "eval_runtime": 222.8863, | |
| "eval_samples_per_second": 0.57, | |
| "eval_steps_per_second": 0.072, | |
| "step": 2736 | |
| }, | |
| { | |
| "epoch": 36.00105263157895, | |
| "grad_norm": 5.967014789581299, | |
| "learning_rate": 1.5497076023391813e-05, | |
| "loss": 0.1196, | |
| "step": 2740 | |
| }, | |
| { | |
| "epoch": 36.003684210526316, | |
| "grad_norm": 35.71928024291992, | |
| "learning_rate": 1.5350877192982457e-05, | |
| "loss": 0.1805, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 36.00631578947368, | |
| "grad_norm": 0.19552262127399445, | |
| "learning_rate": 1.5204678362573099e-05, | |
| "loss": 0.069, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 36.008947368421055, | |
| "grad_norm": 0.09623876214027405, | |
| "learning_rate": 1.5058479532163744e-05, | |
| "loss": 0.1688, | |
| "step": 2770 | |
| }, | |
| { | |
| "epoch": 36.01157894736842, | |
| "grad_norm": 4.143298625946045, | |
| "learning_rate": 1.4912280701754386e-05, | |
| "loss": 0.1548, | |
| "step": 2780 | |
| }, | |
| { | |
| "epoch": 36.01421052631579, | |
| "grad_norm": 0.08483076095581055, | |
| "learning_rate": 1.4766081871345031e-05, | |
| "loss": 0.1321, | |
| "step": 2790 | |
| }, | |
| { | |
| "epoch": 36.01684210526316, | |
| "grad_norm": 27.361652374267578, | |
| "learning_rate": 1.4619883040935673e-05, | |
| "loss": 0.1274, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 36.019473684210524, | |
| "grad_norm": 0.30415764451026917, | |
| "learning_rate": 1.4473684210526317e-05, | |
| "loss": 0.1628, | |
| "step": 2810 | |
| }, | |
| { | |
| "epoch": 36.02, | |
| "eval_accuracy": 0.8031496062992126, | |
| "eval_loss": 0.7345768809318542, | |
| "eval_runtime": 218.0492, | |
| "eval_samples_per_second": 0.582, | |
| "eval_steps_per_second": 0.073, | |
| "step": 2812 | |
| }, | |
| { | |
| "epoch": 37.002105263157894, | |
| "grad_norm": 1.5019361972808838, | |
| "learning_rate": 1.4327485380116959e-05, | |
| "loss": 0.111, | |
| "step": 2820 | |
| }, | |
| { | |
| "epoch": 37.00473684210527, | |
| "grad_norm": 20.669527053833008, | |
| "learning_rate": 1.4181286549707604e-05, | |
| "loss": 0.0966, | |
| "step": 2830 | |
| }, | |
| { | |
| "epoch": 37.00736842105263, | |
| "grad_norm": 8.103653907775879, | |
| "learning_rate": 1.4035087719298246e-05, | |
| "loss": 0.1641, | |
| "step": 2840 | |
| }, | |
| { | |
| "epoch": 37.01, | |
| "grad_norm": 19.656436920166016, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.0822, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 37.01263157894737, | |
| "grad_norm": 4.070357799530029, | |
| "learning_rate": 1.3742690058479531e-05, | |
| "loss": 0.1132, | |
| "step": 2860 | |
| }, | |
| { | |
| "epoch": 37.015263157894736, | |
| "grad_norm": 2.3238236904144287, | |
| "learning_rate": 1.3596491228070177e-05, | |
| "loss": 0.1275, | |
| "step": 2870 | |
| }, | |
| { | |
| "epoch": 37.0178947368421, | |
| "grad_norm": 0.8202667236328125, | |
| "learning_rate": 1.3450292397660819e-05, | |
| "loss": 0.0584, | |
| "step": 2880 | |
| }, | |
| { | |
| "epoch": 37.02, | |
| "eval_accuracy": 0.7322834645669292, | |
| "eval_loss": 1.0111991167068481, | |
| "eval_runtime": 215.1627, | |
| "eval_samples_per_second": 0.59, | |
| "eval_steps_per_second": 0.074, | |
| "step": 2888 | |
| }, | |
| { | |
| "epoch": 38.00052631578947, | |
| "grad_norm": 8.352115631103516, | |
| "learning_rate": 1.3304093567251464e-05, | |
| "loss": 0.1348, | |
| "step": 2890 | |
| }, | |
| { | |
| "epoch": 38.003157894736844, | |
| "grad_norm": 3.482306480407715, | |
| "learning_rate": 1.3157894736842106e-05, | |
| "loss": 0.1314, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 38.00578947368421, | |
| "grad_norm": 3.1589252948760986, | |
| "learning_rate": 1.301169590643275e-05, | |
| "loss": 0.0665, | |
| "step": 2910 | |
| }, | |
| { | |
| "epoch": 38.008421052631576, | |
| "grad_norm": 2.0633227825164795, | |
| "learning_rate": 1.2865497076023392e-05, | |
| "loss": 0.0588, | |
| "step": 2920 | |
| }, | |
| { | |
| "epoch": 38.01105263157895, | |
| "grad_norm": 15.660873413085938, | |
| "learning_rate": 1.2719298245614037e-05, | |
| "loss": 0.0872, | |
| "step": 2930 | |
| }, | |
| { | |
| "epoch": 38.013684210526314, | |
| "grad_norm": 0.21239978075027466, | |
| "learning_rate": 1.2573099415204679e-05, | |
| "loss": 0.0914, | |
| "step": 2940 | |
| }, | |
| { | |
| "epoch": 38.01631578947369, | |
| "grad_norm": 3.8502485752105713, | |
| "learning_rate": 1.242690058479532e-05, | |
| "loss": 0.0803, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 38.01894736842105, | |
| "grad_norm": 0.7816948890686035, | |
| "learning_rate": 1.2280701754385964e-05, | |
| "loss": 0.0567, | |
| "step": 2960 | |
| }, | |
| { | |
| "epoch": 38.02, | |
| "eval_accuracy": 0.7322834645669292, | |
| "eval_loss": 1.058438777923584, | |
| "eval_runtime": 210.151, | |
| "eval_samples_per_second": 0.604, | |
| "eval_steps_per_second": 0.076, | |
| "step": 2964 | |
| }, | |
| { | |
| "epoch": 39.00157894736842, | |
| "grad_norm": 1.5109614133834839, | |
| "learning_rate": 1.2134502923976608e-05, | |
| "loss": 0.1519, | |
| "step": 2970 | |
| }, | |
| { | |
| "epoch": 39.00421052631579, | |
| "grad_norm": 23.389062881469727, | |
| "learning_rate": 1.1988304093567252e-05, | |
| "loss": 0.1978, | |
| "step": 2980 | |
| }, | |
| { | |
| "epoch": 39.00684210526316, | |
| "grad_norm": 2.617483139038086, | |
| "learning_rate": 1.1842105263157895e-05, | |
| "loss": 0.3215, | |
| "step": 2990 | |
| }, | |
| { | |
| "epoch": 39.009473684210526, | |
| "grad_norm": 2.0149471759796143, | |
| "learning_rate": 1.1695906432748537e-05, | |
| "loss": 0.0918, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 39.01210526315789, | |
| "grad_norm": 0.3655383288860321, | |
| "learning_rate": 1.154970760233918e-05, | |
| "loss": 0.0587, | |
| "step": 3010 | |
| }, | |
| { | |
| "epoch": 39.014736842105265, | |
| "grad_norm": 23.13666343688965, | |
| "learning_rate": 1.1403508771929824e-05, | |
| "loss": 0.1863, | |
| "step": 3020 | |
| }, | |
| { | |
| "epoch": 39.01736842105263, | |
| "grad_norm": 0.5490059852600098, | |
| "learning_rate": 1.1257309941520468e-05, | |
| "loss": 0.0516, | |
| "step": 3030 | |
| }, | |
| { | |
| "epoch": 39.02, | |
| "grad_norm": 6.660675048828125, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.1358, | |
| "step": 3040 | |
| }, | |
| { | |
| "epoch": 39.02, | |
| "eval_accuracy": 0.7322834645669292, | |
| "eval_loss": 1.05661940574646, | |
| "eval_runtime": 208.362, | |
| "eval_samples_per_second": 0.61, | |
| "eval_steps_per_second": 0.077, | |
| "step": 3040 | |
| }, | |
| { | |
| "epoch": 40.002631578947366, | |
| "grad_norm": 0.15694499015808105, | |
| "learning_rate": 1.0964912280701754e-05, | |
| "loss": 0.2102, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 40.00526315789474, | |
| "grad_norm": 21.509859085083008, | |
| "learning_rate": 1.0818713450292397e-05, | |
| "loss": 0.107, | |
| "step": 3060 | |
| }, | |
| { | |
| "epoch": 40.007894736842104, | |
| "grad_norm": 13.848420143127441, | |
| "learning_rate": 1.067251461988304e-05, | |
| "loss": 0.118, | |
| "step": 3070 | |
| }, | |
| { | |
| "epoch": 40.01052631578948, | |
| "grad_norm": 0.11400782316923141, | |
| "learning_rate": 1.0526315789473684e-05, | |
| "loss": 0.1119, | |
| "step": 3080 | |
| }, | |
| { | |
| "epoch": 40.01315789473684, | |
| "grad_norm": 27.989046096801758, | |
| "learning_rate": 1.0380116959064328e-05, | |
| "loss": 0.2483, | |
| "step": 3090 | |
| }, | |
| { | |
| "epoch": 40.01578947368421, | |
| "grad_norm": 3.2790133953094482, | |
| "learning_rate": 1.023391812865497e-05, | |
| "loss": 0.0518, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 40.01842105263158, | |
| "grad_norm": 1.107125163078308, | |
| "learning_rate": 1.0087719298245614e-05, | |
| "loss": 0.0796, | |
| "step": 3110 | |
| }, | |
| { | |
| "epoch": 40.02, | |
| "eval_accuracy": 0.7480314960629921, | |
| "eval_loss": 0.9323013424873352, | |
| "eval_runtime": 206.0517, | |
| "eval_samples_per_second": 0.616, | |
| "eval_steps_per_second": 0.078, | |
| "step": 3116 | |
| }, | |
| { | |
| "epoch": 41.00105263157895, | |
| "grad_norm": 0.9045488238334656, | |
| "learning_rate": 9.941520467836257e-06, | |
| "loss": 0.1818, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 41.003684210526316, | |
| "grad_norm": 6.538638114929199, | |
| "learning_rate": 9.795321637426901e-06, | |
| "loss": 0.1703, | |
| "step": 3130 | |
| }, | |
| { | |
| "epoch": 41.00631578947368, | |
| "grad_norm": 4.6542277336120605, | |
| "learning_rate": 9.649122807017545e-06, | |
| "loss": 0.1232, | |
| "step": 3140 | |
| }, | |
| { | |
| "epoch": 41.008947368421055, | |
| "grad_norm": 0.4844169318675995, | |
| "learning_rate": 9.502923976608186e-06, | |
| "loss": 0.1386, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 41.01157894736842, | |
| "grad_norm": 0.5313725471496582, | |
| "learning_rate": 9.35672514619883e-06, | |
| "loss": 0.0692, | |
| "step": 3160 | |
| }, | |
| { | |
| "epoch": 41.01421052631579, | |
| "grad_norm": 0.22919414937496185, | |
| "learning_rate": 9.210526315789474e-06, | |
| "loss": 0.0188, | |
| "step": 3170 | |
| }, | |
| { | |
| "epoch": 41.01684210526316, | |
| "grad_norm": 2.425123929977417, | |
| "learning_rate": 9.064327485380117e-06, | |
| "loss": 0.0582, | |
| "step": 3180 | |
| }, | |
| { | |
| "epoch": 41.019473684210524, | |
| "grad_norm": 0.15257947146892548, | |
| "learning_rate": 8.918128654970761e-06, | |
| "loss": 0.0828, | |
| "step": 3190 | |
| }, | |
| { | |
| "epoch": 41.02, | |
| "eval_accuracy": 0.7952755905511811, | |
| "eval_loss": 0.7611303925514221, | |
| "eval_runtime": 218.1076, | |
| "eval_samples_per_second": 0.582, | |
| "eval_steps_per_second": 0.073, | |
| "step": 3192 | |
| }, | |
| { | |
| "epoch": 42.002105263157894, | |
| "grad_norm": 0.6726482510566711, | |
| "learning_rate": 8.771929824561403e-06, | |
| "loss": 0.1402, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 42.00473684210527, | |
| "grad_norm": 15.30670166015625, | |
| "learning_rate": 8.625730994152046e-06, | |
| "loss": 0.0809, | |
| "step": 3210 | |
| }, | |
| { | |
| "epoch": 42.00736842105263, | |
| "grad_norm": 3.397010326385498, | |
| "learning_rate": 8.47953216374269e-06, | |
| "loss": 0.1602, | |
| "step": 3220 | |
| }, | |
| { | |
| "epoch": 42.01, | |
| "grad_norm": 19.94878387451172, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.086, | |
| "step": 3230 | |
| }, | |
| { | |
| "epoch": 42.01263157894737, | |
| "grad_norm": 8.212011337280273, | |
| "learning_rate": 8.187134502923977e-06, | |
| "loss": 0.143, | |
| "step": 3240 | |
| }, | |
| { | |
| "epoch": 42.015263157894736, | |
| "grad_norm": 4.16144323348999, | |
| "learning_rate": 8.04093567251462e-06, | |
| "loss": 0.0462, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 42.0178947368421, | |
| "grad_norm": 1.0098721981048584, | |
| "learning_rate": 7.894736842105263e-06, | |
| "loss": 0.0661, | |
| "step": 3260 | |
| }, | |
| { | |
| "epoch": 42.02, | |
| "eval_accuracy": 0.7874015748031497, | |
| "eval_loss": 0.7284496426582336, | |
| "eval_runtime": 213.0991, | |
| "eval_samples_per_second": 0.596, | |
| "eval_steps_per_second": 0.075, | |
| "step": 3268 | |
| }, | |
| { | |
| "epoch": 43.00052631578947, | |
| "grad_norm": 3.563675880432129, | |
| "learning_rate": 7.748538011695907e-06, | |
| "loss": 0.0854, | |
| "step": 3270 | |
| }, | |
| { | |
| "epoch": 43.003157894736844, | |
| "grad_norm": 23.9377384185791, | |
| "learning_rate": 7.602339181286549e-06, | |
| "loss": 0.1174, | |
| "step": 3280 | |
| }, | |
| { | |
| "epoch": 43.00578947368421, | |
| "grad_norm": 5.841782093048096, | |
| "learning_rate": 7.456140350877193e-06, | |
| "loss": 0.0652, | |
| "step": 3290 | |
| }, | |
| { | |
| "epoch": 43.008421052631576, | |
| "grad_norm": 3.5263381004333496, | |
| "learning_rate": 7.3099415204678366e-06, | |
| "loss": 0.107, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 43.01105263157895, | |
| "grad_norm": 0.21935203671455383, | |
| "learning_rate": 7.163742690058479e-06, | |
| "loss": 0.1081, | |
| "step": 3310 | |
| }, | |
| { | |
| "epoch": 43.013684210526314, | |
| "grad_norm": 0.27123141288757324, | |
| "learning_rate": 7.017543859649123e-06, | |
| "loss": 0.1304, | |
| "step": 3320 | |
| }, | |
| { | |
| "epoch": 43.01631578947369, | |
| "grad_norm": 0.10241150110960007, | |
| "learning_rate": 6.871345029239766e-06, | |
| "loss": 0.0419, | |
| "step": 3330 | |
| }, | |
| { | |
| "epoch": 43.01894736842105, | |
| "grad_norm": 0.14180737733840942, | |
| "learning_rate": 6.725146198830409e-06, | |
| "loss": 0.0882, | |
| "step": 3340 | |
| }, | |
| { | |
| "epoch": 43.02, | |
| "eval_accuracy": 0.7952755905511811, | |
| "eval_loss": 0.6982414722442627, | |
| "eval_runtime": 211.2161, | |
| "eval_samples_per_second": 0.601, | |
| "eval_steps_per_second": 0.076, | |
| "step": 3344 | |
| }, | |
| { | |
| "epoch": 44.00157894736842, | |
| "grad_norm": 3.263852834701538, | |
| "learning_rate": 6.578947368421053e-06, | |
| "loss": 0.0781, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 44.00421052631579, | |
| "grad_norm": 2.984760284423828, | |
| "learning_rate": 6.432748538011696e-06, | |
| "loss": 0.0315, | |
| "step": 3360 | |
| }, | |
| { | |
| "epoch": 44.00684210526316, | |
| "grad_norm": 0.09511619806289673, | |
| "learning_rate": 6.286549707602339e-06, | |
| "loss": 0.0384, | |
| "step": 3370 | |
| }, | |
| { | |
| "epoch": 44.009473684210526, | |
| "grad_norm": 2.247210741043091, | |
| "learning_rate": 6.140350877192982e-06, | |
| "loss": 0.0445, | |
| "step": 3380 | |
| }, | |
| { | |
| "epoch": 44.01210526315789, | |
| "grad_norm": 31.92279052734375, | |
| "learning_rate": 5.994152046783626e-06, | |
| "loss": 0.1253, | |
| "step": 3390 | |
| }, | |
| { | |
| "epoch": 44.014736842105265, | |
| "grad_norm": 2.3949315547943115, | |
| "learning_rate": 5.8479532163742686e-06, | |
| "loss": 0.077, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 44.01736842105263, | |
| "grad_norm": 0.148012176156044, | |
| "learning_rate": 5.701754385964912e-06, | |
| "loss": 0.0954, | |
| "step": 3410 | |
| }, | |
| { | |
| "epoch": 44.02, | |
| "grad_norm": 0.5894277691841125, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.0398, | |
| "step": 3420 | |
| }, | |
| { | |
| "epoch": 44.02, | |
| "eval_accuracy": 0.7716535433070866, | |
| "eval_loss": 0.8585867285728455, | |
| "eval_runtime": 217.9943, | |
| "eval_samples_per_second": 0.583, | |
| "eval_steps_per_second": 0.073, | |
| "step": 3420 | |
| }, | |
| { | |
| "epoch": 45.002631578947366, | |
| "grad_norm": 1.641733169555664, | |
| "learning_rate": 5.409356725146199e-06, | |
| "loss": 0.0984, | |
| "step": 3430 | |
| }, | |
| { | |
| "epoch": 45.00526315789474, | |
| "grad_norm": 1.107046127319336, | |
| "learning_rate": 5.263157894736842e-06, | |
| "loss": 0.0461, | |
| "step": 3440 | |
| }, | |
| { | |
| "epoch": 45.007894736842104, | |
| "grad_norm": 4.157370567321777, | |
| "learning_rate": 5.116959064327485e-06, | |
| "loss": 0.0995, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 45.01052631578948, | |
| "grad_norm": 2.8948891162872314, | |
| "learning_rate": 4.970760233918129e-06, | |
| "loss": 0.0663, | |
| "step": 3460 | |
| }, | |
| { | |
| "epoch": 45.01315789473684, | |
| "grad_norm": 1.7566790580749512, | |
| "learning_rate": 4.824561403508772e-06, | |
| "loss": 0.051, | |
| "step": 3470 | |
| }, | |
| { | |
| "epoch": 45.01578947368421, | |
| "grad_norm": 20.800312042236328, | |
| "learning_rate": 4.678362573099415e-06, | |
| "loss": 0.0804, | |
| "step": 3480 | |
| }, | |
| { | |
| "epoch": 45.01842105263158, | |
| "grad_norm": 0.07893835008144379, | |
| "learning_rate": 4.532163742690059e-06, | |
| "loss": 0.2085, | |
| "step": 3490 | |
| }, | |
| { | |
| "epoch": 45.02, | |
| "eval_accuracy": 0.7716535433070866, | |
| "eval_loss": 0.7989789247512817, | |
| "eval_runtime": 208.7877, | |
| "eval_samples_per_second": 0.608, | |
| "eval_steps_per_second": 0.077, | |
| "step": 3496 | |
| }, | |
| { | |
| "epoch": 46.00105263157895, | |
| "grad_norm": 22.98341941833496, | |
| "learning_rate": 4.3859649122807014e-06, | |
| "loss": 0.0652, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 46.003684210526316, | |
| "grad_norm": 2.6828200817108154, | |
| "learning_rate": 4.239766081871345e-06, | |
| "loss": 0.0426, | |
| "step": 3510 | |
| }, | |
| { | |
| "epoch": 46.00631578947368, | |
| "grad_norm": 7.188971996307373, | |
| "learning_rate": 4.093567251461989e-06, | |
| "loss": 0.0808, | |
| "step": 3520 | |
| }, | |
| { | |
| "epoch": 46.008947368421055, | |
| "grad_norm": 0.5306025147438049, | |
| "learning_rate": 3.9473684210526315e-06, | |
| "loss": 0.0881, | |
| "step": 3530 | |
| }, | |
| { | |
| "epoch": 46.01157894736842, | |
| "grad_norm": 1.7458972930908203, | |
| "learning_rate": 3.8011695906432747e-06, | |
| "loss": 0.0808, | |
| "step": 3540 | |
| }, | |
| { | |
| "epoch": 46.01421052631579, | |
| "grad_norm": 0.7338550686836243, | |
| "learning_rate": 3.6549707602339183e-06, | |
| "loss": 0.0545, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 46.01684210526316, | |
| "grad_norm": 1.4630274772644043, | |
| "learning_rate": 3.5087719298245615e-06, | |
| "loss": 0.119, | |
| "step": 3560 | |
| }, | |
| { | |
| "epoch": 46.019473684210524, | |
| "grad_norm": 0.12060370296239853, | |
| "learning_rate": 3.3625730994152047e-06, | |
| "loss": 0.0509, | |
| "step": 3570 | |
| }, | |
| { | |
| "epoch": 46.02, | |
| "eval_accuracy": 0.8267716535433071, | |
| "eval_loss": 0.7133814096450806, | |
| "eval_runtime": 215.9384, | |
| "eval_samples_per_second": 0.588, | |
| "eval_steps_per_second": 0.074, | |
| "step": 3572 | |
| }, | |
| { | |
| "epoch": 47.002105263157894, | |
| "grad_norm": 4.411724090576172, | |
| "learning_rate": 3.216374269005848e-06, | |
| "loss": 0.1346, | |
| "step": 3580 | |
| }, | |
| { | |
| "epoch": 47.00473684210527, | |
| "grad_norm": 0.7380800843238831, | |
| "learning_rate": 3.070175438596491e-06, | |
| "loss": 0.0715, | |
| "step": 3590 | |
| }, | |
| { | |
| "epoch": 47.00736842105263, | |
| "grad_norm": 3.584591865539551, | |
| "learning_rate": 2.9239766081871343e-06, | |
| "loss": 0.1098, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 47.01, | |
| "grad_norm": 20.62811851501465, | |
| "learning_rate": 2.777777777777778e-06, | |
| "loss": 0.0818, | |
| "step": 3610 | |
| }, | |
| { | |
| "epoch": 47.01263157894737, | |
| "grad_norm": 12.115966796875, | |
| "learning_rate": 2.631578947368421e-06, | |
| "loss": 0.0649, | |
| "step": 3620 | |
| }, | |
| { | |
| "epoch": 47.015263157894736, | |
| "grad_norm": 0.42892947793006897, | |
| "learning_rate": 2.4853801169590643e-06, | |
| "loss": 0.0611, | |
| "step": 3630 | |
| }, | |
| { | |
| "epoch": 47.0178947368421, | |
| "grad_norm": 29.426136016845703, | |
| "learning_rate": 2.3391812865497075e-06, | |
| "loss": 0.0791, | |
| "step": 3640 | |
| }, | |
| { | |
| "epoch": 47.02, | |
| "eval_accuracy": 0.8188976377952756, | |
| "eval_loss": 0.6887122988700867, | |
| "eval_runtime": 227.1782, | |
| "eval_samples_per_second": 0.559, | |
| "eval_steps_per_second": 0.07, | |
| "step": 3648 | |
| }, | |
| { | |
| "epoch": 48.00052631578947, | |
| "grad_norm": 0.11056993901729584, | |
| "learning_rate": 2.1929824561403507e-06, | |
| "loss": 0.1203, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 48.003157894736844, | |
| "grad_norm": 3.5159361362457275, | |
| "learning_rate": 2.0467836257309943e-06, | |
| "loss": 0.0703, | |
| "step": 3660 | |
| }, | |
| { | |
| "epoch": 48.00578947368421, | |
| "grad_norm": 0.193506121635437, | |
| "learning_rate": 1.9005847953216373e-06, | |
| "loss": 0.1127, | |
| "step": 3670 | |
| }, | |
| { | |
| "epoch": 48.008421052631576, | |
| "grad_norm": 0.2963356673717499, | |
| "learning_rate": 1.7543859649122807e-06, | |
| "loss": 0.0622, | |
| "step": 3680 | |
| }, | |
| { | |
| "epoch": 48.01105263157895, | |
| "grad_norm": 0.357128381729126, | |
| "learning_rate": 1.608187134502924e-06, | |
| "loss": 0.0157, | |
| "step": 3690 | |
| }, | |
| { | |
| "epoch": 48.013684210526314, | |
| "grad_norm": 5.457529067993164, | |
| "learning_rate": 1.4619883040935671e-06, | |
| "loss": 0.0483, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 48.01631578947369, | |
| "grad_norm": 32.87456512451172, | |
| "learning_rate": 1.3157894736842106e-06, | |
| "loss": 0.0877, | |
| "step": 3710 | |
| }, | |
| { | |
| "epoch": 48.01894736842105, | |
| "grad_norm": 12.573355674743652, | |
| "learning_rate": 1.1695906432748538e-06, | |
| "loss": 0.0469, | |
| "step": 3720 | |
| }, | |
| { | |
| "epoch": 48.02, | |
| "eval_accuracy": 0.8031496062992126, | |
| "eval_loss": 0.7159085273742676, | |
| "eval_runtime": 224.784, | |
| "eval_samples_per_second": 0.565, | |
| "eval_steps_per_second": 0.071, | |
| "step": 3724 | |
| }, | |
| { | |
| "epoch": 49.00157894736842, | |
| "grad_norm": 0.6939800381660461, | |
| "learning_rate": 1.0233918128654972e-06, | |
| "loss": 0.1366, | |
| "step": 3730 | |
| }, | |
| { | |
| "epoch": 49.00421052631579, | |
| "grad_norm": 1.9841969013214111, | |
| "learning_rate": 8.771929824561404e-07, | |
| "loss": 0.0566, | |
| "step": 3740 | |
| }, | |
| { | |
| "epoch": 49.00684210526316, | |
| "grad_norm": 24.58515739440918, | |
| "learning_rate": 7.309941520467836e-07, | |
| "loss": 0.1208, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 49.009473684210526, | |
| "grad_norm": 0.08509305864572525, | |
| "learning_rate": 5.847953216374269e-07, | |
| "loss": 0.071, | |
| "step": 3760 | |
| }, | |
| { | |
| "epoch": 49.01210526315789, | |
| "grad_norm": 3.9984312057495117, | |
| "learning_rate": 4.385964912280702e-07, | |
| "loss": 0.058, | |
| "step": 3770 | |
| }, | |
| { | |
| "epoch": 49.014736842105265, | |
| "grad_norm": 0.45436879992485046, | |
| "learning_rate": 2.9239766081871344e-07, | |
| "loss": 0.1048, | |
| "step": 3780 | |
| }, | |
| { | |
| "epoch": 49.01736842105263, | |
| "grad_norm": 0.8313738703727722, | |
| "learning_rate": 1.4619883040935672e-07, | |
| "loss": 0.0478, | |
| "step": 3790 | |
| }, | |
| { | |
| "epoch": 49.02, | |
| "grad_norm": 0.8988305926322937, | |
| "learning_rate": 0.0, | |
| "loss": 0.0621, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 49.02, | |
| "eval_accuracy": 0.8031496062992126, | |
| "eval_loss": 0.7061663866043091, | |
| "eval_runtime": 216.7444, | |
| "eval_samples_per_second": 0.586, | |
| "eval_steps_per_second": 0.074, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 49.02, | |
| "step": 3800, | |
| "total_flos": 3.7892187304073626e+19, | |
| "train_loss": 0.6421483513399174, | |
| "train_runtime": 65613.5675, | |
| "train_samples_per_second": 0.463, | |
| "train_steps_per_second": 0.058 | |
| }, | |
| { | |
| "epoch": 49.02, | |
| "eval_accuracy": 0.6838709677419355, | |
| "eval_loss": 1.2759393453598022, | |
| "eval_runtime": 270.5851, | |
| "eval_samples_per_second": 0.573, | |
| "eval_steps_per_second": 0.074, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 49.02, | |
| "eval_accuracy": 0.6838709677419355, | |
| "eval_loss": 1.2759393453598022, | |
| "eval_runtime": 271.1828, | |
| "eval_samples_per_second": 0.572, | |
| "eval_steps_per_second": 0.074, | |
| "step": 3800 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 3800, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "total_flos": 3.7892187304073626e+19, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |