| { | |
| "best_metric": 0.8672510958046337, | |
| "best_model_checkpoint": "convnextv2-tiny-1k-224-finetuned-galaxy10-decals/checkpoint-785", | |
| "epoch": 9.977728285077951, | |
| "eval_steps": 500, | |
| "global_step": 1120, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08908685968819599, | |
| "grad_norm": 2.9069173336029053, | |
| "learning_rate": 4.464285714285715e-06, | |
| "loss": 2.3159, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.17817371937639198, | |
| "grad_norm": 4.669096946716309, | |
| "learning_rate": 8.92857142857143e-06, | |
| "loss": 2.2657, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.267260579064588, | |
| "grad_norm": 5.892533302307129, | |
| "learning_rate": 1.3392857142857144e-05, | |
| "loss": 2.1852, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.35634743875278396, | |
| "grad_norm": 4.657855033874512, | |
| "learning_rate": 1.785714285714286e-05, | |
| "loss": 2.0843, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.44543429844098, | |
| "grad_norm": 4.414278030395508, | |
| "learning_rate": 2.2321428571428575e-05, | |
| "loss": 1.9644, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.534521158129176, | |
| "grad_norm": 6.150153636932373, | |
| "learning_rate": 2.6785714285714288e-05, | |
| "loss": 1.7921, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.623608017817372, | |
| "grad_norm": 7.54302978515625, | |
| "learning_rate": 3.125e-05, | |
| "loss": 1.5743, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.7126948775055679, | |
| "grad_norm": 11.624669075012207, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 1.42, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.8017817371937639, | |
| "grad_norm": 11.175118446350098, | |
| "learning_rate": 4.017857142857143e-05, | |
| "loss": 1.301, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.89086859688196, | |
| "grad_norm": 13.484392166137695, | |
| "learning_rate": 4.464285714285715e-05, | |
| "loss": 1.2246, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.9799554565701559, | |
| "grad_norm": 22.607799530029297, | |
| "learning_rate": 4.910714285714286e-05, | |
| "loss": 1.0664, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.9977728285077951, | |
| "eval_accuracy": 0.6725109580463369, | |
| "eval_loss": 0.9818494915962219, | |
| "eval_runtime": 12.2808, | |
| "eval_samples_per_second": 130.041, | |
| "eval_steps_per_second": 4.071, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 1.069042316258352, | |
| "grad_norm": 11.895453453063965, | |
| "learning_rate": 4.960317460317461e-05, | |
| "loss": 1.0415, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.158129175946548, | |
| "grad_norm": 15.956673622131348, | |
| "learning_rate": 4.910714285714286e-05, | |
| "loss": 1.0021, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.247216035634744, | |
| "grad_norm": 12.564276695251465, | |
| "learning_rate": 4.8611111111111115e-05, | |
| "loss": 0.9518, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.3363028953229399, | |
| "grad_norm": 15.347941398620605, | |
| "learning_rate": 4.811507936507937e-05, | |
| "loss": 0.9316, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.4253897550111359, | |
| "grad_norm": 13.298521041870117, | |
| "learning_rate": 4.761904761904762e-05, | |
| "loss": 0.8316, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.5144766146993318, | |
| "grad_norm": 12.444368362426758, | |
| "learning_rate": 4.7123015873015876e-05, | |
| "loss": 0.8863, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.6035634743875278, | |
| "grad_norm": 10.18455982208252, | |
| "learning_rate": 4.662698412698413e-05, | |
| "loss": 0.8304, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.692650334075724, | |
| "grad_norm": 13.635807991027832, | |
| "learning_rate": 4.613095238095239e-05, | |
| "loss": 0.7748, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.7817371937639197, | |
| "grad_norm": 9.778557777404785, | |
| "learning_rate": 4.563492063492064e-05, | |
| "loss": 0.7561, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.8708240534521159, | |
| "grad_norm": 16.08139419555664, | |
| "learning_rate": 4.5138888888888894e-05, | |
| "loss": 0.8051, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.9599109131403119, | |
| "grad_norm": 12.761980056762695, | |
| "learning_rate": 4.464285714285715e-05, | |
| "loss": 0.8019, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.9955456570155903, | |
| "eval_accuracy": 0.7989981214777708, | |
| "eval_loss": 0.6332681775093079, | |
| "eval_runtime": 12.3035, | |
| "eval_samples_per_second": 129.8, | |
| "eval_steps_per_second": 4.064, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 2.048997772828508, | |
| "grad_norm": 10.136615753173828, | |
| "learning_rate": 4.41468253968254e-05, | |
| "loss": 0.7221, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.138084632516704, | |
| "grad_norm": 13.145553588867188, | |
| "learning_rate": 4.3650793650793655e-05, | |
| "loss": 0.6877, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.2271714922048997, | |
| "grad_norm": 14.941648483276367, | |
| "learning_rate": 4.315476190476191e-05, | |
| "loss": 0.7462, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.316258351893096, | |
| "grad_norm": 9.842031478881836, | |
| "learning_rate": 4.265873015873016e-05, | |
| "loss": 0.6886, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.4053452115812917, | |
| "grad_norm": 11.978471755981445, | |
| "learning_rate": 4.2162698412698416e-05, | |
| "loss": 0.6843, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.494432071269488, | |
| "grad_norm": 12.24606990814209, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.7038, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.5835189309576836, | |
| "grad_norm": 12.357331275939941, | |
| "learning_rate": 4.117063492063492e-05, | |
| "loss": 0.696, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.6726057906458798, | |
| "grad_norm": 16.765913009643555, | |
| "learning_rate": 4.067460317460318e-05, | |
| "loss": 0.7174, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.7616926503340755, | |
| "grad_norm": 13.57442569732666, | |
| "learning_rate": 4.017857142857143e-05, | |
| "loss": 0.7305, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.8507795100222717, | |
| "grad_norm": 13.48181438446045, | |
| "learning_rate": 3.968253968253968e-05, | |
| "loss": 0.6957, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.939866369710468, | |
| "grad_norm": 9.809232711791992, | |
| "learning_rate": 3.918650793650794e-05, | |
| "loss": 0.6524, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.9933184855233854, | |
| "eval_accuracy": 0.8340638697557922, | |
| "eval_loss": 0.5247990489006042, | |
| "eval_runtime": 12.2883, | |
| "eval_samples_per_second": 129.961, | |
| "eval_steps_per_second": 4.069, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 3.0289532293986636, | |
| "grad_norm": 16.003482818603516, | |
| "learning_rate": 3.8690476190476195e-05, | |
| "loss": 0.6928, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 3.11804008908686, | |
| "grad_norm": 23.448598861694336, | |
| "learning_rate": 3.8194444444444444e-05, | |
| "loss": 0.6696, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 3.2071269487750556, | |
| "grad_norm": 12.516254425048828, | |
| "learning_rate": 3.76984126984127e-05, | |
| "loss": 0.6665, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 3.2962138084632517, | |
| "grad_norm": 13.503238677978516, | |
| "learning_rate": 3.7202380952380956e-05, | |
| "loss": 0.6189, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 3.3853006681514475, | |
| "grad_norm": 14.721902847290039, | |
| "learning_rate": 3.6706349206349205e-05, | |
| "loss": 0.6405, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.4743875278396437, | |
| "grad_norm": 11.428637504577637, | |
| "learning_rate": 3.621031746031746e-05, | |
| "loss": 0.6502, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.5634743875278394, | |
| "grad_norm": 8.628026008605957, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 0.6335, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.6525612472160356, | |
| "grad_norm": 12.637211799621582, | |
| "learning_rate": 3.521825396825397e-05, | |
| "loss": 0.6457, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 3.7416481069042318, | |
| "grad_norm": 13.72917652130127, | |
| "learning_rate": 3.472222222222222e-05, | |
| "loss": 0.6338, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.8307349665924275, | |
| "grad_norm": 14.159635543823242, | |
| "learning_rate": 3.422619047619048e-05, | |
| "loss": 0.6318, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.9198218262806237, | |
| "grad_norm": 12.676724433898926, | |
| "learning_rate": 3.3730158730158734e-05, | |
| "loss": 0.6339, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8447088290544772, | |
| "eval_loss": 0.4730662703514099, | |
| "eval_runtime": 12.4926, | |
| "eval_samples_per_second": 127.835, | |
| "eval_steps_per_second": 4.002, | |
| "step": 449 | |
| }, | |
| { | |
| "epoch": 4.008908685968819, | |
| "grad_norm": 11.317361831665039, | |
| "learning_rate": 3.3234126984126983e-05, | |
| "loss": 0.613, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 4.097995545657016, | |
| "grad_norm": 14.402922630310059, | |
| "learning_rate": 3.273809523809524e-05, | |
| "loss": 0.6124, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 4.187082405345212, | |
| "grad_norm": 9.939033508300781, | |
| "learning_rate": 3.2242063492063495e-05, | |
| "loss": 0.5868, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 4.276169265033408, | |
| "grad_norm": 10.611005783081055, | |
| "learning_rate": 3.1746031746031745e-05, | |
| "loss": 0.5786, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 4.365256124721603, | |
| "grad_norm": 11.104296684265137, | |
| "learning_rate": 3.125e-05, | |
| "loss": 0.544, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 4.4543429844097995, | |
| "grad_norm": 14.008048057556152, | |
| "learning_rate": 3.075396825396826e-05, | |
| "loss": 0.6175, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.543429844097996, | |
| "grad_norm": 9.320144653320312, | |
| "learning_rate": 3.0257936507936506e-05, | |
| "loss": 0.5999, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 4.632516703786192, | |
| "grad_norm": 9.274946212768555, | |
| "learning_rate": 2.9761904761904762e-05, | |
| "loss": 0.5709, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.721603563474387, | |
| "grad_norm": 12.640064239501953, | |
| "learning_rate": 2.9265873015873018e-05, | |
| "loss": 0.6231, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 4.810690423162583, | |
| "grad_norm": 11.968724250793457, | |
| "learning_rate": 2.876984126984127e-05, | |
| "loss": 0.6206, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 4.8997772828507795, | |
| "grad_norm": 11.681157112121582, | |
| "learning_rate": 2.8273809523809523e-05, | |
| "loss": 0.6031, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 4.988864142538976, | |
| "grad_norm": 13.320256233215332, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.5178, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 4.997772828507795, | |
| "eval_accuracy": 0.8503443957420163, | |
| "eval_loss": 0.4537416100502014, | |
| "eval_runtime": 12.2913, | |
| "eval_samples_per_second": 129.93, | |
| "eval_steps_per_second": 4.068, | |
| "step": 561 | |
| }, | |
| { | |
| "epoch": 5.077951002227172, | |
| "grad_norm": 15.332996368408203, | |
| "learning_rate": 2.7281746031746032e-05, | |
| "loss": 0.5617, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 5.167037861915367, | |
| "grad_norm": 14.994087219238281, | |
| "learning_rate": 2.6785714285714288e-05, | |
| "loss": 0.5797, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 5.256124721603563, | |
| "grad_norm": 13.461969375610352, | |
| "learning_rate": 2.628968253968254e-05, | |
| "loss": 0.5524, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 5.3452115812917596, | |
| "grad_norm": 12.29080581665039, | |
| "learning_rate": 2.5793650793650796e-05, | |
| "loss": 0.5824, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.434298440979956, | |
| "grad_norm": 11.07197380065918, | |
| "learning_rate": 2.529761904761905e-05, | |
| "loss": 0.554, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.523385300668151, | |
| "grad_norm": 9.797560691833496, | |
| "learning_rate": 2.4801587301587305e-05, | |
| "loss": 0.5108, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 5.612472160356347, | |
| "grad_norm": 10.469209671020508, | |
| "learning_rate": 2.4305555555555558e-05, | |
| "loss": 0.5586, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 5.701559020044543, | |
| "grad_norm": 13.22735595703125, | |
| "learning_rate": 2.380952380952381e-05, | |
| "loss": 0.5358, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 5.79064587973274, | |
| "grad_norm": 8.305063247680664, | |
| "learning_rate": 2.3313492063492066e-05, | |
| "loss": 0.5295, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.879732739420936, | |
| "grad_norm": 18.399051666259766, | |
| "learning_rate": 2.281746031746032e-05, | |
| "loss": 0.5442, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 5.968819599109131, | |
| "grad_norm": 8.103595733642578, | |
| "learning_rate": 2.2321428571428575e-05, | |
| "loss": 0.5907, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 5.99554565701559, | |
| "eval_accuracy": 0.8472135253600501, | |
| "eval_loss": 0.4555535316467285, | |
| "eval_runtime": 12.2927, | |
| "eval_samples_per_second": 129.914, | |
| "eval_steps_per_second": 4.067, | |
| "step": 673 | |
| }, | |
| { | |
| "epoch": 6.057906458797327, | |
| "grad_norm": 10.681763648986816, | |
| "learning_rate": 2.1825396825396827e-05, | |
| "loss": 0.5332, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 6.146993318485523, | |
| "grad_norm": 10.129424095153809, | |
| "learning_rate": 2.132936507936508e-05, | |
| "loss": 0.4747, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 6.23608017817372, | |
| "grad_norm": 16.834814071655273, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 0.5576, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 6.325167037861915, | |
| "grad_norm": 11.258397102355957, | |
| "learning_rate": 2.033730158730159e-05, | |
| "loss": 0.5063, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 6.414253897550111, | |
| "grad_norm": 15.159914016723633, | |
| "learning_rate": 1.984126984126984e-05, | |
| "loss": 0.5385, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 6.503340757238307, | |
| "grad_norm": 10.242027282714844, | |
| "learning_rate": 1.9345238095238097e-05, | |
| "loss": 0.5046, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 6.5924276169265035, | |
| "grad_norm": 10.377813339233398, | |
| "learning_rate": 1.884920634920635e-05, | |
| "loss": 0.5247, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 6.6815144766147, | |
| "grad_norm": 12.55459213256836, | |
| "learning_rate": 1.8353174603174602e-05, | |
| "loss": 0.529, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 6.770601336302895, | |
| "grad_norm": 16.02656364440918, | |
| "learning_rate": 1.785714285714286e-05, | |
| "loss": 0.5073, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 6.859688195991091, | |
| "grad_norm": 16.140487670898438, | |
| "learning_rate": 1.736111111111111e-05, | |
| "loss": 0.5414, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 6.948775055679287, | |
| "grad_norm": 17.321931838989258, | |
| "learning_rate": 1.6865079365079367e-05, | |
| "loss": 0.5292, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 6.993318485523385, | |
| "eval_accuracy": 0.8672510958046337, | |
| "eval_loss": 0.41685062646865845, | |
| "eval_runtime": 12.3673, | |
| "eval_samples_per_second": 129.131, | |
| "eval_steps_per_second": 4.043, | |
| "step": 785 | |
| }, | |
| { | |
| "epoch": 7.0378619153674835, | |
| "grad_norm": 9.479693412780762, | |
| "learning_rate": 1.636904761904762e-05, | |
| "loss": 0.4586, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 7.12694877505568, | |
| "grad_norm": 11.711000442504883, | |
| "learning_rate": 1.5873015873015872e-05, | |
| "loss": 0.5188, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 7.216035634743875, | |
| "grad_norm": 11.616864204406738, | |
| "learning_rate": 1.537698412698413e-05, | |
| "loss": 0.5024, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 7.305122494432071, | |
| "grad_norm": 10.370725631713867, | |
| "learning_rate": 1.4880952380952381e-05, | |
| "loss": 0.4902, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 7.394209354120267, | |
| "grad_norm": 14.04218864440918, | |
| "learning_rate": 1.4384920634920635e-05, | |
| "loss": 0.5149, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 7.4832962138084635, | |
| "grad_norm": 13.194646835327148, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.5562, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 7.57238307349666, | |
| "grad_norm": 9.960190773010254, | |
| "learning_rate": 1.3392857142857144e-05, | |
| "loss": 0.4921, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 7.661469933184855, | |
| "grad_norm": 15.14493465423584, | |
| "learning_rate": 1.2896825396825398e-05, | |
| "loss": 0.471, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 7.750556792873051, | |
| "grad_norm": 11.185235977172852, | |
| "learning_rate": 1.2400793650793652e-05, | |
| "loss": 0.4963, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 7.839643652561247, | |
| "grad_norm": 12.782095909118652, | |
| "learning_rate": 1.1904761904761905e-05, | |
| "loss": 0.4915, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 7.928730512249444, | |
| "grad_norm": 11.89919376373291, | |
| "learning_rate": 1.140873015873016e-05, | |
| "loss": 0.5017, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8597370068879149, | |
| "eval_loss": 0.4106651544570923, | |
| "eval_runtime": 12.3902, | |
| "eval_samples_per_second": 128.892, | |
| "eval_steps_per_second": 4.035, | |
| "step": 898 | |
| }, | |
| { | |
| "epoch": 8.017817371937639, | |
| "grad_norm": 12.601805686950684, | |
| "learning_rate": 1.0912698412698414e-05, | |
| "loss": 0.5064, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 8.106904231625835, | |
| "grad_norm": 8.723831176757812, | |
| "learning_rate": 1.0416666666666668e-05, | |
| "loss": 0.4181, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 8.195991091314031, | |
| "grad_norm": 12.781538963317871, | |
| "learning_rate": 9.92063492063492e-06, | |
| "loss": 0.4427, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 8.285077951002227, | |
| "grad_norm": 12.263012886047363, | |
| "learning_rate": 9.424603174603175e-06, | |
| "loss": 0.5087, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 8.374164810690424, | |
| "grad_norm": 17.41984748840332, | |
| "learning_rate": 8.92857142857143e-06, | |
| "loss": 0.5301, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 8.46325167037862, | |
| "grad_norm": 10.731024742126465, | |
| "learning_rate": 8.432539682539684e-06, | |
| "loss": 0.4987, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 8.552338530066816, | |
| "grad_norm": 15.722013473510742, | |
| "learning_rate": 7.936507936507936e-06, | |
| "loss": 0.4613, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 8.64142538975501, | |
| "grad_norm": 11.301126480102539, | |
| "learning_rate": 7.4404761904761905e-06, | |
| "loss": 0.5136, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 8.730512249443207, | |
| "grad_norm": 21.23493194580078, | |
| "learning_rate": 6.944444444444445e-06, | |
| "loss": 0.494, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 8.819599109131403, | |
| "grad_norm": 10.211363792419434, | |
| "learning_rate": 6.448412698412699e-06, | |
| "loss": 0.4619, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 8.908685968819599, | |
| "grad_norm": 12.277856826782227, | |
| "learning_rate": 5.9523809523809525e-06, | |
| "loss": 0.4361, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.997772828507795, | |
| "grad_norm": 12.28085708618164, | |
| "learning_rate": 5.456349206349207e-06, | |
| "loss": 0.4605, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 8.997772828507795, | |
| "eval_accuracy": 0.8634940513462742, | |
| "eval_loss": 0.40621063113212585, | |
| "eval_runtime": 12.3051, | |
| "eval_samples_per_second": 129.783, | |
| "eval_steps_per_second": 4.063, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 9.086859688195991, | |
| "grad_norm": 8.358485221862793, | |
| "learning_rate": 4.96031746031746e-06, | |
| "loss": 0.4403, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 9.175946547884188, | |
| "grad_norm": 8.35409164428711, | |
| "learning_rate": 4.464285714285715e-06, | |
| "loss": 0.4514, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 9.265033407572384, | |
| "grad_norm": 10.057600021362305, | |
| "learning_rate": 3.968253968253968e-06, | |
| "loss": 0.427, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 9.35412026726058, | |
| "grad_norm": 7.57137393951416, | |
| "learning_rate": 3.4722222222222224e-06, | |
| "loss": 0.4619, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 9.443207126948774, | |
| "grad_norm": 9.249728202819824, | |
| "learning_rate": 2.9761904761904763e-06, | |
| "loss": 0.4606, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 9.53229398663697, | |
| "grad_norm": 10.303194046020508, | |
| "learning_rate": 2.48015873015873e-06, | |
| "loss": 0.508, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 9.621380846325167, | |
| "grad_norm": 11.307740211486816, | |
| "learning_rate": 1.984126984126984e-06, | |
| "loss": 0.4667, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 9.710467706013363, | |
| "grad_norm": 11.34073543548584, | |
| "learning_rate": 1.4880952380952381e-06, | |
| "loss": 0.4803, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 9.799554565701559, | |
| "grad_norm": 12.684574127197266, | |
| "learning_rate": 9.92063492063492e-07, | |
| "loss": 0.4613, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 9.888641425389755, | |
| "grad_norm": 12.430156707763672, | |
| "learning_rate": 4.96031746031746e-07, | |
| "loss": 0.434, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 9.977728285077951, | |
| "grad_norm": 19.223121643066406, | |
| "learning_rate": 0.0, | |
| "loss": 0.4765, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 9.977728285077951, | |
| "eval_accuracy": 0.8647463994990607, | |
| "eval_loss": 0.3980247676372528, | |
| "eval_runtime": 12.312, | |
| "eval_samples_per_second": 129.711, | |
| "eval_steps_per_second": 4.061, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 9.977728285077951, | |
| "step": 1120, | |
| "total_flos": 3.6084187126879396e+18, | |
| "train_loss": 0.7042633635657174, | |
| "train_runtime": 2366.3213, | |
| "train_samples_per_second": 60.706, | |
| "train_steps_per_second": 0.473 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1120, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 3.6084187126879396e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |