| { | |
| "best_metric": 0.007947824895381927, | |
| "best_model_checkpoint": "./beans_outputs/checkpoint-3250", | |
| "epoch": 50.0, | |
| "eval_steps": 500, | |
| "global_step": 3250, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07692307692307693, | |
| "grad_norm": 2.198366403579712, | |
| "learning_rate": 1.9692307692307696e-05, | |
| "loss": 1.0245, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.15384615384615385, | |
| "grad_norm": 1.8460220098495483, | |
| "learning_rate": 1.9384615384615386e-05, | |
| "loss": 0.9453, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.23076923076923078, | |
| "grad_norm": 2.117300271987915, | |
| "learning_rate": 1.907692307692308e-05, | |
| "loss": 0.8406, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.3076923076923077, | |
| "grad_norm": 2.645448684692383, | |
| "learning_rate": 1.876923076923077e-05, | |
| "loss": 0.6951, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.38461538461538464, | |
| "grad_norm": 4.24210262298584, | |
| "learning_rate": 1.8461538461538465e-05, | |
| "loss": 0.662, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.46153846153846156, | |
| "grad_norm": 2.095567464828491, | |
| "learning_rate": 1.8153846153846155e-05, | |
| "loss": 0.5811, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.5384615384615384, | |
| "grad_norm": 2.079636573791504, | |
| "learning_rate": 1.784615384615385e-05, | |
| "loss": 0.5193, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.6153846153846154, | |
| "grad_norm": 1.218340277671814, | |
| "learning_rate": 1.753846153846154e-05, | |
| "loss": 0.3945, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.6923076923076923, | |
| "grad_norm": 2.4416184425354004, | |
| "learning_rate": 1.7230769230769234e-05, | |
| "loss": 0.3528, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 2.8852720260620117, | |
| "learning_rate": 1.6923076923076924e-05, | |
| "loss": 0.3689, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.8461538461538461, | |
| "grad_norm": 1.9503966569900513, | |
| "learning_rate": 1.6615384615384618e-05, | |
| "loss": 0.3368, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.9230769230769231, | |
| "grad_norm": 2.335829496383667, | |
| "learning_rate": 1.630769230769231e-05, | |
| "loss": 0.2468, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 7.0948638916015625, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.2859, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9624060150375939, | |
| "eval_loss": 0.21892516314983368, | |
| "eval_runtime": 0.9855, | |
| "eval_samples_per_second": 134.957, | |
| "eval_steps_per_second": 17.25, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.0769230769230769, | |
| "grad_norm": 1.765810251235962, | |
| "learning_rate": 1.5692307692307693e-05, | |
| "loss": 0.2006, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.1538461538461537, | |
| "grad_norm": 3.1403791904449463, | |
| "learning_rate": 1.5384615384615387e-05, | |
| "loss": 0.2243, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.2307692307692308, | |
| "grad_norm": 1.4933677911758423, | |
| "learning_rate": 1.5076923076923078e-05, | |
| "loss": 0.2035, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.3076923076923077, | |
| "grad_norm": 0.5171657800674438, | |
| "learning_rate": 1.4769230769230772e-05, | |
| "loss": 0.2208, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.3846153846153846, | |
| "grad_norm": 0.600874662399292, | |
| "learning_rate": 1.4461538461538462e-05, | |
| "loss": 0.211, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.4615384615384617, | |
| "grad_norm": 3.5654282569885254, | |
| "learning_rate": 1.4153846153846156e-05, | |
| "loss": 0.1607, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 0.47061973810195923, | |
| "learning_rate": 1.3846153846153847e-05, | |
| "loss": 0.2526, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.6153846153846154, | |
| "grad_norm": 0.6342479586601257, | |
| "learning_rate": 1.353846153846154e-05, | |
| "loss": 0.1804, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.6923076923076923, | |
| "grad_norm": 2.391136884689331, | |
| "learning_rate": 1.3230769230769231e-05, | |
| "loss": 0.2121, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.7692307692307692, | |
| "grad_norm": 1.8535972833633423, | |
| "learning_rate": 1.2923076923076925e-05, | |
| "loss": 0.1678, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.8461538461538463, | |
| "grad_norm": 11.256393432617188, | |
| "learning_rate": 1.2615384615384616e-05, | |
| "loss": 0.2574, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.9230769230769231, | |
| "grad_norm": 0.4076194167137146, | |
| "learning_rate": 1.230769230769231e-05, | |
| "loss": 0.1442, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.3490784168243408, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.1316, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9699248120300752, | |
| "eval_loss": 0.13337133824825287, | |
| "eval_runtime": 0.9938, | |
| "eval_samples_per_second": 133.827, | |
| "eval_steps_per_second": 17.106, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.076923076923077, | |
| "grad_norm": 0.5328232645988464, | |
| "learning_rate": 1.1692307692307694e-05, | |
| "loss": 0.0897, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.1538461538461537, | |
| "grad_norm": 0.3375113904476166, | |
| "learning_rate": 1.1384615384615385e-05, | |
| "loss": 0.1475, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.230769230769231, | |
| "grad_norm": 7.195676326751709, | |
| "learning_rate": 1.1076923076923079e-05, | |
| "loss": 0.1321, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 8.226975440979004, | |
| "learning_rate": 1.076923076923077e-05, | |
| "loss": 0.1375, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.3846153846153846, | |
| "grad_norm": 5.051274299621582, | |
| "learning_rate": 1.0461538461538463e-05, | |
| "loss": 0.1755, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.4615384615384617, | |
| "grad_norm": 0.30559489130973816, | |
| "learning_rate": 1.0153846153846154e-05, | |
| "loss": 0.1542, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.5384615384615383, | |
| "grad_norm": 6.551879405975342, | |
| "learning_rate": 9.846153846153848e-06, | |
| "loss": 0.1872, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.6153846153846154, | |
| "grad_norm": 1.2321629524230957, | |
| "learning_rate": 9.53846153846154e-06, | |
| "loss": 0.0714, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.6923076923076925, | |
| "grad_norm": 0.3684569299221039, | |
| "learning_rate": 9.230769230769232e-06, | |
| "loss": 0.1325, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.769230769230769, | |
| "grad_norm": 9.62742805480957, | |
| "learning_rate": 8.923076923076925e-06, | |
| "loss": 0.1203, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.8461538461538463, | |
| "grad_norm": 0.3335096836090088, | |
| "learning_rate": 8.615384615384617e-06, | |
| "loss": 0.154, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.9230769230769234, | |
| "grad_norm": 1.1392183303833008, | |
| "learning_rate": 8.307692307692309e-06, | |
| "loss": 0.1306, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.29587361216545105, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.1438, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9699248120300752, | |
| "eval_loss": 0.09808181971311569, | |
| "eval_runtime": 0.9838, | |
| "eval_samples_per_second": 135.192, | |
| "eval_steps_per_second": 17.28, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 2.2563395500183105, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 0.1325, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.1538461538461537, | |
| "grad_norm": 6.758065223693848, | |
| "learning_rate": 7.384615384615386e-06, | |
| "loss": 0.154, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 3.230769230769231, | |
| "grad_norm": 8.712298393249512, | |
| "learning_rate": 7.076923076923078e-06, | |
| "loss": 0.1422, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.3076923076923075, | |
| "grad_norm": 0.23161759972572327, | |
| "learning_rate": 6.76923076923077e-06, | |
| "loss": 0.0768, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.3846153846153846, | |
| "grad_norm": 5.164632797241211, | |
| "learning_rate": 6.461538461538463e-06, | |
| "loss": 0.1136, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.4615384615384617, | |
| "grad_norm": 12.590032577514648, | |
| "learning_rate": 6.153846153846155e-06, | |
| "loss": 0.0854, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.5384615384615383, | |
| "grad_norm": 4.483827590942383, | |
| "learning_rate": 5.846153846153847e-06, | |
| "loss": 0.098, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.6153846153846154, | |
| "grad_norm": 7.863719940185547, | |
| "learning_rate": 5.538461538461539e-06, | |
| "loss": 0.0778, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.6923076923076925, | |
| "grad_norm": 0.21892791986465454, | |
| "learning_rate": 5.230769230769232e-06, | |
| "loss": 0.0957, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.769230769230769, | |
| "grad_norm": 0.2107536494731903, | |
| "learning_rate": 4.923076923076924e-06, | |
| "loss": 0.0775, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "grad_norm": 0.9603075981140137, | |
| "learning_rate": 4.615384615384616e-06, | |
| "loss": 0.1357, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.9230769230769234, | |
| "grad_norm": 0.21214883029460907, | |
| "learning_rate": 4.307692307692308e-06, | |
| "loss": 0.1225, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.2996317744255066, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.0833, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9924812030075187, | |
| "eval_loss": 0.06558172404766083, | |
| "eval_runtime": 1.0412, | |
| "eval_samples_per_second": 127.735, | |
| "eval_steps_per_second": 16.327, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.076923076923077, | |
| "grad_norm": 9.374862670898438, | |
| "learning_rate": 3.692307692307693e-06, | |
| "loss": 0.1009, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 4.153846153846154, | |
| "grad_norm": 1.5623271465301514, | |
| "learning_rate": 3.384615384615385e-06, | |
| "loss": 0.1658, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 4.230769230769231, | |
| "grad_norm": 0.3211674094200134, | |
| "learning_rate": 3.0769230769230774e-06, | |
| "loss": 0.1534, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 4.3076923076923075, | |
| "grad_norm": 1.208674430847168, | |
| "learning_rate": 2.7692307692307697e-06, | |
| "loss": 0.0504, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 4.384615384615385, | |
| "grad_norm": 0.3158506751060486, | |
| "learning_rate": 2.461538461538462e-06, | |
| "loss": 0.1345, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 4.461538461538462, | |
| "grad_norm": 0.2690170705318451, | |
| "learning_rate": 2.153846153846154e-06, | |
| "loss": 0.0816, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 4.538461538461538, | |
| "grad_norm": 9.395990371704102, | |
| "learning_rate": 1.8461538461538465e-06, | |
| "loss": 0.0714, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 4.615384615384615, | |
| "grad_norm": 3.067613124847412, | |
| "learning_rate": 1.5384615384615387e-06, | |
| "loss": 0.0892, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.6923076923076925, | |
| "grad_norm": 5.167181491851807, | |
| "learning_rate": 1.230769230769231e-06, | |
| "loss": 0.087, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 4.769230769230769, | |
| "grad_norm": 0.2610948383808136, | |
| "learning_rate": 9.230769230769232e-07, | |
| "loss": 0.0807, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 4.846153846153846, | |
| "grad_norm": 0.22187209129333496, | |
| "learning_rate": 6.153846153846155e-07, | |
| "loss": 0.0668, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 4.923076923076923, | |
| "grad_norm": 0.23086942732334137, | |
| "learning_rate": 3.0769230769230774e-07, | |
| "loss": 0.0784, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.6445406675338745, | |
| "learning_rate": 0.0, | |
| "loss": 0.1107, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9774436090225563, | |
| "eval_loss": 0.0816693976521492, | |
| "eval_runtime": 1.9096, | |
| "eval_samples_per_second": 69.648, | |
| "eval_steps_per_second": 8.902, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 10.153846153846153, | |
| "grad_norm": 2.079591751098633, | |
| "learning_rate": 1.593846153846154e-05, | |
| "loss": 0.0956, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 10.307692307692308, | |
| "grad_norm": 0.22365273535251617, | |
| "learning_rate": 1.587692307692308e-05, | |
| "loss": 0.0704, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 10.461538461538462, | |
| "grad_norm": 1.2620346546173096, | |
| "learning_rate": 1.5815384615384616e-05, | |
| "loss": 0.0927, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 10.615384615384615, | |
| "grad_norm": 0.17169518768787384, | |
| "learning_rate": 1.5753846153846154e-05, | |
| "loss": 0.1318, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 10.76923076923077, | |
| "grad_norm": 1.01598060131073, | |
| "learning_rate": 1.5692307692307693e-05, | |
| "loss": 0.1102, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 10.923076923076923, | |
| "grad_norm": 1.3003897666931152, | |
| "learning_rate": 1.5630769230769232e-05, | |
| "loss": 0.098, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.9924812030075187, | |
| "eval_loss": 0.056982800364494324, | |
| "eval_runtime": 0.6404, | |
| "eval_samples_per_second": 207.697, | |
| "eval_steps_per_second": 14.055, | |
| "step": 715 | |
| }, | |
| { | |
| "epoch": 11.076923076923077, | |
| "grad_norm": 2.616173505783081, | |
| "learning_rate": 1.556923076923077e-05, | |
| "loss": 0.1061, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 11.23076923076923, | |
| "grad_norm": 0.20168966054916382, | |
| "learning_rate": 1.550769230769231e-05, | |
| "loss": 0.1275, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 11.384615384615385, | |
| "grad_norm": 2.059192419052124, | |
| "learning_rate": 1.544615384615385e-05, | |
| "loss": 0.0463, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 11.538461538461538, | |
| "grad_norm": 0.1697787493467331, | |
| "learning_rate": 1.5384615384615387e-05, | |
| "loss": 0.0759, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 11.692307692307692, | |
| "grad_norm": 0.8658211827278137, | |
| "learning_rate": 1.5323076923076926e-05, | |
| "loss": 0.1983, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 11.846153846153847, | |
| "grad_norm": 0.4407813251018524, | |
| "learning_rate": 1.5261538461538465e-05, | |
| "loss": 0.1181, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 6.601933479309082, | |
| "learning_rate": 1.5200000000000002e-05, | |
| "loss": 0.0935, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.04177865758538246, | |
| "eval_runtime": 0.6316, | |
| "eval_samples_per_second": 210.569, | |
| "eval_steps_per_second": 14.249, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 12.153846153846153, | |
| "grad_norm": 1.3233654499053955, | |
| "learning_rate": 1.5138461538461539e-05, | |
| "loss": 0.0949, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 12.307692307692308, | |
| "grad_norm": 10.496395111083984, | |
| "learning_rate": 1.5076923076923078e-05, | |
| "loss": 0.1141, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 12.461538461538462, | |
| "grad_norm": 0.40288811922073364, | |
| "learning_rate": 1.5015384615384617e-05, | |
| "loss": 0.0662, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 12.615384615384615, | |
| "grad_norm": 1.3594541549682617, | |
| "learning_rate": 1.4953846153846154e-05, | |
| "loss": 0.0672, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 12.76923076923077, | |
| "grad_norm": 0.15957416594028473, | |
| "learning_rate": 1.4892307692307692e-05, | |
| "loss": 0.1198, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 12.923076923076923, | |
| "grad_norm": 3.9073538780212402, | |
| "learning_rate": 1.4830769230769233e-05, | |
| "loss": 0.0907, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.9699248120300752, | |
| "eval_loss": 0.10930211842060089, | |
| "eval_runtime": 0.6314, | |
| "eval_samples_per_second": 210.658, | |
| "eval_steps_per_second": 14.255, | |
| "step": 845 | |
| }, | |
| { | |
| "epoch": 13.076923076923077, | |
| "grad_norm": 4.819633483886719, | |
| "learning_rate": 1.4769230769230772e-05, | |
| "loss": 0.0678, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 13.23076923076923, | |
| "grad_norm": 0.1750306636095047, | |
| "learning_rate": 1.4707692307692309e-05, | |
| "loss": 0.0498, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 13.384615384615385, | |
| "grad_norm": 0.7398102879524231, | |
| "learning_rate": 1.4646153846153848e-05, | |
| "loss": 0.112, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 13.538461538461538, | |
| "grad_norm": 1.1426323652267456, | |
| "learning_rate": 1.4584615384615386e-05, | |
| "loss": 0.061, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 13.692307692307692, | |
| "grad_norm": 0.16307789087295532, | |
| "learning_rate": 1.4523076923076923e-05, | |
| "loss": 0.0405, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 13.846153846153847, | |
| "grad_norm": 3.81508207321167, | |
| "learning_rate": 1.4461538461538462e-05, | |
| "loss": 0.0768, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 0.482637494802475, | |
| "learning_rate": 1.4400000000000001e-05, | |
| "loss": 0.0947, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.03473825752735138, | |
| "eval_runtime": 0.6324, | |
| "eval_samples_per_second": 210.312, | |
| "eval_steps_per_second": 14.232, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 14.153846153846153, | |
| "grad_norm": 0.18653550744056702, | |
| "learning_rate": 1.4338461538461538e-05, | |
| "loss": 0.0853, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 14.307692307692308, | |
| "grad_norm": 1.822533130645752, | |
| "learning_rate": 1.4276923076923077e-05, | |
| "loss": 0.1008, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 14.461538461538462, | |
| "grad_norm": 0.32205572724342346, | |
| "learning_rate": 1.4215384615384617e-05, | |
| "loss": 0.0638, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 14.615384615384615, | |
| "grad_norm": 0.5714139342308044, | |
| "learning_rate": 1.4153846153846156e-05, | |
| "loss": 0.1159, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 14.76923076923077, | |
| "grad_norm": 5.901656627655029, | |
| "learning_rate": 1.4092307692307693e-05, | |
| "loss": 0.091, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 14.923076923076923, | |
| "grad_norm": 2.119544744491577, | |
| "learning_rate": 1.4030769230769232e-05, | |
| "loss": 0.1259, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.9849624060150376, | |
| "eval_loss": 0.07099475711584091, | |
| "eval_runtime": 0.6277, | |
| "eval_samples_per_second": 211.88, | |
| "eval_steps_per_second": 14.338, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 15.076923076923077, | |
| "grad_norm": 0.14739356935024261, | |
| "learning_rate": 1.3969230769230771e-05, | |
| "loss": 0.082, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 15.23076923076923, | |
| "grad_norm": 0.12192496657371521, | |
| "learning_rate": 1.3907692307692308e-05, | |
| "loss": 0.0558, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 15.384615384615385, | |
| "grad_norm": 0.11841657012701035, | |
| "learning_rate": 1.3846153846153847e-05, | |
| "loss": 0.0852, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 15.538461538461538, | |
| "grad_norm": 0.8062542080879211, | |
| "learning_rate": 1.3784615384615386e-05, | |
| "loss": 0.0793, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 15.692307692307692, | |
| "grad_norm": 0.11781653016805649, | |
| "learning_rate": 1.3723076923076923e-05, | |
| "loss": 0.0436, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 15.846153846153847, | |
| "grad_norm": 0.4233933389186859, | |
| "learning_rate": 1.3661538461538461e-05, | |
| "loss": 0.0448, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.3999457061290741, | |
| "learning_rate": 1.3600000000000002e-05, | |
| "loss": 0.0325, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.9774436090225563, | |
| "eval_loss": 0.05867745727300644, | |
| "eval_runtime": 0.6418, | |
| "eval_samples_per_second": 207.244, | |
| "eval_steps_per_second": 14.024, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 16.153846153846153, | |
| "grad_norm": 0.36126333475112915, | |
| "learning_rate": 1.353846153846154e-05, | |
| "loss": 0.0373, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 16.307692307692307, | |
| "grad_norm": 2.5870630741119385, | |
| "learning_rate": 1.3476923076923078e-05, | |
| "loss": 0.0746, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 16.46153846153846, | |
| "grad_norm": 10.347090721130371, | |
| "learning_rate": 1.3415384615384617e-05, | |
| "loss": 0.047, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 16.615384615384617, | |
| "grad_norm": 0.5827314257621765, | |
| "learning_rate": 1.3353846153846155e-05, | |
| "loss": 0.0832, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 16.76923076923077, | |
| "grad_norm": 0.1550850123167038, | |
| "learning_rate": 1.3292307692307692e-05, | |
| "loss": 0.1211, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 16.923076923076923, | |
| "grad_norm": 0.8602961897850037, | |
| "learning_rate": 1.3230769230769231e-05, | |
| "loss": 0.1397, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.9924812030075187, | |
| "eval_loss": 0.049453094601631165, | |
| "eval_runtime": 0.6291, | |
| "eval_samples_per_second": 211.407, | |
| "eval_steps_per_second": 14.306, | |
| "step": 1105 | |
| }, | |
| { | |
| "epoch": 17.076923076923077, | |
| "grad_norm": 1.7802873849868774, | |
| "learning_rate": 1.316923076923077e-05, | |
| "loss": 0.0908, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 17.23076923076923, | |
| "grad_norm": 0.10594117641448975, | |
| "learning_rate": 1.3107692307692307e-05, | |
| "loss": 0.0698, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 17.384615384615383, | |
| "grad_norm": 0.21712522208690643, | |
| "learning_rate": 1.3046153846153846e-05, | |
| "loss": 0.1116, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 17.53846153846154, | |
| "grad_norm": 6.615525245666504, | |
| "learning_rate": 1.2984615384615386e-05, | |
| "loss": 0.0767, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 17.692307692307693, | |
| "grad_norm": 0.10564947128295898, | |
| "learning_rate": 1.2923076923076925e-05, | |
| "loss": 0.0522, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 17.846153846153847, | |
| "grad_norm": 1.2303924560546875, | |
| "learning_rate": 1.2861538461538462e-05, | |
| "loss": 0.0558, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 0.10486020892858505, | |
| "learning_rate": 1.2800000000000001e-05, | |
| "loss": 0.0456, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.9774436090225563, | |
| "eval_loss": 0.051864467561244965, | |
| "eval_runtime": 0.6367, | |
| "eval_samples_per_second": 208.899, | |
| "eval_steps_per_second": 14.136, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 18.153846153846153, | |
| "grad_norm": 0.09572620689868927, | |
| "learning_rate": 1.273846153846154e-05, | |
| "loss": 0.0446, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 18.307692307692307, | |
| "grad_norm": 0.12069143354892731, | |
| "learning_rate": 1.2676923076923077e-05, | |
| "loss": 0.0612, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 18.46153846153846, | |
| "grad_norm": 3.15175724029541, | |
| "learning_rate": 1.2615384615384616e-05, | |
| "loss": 0.1213, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 18.615384615384617, | |
| "grad_norm": 0.6527738571166992, | |
| "learning_rate": 1.2553846153846155e-05, | |
| "loss": 0.1027, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 18.76923076923077, | |
| "grad_norm": 0.6189332604408264, | |
| "learning_rate": 1.2492307692307692e-05, | |
| "loss": 0.0262, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 18.923076923076923, | |
| "grad_norm": 0.1209394633769989, | |
| "learning_rate": 1.243076923076923e-05, | |
| "loss": 0.0439, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.021639494225382805, | |
| "eval_runtime": 0.6265, | |
| "eval_samples_per_second": 212.296, | |
| "eval_steps_per_second": 14.366, | |
| "step": 1235 | |
| }, | |
| { | |
| "epoch": 19.076923076923077, | |
| "grad_norm": 6.580589771270752, | |
| "learning_rate": 1.2369230769230771e-05, | |
| "loss": 0.0469, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 19.23076923076923, | |
| "grad_norm": 8.734590530395508, | |
| "learning_rate": 1.230769230769231e-05, | |
| "loss": 0.0297, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 19.384615384615383, | |
| "grad_norm": 1.7616609334945679, | |
| "learning_rate": 1.2246153846153847e-05, | |
| "loss": 0.1147, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 19.53846153846154, | |
| "grad_norm": 0.3767222464084625, | |
| "learning_rate": 1.2184615384615386e-05, | |
| "loss": 0.0499, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 19.692307692307693, | |
| "grad_norm": 0.1071651503443718, | |
| "learning_rate": 1.2123076923076924e-05, | |
| "loss": 0.0623, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 19.846153846153847, | |
| "grad_norm": 4.712902545928955, | |
| "learning_rate": 1.2061538461538462e-05, | |
| "loss": 0.0365, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 0.08926769345998764, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.0484, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9924812030075187, | |
| "eval_loss": 0.03160810098052025, | |
| "eval_runtime": 0.6412, | |
| "eval_samples_per_second": 207.432, | |
| "eval_steps_per_second": 14.037, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 20.153846153846153, | |
| "grad_norm": 1.210523009300232, | |
| "learning_rate": 1.1938461538461539e-05, | |
| "loss": 0.0789, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 20.307692307692307, | |
| "grad_norm": 0.21540193259716034, | |
| "learning_rate": 1.1876923076923076e-05, | |
| "loss": 0.0214, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 20.46153846153846, | |
| "grad_norm": 0.08566620200872421, | |
| "learning_rate": 1.1815384615384617e-05, | |
| "loss": 0.0308, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 20.615384615384617, | |
| "grad_norm": 1.3403387069702148, | |
| "learning_rate": 1.1753846153846155e-05, | |
| "loss": 0.0656, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 20.76923076923077, | |
| "grad_norm": 1.3895039558410645, | |
| "learning_rate": 1.1692307692307694e-05, | |
| "loss": 0.0651, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 20.923076923076923, | |
| "grad_norm": 2.5756139755249023, | |
| "learning_rate": 1.1630769230769231e-05, | |
| "loss": 0.0276, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.019228629767894745, | |
| "eval_runtime": 0.6415, | |
| "eval_samples_per_second": 207.316, | |
| "eval_steps_per_second": 14.029, | |
| "step": 1365 | |
| }, | |
| { | |
| "epoch": 21.076923076923077, | |
| "grad_norm": 0.11552825570106506, | |
| "learning_rate": 1.156923076923077e-05, | |
| "loss": 0.0494, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 21.23076923076923, | |
| "grad_norm": 0.08301220834255219, | |
| "learning_rate": 1.1507692307692309e-05, | |
| "loss": 0.1079, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 21.384615384615383, | |
| "grad_norm": 0.08622205257415771, | |
| "learning_rate": 1.1446153846153846e-05, | |
| "loss": 0.0397, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 21.53846153846154, | |
| "grad_norm": 1.3071388006210327, | |
| "learning_rate": 1.1384615384615385e-05, | |
| "loss": 0.1094, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 21.692307692307693, | |
| "grad_norm": 17.1097469329834, | |
| "learning_rate": 1.1323076923076924e-05, | |
| "loss": 0.0298, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 21.846153846153847, | |
| "grad_norm": 0.08082272112369537, | |
| "learning_rate": 1.126153846153846e-05, | |
| "loss": 0.0196, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "grad_norm": 0.08441055566072464, | |
| "learning_rate": 1.1200000000000001e-05, | |
| "loss": 0.0348, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0177127867937088, | |
| "eval_runtime": 0.5805, | |
| "eval_samples_per_second": 229.097, | |
| "eval_steps_per_second": 15.503, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 22.153846153846153, | |
| "grad_norm": 0.6167936325073242, | |
| "learning_rate": 1.113846153846154e-05, | |
| "loss": 0.0675, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 22.307692307692307, | |
| "grad_norm": 6.9912638664245605, | |
| "learning_rate": 1.1076923076923079e-05, | |
| "loss": 0.0644, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 22.46153846153846, | |
| "grad_norm": 0.07652874290943146, | |
| "learning_rate": 1.1015384615384616e-05, | |
| "loss": 0.0572, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 22.615384615384617, | |
| "grad_norm": 0.08351747691631317, | |
| "learning_rate": 1.0953846153846155e-05, | |
| "loss": 0.051, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 22.76923076923077, | |
| "grad_norm": 0.11051066219806671, | |
| "learning_rate": 1.0892307692307693e-05, | |
| "loss": 0.032, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 22.923076923076923, | |
| "grad_norm": 10.532815933227539, | |
| "learning_rate": 1.083076923076923e-05, | |
| "loss": 0.0326, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.01754232682287693, | |
| "eval_runtime": 0.6431, | |
| "eval_samples_per_second": 206.82, | |
| "eval_steps_per_second": 13.995, | |
| "step": 1495 | |
| }, | |
| { | |
| "epoch": 23.076923076923077, | |
| "grad_norm": 0.08222197741270065, | |
| "learning_rate": 1.076923076923077e-05, | |
| "loss": 0.0462, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 23.23076923076923, | |
| "grad_norm": 0.0753277987241745, | |
| "learning_rate": 1.0707692307692308e-05, | |
| "loss": 0.0516, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 23.384615384615383, | |
| "grad_norm": 0.9446002840995789, | |
| "learning_rate": 1.0646153846153845e-05, | |
| "loss": 0.0214, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 23.53846153846154, | |
| "grad_norm": 0.0864008441567421, | |
| "learning_rate": 1.0584615384615386e-05, | |
| "loss": 0.0185, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 23.692307692307693, | |
| "grad_norm": 7.382317543029785, | |
| "learning_rate": 1.0523076923076924e-05, | |
| "loss": 0.0674, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 23.846153846153847, | |
| "grad_norm": 0.18953648209571838, | |
| "learning_rate": 1.0461538461538463e-05, | |
| "loss": 0.02, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "grad_norm": 2.4876315593719482, | |
| "learning_rate": 1.04e-05, | |
| "loss": 0.1014, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.9924812030075187, | |
| "eval_loss": 0.02354033850133419, | |
| "eval_runtime": 0.6442, | |
| "eval_samples_per_second": 206.452, | |
| "eval_steps_per_second": 13.97, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 24.153846153846153, | |
| "grad_norm": 0.08115736395120621, | |
| "learning_rate": 1.033846153846154e-05, | |
| "loss": 0.014, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 24.307692307692307, | |
| "grad_norm": 0.06984913349151611, | |
| "learning_rate": 1.0276923076923078e-05, | |
| "loss": 0.0328, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 24.46153846153846, | |
| "grad_norm": 7.471628665924072, | |
| "learning_rate": 1.0215384615384615e-05, | |
| "loss": 0.0321, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 24.615384615384617, | |
| "grad_norm": 0.0844321921467781, | |
| "learning_rate": 1.0153846153846154e-05, | |
| "loss": 0.051, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 24.76923076923077, | |
| "grad_norm": 0.0683414489030838, | |
| "learning_rate": 1.0092307692307693e-05, | |
| "loss": 0.0428, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 24.923076923076923, | |
| "grad_norm": 0.07370961457490921, | |
| "learning_rate": 1.0030769230769231e-05, | |
| "loss": 0.0395, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.9849624060150376, | |
| "eval_loss": 0.04511820524930954, | |
| "eval_runtime": 0.6443, | |
| "eval_samples_per_second": 206.419, | |
| "eval_steps_per_second": 13.968, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 25.076923076923077, | |
| "grad_norm": 0.7060135006904602, | |
| "learning_rate": 9.96923076923077e-06, | |
| "loss": 0.0197, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 25.23076923076923, | |
| "grad_norm": 0.5647442936897278, | |
| "learning_rate": 9.907692307692309e-06, | |
| "loss": 0.0636, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 25.384615384615383, | |
| "grad_norm": 0.06951487809419632, | |
| "learning_rate": 9.846153846153848e-06, | |
| "loss": 0.0338, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 25.53846153846154, | |
| "grad_norm": 0.9578651785850525, | |
| "learning_rate": 9.784615384615387e-06, | |
| "loss": 0.049, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 25.692307692307693, | |
| "grad_norm": 0.1774640828371048, | |
| "learning_rate": 9.723076923076924e-06, | |
| "loss": 0.0135, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 25.846153846153847, | |
| "grad_norm": 0.06652193516492844, | |
| "learning_rate": 9.661538461538462e-06, | |
| "loss": 0.046, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "grad_norm": 0.06518968194723129, | |
| "learning_rate": 9.600000000000001e-06, | |
| "loss": 0.0265, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.9924812030075187, | |
| "eval_loss": 0.0296646561473608, | |
| "eval_runtime": 0.5911, | |
| "eval_samples_per_second": 225.018, | |
| "eval_steps_per_second": 15.227, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 26.153846153846153, | |
| "grad_norm": 0.07430905103683472, | |
| "learning_rate": 9.53846153846154e-06, | |
| "loss": 0.0326, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 26.307692307692307, | |
| "grad_norm": 0.07437779009342194, | |
| "learning_rate": 9.476923076923079e-06, | |
| "loss": 0.0205, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 26.46153846153846, | |
| "grad_norm": 5.995608329772949, | |
| "learning_rate": 9.415384615384616e-06, | |
| "loss": 0.0725, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 26.615384615384617, | |
| "grad_norm": 0.09473489224910736, | |
| "learning_rate": 9.353846153846155e-06, | |
| "loss": 0.0142, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 26.76923076923077, | |
| "grad_norm": 0.05937571823596954, | |
| "learning_rate": 9.292307692307694e-06, | |
| "loss": 0.0212, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 26.923076923076923, | |
| "grad_norm": 19.040122985839844, | |
| "learning_rate": 9.230769230769232e-06, | |
| "loss": 0.0569, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 0.9924812030075187, | |
| "eval_loss": 0.026294343173503876, | |
| "eval_runtime": 0.642, | |
| "eval_samples_per_second": 207.155, | |
| "eval_steps_per_second": 14.018, | |
| "step": 1755 | |
| }, | |
| { | |
| "epoch": 27.076923076923077, | |
| "grad_norm": 0.5836367011070251, | |
| "learning_rate": 9.169230769230771e-06, | |
| "loss": 0.1035, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 27.23076923076923, | |
| "grad_norm": 0.06087055802345276, | |
| "learning_rate": 9.107692307692308e-06, | |
| "loss": 0.0518, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 27.384615384615383, | |
| "grad_norm": 0.06178651750087738, | |
| "learning_rate": 9.046153846153847e-06, | |
| "loss": 0.0477, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 27.53846153846154, | |
| "grad_norm": 0.0936984047293663, | |
| "learning_rate": 8.984615384615386e-06, | |
| "loss": 0.012, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 27.692307692307693, | |
| "grad_norm": 0.056670334190130234, | |
| "learning_rate": 8.923076923076925e-06, | |
| "loss": 0.0363, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 27.846153846153847, | |
| "grad_norm": 0.13299456238746643, | |
| "learning_rate": 8.861538461538463e-06, | |
| "loss": 0.013, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "grad_norm": 0.0713280737400055, | |
| "learning_rate": 8.8e-06, | |
| "loss": 0.0666, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.9849624060150376, | |
| "eval_loss": 0.02451479434967041, | |
| "eval_runtime": 0.6311, | |
| "eval_samples_per_second": 210.727, | |
| "eval_steps_per_second": 14.26, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 28.153846153846153, | |
| "grad_norm": 0.08556320518255234, | |
| "learning_rate": 8.73846153846154e-06, | |
| "loss": 0.0119, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 28.307692307692307, | |
| "grad_norm": 0.05891846865415573, | |
| "learning_rate": 8.676923076923078e-06, | |
| "loss": 0.0456, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 28.46153846153846, | |
| "grad_norm": 0.05940578132867813, | |
| "learning_rate": 8.615384615384617e-06, | |
| "loss": 0.0541, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 28.615384615384617, | |
| "grad_norm": 0.06775704026222229, | |
| "learning_rate": 8.553846153846156e-06, | |
| "loss": 0.0162, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 28.76923076923077, | |
| "grad_norm": 0.058256104588508606, | |
| "learning_rate": 8.492307692307693e-06, | |
| "loss": 0.0354, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 28.923076923076923, | |
| "grad_norm": 1.286210060119629, | |
| "learning_rate": 8.430769230769231e-06, | |
| "loss": 0.0285, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 0.9774436090225563, | |
| "eval_loss": 0.041793130338191986, | |
| "eval_runtime": 0.6391, | |
| "eval_samples_per_second": 208.111, | |
| "eval_steps_per_second": 14.083, | |
| "step": 1885 | |
| }, | |
| { | |
| "epoch": 29.076923076923077, | |
| "grad_norm": 2.135648727416992, | |
| "learning_rate": 8.36923076923077e-06, | |
| "loss": 0.0197, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 29.23076923076923, | |
| "grad_norm": 1.5485310554504395, | |
| "learning_rate": 8.307692307692309e-06, | |
| "loss": 0.0129, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 29.384615384615383, | |
| "grad_norm": 1.2594960927963257, | |
| "learning_rate": 8.246153846153848e-06, | |
| "loss": 0.0964, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 29.53846153846154, | |
| "grad_norm": 0.13048137724399567, | |
| "learning_rate": 8.184615384615385e-06, | |
| "loss": 0.0111, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 29.692307692307693, | |
| "grad_norm": 0.38255247473716736, | |
| "learning_rate": 8.123076923076924e-06, | |
| "loss": 0.0292, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 29.846153846153847, | |
| "grad_norm": 2.1401822566986084, | |
| "learning_rate": 8.061538461538463e-06, | |
| "loss": 0.0417, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 0.2889564633369446, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.0892, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.9924812030075187, | |
| "eval_loss": 0.020448315888643265, | |
| "eval_runtime": 0.5776, | |
| "eval_samples_per_second": 230.273, | |
| "eval_steps_per_second": 15.582, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 30.153846153846153, | |
| "grad_norm": 3.4330976009368896, | |
| "learning_rate": 7.93846153846154e-06, | |
| "loss": 0.0466, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 30.307692307692307, | |
| "grad_norm": 0.05678678676486015, | |
| "learning_rate": 7.876923076923077e-06, | |
| "loss": 0.0701, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 30.46153846153846, | |
| "grad_norm": 0.05310628563165665, | |
| "learning_rate": 7.815384615384616e-06, | |
| "loss": 0.0118, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 30.615384615384617, | |
| "grad_norm": 0.07134439796209335, | |
| "learning_rate": 7.753846153846155e-06, | |
| "loss": 0.0412, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 30.76923076923077, | |
| "grad_norm": 0.06206020340323448, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 0.0254, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 30.923076923076923, | |
| "grad_norm": 0.0549427792429924, | |
| "learning_rate": 7.630769230769232e-06, | |
| "loss": 0.0371, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_accuracy": 0.9849624060150376, | |
| "eval_loss": 0.03390338271856308, | |
| "eval_runtime": 0.6391, | |
| "eval_samples_per_second": 208.12, | |
| "eval_steps_per_second": 14.083, | |
| "step": 2015 | |
| }, | |
| { | |
| "epoch": 31.076923076923077, | |
| "grad_norm": 0.05257405340671539, | |
| "learning_rate": 7.5692307692307695e-06, | |
| "loss": 0.0119, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 31.23076923076923, | |
| "grad_norm": 1.0072262287139893, | |
| "learning_rate": 7.507692307692308e-06, | |
| "loss": 0.0131, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 31.384615384615383, | |
| "grad_norm": 0.054051704704761505, | |
| "learning_rate": 7.446153846153846e-06, | |
| "loss": 0.0309, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 31.53846153846154, | |
| "grad_norm": 0.059842586517333984, | |
| "learning_rate": 7.384615384615386e-06, | |
| "loss": 0.0699, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 31.692307692307693, | |
| "grad_norm": 0.4310505986213684, | |
| "learning_rate": 7.323076923076924e-06, | |
| "loss": 0.0105, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 31.846153846153847, | |
| "grad_norm": 0.05525004491209984, | |
| "learning_rate": 7.261538461538462e-06, | |
| "loss": 0.0654, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "grad_norm": 0.06051107123494148, | |
| "learning_rate": 7.2000000000000005e-06, | |
| "loss": 0.0105, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.01434730738401413, | |
| "eval_runtime": 0.5855, | |
| "eval_samples_per_second": 227.144, | |
| "eval_steps_per_second": 15.371, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 32.15384615384615, | |
| "grad_norm": 0.04868883639574051, | |
| "learning_rate": 7.1384615384615385e-06, | |
| "loss": 0.032, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 32.30769230769231, | |
| "grad_norm": 0.10859151929616928, | |
| "learning_rate": 7.076923076923078e-06, | |
| "loss": 0.084, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 32.46153846153846, | |
| "grad_norm": 0.05709298700094223, | |
| "learning_rate": 7.015384615384616e-06, | |
| "loss": 0.0189, | |
| "step": 2110 | |
| }, | |
| { | |
| "epoch": 32.61538461538461, | |
| "grad_norm": 0.08583523333072662, | |
| "learning_rate": 6.953846153846154e-06, | |
| "loss": 0.0124, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 32.76923076923077, | |
| "grad_norm": 0.7491576671600342, | |
| "learning_rate": 6.892307692307693e-06, | |
| "loss": 0.0213, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 32.92307692307692, | |
| "grad_norm": 0.3305934965610504, | |
| "learning_rate": 6.830769230769231e-06, | |
| "loss": 0.0563, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.014035705476999283, | |
| "eval_runtime": 0.6445, | |
| "eval_samples_per_second": 206.373, | |
| "eval_steps_per_second": 13.965, | |
| "step": 2145 | |
| }, | |
| { | |
| "epoch": 33.07692307692308, | |
| "grad_norm": 0.049748744815588, | |
| "learning_rate": 6.76923076923077e-06, | |
| "loss": 0.0104, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 33.23076923076923, | |
| "grad_norm": 0.05033630132675171, | |
| "learning_rate": 6.707692307692308e-06, | |
| "loss": 0.0548, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 33.38461538461539, | |
| "grad_norm": 0.054612692445516586, | |
| "learning_rate": 6.646153846153846e-06, | |
| "loss": 0.0356, | |
| "step": 2170 | |
| }, | |
| { | |
| "epoch": 33.53846153846154, | |
| "grad_norm": 0.05341866612434387, | |
| "learning_rate": 6.584615384615385e-06, | |
| "loss": 0.0133, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 33.69230769230769, | |
| "grad_norm": 0.04895515367388725, | |
| "learning_rate": 6.523076923076923e-06, | |
| "loss": 0.0213, | |
| "step": 2190 | |
| }, | |
| { | |
| "epoch": 33.84615384615385, | |
| "grad_norm": 0.063107430934906, | |
| "learning_rate": 6.461538461538463e-06, | |
| "loss": 0.0112, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "grad_norm": 7.929271221160889, | |
| "learning_rate": 6.4000000000000006e-06, | |
| "loss": 0.0573, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.010156131349503994, | |
| "eval_runtime": 0.6308, | |
| "eval_samples_per_second": 210.855, | |
| "eval_steps_per_second": 14.268, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 34.15384615384615, | |
| "grad_norm": 0.11311448365449905, | |
| "learning_rate": 6.3384615384615385e-06, | |
| "loss": 0.0271, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 34.30769230769231, | |
| "grad_norm": 0.17935284972190857, | |
| "learning_rate": 6.276923076923077e-06, | |
| "loss": 0.0444, | |
| "step": 2230 | |
| }, | |
| { | |
| "epoch": 34.46153846153846, | |
| "grad_norm": 0.07903819531202316, | |
| "learning_rate": 6.215384615384615e-06, | |
| "loss": 0.039, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 34.61538461538461, | |
| "grad_norm": 0.07042822986841202, | |
| "learning_rate": 6.153846153846155e-06, | |
| "loss": 0.0617, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 34.76923076923077, | |
| "grad_norm": 0.05035420507192612, | |
| "learning_rate": 6.092307692307693e-06, | |
| "loss": 0.0505, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 34.92307692307692, | |
| "grad_norm": 0.04776820167899132, | |
| "learning_rate": 6.030769230769231e-06, | |
| "loss": 0.0409, | |
| "step": 2270 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.009572061710059643, | |
| "eval_runtime": 0.6399, | |
| "eval_samples_per_second": 207.852, | |
| "eval_steps_per_second": 14.065, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 35.07692307692308, | |
| "grad_norm": 20.547210693359375, | |
| "learning_rate": 5.9692307692307695e-06, | |
| "loss": 0.0212, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 35.23076923076923, | |
| "grad_norm": 16.86725616455078, | |
| "learning_rate": 5.907692307692308e-06, | |
| "loss": 0.0421, | |
| "step": 2290 | |
| }, | |
| { | |
| "epoch": 35.38461538461539, | |
| "grad_norm": 1.5179917812347412, | |
| "learning_rate": 5.846153846153847e-06, | |
| "loss": 0.1226, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 35.53846153846154, | |
| "grad_norm": 0.045909151434898376, | |
| "learning_rate": 5.784615384615385e-06, | |
| "loss": 0.0092, | |
| "step": 2310 | |
| }, | |
| { | |
| "epoch": 35.69230769230769, | |
| "grad_norm": 0.04946780204772949, | |
| "learning_rate": 5.723076923076923e-06, | |
| "loss": 0.0505, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 35.84615384615385, | |
| "grad_norm": 0.055896684527397156, | |
| "learning_rate": 5.661538461538462e-06, | |
| "loss": 0.0159, | |
| "step": 2330 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "grad_norm": 0.04570392891764641, | |
| "learning_rate": 5.600000000000001e-06, | |
| "loss": 0.0523, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.9924812030075187, | |
| "eval_loss": 0.01487450860440731, | |
| "eval_runtime": 0.6368, | |
| "eval_samples_per_second": 208.852, | |
| "eval_steps_per_second": 14.133, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 36.15384615384615, | |
| "grad_norm": 0.06980939954519272, | |
| "learning_rate": 5.538461538461539e-06, | |
| "loss": 0.0219, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 36.30769230769231, | |
| "grad_norm": 0.05014393478631973, | |
| "learning_rate": 5.476923076923077e-06, | |
| "loss": 0.0352, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 36.46153846153846, | |
| "grad_norm": 0.04635272175073624, | |
| "learning_rate": 5.415384615384615e-06, | |
| "loss": 0.0328, | |
| "step": 2370 | |
| }, | |
| { | |
| "epoch": 36.61538461538461, | |
| "grad_norm": 0.04787183925509453, | |
| "learning_rate": 5.353846153846154e-06, | |
| "loss": 0.0106, | |
| "step": 2380 | |
| }, | |
| { | |
| "epoch": 36.76923076923077, | |
| "grad_norm": 0.05444607138633728, | |
| "learning_rate": 5.292307692307693e-06, | |
| "loss": 0.0443, | |
| "step": 2390 | |
| }, | |
| { | |
| "epoch": 36.92307692307692, | |
| "grad_norm": 0.046256501227617264, | |
| "learning_rate": 5.230769230769232e-06, | |
| "loss": 0.0131, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_accuracy": 0.9924812030075187, | |
| "eval_loss": 0.0196556244045496, | |
| "eval_runtime": 0.641, | |
| "eval_samples_per_second": 207.498, | |
| "eval_steps_per_second": 14.041, | |
| "step": 2405 | |
| }, | |
| { | |
| "epoch": 37.07692307692308, | |
| "grad_norm": 0.045623164623975754, | |
| "learning_rate": 5.16923076923077e-06, | |
| "loss": 0.0112, | |
| "step": 2410 | |
| }, | |
| { | |
| "epoch": 37.23076923076923, | |
| "grad_norm": 0.04630188271403313, | |
| "learning_rate": 5.1076923076923075e-06, | |
| "loss": 0.0129, | |
| "step": 2420 | |
| }, | |
| { | |
| "epoch": 37.38461538461539, | |
| "grad_norm": 1.0303974151611328, | |
| "learning_rate": 5.046153846153846e-06, | |
| "loss": 0.0523, | |
| "step": 2430 | |
| }, | |
| { | |
| "epoch": 37.53846153846154, | |
| "grad_norm": 0.8218058347702026, | |
| "learning_rate": 4.984615384615385e-06, | |
| "loss": 0.0532, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 37.69230769230769, | |
| "grad_norm": 1.5203208923339844, | |
| "learning_rate": 4.923076923076924e-06, | |
| "loss": 0.0345, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 37.84615384615385, | |
| "grad_norm": 0.05802327021956444, | |
| "learning_rate": 4.861538461538462e-06, | |
| "loss": 0.0251, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "grad_norm": 0.06536999344825745, | |
| "learning_rate": 4.800000000000001e-06, | |
| "loss": 0.0329, | |
| "step": 2470 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.010934116318821907, | |
| "eval_runtime": 0.6397, | |
| "eval_samples_per_second": 207.921, | |
| "eval_steps_per_second": 14.07, | |
| "step": 2470 | |
| }, | |
| { | |
| "epoch": 38.15384615384615, | |
| "grad_norm": 0.07708264887332916, | |
| "learning_rate": 4.738461538461539e-06, | |
| "loss": 0.0339, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 38.30769230769231, | |
| "grad_norm": 0.05018337070941925, | |
| "learning_rate": 4.676923076923077e-06, | |
| "loss": 0.0371, | |
| "step": 2490 | |
| }, | |
| { | |
| "epoch": 38.46153846153846, | |
| "grad_norm": 2.005122423171997, | |
| "learning_rate": 4.615384615384616e-06, | |
| "loss": 0.0493, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 38.61538461538461, | |
| "grad_norm": 0.04191539064049721, | |
| "learning_rate": 4.553846153846154e-06, | |
| "loss": 0.056, | |
| "step": 2510 | |
| }, | |
| { | |
| "epoch": 38.76923076923077, | |
| "grad_norm": 0.08912540227174759, | |
| "learning_rate": 4.492307692307693e-06, | |
| "loss": 0.0675, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 38.92307692307692, | |
| "grad_norm": 4.123304843902588, | |
| "learning_rate": 4.430769230769232e-06, | |
| "loss": 0.0577, | |
| "step": 2530 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.00963473692536354, | |
| "eval_runtime": 0.6269, | |
| "eval_samples_per_second": 212.155, | |
| "eval_steps_per_second": 14.356, | |
| "step": 2535 | |
| }, | |
| { | |
| "epoch": 39.07692307692308, | |
| "grad_norm": 0.12956681847572327, | |
| "learning_rate": 4.36923076923077e-06, | |
| "loss": 0.0348, | |
| "step": 2540 | |
| }, | |
| { | |
| "epoch": 39.23076923076923, | |
| "grad_norm": 0.0469551756978035, | |
| "learning_rate": 4.307692307692308e-06, | |
| "loss": 0.047, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 39.38461538461539, | |
| "grad_norm": 0.0566897876560688, | |
| "learning_rate": 4.246153846153846e-06, | |
| "loss": 0.0305, | |
| "step": 2560 | |
| }, | |
| { | |
| "epoch": 39.53846153846154, | |
| "grad_norm": 0.04538924619555473, | |
| "learning_rate": 4.184615384615385e-06, | |
| "loss": 0.0083, | |
| "step": 2570 | |
| }, | |
| { | |
| "epoch": 39.69230769230769, | |
| "grad_norm": 0.1393657773733139, | |
| "learning_rate": 4.123076923076924e-06, | |
| "loss": 0.0087, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 39.84615384615385, | |
| "grad_norm": 0.04170211777091026, | |
| "learning_rate": 4.061538461538462e-06, | |
| "loss": 0.008, | |
| "step": 2590 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 0.04205217584967613, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.0085, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.9924812030075187, | |
| "eval_loss": 0.014666187576949596, | |
| "eval_runtime": 0.5786, | |
| "eval_samples_per_second": 229.849, | |
| "eval_steps_per_second": 15.554, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 40.15384615384615, | |
| "grad_norm": 0.04466895014047623, | |
| "learning_rate": 3.938461538461539e-06, | |
| "loss": 0.0376, | |
| "step": 2610 | |
| }, | |
| { | |
| "epoch": 40.30769230769231, | |
| "grad_norm": 0.04949569329619408, | |
| "learning_rate": 3.876923076923077e-06, | |
| "loss": 0.0621, | |
| "step": 2620 | |
| }, | |
| { | |
| "epoch": 40.46153846153846, | |
| "grad_norm": 0.0461997464299202, | |
| "learning_rate": 3.815384615384616e-06, | |
| "loss": 0.0107, | |
| "step": 2630 | |
| }, | |
| { | |
| "epoch": 40.61538461538461, | |
| "grad_norm": 0.048004575073719025, | |
| "learning_rate": 3.753846153846154e-06, | |
| "loss": 0.0093, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 40.76923076923077, | |
| "grad_norm": 0.04209740087389946, | |
| "learning_rate": 3.692307692307693e-06, | |
| "loss": 0.0342, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 40.92307692307692, | |
| "grad_norm": 0.15323477983474731, | |
| "learning_rate": 3.630769230769231e-06, | |
| "loss": 0.0618, | |
| "step": 2660 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.009433195926249027, | |
| "eval_runtime": 0.6376, | |
| "eval_samples_per_second": 208.608, | |
| "eval_steps_per_second": 14.116, | |
| "step": 2665 | |
| }, | |
| { | |
| "epoch": 41.07692307692308, | |
| "grad_norm": 0.040465518832206726, | |
| "learning_rate": 3.5692307692307692e-06, | |
| "loss": 0.0079, | |
| "step": 2670 | |
| }, | |
| { | |
| "epoch": 41.23076923076923, | |
| "grad_norm": 0.06956275552511215, | |
| "learning_rate": 3.507692307692308e-06, | |
| "loss": 0.0278, | |
| "step": 2680 | |
| }, | |
| { | |
| "epoch": 41.38461538461539, | |
| "grad_norm": 0.04409582540392876, | |
| "learning_rate": 3.4461538461538464e-06, | |
| "loss": 0.0079, | |
| "step": 2690 | |
| }, | |
| { | |
| "epoch": 41.53846153846154, | |
| "grad_norm": 13.665828704833984, | |
| "learning_rate": 3.384615384615385e-06, | |
| "loss": 0.0187, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 41.69230769230769, | |
| "grad_norm": 1.3187448978424072, | |
| "learning_rate": 3.323076923076923e-06, | |
| "loss": 0.1204, | |
| "step": 2710 | |
| }, | |
| { | |
| "epoch": 41.84615384615385, | |
| "grad_norm": 0.04300126060843468, | |
| "learning_rate": 3.2615384615384615e-06, | |
| "loss": 0.0198, | |
| "step": 2720 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "grad_norm": 0.0438438281416893, | |
| "learning_rate": 3.2000000000000003e-06, | |
| "loss": 0.0847, | |
| "step": 2730 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_accuracy": 0.9924812030075187, | |
| "eval_loss": 0.019689705222845078, | |
| "eval_runtime": 0.634, | |
| "eval_samples_per_second": 209.767, | |
| "eval_steps_per_second": 14.195, | |
| "step": 2730 | |
| }, | |
| { | |
| "epoch": 42.15384615384615, | |
| "grad_norm": 0.055344920605421066, | |
| "learning_rate": 3.1384615384615386e-06, | |
| "loss": 0.0343, | |
| "step": 2740 | |
| }, | |
| { | |
| "epoch": 42.30769230769231, | |
| "grad_norm": 0.04840540513396263, | |
| "learning_rate": 3.0769230769230774e-06, | |
| "loss": 0.0091, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 42.46153846153846, | |
| "grad_norm": 0.0416325181722641, | |
| "learning_rate": 3.0153846153846154e-06, | |
| "loss": 0.0379, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 42.61538461538461, | |
| "grad_norm": 0.04047630727291107, | |
| "learning_rate": 2.953846153846154e-06, | |
| "loss": 0.0344, | |
| "step": 2770 | |
| }, | |
| { | |
| "epoch": 42.76923076923077, | |
| "grad_norm": 0.039694271981716156, | |
| "learning_rate": 2.8923076923076925e-06, | |
| "loss": 0.0556, | |
| "step": 2780 | |
| }, | |
| { | |
| "epoch": 42.92307692307692, | |
| "grad_norm": 0.0425509512424469, | |
| "learning_rate": 2.830769230769231e-06, | |
| "loss": 0.0291, | |
| "step": 2790 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.008893251419067383, | |
| "eval_runtime": 0.6271, | |
| "eval_samples_per_second": 212.079, | |
| "eval_steps_per_second": 14.351, | |
| "step": 2795 | |
| }, | |
| { | |
| "epoch": 43.07692307692308, | |
| "grad_norm": 0.04988468438386917, | |
| "learning_rate": 2.7692307692307697e-06, | |
| "loss": 0.0369, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 43.23076923076923, | |
| "grad_norm": 0.07137361913919449, | |
| "learning_rate": 2.7076923076923076e-06, | |
| "loss": 0.0434, | |
| "step": 2810 | |
| }, | |
| { | |
| "epoch": 43.38461538461539, | |
| "grad_norm": 7.0051679611206055, | |
| "learning_rate": 2.6461538461538464e-06, | |
| "loss": 0.0291, | |
| "step": 2820 | |
| }, | |
| { | |
| "epoch": 43.53846153846154, | |
| "grad_norm": 0.045469850301742554, | |
| "learning_rate": 2.584615384615385e-06, | |
| "loss": 0.0338, | |
| "step": 2830 | |
| }, | |
| { | |
| "epoch": 43.69230769230769, | |
| "grad_norm": 0.08003593236207962, | |
| "learning_rate": 2.523076923076923e-06, | |
| "loss": 0.0099, | |
| "step": 2840 | |
| }, | |
| { | |
| "epoch": 43.84615384615385, | |
| "grad_norm": 0.04380409047007561, | |
| "learning_rate": 2.461538461538462e-06, | |
| "loss": 0.0111, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "grad_norm": 13.31029224395752, | |
| "learning_rate": 2.4000000000000003e-06, | |
| "loss": 0.0568, | |
| "step": 2860 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.008692615665495396, | |
| "eval_runtime": 0.585, | |
| "eval_samples_per_second": 227.347, | |
| "eval_steps_per_second": 15.384, | |
| "step": 2860 | |
| }, | |
| { | |
| "epoch": 44.15384615384615, | |
| "grad_norm": 0.04654600843787193, | |
| "learning_rate": 2.3384615384615387e-06, | |
| "loss": 0.0087, | |
| "step": 2870 | |
| }, | |
| { | |
| "epoch": 44.30769230769231, | |
| "grad_norm": 7.452500343322754, | |
| "learning_rate": 2.276923076923077e-06, | |
| "loss": 0.0108, | |
| "step": 2880 | |
| }, | |
| { | |
| "epoch": 44.46153846153846, | |
| "grad_norm": 13.458589553833008, | |
| "learning_rate": 2.215384615384616e-06, | |
| "loss": 0.0274, | |
| "step": 2890 | |
| }, | |
| { | |
| "epoch": 44.61538461538461, | |
| "grad_norm": 0.044014327228069305, | |
| "learning_rate": 2.153846153846154e-06, | |
| "loss": 0.0299, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 44.76923076923077, | |
| "grad_norm": 0.041860181838274, | |
| "learning_rate": 2.0923076923076926e-06, | |
| "loss": 0.0112, | |
| "step": 2910 | |
| }, | |
| { | |
| "epoch": 44.92307692307692, | |
| "grad_norm": 0.04078350216150284, | |
| "learning_rate": 2.030769230769231e-06, | |
| "loss": 0.0077, | |
| "step": 2920 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.010402214713394642, | |
| "eval_runtime": 0.6383, | |
| "eval_samples_per_second": 208.376, | |
| "eval_steps_per_second": 14.101, | |
| "step": 2925 | |
| }, | |
| { | |
| "epoch": 45.07692307692308, | |
| "grad_norm": 0.7966273427009583, | |
| "learning_rate": 1.9692307692307693e-06, | |
| "loss": 0.0432, | |
| "step": 2930 | |
| }, | |
| { | |
| "epoch": 45.23076923076923, | |
| "grad_norm": 0.04070662334561348, | |
| "learning_rate": 1.907692307692308e-06, | |
| "loss": 0.0549, | |
| "step": 2940 | |
| }, | |
| { | |
| "epoch": 45.38461538461539, | |
| "grad_norm": 0.042289331555366516, | |
| "learning_rate": 1.8461538461538465e-06, | |
| "loss": 0.0364, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 45.53846153846154, | |
| "grad_norm": 0.04655339941382408, | |
| "learning_rate": 1.7846153846153846e-06, | |
| "loss": 0.0114, | |
| "step": 2960 | |
| }, | |
| { | |
| "epoch": 45.69230769230769, | |
| "grad_norm": 0.04026507958769798, | |
| "learning_rate": 1.7230769230769232e-06, | |
| "loss": 0.0078, | |
| "step": 2970 | |
| }, | |
| { | |
| "epoch": 45.84615384615385, | |
| "grad_norm": 0.048073675483465195, | |
| "learning_rate": 1.6615384615384616e-06, | |
| "loss": 0.0155, | |
| "step": 2980 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "grad_norm": 0.1250167340040207, | |
| "learning_rate": 1.6000000000000001e-06, | |
| "loss": 0.008, | |
| "step": 2990 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.013788605108857155, | |
| "eval_runtime": 0.5799, | |
| "eval_samples_per_second": 229.355, | |
| "eval_steps_per_second": 15.52, | |
| "step": 2990 | |
| }, | |
| { | |
| "epoch": 46.15384615384615, | |
| "grad_norm": 0.3202461302280426, | |
| "learning_rate": 1.5384615384615387e-06, | |
| "loss": 0.0352, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 46.30769230769231, | |
| "grad_norm": 2.8946588039398193, | |
| "learning_rate": 1.476923076923077e-06, | |
| "loss": 0.0196, | |
| "step": 3010 | |
| }, | |
| { | |
| "epoch": 46.46153846153846, | |
| "grad_norm": 0.0777769535779953, | |
| "learning_rate": 1.4153846153846155e-06, | |
| "loss": 0.0079, | |
| "step": 3020 | |
| }, | |
| { | |
| "epoch": 46.61538461538461, | |
| "grad_norm": 1.6607468128204346, | |
| "learning_rate": 1.3538461538461538e-06, | |
| "loss": 0.0102, | |
| "step": 3030 | |
| }, | |
| { | |
| "epoch": 46.76923076923077, | |
| "grad_norm": 0.04541005194187164, | |
| "learning_rate": 1.2923076923076924e-06, | |
| "loss": 0.0085, | |
| "step": 3040 | |
| }, | |
| { | |
| "epoch": 46.92307692307692, | |
| "grad_norm": 0.041475191712379456, | |
| "learning_rate": 1.230769230769231e-06, | |
| "loss": 0.0272, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.00810349639505148, | |
| "eval_runtime": 0.6353, | |
| "eval_samples_per_second": 209.365, | |
| "eval_steps_per_second": 14.168, | |
| "step": 3055 | |
| }, | |
| { | |
| "epoch": 47.07692307692308, | |
| "grad_norm": 0.048281800001859665, | |
| "learning_rate": 1.1692307692307693e-06, | |
| "loss": 0.008, | |
| "step": 3060 | |
| }, | |
| { | |
| "epoch": 47.23076923076923, | |
| "grad_norm": 0.03975387290120125, | |
| "learning_rate": 1.107692307692308e-06, | |
| "loss": 0.0431, | |
| "step": 3070 | |
| }, | |
| { | |
| "epoch": 47.38461538461539, | |
| "grad_norm": 0.040405042469501495, | |
| "learning_rate": 1.0461538461538463e-06, | |
| "loss": 0.0355, | |
| "step": 3080 | |
| }, | |
| { | |
| "epoch": 47.53846153846154, | |
| "grad_norm": 0.04081344977021217, | |
| "learning_rate": 9.846153846153847e-07, | |
| "loss": 0.0078, | |
| "step": 3090 | |
| }, | |
| { | |
| "epoch": 47.69230769230769, | |
| "grad_norm": 0.045139458030462265, | |
| "learning_rate": 9.230769230769232e-07, | |
| "loss": 0.0096, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 47.84615384615385, | |
| "grad_norm": 0.06699339300394058, | |
| "learning_rate": 8.615384615384616e-07, | |
| "loss": 0.0402, | |
| "step": 3110 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "grad_norm": 0.04077847674489021, | |
| "learning_rate": 8.000000000000001e-07, | |
| "loss": 0.008, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.008442863821983337, | |
| "eval_runtime": 0.6316, | |
| "eval_samples_per_second": 210.576, | |
| "eval_steps_per_second": 14.25, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 48.15384615384615, | |
| "grad_norm": 0.14330124855041504, | |
| "learning_rate": 7.384615384615385e-07, | |
| "loss": 0.0396, | |
| "step": 3130 | |
| }, | |
| { | |
| "epoch": 48.30769230769231, | |
| "grad_norm": 0.04051917791366577, | |
| "learning_rate": 6.769230769230769e-07, | |
| "loss": 0.0613, | |
| "step": 3140 | |
| }, | |
| { | |
| "epoch": 48.46153846153846, | |
| "grad_norm": 0.03945121914148331, | |
| "learning_rate": 6.153846153846155e-07, | |
| "loss": 0.0092, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 48.61538461538461, | |
| "grad_norm": 0.04850227013230324, | |
| "learning_rate": 5.53846153846154e-07, | |
| "loss": 0.0395, | |
| "step": 3160 | |
| }, | |
| { | |
| "epoch": 48.76923076923077, | |
| "grad_norm": 0.611132800579071, | |
| "learning_rate": 4.923076923076923e-07, | |
| "loss": 0.015, | |
| "step": 3170 | |
| }, | |
| { | |
| "epoch": 48.92307692307692, | |
| "grad_norm": 0.039736129343509674, | |
| "learning_rate": 4.307692307692308e-07, | |
| "loss": 0.0112, | |
| "step": 3180 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.008192874491214752, | |
| "eval_runtime": 0.594, | |
| "eval_samples_per_second": 223.907, | |
| "eval_steps_per_second": 15.152, | |
| "step": 3185 | |
| }, | |
| { | |
| "epoch": 49.07692307692308, | |
| "grad_norm": 0.0747324600815773, | |
| "learning_rate": 3.6923076923076927e-07, | |
| "loss": 0.018, | |
| "step": 3190 | |
| }, | |
| { | |
| "epoch": 49.23076923076923, | |
| "grad_norm": 0.03965931013226509, | |
| "learning_rate": 3.0769230769230774e-07, | |
| "loss": 0.0948, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 49.38461538461539, | |
| "grad_norm": 3.6046483516693115, | |
| "learning_rate": 2.4615384615384616e-07, | |
| "loss": 0.0302, | |
| "step": 3210 | |
| }, | |
| { | |
| "epoch": 49.53846153846154, | |
| "grad_norm": 0.03949074074625969, | |
| "learning_rate": 1.8461538461538464e-07, | |
| "loss": 0.0094, | |
| "step": 3220 | |
| }, | |
| { | |
| "epoch": 49.69230769230769, | |
| "grad_norm": 0.03980456292629242, | |
| "learning_rate": 1.2307692307692308e-07, | |
| "loss": 0.0388, | |
| "step": 3230 | |
| }, | |
| { | |
| "epoch": 49.84615384615385, | |
| "grad_norm": 8.264440536499023, | |
| "learning_rate": 6.153846153846154e-08, | |
| "loss": 0.0143, | |
| "step": 3240 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "grad_norm": 0.043733663856983185, | |
| "learning_rate": 0.0, | |
| "loss": 0.013, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.007947824895381927, | |
| "eval_runtime": 0.6285, | |
| "eval_samples_per_second": 211.608, | |
| "eval_steps_per_second": 14.319, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "step": 3250, | |
| "total_flos": 3.6243328994998477e+18, | |
| "train_loss": 0.0, | |
| "train_runtime": 0.0791, | |
| "train_samples_per_second": 65335.775, | |
| "train_steps_per_second": 4107.181 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 325, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.6243328994998477e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |