| { | |
| "best_metric": 0.5957671957671957, | |
| "best_model_checkpoint": "vivit-b-16x2-kinetics400-ft-3620\\checkpoint-2442", | |
| "epoch": 49.01109090909091, | |
| "eval_steps": 500, | |
| "global_step": 5500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0018181818181818182, | |
| "grad_norm": 3.741217613220215, | |
| "learning_rate": 9.09090909090909e-07, | |
| "loss": 1.1478, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0036363636363636364, | |
| "grad_norm": 10.490788459777832, | |
| "learning_rate": 1.818181818181818e-06, | |
| "loss": 1.1191, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.005454545454545455, | |
| "grad_norm": 5.955273151397705, | |
| "learning_rate": 2.7272727272727272e-06, | |
| "loss": 1.1088, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.007272727272727273, | |
| "grad_norm": 7.072606086730957, | |
| "learning_rate": 3.636363636363636e-06, | |
| "loss": 1.1015, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.00909090909090909, | |
| "grad_norm": 8.753866195678711, | |
| "learning_rate": 4.5454545454545455e-06, | |
| "loss": 1.0891, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.01090909090909091, | |
| "grad_norm": 4.427119255065918, | |
| "learning_rate": 5.4545454545454545e-06, | |
| "loss": 1.1044, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.012727272727272728, | |
| "grad_norm": 14.78829288482666, | |
| "learning_rate": 6.363636363636363e-06, | |
| "loss": 1.1065, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.014545454545454545, | |
| "grad_norm": 3.910771131515503, | |
| "learning_rate": 7.272727272727272e-06, | |
| "loss": 1.0878, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.016363636363636365, | |
| "grad_norm": 9.14891529083252, | |
| "learning_rate": 8.181818181818183e-06, | |
| "loss": 1.0748, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.01818181818181818, | |
| "grad_norm": 5.1411004066467285, | |
| "learning_rate": 9.090909090909091e-06, | |
| "loss": 1.1173, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 8.358501434326172, | |
| "learning_rate": 1e-05, | |
| "loss": 1.0684, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.020181818181818183, | |
| "eval_accuracy": 0.3798941798941799, | |
| "eval_loss": 1.1114321947097778, | |
| "eval_runtime": 131.7711, | |
| "eval_samples_per_second": 7.172, | |
| "eval_steps_per_second": 0.903, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 1.0016363636363637, | |
| "grad_norm": 4.467060565948486, | |
| "learning_rate": 1.0909090909090909e-05, | |
| "loss": 1.0851, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.0034545454545454, | |
| "grad_norm": 4.482585430145264, | |
| "learning_rate": 1.1818181818181819e-05, | |
| "loss": 1.1045, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.0052727272727273, | |
| "grad_norm": 10.72672176361084, | |
| "learning_rate": 1.2727272727272727e-05, | |
| "loss": 1.0377, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.007090909090909, | |
| "grad_norm": 5.500144958496094, | |
| "learning_rate": 1.3636363636363637e-05, | |
| "loss": 0.9757, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.008909090909091, | |
| "grad_norm": 10.576444625854492, | |
| "learning_rate": 1.4545454545454545e-05, | |
| "loss": 1.006, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.0107272727272727, | |
| "grad_norm": 11.98294734954834, | |
| "learning_rate": 1.5454545454545454e-05, | |
| "loss": 1.0666, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.0125454545454546, | |
| "grad_norm": 7.530726909637451, | |
| "learning_rate": 1.6363636363636366e-05, | |
| "loss": 1.1369, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.0143636363636364, | |
| "grad_norm": 7.2719221115112305, | |
| "learning_rate": 1.7272727272727274e-05, | |
| "loss": 1.0945, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.016181818181818, | |
| "grad_norm": 7.152905464172363, | |
| "learning_rate": 1.8181818181818182e-05, | |
| "loss": 0.9561, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.018, | |
| "grad_norm": 7.458053112030029, | |
| "learning_rate": 1.9090909090909094e-05, | |
| "loss": 1.1724, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.0198181818181817, | |
| "grad_norm": 8.600687980651855, | |
| "learning_rate": 2e-05, | |
| "loss": 1.0415, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.020181818181818, | |
| "eval_accuracy": 0.5248677248677248, | |
| "eval_loss": 1.0135376453399658, | |
| "eval_runtime": 134.43, | |
| "eval_samples_per_second": 7.03, | |
| "eval_steps_per_second": 0.885, | |
| "step": 222 | |
| }, | |
| { | |
| "epoch": 2.0014545454545454, | |
| "grad_norm": 4.647419452667236, | |
| "learning_rate": 2.090909090909091e-05, | |
| "loss": 0.9994, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.0032727272727273, | |
| "grad_norm": 3.433703660964966, | |
| "learning_rate": 2.1818181818181818e-05, | |
| "loss": 1.0031, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.0050909090909093, | |
| "grad_norm": 8.593055725097656, | |
| "learning_rate": 2.272727272727273e-05, | |
| "loss": 1.0471, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.0069090909090908, | |
| "grad_norm": 10.779848098754883, | |
| "learning_rate": 2.3636363636363637e-05, | |
| "loss": 1.015, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.0087272727272727, | |
| "grad_norm": 5.282721042633057, | |
| "learning_rate": 2.4545454545454545e-05, | |
| "loss": 1.1351, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.0105454545454546, | |
| "grad_norm": 7.395419597625732, | |
| "learning_rate": 2.5454545454545454e-05, | |
| "loss": 1.0316, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.0123636363636366, | |
| "grad_norm": 7.214972019195557, | |
| "learning_rate": 2.636363636363636e-05, | |
| "loss": 1.0889, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.014181818181818, | |
| "grad_norm": 8.676461219787598, | |
| "learning_rate": 2.7272727272727273e-05, | |
| "loss": 0.9877, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 15.418935775756836, | |
| "learning_rate": 2.818181818181818e-05, | |
| "loss": 0.9833, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.017818181818182, | |
| "grad_norm": 12.667856216430664, | |
| "learning_rate": 2.909090909090909e-05, | |
| "loss": 0.9863, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.0196363636363635, | |
| "grad_norm": 9.46337890625, | |
| "learning_rate": 3e-05, | |
| "loss": 1.0271, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.0201818181818183, | |
| "eval_accuracy": 0.4857142857142857, | |
| "eval_loss": 1.0630271434783936, | |
| "eval_runtime": 135.405, | |
| "eval_samples_per_second": 6.979, | |
| "eval_steps_per_second": 0.879, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 3.001272727272727, | |
| "grad_norm": 7.892024993896484, | |
| "learning_rate": 3.090909090909091e-05, | |
| "loss": 1.0633, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 3.003090909090909, | |
| "grad_norm": 6.610105037689209, | |
| "learning_rate": 3.181818181818182e-05, | |
| "loss": 0.9679, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 3.004909090909091, | |
| "grad_norm": 10.291769981384277, | |
| "learning_rate": 3.272727272727273e-05, | |
| "loss": 1.0005, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 3.006727272727273, | |
| "grad_norm": 15.271839141845703, | |
| "learning_rate": 3.3636363636363636e-05, | |
| "loss": 1.1185, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 3.0085454545454544, | |
| "grad_norm": 4.6768879890441895, | |
| "learning_rate": 3.454545454545455e-05, | |
| "loss": 0.9465, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.0103636363636364, | |
| "grad_norm": 16.15147590637207, | |
| "learning_rate": 3.545454545454546e-05, | |
| "loss": 1.0068, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.0121818181818183, | |
| "grad_norm": 14.014873504638672, | |
| "learning_rate": 3.6363636363636364e-05, | |
| "loss": 1.1818, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.014, | |
| "grad_norm": 7.938851356506348, | |
| "learning_rate": 3.7272727272727276e-05, | |
| "loss": 0.9357, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 3.0158181818181817, | |
| "grad_norm": 7.2831807136535645, | |
| "learning_rate": 3.818181818181819e-05, | |
| "loss": 1.0555, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.0176363636363637, | |
| "grad_norm": 14.499138832092285, | |
| "learning_rate": 3.909090909090909e-05, | |
| "loss": 1.0117, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.0194545454545456, | |
| "grad_norm": 5.862173080444336, | |
| "learning_rate": 4e-05, | |
| "loss": 1.1609, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.0201818181818183, | |
| "eval_accuracy": 0.4222222222222222, | |
| "eval_loss": 1.0202808380126953, | |
| "eval_runtime": 134.3297, | |
| "eval_samples_per_second": 7.035, | |
| "eval_steps_per_second": 0.886, | |
| "step": 444 | |
| }, | |
| { | |
| "epoch": 4.001090909090909, | |
| "grad_norm": 12.035215377807617, | |
| "learning_rate": 4.0909090909090915e-05, | |
| "loss": 1.065, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 4.002909090909091, | |
| "grad_norm": 14.393060684204102, | |
| "learning_rate": 4.181818181818182e-05, | |
| "loss": 1.1474, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 4.004727272727273, | |
| "grad_norm": 13.286993026733398, | |
| "learning_rate": 4.2727272727272724e-05, | |
| "loss": 1.1265, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 4.006545454545455, | |
| "grad_norm": 6.933895587921143, | |
| "learning_rate": 4.3636363636363636e-05, | |
| "loss": 1.0134, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 4.008363636363637, | |
| "grad_norm": 3.2165865898132324, | |
| "learning_rate": 4.454545454545455e-05, | |
| "loss": 1.0298, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 4.0101818181818185, | |
| "grad_norm": 14.604901313781738, | |
| "learning_rate": 4.545454545454546e-05, | |
| "loss": 0.9201, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.012, | |
| "grad_norm": 3.9145591259002686, | |
| "learning_rate": 4.636363636363636e-05, | |
| "loss": 1.1511, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 4.0138181818181815, | |
| "grad_norm": 10.365116119384766, | |
| "learning_rate": 4.7272727272727275e-05, | |
| "loss": 1.0987, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.0156363636363634, | |
| "grad_norm": 4.58921480178833, | |
| "learning_rate": 4.8181818181818186e-05, | |
| "loss": 1.0293, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 4.017454545454545, | |
| "grad_norm": 4.619180679321289, | |
| "learning_rate": 4.909090909090909e-05, | |
| "loss": 0.9006, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 4.019272727272727, | |
| "grad_norm": 10.984779357910156, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9824, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 4.020181818181818, | |
| "eval_accuracy": 0.5248677248677248, | |
| "eval_loss": 1.0219199657440186, | |
| "eval_runtime": 132.6767, | |
| "eval_samples_per_second": 7.123, | |
| "eval_steps_per_second": 0.897, | |
| "step": 555 | |
| }, | |
| { | |
| "epoch": 5.000909090909091, | |
| "grad_norm": 5.827091217041016, | |
| "learning_rate": 4.98989898989899e-05, | |
| "loss": 1.067, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 5.002727272727273, | |
| "grad_norm": 12.666735649108887, | |
| "learning_rate": 4.97979797979798e-05, | |
| "loss": 1.0333, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 5.004545454545455, | |
| "grad_norm": 10.529485702514648, | |
| "learning_rate": 4.9696969696969694e-05, | |
| "loss": 1.1087, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 5.006363636363637, | |
| "grad_norm": 5.182685852050781, | |
| "learning_rate": 4.9595959595959594e-05, | |
| "loss": 0.9791, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 5.008181818181818, | |
| "grad_norm": 9.57801628112793, | |
| "learning_rate": 4.94949494949495e-05, | |
| "loss": 0.9401, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "grad_norm": 9.004761695861816, | |
| "learning_rate": 4.93939393939394e-05, | |
| "loss": 0.9347, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.011818181818182, | |
| "grad_norm": 7.18081521987915, | |
| "learning_rate": 4.92929292929293e-05, | |
| "loss": 1.0563, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 5.013636363636364, | |
| "grad_norm": 4.530564785003662, | |
| "learning_rate": 4.919191919191919e-05, | |
| "loss": 1.1042, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 5.015454545454546, | |
| "grad_norm": 8.461360931396484, | |
| "learning_rate": 4.909090909090909e-05, | |
| "loss": 1.073, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 5.0172727272727276, | |
| "grad_norm": 3.762699842453003, | |
| "learning_rate": 4.898989898989899e-05, | |
| "loss": 1.0176, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.0190909090909095, | |
| "grad_norm": 5.622332572937012, | |
| "learning_rate": 4.888888888888889e-05, | |
| "loss": 1.0247, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 5.020181818181818, | |
| "eval_accuracy": 0.5026455026455027, | |
| "eval_loss": 1.020993947982788, | |
| "eval_runtime": 132.4215, | |
| "eval_samples_per_second": 7.136, | |
| "eval_steps_per_second": 0.899, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 6.000727272727273, | |
| "grad_norm": 10.156366348266602, | |
| "learning_rate": 4.878787878787879e-05, | |
| "loss": 0.9959, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 6.002545454545454, | |
| "grad_norm": 7.045960426330566, | |
| "learning_rate": 4.868686868686869e-05, | |
| "loss": 0.9444, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 6.004363636363636, | |
| "grad_norm": 16.915678024291992, | |
| "learning_rate": 4.858585858585859e-05, | |
| "loss": 0.9367, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 6.006181818181818, | |
| "grad_norm": 6.158443927764893, | |
| "learning_rate": 4.848484848484849e-05, | |
| "loss": 1.05, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 6.008, | |
| "grad_norm": 13.542768478393555, | |
| "learning_rate": 4.838383838383839e-05, | |
| "loss": 0.9787, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 6.009818181818182, | |
| "grad_norm": 5.779530048370361, | |
| "learning_rate": 4.828282828282829e-05, | |
| "loss": 0.9801, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 6.011636363636364, | |
| "grad_norm": 5.2666192054748535, | |
| "learning_rate": 4.8181818181818186e-05, | |
| "loss": 1.0266, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 6.013454545454546, | |
| "grad_norm": 5.806363105773926, | |
| "learning_rate": 4.808080808080808e-05, | |
| "loss": 0.9121, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 6.015272727272727, | |
| "grad_norm": 7.450347423553467, | |
| "learning_rate": 4.797979797979798e-05, | |
| "loss": 1.079, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 6.017090909090909, | |
| "grad_norm": 8.759308815002441, | |
| "learning_rate": 4.787878787878788e-05, | |
| "loss": 0.8933, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 6.018909090909091, | |
| "grad_norm": 7.260105133056641, | |
| "learning_rate": 4.7777777777777784e-05, | |
| "loss": 1.0824, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 6.020181818181818, | |
| "eval_accuracy": 0.47195767195767196, | |
| "eval_loss": 0.9947441220283508, | |
| "eval_runtime": 134.8662, | |
| "eval_samples_per_second": 7.007, | |
| "eval_steps_per_second": 0.882, | |
| "step": 777 | |
| }, | |
| { | |
| "epoch": 7.000545454545454, | |
| "grad_norm": 7.011205196380615, | |
| "learning_rate": 4.7676767676767684e-05, | |
| "loss": 1.0872, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 7.002363636363636, | |
| "grad_norm": 5.094061851501465, | |
| "learning_rate": 4.7575757575757576e-05, | |
| "loss": 0.8856, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 7.004181818181818, | |
| "grad_norm": 8.960957527160645, | |
| "learning_rate": 4.7474747474747476e-05, | |
| "loss": 0.8666, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 7.006, | |
| "grad_norm": 7.651601314544678, | |
| "learning_rate": 4.7373737373737375e-05, | |
| "loss": 1.1694, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 7.007818181818182, | |
| "grad_norm": 14.146735191345215, | |
| "learning_rate": 4.7272727272727275e-05, | |
| "loss": 1.05, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 7.009636363636363, | |
| "grad_norm": 14.381610870361328, | |
| "learning_rate": 4.7171717171717174e-05, | |
| "loss": 0.9629, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 7.011454545454545, | |
| "grad_norm": 10.896899223327637, | |
| "learning_rate": 4.7070707070707074e-05, | |
| "loss": 0.9518, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 7.013272727272727, | |
| "grad_norm": 2.556317090988159, | |
| "learning_rate": 4.696969696969697e-05, | |
| "loss": 1.0336, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 7.015090909090909, | |
| "grad_norm": 8.311176300048828, | |
| "learning_rate": 4.686868686868687e-05, | |
| "loss": 1.0507, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 7.016909090909091, | |
| "grad_norm": 12.525547981262207, | |
| "learning_rate": 4.676767676767677e-05, | |
| "loss": 0.8741, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 7.018727272727273, | |
| "grad_norm": 8.925640106201172, | |
| "learning_rate": 4.666666666666667e-05, | |
| "loss": 0.943, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 7.020181818181818, | |
| "eval_accuracy": 0.4380952380952381, | |
| "eval_loss": 1.1085319519042969, | |
| "eval_runtime": 134.7633, | |
| "eval_samples_per_second": 7.012, | |
| "eval_steps_per_second": 0.883, | |
| "step": 888 | |
| }, | |
| { | |
| "epoch": 8.000363636363636, | |
| "grad_norm": 7.20665168762207, | |
| "learning_rate": 4.656565656565657e-05, | |
| "loss": 1.0359, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 8.002181818181818, | |
| "grad_norm": 8.184736251831055, | |
| "learning_rate": 4.6464646464646464e-05, | |
| "loss": 0.9215, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 8.004, | |
| "grad_norm": 6.529689788818359, | |
| "learning_rate": 4.636363636363636e-05, | |
| "loss": 1.1004, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 8.005818181818182, | |
| "grad_norm": 9.363311767578125, | |
| "learning_rate": 4.626262626262626e-05, | |
| "loss": 0.9776, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 8.007636363636363, | |
| "grad_norm": 7.587140083312988, | |
| "learning_rate": 4.616161616161616e-05, | |
| "loss": 0.9948, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 8.009454545454545, | |
| "grad_norm": 3.559080123901367, | |
| "learning_rate": 4.606060606060607e-05, | |
| "loss": 1.1077, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 8.011272727272727, | |
| "grad_norm": 6.365655422210693, | |
| "learning_rate": 4.595959595959596e-05, | |
| "loss": 0.9794, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 8.01309090909091, | |
| "grad_norm": 6.0974040031433105, | |
| "learning_rate": 4.585858585858586e-05, | |
| "loss": 1.0501, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 8.014909090909091, | |
| "grad_norm": 9.754732131958008, | |
| "learning_rate": 4.575757575757576e-05, | |
| "loss": 0.9277, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 8.016727272727273, | |
| "grad_norm": 6.327628135681152, | |
| "learning_rate": 4.565656565656566e-05, | |
| "loss": 0.9875, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 8.018545454545455, | |
| "grad_norm": 5.392932415008545, | |
| "learning_rate": 4.555555555555556e-05, | |
| "loss": 0.8807, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 8.020181818181818, | |
| "eval_accuracy": 0.5767195767195767, | |
| "eval_loss": 0.9345303773880005, | |
| "eval_runtime": 135.6213, | |
| "eval_samples_per_second": 6.968, | |
| "eval_steps_per_second": 0.877, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 9.000181818181819, | |
| "grad_norm": 6.117162704467773, | |
| "learning_rate": 4.545454545454546e-05, | |
| "loss": 1.0491, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 9.002, | |
| "grad_norm": 6.519684791564941, | |
| "learning_rate": 4.535353535353535e-05, | |
| "loss": 0.927, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 9.003818181818183, | |
| "grad_norm": 3.578704595565796, | |
| "learning_rate": 4.525252525252526e-05, | |
| "loss": 1.0444, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 9.005636363636363, | |
| "grad_norm": 3.731860637664795, | |
| "learning_rate": 4.515151515151516e-05, | |
| "loss": 1.0397, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 9.007454545454545, | |
| "grad_norm": 6.1547064781188965, | |
| "learning_rate": 4.5050505050505056e-05, | |
| "loss": 0.9507, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 9.009272727272727, | |
| "grad_norm": 6.028125286102295, | |
| "learning_rate": 4.494949494949495e-05, | |
| "loss": 0.8981, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 9.011090909090909, | |
| "grad_norm": 5.517436504364014, | |
| "learning_rate": 4.484848484848485e-05, | |
| "loss": 0.9094, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 9.01290909090909, | |
| "grad_norm": 3.5511577129364014, | |
| "learning_rate": 4.474747474747475e-05, | |
| "loss": 1.0594, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 9.014727272727272, | |
| "grad_norm": 3.742021322250366, | |
| "learning_rate": 4.464646464646465e-05, | |
| "loss": 0.9379, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 9.016545454545454, | |
| "grad_norm": 13.917617797851562, | |
| "learning_rate": 4.454545454545455e-05, | |
| "loss": 1.1228, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 9.018363636363636, | |
| "grad_norm": 4.444940090179443, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.9285, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 9.020181818181818, | |
| "grad_norm": 20.118894577026367, | |
| "learning_rate": 4.4343434343434346e-05, | |
| "loss": 1.1009, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 9.020181818181818, | |
| "eval_accuracy": 0.5164021164021164, | |
| "eval_loss": 0.9854611158370972, | |
| "eval_runtime": 133.8494, | |
| "eval_samples_per_second": 7.06, | |
| "eval_steps_per_second": 0.889, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 10.001818181818182, | |
| "grad_norm": 4.6963090896606445, | |
| "learning_rate": 4.4242424242424246e-05, | |
| "loss": 1.053, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 10.003636363636364, | |
| "grad_norm": 7.05235481262207, | |
| "learning_rate": 4.4141414141414145e-05, | |
| "loss": 1.0504, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 10.005454545454546, | |
| "grad_norm": 6.721865177154541, | |
| "learning_rate": 4.4040404040404044e-05, | |
| "loss": 1.0168, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 10.007272727272728, | |
| "grad_norm": 2.274195909500122, | |
| "learning_rate": 4.3939393939393944e-05, | |
| "loss": 0.8676, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 10.00909090909091, | |
| "grad_norm": 3.7177436351776123, | |
| "learning_rate": 4.383838383838384e-05, | |
| "loss": 1.0794, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 10.010909090909092, | |
| "grad_norm": 4.672595024108887, | |
| "learning_rate": 4.3737373737373736e-05, | |
| "loss": 0.9912, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 10.012727272727274, | |
| "grad_norm": 8.199527740478516, | |
| "learning_rate": 4.3636363636363636e-05, | |
| "loss": 0.9819, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 10.014545454545454, | |
| "grad_norm": 8.785588264465332, | |
| "learning_rate": 4.3535353535353535e-05, | |
| "loss": 0.9493, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 10.016363636363636, | |
| "grad_norm": 5.248023509979248, | |
| "learning_rate": 4.343434343434344e-05, | |
| "loss": 0.8966, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 10.018181818181818, | |
| "grad_norm": 8.876505851745605, | |
| "learning_rate": 4.3333333333333334e-05, | |
| "loss": 0.9195, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 10.02, | |
| "grad_norm": 9.349332809448242, | |
| "learning_rate": 4.3232323232323234e-05, | |
| "loss": 1.0292, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.020181818181818, | |
| "eval_accuracy": 0.43386243386243384, | |
| "eval_loss": 1.0506019592285156, | |
| "eval_runtime": 134.4694, | |
| "eval_samples_per_second": 7.028, | |
| "eval_steps_per_second": 0.885, | |
| "step": 1221 | |
| }, | |
| { | |
| "epoch": 11.001636363636363, | |
| "grad_norm": 2.947324752807617, | |
| "learning_rate": 4.313131313131313e-05, | |
| "loss": 1.033, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 11.003454545454545, | |
| "grad_norm": 5.121668815612793, | |
| "learning_rate": 4.303030303030303e-05, | |
| "loss": 0.941, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 11.005272727272727, | |
| "grad_norm": 9.943852424621582, | |
| "learning_rate": 4.292929292929293e-05, | |
| "loss": 0.8642, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 11.007090909090909, | |
| "grad_norm": 9.663615226745605, | |
| "learning_rate": 4.282828282828283e-05, | |
| "loss": 0.903, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 11.008909090909091, | |
| "grad_norm": 6.816159248352051, | |
| "learning_rate": 4.2727272727272724e-05, | |
| "loss": 1.0538, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 11.010727272727273, | |
| "grad_norm": 4.542863845825195, | |
| "learning_rate": 4.262626262626263e-05, | |
| "loss": 0.9424, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 11.012545454545455, | |
| "grad_norm": 6.401607036590576, | |
| "learning_rate": 4.252525252525253e-05, | |
| "loss": 0.9769, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 11.014363636363637, | |
| "grad_norm": 4.389651775360107, | |
| "learning_rate": 4.242424242424243e-05, | |
| "loss": 0.9633, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 11.016181818181819, | |
| "grad_norm": 3.288135051727295, | |
| "learning_rate": 4.232323232323233e-05, | |
| "loss": 1.1248, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 11.018, | |
| "grad_norm": 6.90913200378418, | |
| "learning_rate": 4.222222222222222e-05, | |
| "loss": 0.9916, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 11.019818181818183, | |
| "grad_norm": 7.67244815826416, | |
| "learning_rate": 4.212121212121212e-05, | |
| "loss": 0.9071, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 11.020181818181818, | |
| "eval_accuracy": 0.5142857142857142, | |
| "eval_loss": 0.9925752878189087, | |
| "eval_runtime": 137.6148, | |
| "eval_samples_per_second": 6.867, | |
| "eval_steps_per_second": 0.865, | |
| "step": 1332 | |
| }, | |
| { | |
| "epoch": 12.001454545454546, | |
| "grad_norm": 9.924761772155762, | |
| "learning_rate": 4.202020202020202e-05, | |
| "loss": 0.9088, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 12.003272727272726, | |
| "grad_norm": 6.308601379394531, | |
| "learning_rate": 4.191919191919192e-05, | |
| "loss": 1.0805, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 12.005090909090908, | |
| "grad_norm": 3.287738561630249, | |
| "learning_rate": 4.181818181818182e-05, | |
| "loss": 0.897, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 12.00690909090909, | |
| "grad_norm": 11.629283905029297, | |
| "learning_rate": 4.171717171717172e-05, | |
| "loss": 0.9188, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 12.008727272727272, | |
| "grad_norm": 15.360862731933594, | |
| "learning_rate": 4.161616161616162e-05, | |
| "loss": 0.9509, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 12.010545454545454, | |
| "grad_norm": 3.7027997970581055, | |
| "learning_rate": 4.151515151515152e-05, | |
| "loss": 0.9146, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 12.012363636363636, | |
| "grad_norm": 2.7823915481567383, | |
| "learning_rate": 4.141414141414142e-05, | |
| "loss": 0.83, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 12.014181818181818, | |
| "grad_norm": 11.256366729736328, | |
| "learning_rate": 4.131313131313132e-05, | |
| "loss": 0.963, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 12.016, | |
| "grad_norm": 8.553262710571289, | |
| "learning_rate": 4.1212121212121216e-05, | |
| "loss": 1.0025, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 12.017818181818182, | |
| "grad_norm": 10.009743690490723, | |
| "learning_rate": 4.111111111111111e-05, | |
| "loss": 0.9265, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 12.019636363636364, | |
| "grad_norm": 14.513312339782715, | |
| "learning_rate": 4.101010101010101e-05, | |
| "loss": 1.0001, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 12.020181818181818, | |
| "eval_accuracy": 0.4931216931216931, | |
| "eval_loss": 1.0406430959701538, | |
| "eval_runtime": 135.3406, | |
| "eval_samples_per_second": 6.982, | |
| "eval_steps_per_second": 0.879, | |
| "step": 1443 | |
| }, | |
| { | |
| "epoch": 13.001272727272728, | |
| "grad_norm": 10.787238121032715, | |
| "learning_rate": 4.0909090909090915e-05, | |
| "loss": 0.9537, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 13.00309090909091, | |
| "grad_norm": 6.097622394561768, | |
| "learning_rate": 4.0808080808080814e-05, | |
| "loss": 0.9546, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 13.004909090909091, | |
| "grad_norm": 3.8867690563201904, | |
| "learning_rate": 4.070707070707071e-05, | |
| "loss": 0.9046, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 13.006727272727273, | |
| "grad_norm": 3.548370599746704, | |
| "learning_rate": 4.0606060606060606e-05, | |
| "loss": 1.0842, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 13.008545454545455, | |
| "grad_norm": 3.0282113552093506, | |
| "learning_rate": 4.0505050505050506e-05, | |
| "loss": 1.0189, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 13.010363636363637, | |
| "grad_norm": 5.710118293762207, | |
| "learning_rate": 4.0404040404040405e-05, | |
| "loss": 0.9354, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 13.012181818181817, | |
| "grad_norm": 8.055477142333984, | |
| "learning_rate": 4.0303030303030305e-05, | |
| "loss": 0.9242, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 13.014, | |
| "grad_norm": 2.122373342514038, | |
| "learning_rate": 4.0202020202020204e-05, | |
| "loss": 0.8736, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 13.015818181818181, | |
| "grad_norm": 3.9817798137664795, | |
| "learning_rate": 4.01010101010101e-05, | |
| "loss": 0.9973, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 13.017636363636363, | |
| "grad_norm": 9.886878967285156, | |
| "learning_rate": 4e-05, | |
| "loss": 1.0104, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 13.019454545454545, | |
| "grad_norm": 3.8867106437683105, | |
| "learning_rate": 3.98989898989899e-05, | |
| "loss": 0.9698, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 13.020181818181818, | |
| "eval_accuracy": 0.5597883597883598, | |
| "eval_loss": 0.9440351128578186, | |
| "eval_runtime": 135.8312, | |
| "eval_samples_per_second": 6.957, | |
| "eval_steps_per_second": 0.876, | |
| "step": 1554 | |
| }, | |
| { | |
| "epoch": 14.001090909090909, | |
| "grad_norm": 5.418400764465332, | |
| "learning_rate": 3.97979797979798e-05, | |
| "loss": 1.0339, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 14.00290909090909, | |
| "grad_norm": 3.685300827026367, | |
| "learning_rate": 3.96969696969697e-05, | |
| "loss": 0.9178, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 14.004727272727273, | |
| "grad_norm": 8.146584510803223, | |
| "learning_rate": 3.9595959595959594e-05, | |
| "loss": 0.9786, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 14.006545454545455, | |
| "grad_norm": 10.083425521850586, | |
| "learning_rate": 3.9494949494949494e-05, | |
| "loss": 0.9445, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 14.008363636363637, | |
| "grad_norm": 11.22440242767334, | |
| "learning_rate": 3.939393939393939e-05, | |
| "loss": 1.0185, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 14.010181818181819, | |
| "grad_norm": 9.001585960388184, | |
| "learning_rate": 3.929292929292929e-05, | |
| "loss": 1.0519, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 14.012, | |
| "grad_norm": 6.203736782073975, | |
| "learning_rate": 3.91919191919192e-05, | |
| "loss": 0.9379, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 14.013818181818182, | |
| "grad_norm": 9.380956649780273, | |
| "learning_rate": 3.909090909090909e-05, | |
| "loss": 0.9295, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 14.015636363636364, | |
| "grad_norm": 7.258037567138672, | |
| "learning_rate": 3.898989898989899e-05, | |
| "loss": 0.993, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 14.017454545454546, | |
| "grad_norm": 4.416772842407227, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.9866, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 14.019272727272726, | |
| "grad_norm": 5.408873081207275, | |
| "learning_rate": 3.878787878787879e-05, | |
| "loss": 0.9405, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 14.020181818181818, | |
| "eval_accuracy": 0.5322751322751322, | |
| "eval_loss": 0.9666908979415894, | |
| "eval_runtime": 136.0788, | |
| "eval_samples_per_second": 6.945, | |
| "eval_steps_per_second": 0.874, | |
| "step": 1665 | |
| }, | |
| { | |
| "epoch": 15.00090909090909, | |
| "grad_norm": 6.522909164428711, | |
| "learning_rate": 3.868686868686869e-05, | |
| "loss": 1.0054, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 15.002727272727272, | |
| "grad_norm": 4.72935676574707, | |
| "learning_rate": 3.858585858585859e-05, | |
| "loss": 0.888, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 15.004545454545454, | |
| "grad_norm": 11.134079933166504, | |
| "learning_rate": 3.848484848484848e-05, | |
| "loss": 1.0401, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 15.006363636363636, | |
| "grad_norm": 3.3430612087249756, | |
| "learning_rate": 3.838383838383838e-05, | |
| "loss": 0.8786, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 15.008181818181818, | |
| "grad_norm": 3.8444104194641113, | |
| "learning_rate": 3.828282828282829e-05, | |
| "loss": 0.8552, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 15.01, | |
| "grad_norm": 8.576009750366211, | |
| "learning_rate": 3.818181818181819e-05, | |
| "loss": 0.865, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 15.011818181818182, | |
| "grad_norm": 11.537275314331055, | |
| "learning_rate": 3.8080808080808087e-05, | |
| "loss": 0.9144, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 15.013636363636364, | |
| "grad_norm": 2.3551039695739746, | |
| "learning_rate": 3.797979797979798e-05, | |
| "loss": 0.8887, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 15.015454545454546, | |
| "grad_norm": 5.525041103363037, | |
| "learning_rate": 3.787878787878788e-05, | |
| "loss": 0.842, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 15.017272727272728, | |
| "grad_norm": 7.051702499389648, | |
| "learning_rate": 3.777777777777778e-05, | |
| "loss": 0.9548, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 15.01909090909091, | |
| "grad_norm": 7.506717681884766, | |
| "learning_rate": 3.767676767676768e-05, | |
| "loss": 0.8802, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 15.020181818181818, | |
| "eval_accuracy": 0.5862433862433862, | |
| "eval_loss": 0.9010878205299377, | |
| "eval_runtime": 132.3568, | |
| "eval_samples_per_second": 7.14, | |
| "eval_steps_per_second": 0.899, | |
| "step": 1776 | |
| }, | |
| { | |
| "epoch": 16.00072727272727, | |
| "grad_norm": 2.738687515258789, | |
| "learning_rate": 3.757575757575758e-05, | |
| "loss": 1.0482, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 16.002545454545455, | |
| "grad_norm": 8.474298477172852, | |
| "learning_rate": 3.747474747474748e-05, | |
| "loss": 0.9304, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 16.004363636363635, | |
| "grad_norm": 5.706143379211426, | |
| "learning_rate": 3.7373737373737376e-05, | |
| "loss": 0.9842, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 16.00618181818182, | |
| "grad_norm": 10.396550178527832, | |
| "learning_rate": 3.7272727272727276e-05, | |
| "loss": 1.125, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 16.008, | |
| "grad_norm": 7.308793544769287, | |
| "learning_rate": 3.7171717171717175e-05, | |
| "loss": 0.9996, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 16.009818181818183, | |
| "grad_norm": 3.592374563217163, | |
| "learning_rate": 3.7070707070707075e-05, | |
| "loss": 0.9729, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.011636363636363, | |
| "grad_norm": 8.693289756774902, | |
| "learning_rate": 3.6969696969696974e-05, | |
| "loss": 1.0341, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 16.013454545454547, | |
| "grad_norm": 5.157698154449463, | |
| "learning_rate": 3.686868686868687e-05, | |
| "loss": 0.9377, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 16.015272727272727, | |
| "grad_norm": 7.736370086669922, | |
| "learning_rate": 3.6767676767676766e-05, | |
| "loss": 0.9455, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 16.01709090909091, | |
| "grad_norm": 6.4704694747924805, | |
| "learning_rate": 3.6666666666666666e-05, | |
| "loss": 0.9546, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 16.01890909090909, | |
| "grad_norm": 7.747092247009277, | |
| "learning_rate": 3.656565656565657e-05, | |
| "loss": 0.9154, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 16.02018181818182, | |
| "eval_accuracy": 0.5597883597883598, | |
| "eval_loss": 0.942902684211731, | |
| "eval_runtime": 130.5549, | |
| "eval_samples_per_second": 7.238, | |
| "eval_steps_per_second": 0.911, | |
| "step": 1887 | |
| }, | |
| { | |
| "epoch": 17.000545454545456, | |
| "grad_norm": 8.326998710632324, | |
| "learning_rate": 3.6464646464646465e-05, | |
| "loss": 0.9206, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 17.002363636363636, | |
| "grad_norm": 9.486376762390137, | |
| "learning_rate": 3.6363636363636364e-05, | |
| "loss": 0.8327, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 17.004181818181817, | |
| "grad_norm": 9.206704139709473, | |
| "learning_rate": 3.6262626262626264e-05, | |
| "loss": 1.0631, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 17.006, | |
| "grad_norm": 9.73511791229248, | |
| "learning_rate": 3.616161616161616e-05, | |
| "loss": 1.0724, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 17.00781818181818, | |
| "grad_norm": 6.431131839752197, | |
| "learning_rate": 3.606060606060606e-05, | |
| "loss": 0.9529, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 17.009636363636364, | |
| "grad_norm": 4.713401794433594, | |
| "learning_rate": 3.595959595959596e-05, | |
| "loss": 0.9926, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 17.011454545454544, | |
| "grad_norm": 7.5310282707214355, | |
| "learning_rate": 3.5858585858585855e-05, | |
| "loss": 0.9691, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 17.013272727272728, | |
| "grad_norm": 5.700111389160156, | |
| "learning_rate": 3.575757575757576e-05, | |
| "loss": 0.9538, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 17.015090909090908, | |
| "grad_norm": 4.151130676269531, | |
| "learning_rate": 3.565656565656566e-05, | |
| "loss": 0.8932, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 17.016909090909092, | |
| "grad_norm": 3.5625922679901123, | |
| "learning_rate": 3.555555555555556e-05, | |
| "loss": 1.0085, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 17.018727272727272, | |
| "grad_norm": 5.718069553375244, | |
| "learning_rate": 3.545454545454546e-05, | |
| "loss": 0.929, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 17.02018181818182, | |
| "eval_accuracy": 0.5132275132275133, | |
| "eval_loss": 0.9948244690895081, | |
| "eval_runtime": 131.1914, | |
| "eval_samples_per_second": 7.203, | |
| "eval_steps_per_second": 0.907, | |
| "step": 1998 | |
| }, | |
| { | |
| "epoch": 18.000363636363637, | |
| "grad_norm": 2.451860189437866, | |
| "learning_rate": 3.535353535353535e-05, | |
| "loss": 0.9883, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 18.002181818181818, | |
| "grad_norm": 10.55557918548584, | |
| "learning_rate": 3.525252525252525e-05, | |
| "loss": 0.9586, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 18.004, | |
| "grad_norm": 3.5876386165618896, | |
| "learning_rate": 3.515151515151515e-05, | |
| "loss": 0.9265, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 18.00581818181818, | |
| "grad_norm": 4.748887538909912, | |
| "learning_rate": 3.505050505050505e-05, | |
| "loss": 1.021, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 18.007636363636365, | |
| "grad_norm": 6.446643352508545, | |
| "learning_rate": 3.494949494949495e-05, | |
| "loss": 0.8324, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 18.009454545454545, | |
| "grad_norm": 5.21691370010376, | |
| "learning_rate": 3.484848484848485e-05, | |
| "loss": 0.8463, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 18.011272727272726, | |
| "grad_norm": 6.048694133758545, | |
| "learning_rate": 3.474747474747475e-05, | |
| "loss": 0.8215, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 18.01309090909091, | |
| "grad_norm": 7.23938512802124, | |
| "learning_rate": 3.464646464646465e-05, | |
| "loss": 0.937, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 18.01490909090909, | |
| "grad_norm": 7.888938903808594, | |
| "learning_rate": 3.454545454545455e-05, | |
| "loss": 0.8632, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 18.016727272727273, | |
| "grad_norm": 5.132527828216553, | |
| "learning_rate": 3.444444444444445e-05, | |
| "loss": 1.0232, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 18.018545454545453, | |
| "grad_norm": 3.099971294403076, | |
| "learning_rate": 3.434343434343435e-05, | |
| "loss": 0.9112, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 18.02018181818182, | |
| "eval_accuracy": 0.5851851851851851, | |
| "eval_loss": 0.9056147933006287, | |
| "eval_runtime": 132.4218, | |
| "eval_samples_per_second": 7.136, | |
| "eval_steps_per_second": 0.899, | |
| "step": 2109 | |
| }, | |
| { | |
| "epoch": 19.00018181818182, | |
| "grad_norm": 6.3844218254089355, | |
| "learning_rate": 3.424242424242424e-05, | |
| "loss": 0.9953, | |
| "step": 2110 | |
| }, | |
| { | |
| "epoch": 19.002, | |
| "grad_norm": 5.968114852905273, | |
| "learning_rate": 3.414141414141414e-05, | |
| "loss": 0.9817, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 19.003818181818183, | |
| "grad_norm": 3.901590585708618, | |
| "learning_rate": 3.4040404040404045e-05, | |
| "loss": 0.9323, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 19.005636363636363, | |
| "grad_norm": 7.752019882202148, | |
| "learning_rate": 3.3939393939393945e-05, | |
| "loss": 0.9498, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 19.007454545454546, | |
| "grad_norm": 4.7573771476745605, | |
| "learning_rate": 3.3838383838383844e-05, | |
| "loss": 1.1847, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 19.009272727272727, | |
| "grad_norm": 7.086313724517822, | |
| "learning_rate": 3.373737373737374e-05, | |
| "loss": 0.9449, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 19.01109090909091, | |
| "grad_norm": 7.621354579925537, | |
| "learning_rate": 3.3636363636363636e-05, | |
| "loss": 0.9435, | |
| "step": 2170 | |
| }, | |
| { | |
| "epoch": 19.01290909090909, | |
| "grad_norm": 3.54822039604187, | |
| "learning_rate": 3.3535353535353536e-05, | |
| "loss": 0.9775, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 19.014727272727274, | |
| "grad_norm": 5.012094497680664, | |
| "learning_rate": 3.3434343434343435e-05, | |
| "loss": 0.9856, | |
| "step": 2190 | |
| }, | |
| { | |
| "epoch": 19.016545454545454, | |
| "grad_norm": 9.499866485595703, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.9401, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 19.018363636363638, | |
| "grad_norm": 5.7213873863220215, | |
| "learning_rate": 3.3232323232323234e-05, | |
| "loss": 0.8776, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 19.02018181818182, | |
| "grad_norm": 11.805944442749023, | |
| "learning_rate": 3.3131313131313134e-05, | |
| "loss": 0.9202, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 19.02018181818182, | |
| "eval_accuracy": 0.5523809523809524, | |
| "eval_loss": 0.9488540887832642, | |
| "eval_runtime": 132.4819, | |
| "eval_samples_per_second": 7.133, | |
| "eval_steps_per_second": 0.898, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 20.00181818181818, | |
| "grad_norm": 3.7236974239349365, | |
| "learning_rate": 3.303030303030303e-05, | |
| "loss": 0.9245, | |
| "step": 2230 | |
| }, | |
| { | |
| "epoch": 20.003636363636364, | |
| "grad_norm": 6.91562032699585, | |
| "learning_rate": 3.292929292929293e-05, | |
| "loss": 0.9469, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 20.005454545454544, | |
| "grad_norm": 4.592532157897949, | |
| "learning_rate": 3.282828282828283e-05, | |
| "loss": 0.9701, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 20.007272727272728, | |
| "grad_norm": 9.3302001953125, | |
| "learning_rate": 3.272727272727273e-05, | |
| "loss": 0.7942, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 20.009090909090908, | |
| "grad_norm": 8.965353012084961, | |
| "learning_rate": 3.2626262626262624e-05, | |
| "loss": 0.9704, | |
| "step": 2270 | |
| }, | |
| { | |
| "epoch": 20.01090909090909, | |
| "grad_norm": 4.075799942016602, | |
| "learning_rate": 3.2525252525252524e-05, | |
| "loss": 1.0357, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 20.012727272727272, | |
| "grad_norm": 4.786118984222412, | |
| "learning_rate": 3.2424242424242423e-05, | |
| "loss": 0.9595, | |
| "step": 2290 | |
| }, | |
| { | |
| "epoch": 20.014545454545456, | |
| "grad_norm": 6.418935298919678, | |
| "learning_rate": 3.232323232323233e-05, | |
| "loss": 0.9382, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 20.016363636363636, | |
| "grad_norm": 5.954792022705078, | |
| "learning_rate": 3.222222222222223e-05, | |
| "loss": 0.9991, | |
| "step": 2310 | |
| }, | |
| { | |
| "epoch": 20.01818181818182, | |
| "grad_norm": 6.49984884262085, | |
| "learning_rate": 3.212121212121212e-05, | |
| "loss": 0.9113, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 20.02, | |
| "grad_norm": 7.011244773864746, | |
| "learning_rate": 3.202020202020202e-05, | |
| "loss": 0.9004, | |
| "step": 2330 | |
| }, | |
| { | |
| "epoch": 20.02018181818182, | |
| "eval_accuracy": 0.582010582010582, | |
| "eval_loss": 0.8995028734207153, | |
| "eval_runtime": 133.4171, | |
| "eval_samples_per_second": 7.083, | |
| "eval_steps_per_second": 0.892, | |
| "step": 2331 | |
| }, | |
| { | |
| "epoch": 21.001636363636365, | |
| "grad_norm": 4.065828800201416, | |
| "learning_rate": 3.191919191919192e-05, | |
| "loss": 0.9496, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 21.003454545454545, | |
| "grad_norm": 8.006439208984375, | |
| "learning_rate": 3.181818181818182e-05, | |
| "loss": 0.9348, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 21.00527272727273, | |
| "grad_norm": 3.660148859024048, | |
| "learning_rate": 3.171717171717172e-05, | |
| "loss": 0.7659, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 21.00709090909091, | |
| "grad_norm": 6.901710033416748, | |
| "learning_rate": 3.161616161616161e-05, | |
| "loss": 0.928, | |
| "step": 2370 | |
| }, | |
| { | |
| "epoch": 21.00890909090909, | |
| "grad_norm": 4.300201892852783, | |
| "learning_rate": 3.151515151515151e-05, | |
| "loss": 0.8376, | |
| "step": 2380 | |
| }, | |
| { | |
| "epoch": 21.010727272727273, | |
| "grad_norm": 7.4214768409729, | |
| "learning_rate": 3.141414141414142e-05, | |
| "loss": 0.8961, | |
| "step": 2390 | |
| }, | |
| { | |
| "epoch": 21.012545454545453, | |
| "grad_norm": 7.769872188568115, | |
| "learning_rate": 3.131313131313132e-05, | |
| "loss": 0.9421, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 21.014363636363637, | |
| "grad_norm": 6.112030029296875, | |
| "learning_rate": 3.121212121212122e-05, | |
| "loss": 1.003, | |
| "step": 2410 | |
| }, | |
| { | |
| "epoch": 21.016181818181817, | |
| "grad_norm": 2.415757656097412, | |
| "learning_rate": 3.111111111111111e-05, | |
| "loss": 0.8616, | |
| "step": 2420 | |
| }, | |
| { | |
| "epoch": 21.018, | |
| "grad_norm": 7.6159772872924805, | |
| "learning_rate": 3.101010101010101e-05, | |
| "loss": 1.0072, | |
| "step": 2430 | |
| }, | |
| { | |
| "epoch": 21.01981818181818, | |
| "grad_norm": 3.9019908905029297, | |
| "learning_rate": 3.090909090909091e-05, | |
| "loss": 0.9318, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 21.02018181818182, | |
| "eval_accuracy": 0.5957671957671957, | |
| "eval_loss": 0.9031988978385925, | |
| "eval_runtime": 133.1845, | |
| "eval_samples_per_second": 7.095, | |
| "eval_steps_per_second": 0.893, | |
| "step": 2442 | |
| }, | |
| { | |
| "epoch": 22.001454545454546, | |
| "grad_norm": 5.528774261474609, | |
| "learning_rate": 3.080808080808081e-05, | |
| "loss": 0.8702, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 22.003272727272726, | |
| "grad_norm": 4.2135467529296875, | |
| "learning_rate": 3.070707070707071e-05, | |
| "loss": 0.8018, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 22.00509090909091, | |
| "grad_norm": 5.549806118011475, | |
| "learning_rate": 3.060606060606061e-05, | |
| "loss": 0.865, | |
| "step": 2470 | |
| }, | |
| { | |
| "epoch": 22.00690909090909, | |
| "grad_norm": 8.080962181091309, | |
| "learning_rate": 3.050505050505051e-05, | |
| "loss": 0.9013, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 22.008727272727274, | |
| "grad_norm": 6.765378475189209, | |
| "learning_rate": 3.0404040404040406e-05, | |
| "loss": 0.7961, | |
| "step": 2490 | |
| }, | |
| { | |
| "epoch": 22.010545454545454, | |
| "grad_norm": 6.4678874015808105, | |
| "learning_rate": 3.0303030303030306e-05, | |
| "loss": 0.9707, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 22.012363636363638, | |
| "grad_norm": 6.30830717086792, | |
| "learning_rate": 3.0202020202020205e-05, | |
| "loss": 0.9456, | |
| "step": 2510 | |
| }, | |
| { | |
| "epoch": 22.014181818181818, | |
| "grad_norm": 7.281170845031738, | |
| "learning_rate": 3.01010101010101e-05, | |
| "loss": 0.9573, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 22.016, | |
| "grad_norm": 2.3276336193084717, | |
| "learning_rate": 3e-05, | |
| "loss": 0.9226, | |
| "step": 2530 | |
| }, | |
| { | |
| "epoch": 22.017818181818182, | |
| "grad_norm": 10.788141250610352, | |
| "learning_rate": 2.98989898989899e-05, | |
| "loss": 0.949, | |
| "step": 2540 | |
| }, | |
| { | |
| "epoch": 22.019636363636362, | |
| "grad_norm": 12.814167022705078, | |
| "learning_rate": 2.9797979797979796e-05, | |
| "loss": 0.8493, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 22.02018181818182, | |
| "eval_accuracy": 0.5238095238095238, | |
| "eval_loss": 0.997473418712616, | |
| "eval_runtime": 134.17, | |
| "eval_samples_per_second": 7.043, | |
| "eval_steps_per_second": 0.887, | |
| "step": 2553 | |
| }, | |
| { | |
| "epoch": 23.001272727272728, | |
| "grad_norm": 7.309125900268555, | |
| "learning_rate": 2.96969696969697e-05, | |
| "loss": 1.0028, | |
| "step": 2560 | |
| }, | |
| { | |
| "epoch": 23.003090909090908, | |
| "grad_norm": 3.8795406818389893, | |
| "learning_rate": 2.95959595959596e-05, | |
| "loss": 0.9076, | |
| "step": 2570 | |
| }, | |
| { | |
| "epoch": 23.00490909090909, | |
| "grad_norm": 11.91686725616455, | |
| "learning_rate": 2.9494949494949498e-05, | |
| "loss": 0.9604, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 23.00672727272727, | |
| "grad_norm": 3.4642341136932373, | |
| "learning_rate": 2.9393939393939394e-05, | |
| "loss": 0.9178, | |
| "step": 2590 | |
| }, | |
| { | |
| "epoch": 23.008545454545455, | |
| "grad_norm": 6.674759864807129, | |
| "learning_rate": 2.9292929292929294e-05, | |
| "loss": 1.0162, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 23.010363636363635, | |
| "grad_norm": 5.000521659851074, | |
| "learning_rate": 2.9191919191919193e-05, | |
| "loss": 0.9169, | |
| "step": 2610 | |
| }, | |
| { | |
| "epoch": 23.01218181818182, | |
| "grad_norm": 4.165314197540283, | |
| "learning_rate": 2.909090909090909e-05, | |
| "loss": 0.9068, | |
| "step": 2620 | |
| }, | |
| { | |
| "epoch": 23.014, | |
| "grad_norm": 2.4558990001678467, | |
| "learning_rate": 2.898989898989899e-05, | |
| "loss": 0.9802, | |
| "step": 2630 | |
| }, | |
| { | |
| "epoch": 23.015818181818183, | |
| "grad_norm": 4.509535312652588, | |
| "learning_rate": 2.8888888888888888e-05, | |
| "loss": 1.0425, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 23.017636363636363, | |
| "grad_norm": 4.437975883483887, | |
| "learning_rate": 2.878787878787879e-05, | |
| "loss": 0.7647, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 23.019454545454547, | |
| "grad_norm": 4.480066299438477, | |
| "learning_rate": 2.868686868686869e-05, | |
| "loss": 0.8587, | |
| "step": 2660 | |
| }, | |
| { | |
| "epoch": 23.02018181818182, | |
| "eval_accuracy": 0.5259259259259259, | |
| "eval_loss": 1.0142112970352173, | |
| "eval_runtime": 134.2476, | |
| "eval_samples_per_second": 7.039, | |
| "eval_steps_per_second": 0.886, | |
| "step": 2664 | |
| }, | |
| { | |
| "epoch": 24.00109090909091, | |
| "grad_norm": 8.571599960327148, | |
| "learning_rate": 2.8585858585858587e-05, | |
| "loss": 0.9351, | |
| "step": 2670 | |
| }, | |
| { | |
| "epoch": 24.002909090909093, | |
| "grad_norm": 6.295083522796631, | |
| "learning_rate": 2.8484848484848486e-05, | |
| "loss": 1.1825, | |
| "step": 2680 | |
| }, | |
| { | |
| "epoch": 24.004727272727273, | |
| "grad_norm": 8.257028579711914, | |
| "learning_rate": 2.8383838383838386e-05, | |
| "loss": 1.003, | |
| "step": 2690 | |
| }, | |
| { | |
| "epoch": 24.006545454545453, | |
| "grad_norm": 3.9263625144958496, | |
| "learning_rate": 2.8282828282828282e-05, | |
| "loss": 0.8946, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 24.008363636363637, | |
| "grad_norm": 5.969736099243164, | |
| "learning_rate": 2.818181818181818e-05, | |
| "loss": 0.8263, | |
| "step": 2710 | |
| }, | |
| { | |
| "epoch": 24.010181818181817, | |
| "grad_norm": 7.632034778594971, | |
| "learning_rate": 2.808080808080808e-05, | |
| "loss": 0.9067, | |
| "step": 2720 | |
| }, | |
| { | |
| "epoch": 24.012, | |
| "grad_norm": 5.329760551452637, | |
| "learning_rate": 2.7979797979797984e-05, | |
| "loss": 0.8835, | |
| "step": 2730 | |
| }, | |
| { | |
| "epoch": 24.01381818181818, | |
| "grad_norm": 5.737499237060547, | |
| "learning_rate": 2.7878787878787883e-05, | |
| "loss": 0.9203, | |
| "step": 2740 | |
| }, | |
| { | |
| "epoch": 24.015636363636364, | |
| "grad_norm": 4.075726509094238, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.945, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 24.017454545454545, | |
| "grad_norm": 5.517560005187988, | |
| "learning_rate": 2.767676767676768e-05, | |
| "loss": 0.9141, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 24.019272727272728, | |
| "grad_norm": 3.141531467437744, | |
| "learning_rate": 2.7575757575757578e-05, | |
| "loss": 0.958, | |
| "step": 2770 | |
| }, | |
| { | |
| "epoch": 24.02018181818182, | |
| "eval_accuracy": 0.5375661375661376, | |
| "eval_loss": 0.9664833545684814, | |
| "eval_runtime": 135.1561, | |
| "eval_samples_per_second": 6.992, | |
| "eval_steps_per_second": 0.88, | |
| "step": 2775 | |
| }, | |
| { | |
| "epoch": 25.00090909090909, | |
| "grad_norm": 5.119720935821533, | |
| "learning_rate": 2.7474747474747474e-05, | |
| "loss": 1.0862, | |
| "step": 2780 | |
| }, | |
| { | |
| "epoch": 25.002727272727274, | |
| "grad_norm": 10.184759140014648, | |
| "learning_rate": 2.7373737373737374e-05, | |
| "loss": 0.9412, | |
| "step": 2790 | |
| }, | |
| { | |
| "epoch": 25.004545454545454, | |
| "grad_norm": 2.972130060195923, | |
| "learning_rate": 2.7272727272727273e-05, | |
| "loss": 0.7545, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 25.006363636363638, | |
| "grad_norm": 2.982419967651367, | |
| "learning_rate": 2.717171717171717e-05, | |
| "loss": 0.93, | |
| "step": 2810 | |
| }, | |
| { | |
| "epoch": 25.008181818181818, | |
| "grad_norm": 3.9260990619659424, | |
| "learning_rate": 2.7070707070707075e-05, | |
| "loss": 0.7816, | |
| "step": 2820 | |
| }, | |
| { | |
| "epoch": 25.01, | |
| "grad_norm": 6.232370853424072, | |
| "learning_rate": 2.696969696969697e-05, | |
| "loss": 0.9919, | |
| "step": 2830 | |
| }, | |
| { | |
| "epoch": 25.01181818181818, | |
| "grad_norm": 5.925095081329346, | |
| "learning_rate": 2.686868686868687e-05, | |
| "loss": 0.8871, | |
| "step": 2840 | |
| }, | |
| { | |
| "epoch": 25.013636363636362, | |
| "grad_norm": 3.562493085861206, | |
| "learning_rate": 2.676767676767677e-05, | |
| "loss": 0.8277, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 25.015454545454546, | |
| "grad_norm": 5.1987199783325195, | |
| "learning_rate": 2.6666666666666667e-05, | |
| "loss": 0.8438, | |
| "step": 2860 | |
| }, | |
| { | |
| "epoch": 25.017272727272726, | |
| "grad_norm": 7.01310920715332, | |
| "learning_rate": 2.6565656565656566e-05, | |
| "loss": 0.9062, | |
| "step": 2870 | |
| }, | |
| { | |
| "epoch": 25.01909090909091, | |
| "grad_norm": 6.749232292175293, | |
| "learning_rate": 2.6464646464646466e-05, | |
| "loss": 0.996, | |
| "step": 2880 | |
| }, | |
| { | |
| "epoch": 25.02018181818182, | |
| "eval_accuracy": 0.5703703703703704, | |
| "eval_loss": 0.9390914440155029, | |
| "eval_runtime": 132.5682, | |
| "eval_samples_per_second": 7.128, | |
| "eval_steps_per_second": 0.898, | |
| "step": 2886 | |
| }, | |
| { | |
| "epoch": 26.00072727272727, | |
| "grad_norm": 2.76141619682312, | |
| "learning_rate": 2.636363636363636e-05, | |
| "loss": 1.0099, | |
| "step": 2890 | |
| }, | |
| { | |
| "epoch": 26.002545454545455, | |
| "grad_norm": 8.156211853027344, | |
| "learning_rate": 2.6262626262626268e-05, | |
| "loss": 0.9647, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 26.004363636363635, | |
| "grad_norm": 6.6141462326049805, | |
| "learning_rate": 2.6161616161616164e-05, | |
| "loss": 0.9456, | |
| "step": 2910 | |
| }, | |
| { | |
| "epoch": 26.00618181818182, | |
| "grad_norm": 4.650730133056641, | |
| "learning_rate": 2.6060606060606063e-05, | |
| "loss": 0.8798, | |
| "step": 2920 | |
| }, | |
| { | |
| "epoch": 26.008, | |
| "grad_norm": 2.213693618774414, | |
| "learning_rate": 2.5959595959595963e-05, | |
| "loss": 0.8274, | |
| "step": 2930 | |
| }, | |
| { | |
| "epoch": 26.009818181818183, | |
| "grad_norm": 9.269267082214355, | |
| "learning_rate": 2.585858585858586e-05, | |
| "loss": 0.9503, | |
| "step": 2940 | |
| }, | |
| { | |
| "epoch": 26.011636363636363, | |
| "grad_norm": 4.464992046356201, | |
| "learning_rate": 2.575757575757576e-05, | |
| "loss": 0.8205, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 26.013454545454547, | |
| "grad_norm": 6.635782718658447, | |
| "learning_rate": 2.5656565656565658e-05, | |
| "loss": 0.9112, | |
| "step": 2960 | |
| }, | |
| { | |
| "epoch": 26.015272727272727, | |
| "grad_norm": 8.023598670959473, | |
| "learning_rate": 2.5555555555555554e-05, | |
| "loss": 0.9634, | |
| "step": 2970 | |
| }, | |
| { | |
| "epoch": 26.01709090909091, | |
| "grad_norm": 11.189812660217285, | |
| "learning_rate": 2.5454545454545454e-05, | |
| "loss": 0.9046, | |
| "step": 2980 | |
| }, | |
| { | |
| "epoch": 26.01890909090909, | |
| "grad_norm": 3.639500856399536, | |
| "learning_rate": 2.5353535353535356e-05, | |
| "loss": 0.823, | |
| "step": 2990 | |
| }, | |
| { | |
| "epoch": 26.02018181818182, | |
| "eval_accuracy": 0.5777777777777777, | |
| "eval_loss": 0.9170966744422913, | |
| "eval_runtime": 133.3341, | |
| "eval_samples_per_second": 7.087, | |
| "eval_steps_per_second": 0.892, | |
| "step": 2997 | |
| }, | |
| { | |
| "epoch": 27.000545454545456, | |
| "grad_norm": 7.965769290924072, | |
| "learning_rate": 2.5252525252525256e-05, | |
| "loss": 0.9709, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 27.002363636363636, | |
| "grad_norm": 6.372323036193848, | |
| "learning_rate": 2.5151515151515155e-05, | |
| "loss": 0.9405, | |
| "step": 3010 | |
| }, | |
| { | |
| "epoch": 27.004181818181817, | |
| "grad_norm": 8.214998245239258, | |
| "learning_rate": 2.505050505050505e-05, | |
| "loss": 0.8592, | |
| "step": 3020 | |
| }, | |
| { | |
| "epoch": 27.006, | |
| "grad_norm": 2.5504863262176514, | |
| "learning_rate": 2.494949494949495e-05, | |
| "loss": 0.8577, | |
| "step": 3030 | |
| }, | |
| { | |
| "epoch": 27.00781818181818, | |
| "grad_norm": 7.738058090209961, | |
| "learning_rate": 2.4848484848484847e-05, | |
| "loss": 0.9339, | |
| "step": 3040 | |
| }, | |
| { | |
| "epoch": 27.009636363636364, | |
| "grad_norm": 4.073580741882324, | |
| "learning_rate": 2.474747474747475e-05, | |
| "loss": 0.8894, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 27.011454545454544, | |
| "grad_norm": 3.962249994277954, | |
| "learning_rate": 2.464646464646465e-05, | |
| "loss": 1.1608, | |
| "step": 3060 | |
| }, | |
| { | |
| "epoch": 27.013272727272728, | |
| "grad_norm": 5.425260066986084, | |
| "learning_rate": 2.4545454545454545e-05, | |
| "loss": 0.9497, | |
| "step": 3070 | |
| }, | |
| { | |
| "epoch": 27.015090909090908, | |
| "grad_norm": 9.002552032470703, | |
| "learning_rate": 2.4444444444444445e-05, | |
| "loss": 0.9151, | |
| "step": 3080 | |
| }, | |
| { | |
| "epoch": 27.016909090909092, | |
| "grad_norm": 4.565546035766602, | |
| "learning_rate": 2.4343434343434344e-05, | |
| "loss": 0.8929, | |
| "step": 3090 | |
| }, | |
| { | |
| "epoch": 27.018727272727272, | |
| "grad_norm": 3.171492338180542, | |
| "learning_rate": 2.4242424242424244e-05, | |
| "loss": 0.8834, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 27.02018181818182, | |
| "eval_accuracy": 0.5873015873015873, | |
| "eval_loss": 0.89227294921875, | |
| "eval_runtime": 133.2881, | |
| "eval_samples_per_second": 7.09, | |
| "eval_steps_per_second": 0.893, | |
| "step": 3108 | |
| }, | |
| { | |
| "epoch": 28.000363636363637, | |
| "grad_norm": 7.9800896644592285, | |
| "learning_rate": 2.4141414141414143e-05, | |
| "loss": 0.8895, | |
| "step": 3110 | |
| }, | |
| { | |
| "epoch": 28.002181818181818, | |
| "grad_norm": 4.383500576019287, | |
| "learning_rate": 2.404040404040404e-05, | |
| "loss": 0.9218, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 28.004, | |
| "grad_norm": 5.249983310699463, | |
| "learning_rate": 2.393939393939394e-05, | |
| "loss": 0.8501, | |
| "step": 3130 | |
| }, | |
| { | |
| "epoch": 28.00581818181818, | |
| "grad_norm": 9.334481239318848, | |
| "learning_rate": 2.3838383838383842e-05, | |
| "loss": 0.8836, | |
| "step": 3140 | |
| }, | |
| { | |
| "epoch": 28.007636363636365, | |
| "grad_norm": 7.620109558105469, | |
| "learning_rate": 2.3737373737373738e-05, | |
| "loss": 0.9437, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 28.009454545454545, | |
| "grad_norm": 6.430835247039795, | |
| "learning_rate": 2.3636363636363637e-05, | |
| "loss": 0.9573, | |
| "step": 3160 | |
| }, | |
| { | |
| "epoch": 28.011272727272726, | |
| "grad_norm": 5.154223442077637, | |
| "learning_rate": 2.3535353535353537e-05, | |
| "loss": 0.9988, | |
| "step": 3170 | |
| }, | |
| { | |
| "epoch": 28.01309090909091, | |
| "grad_norm": 3.8441991806030273, | |
| "learning_rate": 2.3434343434343436e-05, | |
| "loss": 0.8807, | |
| "step": 3180 | |
| }, | |
| { | |
| "epoch": 28.01490909090909, | |
| "grad_norm": 12.257723808288574, | |
| "learning_rate": 2.3333333333333336e-05, | |
| "loss": 0.8627, | |
| "step": 3190 | |
| }, | |
| { | |
| "epoch": 28.016727272727273, | |
| "grad_norm": 2.526041030883789, | |
| "learning_rate": 2.3232323232323232e-05, | |
| "loss": 0.8983, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 28.018545454545453, | |
| "grad_norm": 5.447187900543213, | |
| "learning_rate": 2.313131313131313e-05, | |
| "loss": 0.8615, | |
| "step": 3210 | |
| }, | |
| { | |
| "epoch": 28.02018181818182, | |
| "eval_accuracy": 0.5470899470899471, | |
| "eval_loss": 0.9576845169067383, | |
| "eval_runtime": 131.1178, | |
| "eval_samples_per_second": 7.207, | |
| "eval_steps_per_second": 0.908, | |
| "step": 3219 | |
| }, | |
| { | |
| "epoch": 29.00018181818182, | |
| "grad_norm": 4.697556972503662, | |
| "learning_rate": 2.3030303030303034e-05, | |
| "loss": 0.9178, | |
| "step": 3220 | |
| }, | |
| { | |
| "epoch": 29.002, | |
| "grad_norm": 7.460103511810303, | |
| "learning_rate": 2.292929292929293e-05, | |
| "loss": 0.901, | |
| "step": 3230 | |
| }, | |
| { | |
| "epoch": 29.003818181818183, | |
| "grad_norm": 4.0857343673706055, | |
| "learning_rate": 2.282828282828283e-05, | |
| "loss": 0.8679, | |
| "step": 3240 | |
| }, | |
| { | |
| "epoch": 29.005636363636363, | |
| "grad_norm": 5.448916912078857, | |
| "learning_rate": 2.272727272727273e-05, | |
| "loss": 0.7971, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 29.007454545454546, | |
| "grad_norm": 9.139603614807129, | |
| "learning_rate": 2.262626262626263e-05, | |
| "loss": 0.9889, | |
| "step": 3260 | |
| }, | |
| { | |
| "epoch": 29.009272727272727, | |
| "grad_norm": 7.882384300231934, | |
| "learning_rate": 2.2525252525252528e-05, | |
| "loss": 0.9392, | |
| "step": 3270 | |
| }, | |
| { | |
| "epoch": 29.01109090909091, | |
| "grad_norm": 4.343461036682129, | |
| "learning_rate": 2.2424242424242424e-05, | |
| "loss": 0.9194, | |
| "step": 3280 | |
| }, | |
| { | |
| "epoch": 29.01290909090909, | |
| "grad_norm": 7.994728088378906, | |
| "learning_rate": 2.2323232323232324e-05, | |
| "loss": 0.9083, | |
| "step": 3290 | |
| }, | |
| { | |
| "epoch": 29.014727272727274, | |
| "grad_norm": 3.5908119678497314, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.987, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 29.016545454545454, | |
| "grad_norm": 4.692817211151123, | |
| "learning_rate": 2.2121212121212123e-05, | |
| "loss": 0.8884, | |
| "step": 3310 | |
| }, | |
| { | |
| "epoch": 29.018363636363638, | |
| "grad_norm": 6.650206565856934, | |
| "learning_rate": 2.2020202020202022e-05, | |
| "loss": 0.8695, | |
| "step": 3320 | |
| }, | |
| { | |
| "epoch": 29.02018181818182, | |
| "grad_norm": 26.35285186767578, | |
| "learning_rate": 2.191919191919192e-05, | |
| "loss": 0.9462, | |
| "step": 3330 | |
| }, | |
| { | |
| "epoch": 29.02018181818182, | |
| "eval_accuracy": 0.562962962962963, | |
| "eval_loss": 0.9467629790306091, | |
| "eval_runtime": 131.6292, | |
| "eval_samples_per_second": 7.179, | |
| "eval_steps_per_second": 0.904, | |
| "step": 3330 | |
| }, | |
| { | |
| "epoch": 30.00181818181818, | |
| "grad_norm": 3.437410593032837, | |
| "learning_rate": 2.1818181818181818e-05, | |
| "loss": 0.8162, | |
| "step": 3340 | |
| }, | |
| { | |
| "epoch": 30.003636363636364, | |
| "grad_norm": 5.97007417678833, | |
| "learning_rate": 2.171717171717172e-05, | |
| "loss": 1.0401, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 30.005454545454544, | |
| "grad_norm": 3.841773509979248, | |
| "learning_rate": 2.1616161616161617e-05, | |
| "loss": 0.7645, | |
| "step": 3360 | |
| }, | |
| { | |
| "epoch": 30.007272727272728, | |
| "grad_norm": 9.196441650390625, | |
| "learning_rate": 2.1515151515151516e-05, | |
| "loss": 0.8248, | |
| "step": 3370 | |
| }, | |
| { | |
| "epoch": 30.009090909090908, | |
| "grad_norm": 8.819738388061523, | |
| "learning_rate": 2.1414141414141416e-05, | |
| "loss": 0.9833, | |
| "step": 3380 | |
| }, | |
| { | |
| "epoch": 30.01090909090909, | |
| "grad_norm": 5.484842777252197, | |
| "learning_rate": 2.1313131313131315e-05, | |
| "loss": 0.9015, | |
| "step": 3390 | |
| }, | |
| { | |
| "epoch": 30.012727272727272, | |
| "grad_norm": 7.970553398132324, | |
| "learning_rate": 2.1212121212121215e-05, | |
| "loss": 0.9099, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 30.014545454545456, | |
| "grad_norm": 5.823868274688721, | |
| "learning_rate": 2.111111111111111e-05, | |
| "loss": 0.9532, | |
| "step": 3410 | |
| }, | |
| { | |
| "epoch": 30.016363636363636, | |
| "grad_norm": 3.9710135459899902, | |
| "learning_rate": 2.101010101010101e-05, | |
| "loss": 0.8651, | |
| "step": 3420 | |
| }, | |
| { | |
| "epoch": 30.01818181818182, | |
| "grad_norm": 4.124075412750244, | |
| "learning_rate": 2.090909090909091e-05, | |
| "loss": 0.955, | |
| "step": 3430 | |
| }, | |
| { | |
| "epoch": 30.02, | |
| "grad_norm": 7.066349983215332, | |
| "learning_rate": 2.080808080808081e-05, | |
| "loss": 0.8909, | |
| "step": 3440 | |
| }, | |
| { | |
| "epoch": 30.02018181818182, | |
| "eval_accuracy": 0.5671957671957671, | |
| "eval_loss": 0.934297502040863, | |
| "eval_runtime": 133.2693, | |
| "eval_samples_per_second": 7.091, | |
| "eval_steps_per_second": 0.893, | |
| "step": 3441 | |
| }, | |
| { | |
| "epoch": 31.001636363636365, | |
| "grad_norm": 8.60042667388916, | |
| "learning_rate": 2.070707070707071e-05, | |
| "loss": 0.9224, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 31.003454545454545, | |
| "grad_norm": 3.15602970123291, | |
| "learning_rate": 2.0606060606060608e-05, | |
| "loss": 0.9395, | |
| "step": 3460 | |
| }, | |
| { | |
| "epoch": 31.00527272727273, | |
| "grad_norm": 2.947406768798828, | |
| "learning_rate": 2.0505050505050504e-05, | |
| "loss": 0.9115, | |
| "step": 3470 | |
| }, | |
| { | |
| "epoch": 31.00709090909091, | |
| "grad_norm": 8.351478576660156, | |
| "learning_rate": 2.0404040404040407e-05, | |
| "loss": 0.9073, | |
| "step": 3480 | |
| }, | |
| { | |
| "epoch": 31.00890909090909, | |
| "grad_norm": 8.569064140319824, | |
| "learning_rate": 2.0303030303030303e-05, | |
| "loss": 0.9537, | |
| "step": 3490 | |
| }, | |
| { | |
| "epoch": 31.010727272727273, | |
| "grad_norm": 5.759737491607666, | |
| "learning_rate": 2.0202020202020203e-05, | |
| "loss": 0.7879, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 31.012545454545453, | |
| "grad_norm": 5.850707530975342, | |
| "learning_rate": 2.0101010101010102e-05, | |
| "loss": 0.9657, | |
| "step": 3510 | |
| }, | |
| { | |
| "epoch": 31.014363636363637, | |
| "grad_norm": 5.242412567138672, | |
| "learning_rate": 2e-05, | |
| "loss": 0.8814, | |
| "step": 3520 | |
| }, | |
| { | |
| "epoch": 31.016181818181817, | |
| "grad_norm": 4.501462936401367, | |
| "learning_rate": 1.98989898989899e-05, | |
| "loss": 0.8106, | |
| "step": 3530 | |
| }, | |
| { | |
| "epoch": 31.018, | |
| "grad_norm": 2.6810784339904785, | |
| "learning_rate": 1.9797979797979797e-05, | |
| "loss": 0.8476, | |
| "step": 3540 | |
| }, | |
| { | |
| "epoch": 31.01981818181818, | |
| "grad_norm": 2.7699146270751953, | |
| "learning_rate": 1.9696969696969697e-05, | |
| "loss": 0.8048, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 31.02018181818182, | |
| "eval_accuracy": 0.5777777777777777, | |
| "eval_loss": 0.9107006192207336, | |
| "eval_runtime": 134.8089, | |
| "eval_samples_per_second": 7.01, | |
| "eval_steps_per_second": 0.883, | |
| "step": 3552 | |
| }, | |
| { | |
| "epoch": 32.00145454545454, | |
| "grad_norm": 6.846524715423584, | |
| "learning_rate": 1.95959595959596e-05, | |
| "loss": 0.8392, | |
| "step": 3560 | |
| }, | |
| { | |
| "epoch": 32.00327272727273, | |
| "grad_norm": 6.047547340393066, | |
| "learning_rate": 1.9494949494949496e-05, | |
| "loss": 0.8437, | |
| "step": 3570 | |
| }, | |
| { | |
| "epoch": 32.00509090909091, | |
| "grad_norm": 4.787844657897949, | |
| "learning_rate": 1.9393939393939395e-05, | |
| "loss": 0.9993, | |
| "step": 3580 | |
| }, | |
| { | |
| "epoch": 32.00690909090909, | |
| "grad_norm": 7.202611923217773, | |
| "learning_rate": 1.9292929292929295e-05, | |
| "loss": 0.8839, | |
| "step": 3590 | |
| }, | |
| { | |
| "epoch": 32.00872727272727, | |
| "grad_norm": 3.939385175704956, | |
| "learning_rate": 1.919191919191919e-05, | |
| "loss": 0.8977, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 32.01054545454546, | |
| "grad_norm": 2.789416790008545, | |
| "learning_rate": 1.9090909090909094e-05, | |
| "loss": 0.841, | |
| "step": 3610 | |
| }, | |
| { | |
| "epoch": 32.01236363636364, | |
| "grad_norm": 6.802464008331299, | |
| "learning_rate": 1.898989898989899e-05, | |
| "loss": 0.8549, | |
| "step": 3620 | |
| }, | |
| { | |
| "epoch": 32.01418181818182, | |
| "grad_norm": 8.131646156311035, | |
| "learning_rate": 1.888888888888889e-05, | |
| "loss": 0.8839, | |
| "step": 3630 | |
| }, | |
| { | |
| "epoch": 32.016, | |
| "grad_norm": 6.163980007171631, | |
| "learning_rate": 1.878787878787879e-05, | |
| "loss": 0.8568, | |
| "step": 3640 | |
| }, | |
| { | |
| "epoch": 32.01781818181818, | |
| "grad_norm": 9.532483100891113, | |
| "learning_rate": 1.8686868686868688e-05, | |
| "loss": 0.8056, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 32.019636363636366, | |
| "grad_norm": 7.999500274658203, | |
| "learning_rate": 1.8585858585858588e-05, | |
| "loss": 0.8109, | |
| "step": 3660 | |
| }, | |
| { | |
| "epoch": 32.02018181818182, | |
| "eval_accuracy": 0.5492063492063493, | |
| "eval_loss": 0.9547345042228699, | |
| "eval_runtime": 135.0614, | |
| "eval_samples_per_second": 6.997, | |
| "eval_steps_per_second": 0.881, | |
| "step": 3663 | |
| }, | |
| { | |
| "epoch": 33.00127272727273, | |
| "grad_norm": 5.978763103485107, | |
| "learning_rate": 1.8484848484848487e-05, | |
| "loss": 0.942, | |
| "step": 3670 | |
| }, | |
| { | |
| "epoch": 33.00309090909091, | |
| "grad_norm": 10.230365753173828, | |
| "learning_rate": 1.8383838383838383e-05, | |
| "loss": 0.8532, | |
| "step": 3680 | |
| }, | |
| { | |
| "epoch": 33.00490909090909, | |
| "grad_norm": 8.178292274475098, | |
| "learning_rate": 1.8282828282828286e-05, | |
| "loss": 0.8999, | |
| "step": 3690 | |
| }, | |
| { | |
| "epoch": 33.006727272727275, | |
| "grad_norm": 9.503693580627441, | |
| "learning_rate": 1.8181818181818182e-05, | |
| "loss": 0.9341, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 33.008545454545455, | |
| "grad_norm": 5.494664192199707, | |
| "learning_rate": 1.808080808080808e-05, | |
| "loss": 0.926, | |
| "step": 3710 | |
| }, | |
| { | |
| "epoch": 33.010363636363635, | |
| "grad_norm": 5.923470973968506, | |
| "learning_rate": 1.797979797979798e-05, | |
| "loss": 0.8584, | |
| "step": 3720 | |
| }, | |
| { | |
| "epoch": 33.012181818181816, | |
| "grad_norm": 6.926611423492432, | |
| "learning_rate": 1.787878787878788e-05, | |
| "loss": 0.7687, | |
| "step": 3730 | |
| }, | |
| { | |
| "epoch": 33.014, | |
| "grad_norm": 5.04495906829834, | |
| "learning_rate": 1.777777777777778e-05, | |
| "loss": 0.9648, | |
| "step": 3740 | |
| }, | |
| { | |
| "epoch": 33.01581818181818, | |
| "grad_norm": 6.400146961212158, | |
| "learning_rate": 1.7676767676767676e-05, | |
| "loss": 0.9959, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 33.01763636363636, | |
| "grad_norm": 8.008678436279297, | |
| "learning_rate": 1.7575757575757576e-05, | |
| "loss": 0.7892, | |
| "step": 3760 | |
| }, | |
| { | |
| "epoch": 33.01945454545454, | |
| "grad_norm": 3.765272378921509, | |
| "learning_rate": 1.7474747474747475e-05, | |
| "loss": 0.9242, | |
| "step": 3770 | |
| }, | |
| { | |
| "epoch": 33.02018181818182, | |
| "eval_accuracy": 0.5597883597883598, | |
| "eval_loss": 0.9275074601173401, | |
| "eval_runtime": 135.4909, | |
| "eval_samples_per_second": 6.975, | |
| "eval_steps_per_second": 0.878, | |
| "step": 3774 | |
| }, | |
| { | |
| "epoch": 34.00109090909091, | |
| "grad_norm": 6.494891166687012, | |
| "learning_rate": 1.7373737373737375e-05, | |
| "loss": 0.7527, | |
| "step": 3780 | |
| }, | |
| { | |
| "epoch": 34.00290909090909, | |
| "grad_norm": 7.506988525390625, | |
| "learning_rate": 1.7272727272727274e-05, | |
| "loss": 0.8774, | |
| "step": 3790 | |
| }, | |
| { | |
| "epoch": 34.00472727272727, | |
| "grad_norm": 6.073853015899658, | |
| "learning_rate": 1.7171717171717173e-05, | |
| "loss": 1.0006, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 34.00654545454545, | |
| "grad_norm": 5.657601833343506, | |
| "learning_rate": 1.707070707070707e-05, | |
| "loss": 0.8368, | |
| "step": 3810 | |
| }, | |
| { | |
| "epoch": 34.00836363636363, | |
| "grad_norm": 9.511337280273438, | |
| "learning_rate": 1.6969696969696972e-05, | |
| "loss": 0.8862, | |
| "step": 3820 | |
| }, | |
| { | |
| "epoch": 34.01018181818182, | |
| "grad_norm": 6.792288303375244, | |
| "learning_rate": 1.686868686868687e-05, | |
| "loss": 0.9277, | |
| "step": 3830 | |
| }, | |
| { | |
| "epoch": 34.012, | |
| "grad_norm": 10.69278335571289, | |
| "learning_rate": 1.6767676767676768e-05, | |
| "loss": 0.9289, | |
| "step": 3840 | |
| }, | |
| { | |
| "epoch": 34.01381818181818, | |
| "grad_norm": 4.471822261810303, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.8878, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 34.01563636363636, | |
| "grad_norm": 8.317084312438965, | |
| "learning_rate": 1.6565656565656567e-05, | |
| "loss": 0.7268, | |
| "step": 3860 | |
| }, | |
| { | |
| "epoch": 34.01745454545455, | |
| "grad_norm": 7.521054744720459, | |
| "learning_rate": 1.6464646464646466e-05, | |
| "loss": 0.8997, | |
| "step": 3870 | |
| }, | |
| { | |
| "epoch": 34.01927272727273, | |
| "grad_norm": 4.9641594886779785, | |
| "learning_rate": 1.6363636363636366e-05, | |
| "loss": 0.9046, | |
| "step": 3880 | |
| }, | |
| { | |
| "epoch": 34.02018181818182, | |
| "eval_accuracy": 0.5830687830687831, | |
| "eval_loss": 0.9289540648460388, | |
| "eval_runtime": 137.5464, | |
| "eval_samples_per_second": 6.87, | |
| "eval_steps_per_second": 0.865, | |
| "step": 3885 | |
| }, | |
| { | |
| "epoch": 35.00090909090909, | |
| "grad_norm": 8.1494722366333, | |
| "learning_rate": 1.6262626262626262e-05, | |
| "loss": 0.9279, | |
| "step": 3890 | |
| }, | |
| { | |
| "epoch": 35.00272727272727, | |
| "grad_norm": 4.9748687744140625, | |
| "learning_rate": 1.6161616161616165e-05, | |
| "loss": 0.6185, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 35.00454545454546, | |
| "grad_norm": 8.476805686950684, | |
| "learning_rate": 1.606060606060606e-05, | |
| "loss": 0.9423, | |
| "step": 3910 | |
| }, | |
| { | |
| "epoch": 35.00636363636364, | |
| "grad_norm": 4.4243974685668945, | |
| "learning_rate": 1.595959595959596e-05, | |
| "loss": 0.8173, | |
| "step": 3920 | |
| }, | |
| { | |
| "epoch": 35.00818181818182, | |
| "grad_norm": 10.399567604064941, | |
| "learning_rate": 1.585858585858586e-05, | |
| "loss": 1.0438, | |
| "step": 3930 | |
| }, | |
| { | |
| "epoch": 35.01, | |
| "grad_norm": 3.4936625957489014, | |
| "learning_rate": 1.5757575757575756e-05, | |
| "loss": 0.8161, | |
| "step": 3940 | |
| }, | |
| { | |
| "epoch": 35.011818181818185, | |
| "grad_norm": 7.2451581954956055, | |
| "learning_rate": 1.565656565656566e-05, | |
| "loss": 0.8241, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 35.013636363636365, | |
| "grad_norm": 3.3511080741882324, | |
| "learning_rate": 1.5555555555555555e-05, | |
| "loss": 0.9642, | |
| "step": 3960 | |
| }, | |
| { | |
| "epoch": 35.015454545454546, | |
| "grad_norm": 2.77311372756958, | |
| "learning_rate": 1.5454545454545454e-05, | |
| "loss": 1.0031, | |
| "step": 3970 | |
| }, | |
| { | |
| "epoch": 35.017272727272726, | |
| "grad_norm": 14.23676872253418, | |
| "learning_rate": 1.5353535353535354e-05, | |
| "loss": 0.9956, | |
| "step": 3980 | |
| }, | |
| { | |
| "epoch": 35.019090909090906, | |
| "grad_norm": 4.298346519470215, | |
| "learning_rate": 1.5252525252525255e-05, | |
| "loss": 0.7677, | |
| "step": 3990 | |
| }, | |
| { | |
| "epoch": 35.02018181818182, | |
| "eval_accuracy": 0.5724867724867725, | |
| "eval_loss": 0.9207894802093506, | |
| "eval_runtime": 139.2805, | |
| "eval_samples_per_second": 6.785, | |
| "eval_steps_per_second": 0.854, | |
| "step": 3996 | |
| }, | |
| { | |
| "epoch": 36.000727272727275, | |
| "grad_norm": 8.253910064697266, | |
| "learning_rate": 1.5151515151515153e-05, | |
| "loss": 0.8331, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 36.002545454545455, | |
| "grad_norm": 12.524259567260742, | |
| "learning_rate": 1.505050505050505e-05, | |
| "loss": 0.8999, | |
| "step": 4010 | |
| }, | |
| { | |
| "epoch": 36.004363636363635, | |
| "grad_norm": 13.681807518005371, | |
| "learning_rate": 1.494949494949495e-05, | |
| "loss": 0.879, | |
| "step": 4020 | |
| }, | |
| { | |
| "epoch": 36.006181818181815, | |
| "grad_norm": 11.588513374328613, | |
| "learning_rate": 1.484848484848485e-05, | |
| "loss": 1.0405, | |
| "step": 4030 | |
| }, | |
| { | |
| "epoch": 36.008, | |
| "grad_norm": 10.12088394165039, | |
| "learning_rate": 1.4747474747474749e-05, | |
| "loss": 0.9501, | |
| "step": 4040 | |
| }, | |
| { | |
| "epoch": 36.00981818181818, | |
| "grad_norm": 7.492605686187744, | |
| "learning_rate": 1.4646464646464647e-05, | |
| "loss": 0.8858, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 36.01163636363636, | |
| "grad_norm": 6.3661394119262695, | |
| "learning_rate": 1.4545454545454545e-05, | |
| "loss": 0.8903, | |
| "step": 4060 | |
| }, | |
| { | |
| "epoch": 36.01345454545454, | |
| "grad_norm": 7.913581848144531, | |
| "learning_rate": 1.4444444444444444e-05, | |
| "loss": 0.8672, | |
| "step": 4070 | |
| }, | |
| { | |
| "epoch": 36.01527272727273, | |
| "grad_norm": 5.75011682510376, | |
| "learning_rate": 1.4343434343434345e-05, | |
| "loss": 0.935, | |
| "step": 4080 | |
| }, | |
| { | |
| "epoch": 36.01709090909091, | |
| "grad_norm": 8.735386848449707, | |
| "learning_rate": 1.4242424242424243e-05, | |
| "loss": 0.9386, | |
| "step": 4090 | |
| }, | |
| { | |
| "epoch": 36.01890909090909, | |
| "grad_norm": 4.150994777679443, | |
| "learning_rate": 1.4141414141414141e-05, | |
| "loss": 0.8501, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 36.02018181818182, | |
| "eval_accuracy": 0.580952380952381, | |
| "eval_loss": 0.912603497505188, | |
| "eval_runtime": 139.8235, | |
| "eval_samples_per_second": 6.759, | |
| "eval_steps_per_second": 0.851, | |
| "step": 4107 | |
| }, | |
| { | |
| "epoch": 37.00054545454545, | |
| "grad_norm": 15.24356460571289, | |
| "learning_rate": 1.404040404040404e-05, | |
| "loss": 0.8454, | |
| "step": 4110 | |
| }, | |
| { | |
| "epoch": 37.00236363636364, | |
| "grad_norm": 5.617706775665283, | |
| "learning_rate": 1.3939393939393942e-05, | |
| "loss": 0.8981, | |
| "step": 4120 | |
| }, | |
| { | |
| "epoch": 37.00418181818182, | |
| "grad_norm": 9.067435264587402, | |
| "learning_rate": 1.383838383838384e-05, | |
| "loss": 0.8377, | |
| "step": 4130 | |
| }, | |
| { | |
| "epoch": 37.006, | |
| "grad_norm": 6.194527626037598, | |
| "learning_rate": 1.3737373737373737e-05, | |
| "loss": 0.8465, | |
| "step": 4140 | |
| }, | |
| { | |
| "epoch": 37.00781818181818, | |
| "grad_norm": 4.81503963470459, | |
| "learning_rate": 1.3636363636363637e-05, | |
| "loss": 0.8916, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 37.00963636363636, | |
| "grad_norm": 7.697696685791016, | |
| "learning_rate": 1.3535353535353538e-05, | |
| "loss": 0.9936, | |
| "step": 4160 | |
| }, | |
| { | |
| "epoch": 37.01145454545455, | |
| "grad_norm": 4.091245651245117, | |
| "learning_rate": 1.3434343434343436e-05, | |
| "loss": 0.8216, | |
| "step": 4170 | |
| }, | |
| { | |
| "epoch": 37.01327272727273, | |
| "grad_norm": 8.986047744750977, | |
| "learning_rate": 1.3333333333333333e-05, | |
| "loss": 0.9964, | |
| "step": 4180 | |
| }, | |
| { | |
| "epoch": 37.01509090909091, | |
| "grad_norm": 3.3684699535369873, | |
| "learning_rate": 1.3232323232323233e-05, | |
| "loss": 0.9505, | |
| "step": 4190 | |
| }, | |
| { | |
| "epoch": 37.01690909090909, | |
| "grad_norm": 5.151857376098633, | |
| "learning_rate": 1.3131313131313134e-05, | |
| "loss": 0.971, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 37.018727272727276, | |
| "grad_norm": 6.723459243774414, | |
| "learning_rate": 1.3030303030303032e-05, | |
| "loss": 0.8468, | |
| "step": 4210 | |
| }, | |
| { | |
| "epoch": 37.02018181818182, | |
| "eval_accuracy": 0.5862433862433862, | |
| "eval_loss": 0.9053242206573486, | |
| "eval_runtime": 139.4496, | |
| "eval_samples_per_second": 6.777, | |
| "eval_steps_per_second": 0.853, | |
| "step": 4218 | |
| }, | |
| { | |
| "epoch": 38.00036363636364, | |
| "grad_norm": 4.479756832122803, | |
| "learning_rate": 1.292929292929293e-05, | |
| "loss": 0.7653, | |
| "step": 4220 | |
| }, | |
| { | |
| "epoch": 38.00218181818182, | |
| "grad_norm": 5.094549655914307, | |
| "learning_rate": 1.2828282828282829e-05, | |
| "loss": 0.9674, | |
| "step": 4230 | |
| }, | |
| { | |
| "epoch": 38.004, | |
| "grad_norm": 5.455288887023926, | |
| "learning_rate": 1.2727272727272727e-05, | |
| "loss": 0.9193, | |
| "step": 4240 | |
| }, | |
| { | |
| "epoch": 38.005818181818185, | |
| "grad_norm": 4.766934871673584, | |
| "learning_rate": 1.2626262626262628e-05, | |
| "loss": 0.8482, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 38.007636363636365, | |
| "grad_norm": 10.875009536743164, | |
| "learning_rate": 1.2525252525252526e-05, | |
| "loss": 0.7313, | |
| "step": 4260 | |
| }, | |
| { | |
| "epoch": 38.009454545454545, | |
| "grad_norm": 4.407993316650391, | |
| "learning_rate": 1.2424242424242424e-05, | |
| "loss": 0.8851, | |
| "step": 4270 | |
| }, | |
| { | |
| "epoch": 38.011272727272726, | |
| "grad_norm": 5.814301490783691, | |
| "learning_rate": 1.2323232323232325e-05, | |
| "loss": 0.8522, | |
| "step": 4280 | |
| }, | |
| { | |
| "epoch": 38.013090909090906, | |
| "grad_norm": 4.259603977203369, | |
| "learning_rate": 1.2222222222222222e-05, | |
| "loss": 0.8599, | |
| "step": 4290 | |
| }, | |
| { | |
| "epoch": 38.01490909090909, | |
| "grad_norm": 9.286781311035156, | |
| "learning_rate": 1.2121212121212122e-05, | |
| "loss": 0.9504, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 38.01672727272727, | |
| "grad_norm": 6.765378952026367, | |
| "learning_rate": 1.202020202020202e-05, | |
| "loss": 0.9203, | |
| "step": 4310 | |
| }, | |
| { | |
| "epoch": 38.01854545454545, | |
| "grad_norm": 5.87042236328125, | |
| "learning_rate": 1.1919191919191921e-05, | |
| "loss": 0.7814, | |
| "step": 4320 | |
| }, | |
| { | |
| "epoch": 38.02018181818182, | |
| "eval_accuracy": 0.5904761904761905, | |
| "eval_loss": 0.8857986330986023, | |
| "eval_runtime": 136.9965, | |
| "eval_samples_per_second": 6.898, | |
| "eval_steps_per_second": 0.869, | |
| "step": 4329 | |
| }, | |
| { | |
| "epoch": 39.000181818181815, | |
| "grad_norm": 4.036645412445068, | |
| "learning_rate": 1.1818181818181819e-05, | |
| "loss": 0.896, | |
| "step": 4330 | |
| }, | |
| { | |
| "epoch": 39.002, | |
| "grad_norm": 6.015566825866699, | |
| "learning_rate": 1.1717171717171718e-05, | |
| "loss": 1.008, | |
| "step": 4340 | |
| }, | |
| { | |
| "epoch": 39.00381818181818, | |
| "grad_norm": 8.852757453918457, | |
| "learning_rate": 1.1616161616161616e-05, | |
| "loss": 0.8159, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 39.00563636363636, | |
| "grad_norm": 6.896280288696289, | |
| "learning_rate": 1.1515151515151517e-05, | |
| "loss": 0.7858, | |
| "step": 4360 | |
| }, | |
| { | |
| "epoch": 39.00745454545454, | |
| "grad_norm": 13.30453109741211, | |
| "learning_rate": 1.1414141414141415e-05, | |
| "loss": 0.7982, | |
| "step": 4370 | |
| }, | |
| { | |
| "epoch": 39.00927272727273, | |
| "grad_norm": 5.01935338973999, | |
| "learning_rate": 1.1313131313131314e-05, | |
| "loss": 0.9107, | |
| "step": 4380 | |
| }, | |
| { | |
| "epoch": 39.01109090909091, | |
| "grad_norm": 7.737326622009277, | |
| "learning_rate": 1.1212121212121212e-05, | |
| "loss": 0.8968, | |
| "step": 4390 | |
| }, | |
| { | |
| "epoch": 39.01290909090909, | |
| "grad_norm": 6.401789665222168, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.881, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 39.01472727272727, | |
| "grad_norm": 5.3346171379089355, | |
| "learning_rate": 1.1010101010101011e-05, | |
| "loss": 0.9034, | |
| "step": 4410 | |
| }, | |
| { | |
| "epoch": 39.01654545454546, | |
| "grad_norm": 4.072237968444824, | |
| "learning_rate": 1.0909090909090909e-05, | |
| "loss": 0.7896, | |
| "step": 4420 | |
| }, | |
| { | |
| "epoch": 39.01836363636364, | |
| "grad_norm": 15.682538986206055, | |
| "learning_rate": 1.0808080808080808e-05, | |
| "loss": 0.8473, | |
| "step": 4430 | |
| }, | |
| { | |
| "epoch": 39.02018181818182, | |
| "grad_norm": 8.316577911376953, | |
| "learning_rate": 1.0707070707070708e-05, | |
| "loss": 0.9354, | |
| "step": 4440 | |
| }, | |
| { | |
| "epoch": 39.02018181818182, | |
| "eval_accuracy": 0.5724867724867725, | |
| "eval_loss": 0.9207160472869873, | |
| "eval_runtime": 137.3805, | |
| "eval_samples_per_second": 6.879, | |
| "eval_steps_per_second": 0.866, | |
| "step": 4440 | |
| }, | |
| { | |
| "epoch": 40.00181818181818, | |
| "grad_norm": 4.132864475250244, | |
| "learning_rate": 1.0606060606060607e-05, | |
| "loss": 0.7832, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 40.00363636363636, | |
| "grad_norm": 3.2783091068267822, | |
| "learning_rate": 1.0505050505050505e-05, | |
| "loss": 0.8667, | |
| "step": 4460 | |
| }, | |
| { | |
| "epoch": 40.00545454545455, | |
| "grad_norm": 4.29857063293457, | |
| "learning_rate": 1.0404040404040405e-05, | |
| "loss": 0.719, | |
| "step": 4470 | |
| }, | |
| { | |
| "epoch": 40.00727272727273, | |
| "grad_norm": 6.1660590171813965, | |
| "learning_rate": 1.0303030303030304e-05, | |
| "loss": 0.912, | |
| "step": 4480 | |
| }, | |
| { | |
| "epoch": 40.00909090909091, | |
| "grad_norm": 5.285783290863037, | |
| "learning_rate": 1.0202020202020204e-05, | |
| "loss": 0.8702, | |
| "step": 4490 | |
| }, | |
| { | |
| "epoch": 40.01090909090909, | |
| "grad_norm": 8.145794868469238, | |
| "learning_rate": 1.0101010101010101e-05, | |
| "loss": 0.9263, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 40.012727272727275, | |
| "grad_norm": 7.457106113433838, | |
| "learning_rate": 1e-05, | |
| "loss": 0.9648, | |
| "step": 4510 | |
| }, | |
| { | |
| "epoch": 40.014545454545456, | |
| "grad_norm": 7.848749160766602, | |
| "learning_rate": 9.898989898989899e-06, | |
| "loss": 0.8373, | |
| "step": 4520 | |
| }, | |
| { | |
| "epoch": 40.016363636363636, | |
| "grad_norm": 8.913769721984863, | |
| "learning_rate": 9.7979797979798e-06, | |
| "loss": 0.8669, | |
| "step": 4530 | |
| }, | |
| { | |
| "epoch": 40.018181818181816, | |
| "grad_norm": 10.52230167388916, | |
| "learning_rate": 9.696969696969698e-06, | |
| "loss": 0.8021, | |
| "step": 4540 | |
| }, | |
| { | |
| "epoch": 40.02, | |
| "grad_norm": 5.463939666748047, | |
| "learning_rate": 9.595959595959595e-06, | |
| "loss": 0.8849, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 40.02018181818182, | |
| "eval_accuracy": 0.5650793650793651, | |
| "eval_loss": 0.9276704788208008, | |
| "eval_runtime": 135.7086, | |
| "eval_samples_per_second": 6.963, | |
| "eval_steps_per_second": 0.877, | |
| "step": 4551 | |
| }, | |
| { | |
| "epoch": 41.001636363636365, | |
| "grad_norm": 8.246631622314453, | |
| "learning_rate": 9.494949494949495e-06, | |
| "loss": 0.8189, | |
| "step": 4560 | |
| }, | |
| { | |
| "epoch": 41.003454545454545, | |
| "grad_norm": 2.8926265239715576, | |
| "learning_rate": 9.393939393939394e-06, | |
| "loss": 1.0385, | |
| "step": 4570 | |
| }, | |
| { | |
| "epoch": 41.005272727272725, | |
| "grad_norm": 9.945596694946289, | |
| "learning_rate": 9.292929292929294e-06, | |
| "loss": 0.8323, | |
| "step": 4580 | |
| }, | |
| { | |
| "epoch": 41.00709090909091, | |
| "grad_norm": 4.7715277671813965, | |
| "learning_rate": 9.191919191919192e-06, | |
| "loss": 0.8293, | |
| "step": 4590 | |
| }, | |
| { | |
| "epoch": 41.00890909090909, | |
| "grad_norm": 6.302114963531494, | |
| "learning_rate": 9.090909090909091e-06, | |
| "loss": 0.7819, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 41.01072727272727, | |
| "grad_norm": 10.754914283752441, | |
| "learning_rate": 8.98989898989899e-06, | |
| "loss": 0.8525, | |
| "step": 4610 | |
| }, | |
| { | |
| "epoch": 41.01254545454545, | |
| "grad_norm": 3.367741346359253, | |
| "learning_rate": 8.88888888888889e-06, | |
| "loss": 0.9213, | |
| "step": 4620 | |
| }, | |
| { | |
| "epoch": 41.01436363636363, | |
| "grad_norm": 4.3797407150268555, | |
| "learning_rate": 8.787878787878788e-06, | |
| "loss": 0.8236, | |
| "step": 4630 | |
| }, | |
| { | |
| "epoch": 41.01618181818182, | |
| "grad_norm": 4.827358722686768, | |
| "learning_rate": 8.686868686868687e-06, | |
| "loss": 0.8398, | |
| "step": 4640 | |
| }, | |
| { | |
| "epoch": 41.018, | |
| "grad_norm": 8.580107688903809, | |
| "learning_rate": 8.585858585858587e-06, | |
| "loss": 0.9439, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 41.01981818181818, | |
| "grad_norm": 8.491392135620117, | |
| "learning_rate": 8.484848484848486e-06, | |
| "loss": 0.7856, | |
| "step": 4660 | |
| }, | |
| { | |
| "epoch": 41.02018181818182, | |
| "eval_accuracy": 0.5915343915343916, | |
| "eval_loss": 0.9130398035049438, | |
| "eval_runtime": 134.1436, | |
| "eval_samples_per_second": 7.045, | |
| "eval_steps_per_second": 0.887, | |
| "step": 4662 | |
| }, | |
| { | |
| "epoch": 42.00145454545454, | |
| "grad_norm": 6.375803470611572, | |
| "learning_rate": 8.383838383838384e-06, | |
| "loss": 0.7439, | |
| "step": 4670 | |
| }, | |
| { | |
| "epoch": 42.00327272727273, | |
| "grad_norm": 3.816681146621704, | |
| "learning_rate": 8.282828282828283e-06, | |
| "loss": 0.8752, | |
| "step": 4680 | |
| }, | |
| { | |
| "epoch": 42.00509090909091, | |
| "grad_norm": 8.415267944335938, | |
| "learning_rate": 8.181818181818183e-06, | |
| "loss": 0.8966, | |
| "step": 4690 | |
| }, | |
| { | |
| "epoch": 42.00690909090909, | |
| "grad_norm": 5.004403114318848, | |
| "learning_rate": 8.080808080808082e-06, | |
| "loss": 0.7916, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 42.00872727272727, | |
| "grad_norm": 8.591304779052734, | |
| "learning_rate": 7.97979797979798e-06, | |
| "loss": 0.8525, | |
| "step": 4710 | |
| }, | |
| { | |
| "epoch": 42.01054545454546, | |
| "grad_norm": 4.411281585693359, | |
| "learning_rate": 7.878787878787878e-06, | |
| "loss": 0.7593, | |
| "step": 4720 | |
| }, | |
| { | |
| "epoch": 42.01236363636364, | |
| "grad_norm": 5.2794718742370605, | |
| "learning_rate": 7.777777777777777e-06, | |
| "loss": 0.984, | |
| "step": 4730 | |
| }, | |
| { | |
| "epoch": 42.01418181818182, | |
| "grad_norm": 5.923099517822266, | |
| "learning_rate": 7.676767676767677e-06, | |
| "loss": 0.8673, | |
| "step": 4740 | |
| }, | |
| { | |
| "epoch": 42.016, | |
| "grad_norm": 7.30630350112915, | |
| "learning_rate": 7.5757575757575764e-06, | |
| "loss": 0.8758, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 42.01781818181818, | |
| "grad_norm": 9.451834678649902, | |
| "learning_rate": 7.474747474747475e-06, | |
| "loss": 0.8842, | |
| "step": 4760 | |
| }, | |
| { | |
| "epoch": 42.019636363636366, | |
| "grad_norm": 5.710024833679199, | |
| "learning_rate": 7.3737373737373745e-06, | |
| "loss": 0.7133, | |
| "step": 4770 | |
| }, | |
| { | |
| "epoch": 42.02018181818182, | |
| "eval_accuracy": 0.5883597883597883, | |
| "eval_loss": 0.9080068469047546, | |
| "eval_runtime": 133.6506, | |
| "eval_samples_per_second": 7.071, | |
| "eval_steps_per_second": 0.89, | |
| "step": 4773 | |
| }, | |
| { | |
| "epoch": 43.00127272727273, | |
| "grad_norm": 5.078426361083984, | |
| "learning_rate": 7.272727272727272e-06, | |
| "loss": 0.7062, | |
| "step": 4780 | |
| }, | |
| { | |
| "epoch": 43.00309090909091, | |
| "grad_norm": 7.484274864196777, | |
| "learning_rate": 7.171717171717173e-06, | |
| "loss": 0.965, | |
| "step": 4790 | |
| }, | |
| { | |
| "epoch": 43.00490909090909, | |
| "grad_norm": 4.4427008628845215, | |
| "learning_rate": 7.0707070707070704e-06, | |
| "loss": 0.7688, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 43.006727272727275, | |
| "grad_norm": 6.1136627197265625, | |
| "learning_rate": 6.969696969696971e-06, | |
| "loss": 0.7731, | |
| "step": 4810 | |
| }, | |
| { | |
| "epoch": 43.008545454545455, | |
| "grad_norm": 8.658069610595703, | |
| "learning_rate": 6.8686868686868685e-06, | |
| "loss": 0.8958, | |
| "step": 4820 | |
| }, | |
| { | |
| "epoch": 43.010363636363635, | |
| "grad_norm": 5.886534214019775, | |
| "learning_rate": 6.767676767676769e-06, | |
| "loss": 0.9499, | |
| "step": 4830 | |
| }, | |
| { | |
| "epoch": 43.012181818181816, | |
| "grad_norm": 10.104177474975586, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.8922, | |
| "step": 4840 | |
| }, | |
| { | |
| "epoch": 43.014, | |
| "grad_norm": 8.859838485717773, | |
| "learning_rate": 6.565656565656567e-06, | |
| "loss": 0.9673, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 43.01581818181818, | |
| "grad_norm": 5.677845001220703, | |
| "learning_rate": 6.464646464646465e-06, | |
| "loss": 0.8809, | |
| "step": 4860 | |
| }, | |
| { | |
| "epoch": 43.01763636363636, | |
| "grad_norm": 6.806046485900879, | |
| "learning_rate": 6.363636363636363e-06, | |
| "loss": 0.831, | |
| "step": 4870 | |
| }, | |
| { | |
| "epoch": 43.01945454545454, | |
| "grad_norm": 7.888843059539795, | |
| "learning_rate": 6.262626262626263e-06, | |
| "loss": 0.932, | |
| "step": 4880 | |
| }, | |
| { | |
| "epoch": 43.02018181818182, | |
| "eval_accuracy": 0.5576719576719577, | |
| "eval_loss": 0.9387907385826111, | |
| "eval_runtime": 137.3927, | |
| "eval_samples_per_second": 6.878, | |
| "eval_steps_per_second": 0.866, | |
| "step": 4884 | |
| }, | |
| { | |
| "epoch": 44.00109090909091, | |
| "grad_norm": 9.031815528869629, | |
| "learning_rate": 6.161616161616162e-06, | |
| "loss": 0.9602, | |
| "step": 4890 | |
| }, | |
| { | |
| "epoch": 44.00290909090909, | |
| "grad_norm": 8.353316307067871, | |
| "learning_rate": 6.060606060606061e-06, | |
| "loss": 0.8246, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 44.00472727272727, | |
| "grad_norm": 4.821561336517334, | |
| "learning_rate": 5.9595959595959605e-06, | |
| "loss": 0.8246, | |
| "step": 4910 | |
| }, | |
| { | |
| "epoch": 44.00654545454545, | |
| "grad_norm": 9.228032112121582, | |
| "learning_rate": 5.858585858585859e-06, | |
| "loss": 0.9177, | |
| "step": 4920 | |
| }, | |
| { | |
| "epoch": 44.00836363636363, | |
| "grad_norm": 6.226870059967041, | |
| "learning_rate": 5.7575757575757586e-06, | |
| "loss": 0.8679, | |
| "step": 4930 | |
| }, | |
| { | |
| "epoch": 44.01018181818182, | |
| "grad_norm": 11.385287284851074, | |
| "learning_rate": 5.656565656565657e-06, | |
| "loss": 0.9084, | |
| "step": 4940 | |
| }, | |
| { | |
| "epoch": 44.012, | |
| "grad_norm": 9.927154541015625, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.7855, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 44.01381818181818, | |
| "grad_norm": 6.668172836303711, | |
| "learning_rate": 5.4545454545454545e-06, | |
| "loss": 0.7826, | |
| "step": 4960 | |
| }, | |
| { | |
| "epoch": 44.01563636363636, | |
| "grad_norm": 4.65692663192749, | |
| "learning_rate": 5.353535353535354e-06, | |
| "loss": 0.8195, | |
| "step": 4970 | |
| }, | |
| { | |
| "epoch": 44.01745454545455, | |
| "grad_norm": 5.107462406158447, | |
| "learning_rate": 5.2525252525252526e-06, | |
| "loss": 0.9274, | |
| "step": 4980 | |
| }, | |
| { | |
| "epoch": 44.01927272727273, | |
| "grad_norm": 7.147744655609131, | |
| "learning_rate": 5.151515151515152e-06, | |
| "loss": 0.6883, | |
| "step": 4990 | |
| }, | |
| { | |
| "epoch": 44.02018181818182, | |
| "eval_accuracy": 0.5936507936507937, | |
| "eval_loss": 0.892484724521637, | |
| "eval_runtime": 137.8997, | |
| "eval_samples_per_second": 6.853, | |
| "eval_steps_per_second": 0.863, | |
| "step": 4995 | |
| }, | |
| { | |
| "epoch": 45.00090909090909, | |
| "grad_norm": 5.661710739135742, | |
| "learning_rate": 5.050505050505051e-06, | |
| "loss": 0.7446, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 45.00272727272727, | |
| "grad_norm": 3.658931016921997, | |
| "learning_rate": 4.949494949494949e-06, | |
| "loss": 0.9806, | |
| "step": 5010 | |
| }, | |
| { | |
| "epoch": 45.00454545454546, | |
| "grad_norm": 4.7141194343566895, | |
| "learning_rate": 4.848484848484849e-06, | |
| "loss": 0.7656, | |
| "step": 5020 | |
| }, | |
| { | |
| "epoch": 45.00636363636364, | |
| "grad_norm": 3.591520309448242, | |
| "learning_rate": 4.747474747474747e-06, | |
| "loss": 0.9483, | |
| "step": 5030 | |
| }, | |
| { | |
| "epoch": 45.00818181818182, | |
| "grad_norm": 9.716415405273438, | |
| "learning_rate": 4.646464646464647e-06, | |
| "loss": 0.8558, | |
| "step": 5040 | |
| }, | |
| { | |
| "epoch": 45.01, | |
| "grad_norm": 7.835188388824463, | |
| "learning_rate": 4.5454545454545455e-06, | |
| "loss": 0.8277, | |
| "step": 5050 | |
| }, | |
| { | |
| "epoch": 45.011818181818185, | |
| "grad_norm": 5.499917984008789, | |
| "learning_rate": 4.444444444444445e-06, | |
| "loss": 0.7901, | |
| "step": 5060 | |
| }, | |
| { | |
| "epoch": 45.013636363636365, | |
| "grad_norm": 6.486261367797852, | |
| "learning_rate": 4.343434343434344e-06, | |
| "loss": 0.6807, | |
| "step": 5070 | |
| }, | |
| { | |
| "epoch": 45.015454545454546, | |
| "grad_norm": 5.808882713317871, | |
| "learning_rate": 4.242424242424243e-06, | |
| "loss": 0.9827, | |
| "step": 5080 | |
| }, | |
| { | |
| "epoch": 45.017272727272726, | |
| "grad_norm": 8.00777530670166, | |
| "learning_rate": 4.141414141414142e-06, | |
| "loss": 0.8919, | |
| "step": 5090 | |
| }, | |
| { | |
| "epoch": 45.019090909090906, | |
| "grad_norm": 9.413071632385254, | |
| "learning_rate": 4.040404040404041e-06, | |
| "loss": 0.9944, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 45.02018181818182, | |
| "eval_accuracy": 0.582010582010582, | |
| "eval_loss": 0.9143127202987671, | |
| "eval_runtime": 137.4856, | |
| "eval_samples_per_second": 6.873, | |
| "eval_steps_per_second": 0.866, | |
| "step": 5106 | |
| }, | |
| { | |
| "epoch": 46.000727272727275, | |
| "grad_norm": 11.19728946685791, | |
| "learning_rate": 3.939393939393939e-06, | |
| "loss": 0.7968, | |
| "step": 5110 | |
| }, | |
| { | |
| "epoch": 46.002545454545455, | |
| "grad_norm": 3.2283504009246826, | |
| "learning_rate": 3.8383838383838385e-06, | |
| "loss": 0.8411, | |
| "step": 5120 | |
| }, | |
| { | |
| "epoch": 46.004363636363635, | |
| "grad_norm": 3.715587854385376, | |
| "learning_rate": 3.7373737373737375e-06, | |
| "loss": 0.926, | |
| "step": 5130 | |
| }, | |
| { | |
| "epoch": 46.006181818181815, | |
| "grad_norm": 6.313870906829834, | |
| "learning_rate": 3.636363636363636e-06, | |
| "loss": 0.9623, | |
| "step": 5140 | |
| }, | |
| { | |
| "epoch": 46.008, | |
| "grad_norm": 6.3307952880859375, | |
| "learning_rate": 3.5353535353535352e-06, | |
| "loss": 0.9009, | |
| "step": 5150 | |
| }, | |
| { | |
| "epoch": 46.00981818181818, | |
| "grad_norm": 5.783283710479736, | |
| "learning_rate": 3.4343434343434343e-06, | |
| "loss": 0.8005, | |
| "step": 5160 | |
| }, | |
| { | |
| "epoch": 46.01163636363636, | |
| "grad_norm": 5.894885063171387, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 0.7556, | |
| "step": 5170 | |
| }, | |
| { | |
| "epoch": 46.01345454545454, | |
| "grad_norm": 6.77593994140625, | |
| "learning_rate": 3.2323232323232324e-06, | |
| "loss": 0.9579, | |
| "step": 5180 | |
| }, | |
| { | |
| "epoch": 46.01527272727273, | |
| "grad_norm": 5.100948333740234, | |
| "learning_rate": 3.1313131313131314e-06, | |
| "loss": 0.8254, | |
| "step": 5190 | |
| }, | |
| { | |
| "epoch": 46.01709090909091, | |
| "grad_norm": 9.13286018371582, | |
| "learning_rate": 3.0303030303030305e-06, | |
| "loss": 0.8354, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 46.01890909090909, | |
| "grad_norm": 4.923421859741211, | |
| "learning_rate": 2.9292929292929295e-06, | |
| "loss": 0.8892, | |
| "step": 5210 | |
| }, | |
| { | |
| "epoch": 46.02018181818182, | |
| "eval_accuracy": 0.5883597883597883, | |
| "eval_loss": 0.9102571606636047, | |
| "eval_runtime": 139.3527, | |
| "eval_samples_per_second": 6.781, | |
| "eval_steps_per_second": 0.854, | |
| "step": 5217 | |
| }, | |
| { | |
| "epoch": 47.00054545454545, | |
| "grad_norm": 6.632534503936768, | |
| "learning_rate": 2.8282828282828286e-06, | |
| "loss": 0.904, | |
| "step": 5220 | |
| }, | |
| { | |
| "epoch": 47.00236363636364, | |
| "grad_norm": 19.992380142211914, | |
| "learning_rate": 2.7272727272727272e-06, | |
| "loss": 0.832, | |
| "step": 5230 | |
| }, | |
| { | |
| "epoch": 47.00418181818182, | |
| "grad_norm": 6.258913516998291, | |
| "learning_rate": 2.6262626262626263e-06, | |
| "loss": 0.9356, | |
| "step": 5240 | |
| }, | |
| { | |
| "epoch": 47.006, | |
| "grad_norm": 6.109975814819336, | |
| "learning_rate": 2.5252525252525253e-06, | |
| "loss": 0.8786, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 47.00781818181818, | |
| "grad_norm": 6.342895030975342, | |
| "learning_rate": 2.4242424242424244e-06, | |
| "loss": 0.9049, | |
| "step": 5260 | |
| }, | |
| { | |
| "epoch": 47.00963636363636, | |
| "grad_norm": 8.066116333007812, | |
| "learning_rate": 2.3232323232323234e-06, | |
| "loss": 0.8481, | |
| "step": 5270 | |
| }, | |
| { | |
| "epoch": 47.01145454545455, | |
| "grad_norm": 9.589644432067871, | |
| "learning_rate": 2.2222222222222225e-06, | |
| "loss": 0.9711, | |
| "step": 5280 | |
| }, | |
| { | |
| "epoch": 47.01327272727273, | |
| "grad_norm": 7.310944080352783, | |
| "learning_rate": 2.1212121212121216e-06, | |
| "loss": 0.8926, | |
| "step": 5290 | |
| }, | |
| { | |
| "epoch": 47.01509090909091, | |
| "grad_norm": 5.914597511291504, | |
| "learning_rate": 2.0202020202020206e-06, | |
| "loss": 0.8469, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 47.01690909090909, | |
| "grad_norm": 7.06303596496582, | |
| "learning_rate": 1.9191919191919192e-06, | |
| "loss": 0.773, | |
| "step": 5310 | |
| }, | |
| { | |
| "epoch": 47.018727272727276, | |
| "grad_norm": 7.100560665130615, | |
| "learning_rate": 1.818181818181818e-06, | |
| "loss": 0.9071, | |
| "step": 5320 | |
| }, | |
| { | |
| "epoch": 47.02018181818182, | |
| "eval_accuracy": 0.5904761904761905, | |
| "eval_loss": 0.9018059968948364, | |
| "eval_runtime": 138.8187, | |
| "eval_samples_per_second": 6.807, | |
| "eval_steps_per_second": 0.857, | |
| "step": 5328 | |
| }, | |
| { | |
| "epoch": 48.00036363636364, | |
| "grad_norm": 7.776629447937012, | |
| "learning_rate": 1.7171717171717171e-06, | |
| "loss": 0.7373, | |
| "step": 5330 | |
| }, | |
| { | |
| "epoch": 48.00218181818182, | |
| "grad_norm": 7.495334148406982, | |
| "learning_rate": 1.6161616161616162e-06, | |
| "loss": 0.814, | |
| "step": 5340 | |
| }, | |
| { | |
| "epoch": 48.004, | |
| "grad_norm": 6.628978729248047, | |
| "learning_rate": 1.5151515151515152e-06, | |
| "loss": 0.8408, | |
| "step": 5350 | |
| }, | |
| { | |
| "epoch": 48.005818181818185, | |
| "grad_norm": 6.9452290534973145, | |
| "learning_rate": 1.4141414141414143e-06, | |
| "loss": 0.8563, | |
| "step": 5360 | |
| }, | |
| { | |
| "epoch": 48.007636363636365, | |
| "grad_norm": 4.360229015350342, | |
| "learning_rate": 1.3131313131313131e-06, | |
| "loss": 0.9381, | |
| "step": 5370 | |
| }, | |
| { | |
| "epoch": 48.009454545454545, | |
| "grad_norm": 4.341993808746338, | |
| "learning_rate": 1.2121212121212122e-06, | |
| "loss": 0.8765, | |
| "step": 5380 | |
| }, | |
| { | |
| "epoch": 48.011272727272726, | |
| "grad_norm": 8.818041801452637, | |
| "learning_rate": 1.1111111111111112e-06, | |
| "loss": 0.8021, | |
| "step": 5390 | |
| }, | |
| { | |
| "epoch": 48.013090909090906, | |
| "grad_norm": 3.769672155380249, | |
| "learning_rate": 1.0101010101010103e-06, | |
| "loss": 1.0232, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 48.01490909090909, | |
| "grad_norm": 4.771209716796875, | |
| "learning_rate": 9.09090909090909e-07, | |
| "loss": 0.878, | |
| "step": 5410 | |
| }, | |
| { | |
| "epoch": 48.01672727272727, | |
| "grad_norm": 3.48793363571167, | |
| "learning_rate": 8.080808080808081e-07, | |
| "loss": 0.8093, | |
| "step": 5420 | |
| }, | |
| { | |
| "epoch": 48.01854545454545, | |
| "grad_norm": 14.258705139160156, | |
| "learning_rate": 7.070707070707071e-07, | |
| "loss": 0.7943, | |
| "step": 5430 | |
| }, | |
| { | |
| "epoch": 48.02018181818182, | |
| "eval_accuracy": 0.5904761904761905, | |
| "eval_loss": 0.9021934270858765, | |
| "eval_runtime": 139.7838, | |
| "eval_samples_per_second": 6.76, | |
| "eval_steps_per_second": 0.851, | |
| "step": 5439 | |
| }, | |
| { | |
| "epoch": 49.000181818181815, | |
| "grad_norm": 5.376201152801514, | |
| "learning_rate": 6.060606060606061e-07, | |
| "loss": 0.9175, | |
| "step": 5440 | |
| }, | |
| { | |
| "epoch": 49.002, | |
| "grad_norm": 7.467062473297119, | |
| "learning_rate": 5.050505050505052e-07, | |
| "loss": 0.7526, | |
| "step": 5450 | |
| }, | |
| { | |
| "epoch": 49.00381818181818, | |
| "grad_norm": 7.232283115386963, | |
| "learning_rate": 4.0404040404040405e-07, | |
| "loss": 0.8092, | |
| "step": 5460 | |
| }, | |
| { | |
| "epoch": 49.00563636363636, | |
| "grad_norm": 3.167647123336792, | |
| "learning_rate": 3.0303030303030305e-07, | |
| "loss": 0.7768, | |
| "step": 5470 | |
| }, | |
| { | |
| "epoch": 49.00745454545454, | |
| "grad_norm": 3.3912150859832764, | |
| "learning_rate": 2.0202020202020202e-07, | |
| "loss": 0.8756, | |
| "step": 5480 | |
| }, | |
| { | |
| "epoch": 49.00927272727273, | |
| "grad_norm": 9.803848266601562, | |
| "learning_rate": 1.0101010101010101e-07, | |
| "loss": 0.8963, | |
| "step": 5490 | |
| }, | |
| { | |
| "epoch": 49.01109090909091, | |
| "grad_norm": 4.264191627502441, | |
| "learning_rate": 0.0, | |
| "loss": 0.8034, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 49.01109090909091, | |
| "eval_accuracy": 0.5947089947089947, | |
| "eval_loss": 0.9004086256027222, | |
| "eval_runtime": 136.9287, | |
| "eval_samples_per_second": 6.901, | |
| "eval_steps_per_second": 0.869, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 49.01109090909091, | |
| "step": 5500, | |
| "total_flos": 4.916643221082378e+19, | |
| "train_loss": 0.930074092431502, | |
| "train_runtime": 13888.1034, | |
| "train_samples_per_second": 3.168, | |
| "train_steps_per_second": 0.396 | |
| }, | |
| { | |
| "epoch": 49.01109090909091, | |
| "eval_accuracy": 0.5566137566137566, | |
| "eval_loss": 0.9280895590782166, | |
| "eval_runtime": 141.6112, | |
| "eval_samples_per_second": 6.673, | |
| "eval_steps_per_second": 0.84, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 49.01109090909091, | |
| "eval_accuracy": 0.5566137566137566, | |
| "eval_loss": 0.9280895590782166, | |
| "eval_runtime": 138.7006, | |
| "eval_samples_per_second": 6.813, | |
| "eval_steps_per_second": 0.858, | |
| "step": 5500 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 5500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "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": 4.916643221082378e+19, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |