| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.2058823529411766, | |
| "eval_steps": 50, | |
| "global_step": 2100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01050420168067227, | |
| "grad_norm": 2.691126585006714, | |
| "learning_rate": 4.98249299719888e-05, | |
| "loss": 1.6756, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.02100840336134454, | |
| "grad_norm": 1.9124071598052979, | |
| "learning_rate": 4.96498599439776e-05, | |
| "loss": 0.5155, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.031512605042016806, | |
| "grad_norm": 1.1180588006973267, | |
| "learning_rate": 4.947478991596639e-05, | |
| "loss": 0.4521, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04201680672268908, | |
| "grad_norm": 0.7634907364845276, | |
| "learning_rate": 4.9299719887955186e-05, | |
| "loss": 0.3732, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.052521008403361345, | |
| "grad_norm": 0.6673978567123413, | |
| "learning_rate": 4.912464985994398e-05, | |
| "loss": 0.3832, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.052521008403361345, | |
| "eval_loss": 0.424630731344223, | |
| "eval_runtime": 13.4948, | |
| "eval_samples_per_second": 35.569, | |
| "eval_steps_per_second": 2.223, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.06302521008403361, | |
| "grad_norm": 0.6445392370223999, | |
| "learning_rate": 4.8949579831932775e-05, | |
| "loss": 0.3386, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.07352941176470588, | |
| "grad_norm": 0.5743226408958435, | |
| "learning_rate": 4.877450980392157e-05, | |
| "loss": 0.3443, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.08403361344537816, | |
| "grad_norm": 0.6735454201698303, | |
| "learning_rate": 4.859943977591036e-05, | |
| "loss": 0.3212, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.09453781512605042, | |
| "grad_norm": 0.660349428653717, | |
| "learning_rate": 4.8424369747899164e-05, | |
| "loss": 0.3379, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.10504201680672269, | |
| "grad_norm": 0.8036162853240967, | |
| "learning_rate": 4.824929971988796e-05, | |
| "loss": 0.3177, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.10504201680672269, | |
| "eval_loss": 0.3798685073852539, | |
| "eval_runtime": 13.5369, | |
| "eval_samples_per_second": 35.459, | |
| "eval_steps_per_second": 2.216, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.11554621848739496, | |
| "grad_norm": 0.6827962398529053, | |
| "learning_rate": 4.807422969187675e-05, | |
| "loss": 0.3405, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.12605042016806722, | |
| "grad_norm": 0.5923526287078857, | |
| "learning_rate": 4.7899159663865554e-05, | |
| "loss": 0.3324, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.13655462184873948, | |
| "grad_norm": 0.6476097106933594, | |
| "learning_rate": 4.772408963585435e-05, | |
| "loss": 0.3216, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.14705882352941177, | |
| "grad_norm": 0.5536680221557617, | |
| "learning_rate": 4.7549019607843135e-05, | |
| "loss": 0.3425, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.15756302521008403, | |
| "grad_norm": 0.5725419521331787, | |
| "learning_rate": 4.7373949579831936e-05, | |
| "loss": 0.3062, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.15756302521008403, | |
| "eval_loss": 0.36464938521385193, | |
| "eval_runtime": 13.5257, | |
| "eval_samples_per_second": 35.488, | |
| "eval_steps_per_second": 2.218, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.16806722689075632, | |
| "grad_norm": 0.6459401249885559, | |
| "learning_rate": 4.719887955182073e-05, | |
| "loss": 0.327, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.17857142857142858, | |
| "grad_norm": 0.7749695181846619, | |
| "learning_rate": 4.7023809523809525e-05, | |
| "loss": 0.3146, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.18907563025210083, | |
| "grad_norm": 0.7452580332756042, | |
| "learning_rate": 4.684873949579832e-05, | |
| "loss": 0.3024, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.19957983193277312, | |
| "grad_norm": 0.632286012172699, | |
| "learning_rate": 4.667366946778712e-05, | |
| "loss": 0.3157, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.21008403361344538, | |
| "grad_norm": 0.6443158984184265, | |
| "learning_rate": 4.6498599439775914e-05, | |
| "loss": 0.3039, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.21008403361344538, | |
| "eval_loss": 0.35029926896095276, | |
| "eval_runtime": 13.5332, | |
| "eval_samples_per_second": 35.468, | |
| "eval_steps_per_second": 2.217, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.22058823529411764, | |
| "grad_norm": 0.6045960783958435, | |
| "learning_rate": 4.632352941176471e-05, | |
| "loss": 0.313, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.23109243697478993, | |
| "grad_norm": 0.5544140338897705, | |
| "learning_rate": 4.61484593837535e-05, | |
| "loss": 0.2992, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.2415966386554622, | |
| "grad_norm": 0.45881810784339905, | |
| "learning_rate": 4.59733893557423e-05, | |
| "loss": 0.2889, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.25210084033613445, | |
| "grad_norm": 0.6315861344337463, | |
| "learning_rate": 4.579831932773109e-05, | |
| "loss": 0.3055, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.26260504201680673, | |
| "grad_norm": 0.7058202624320984, | |
| "learning_rate": 4.562324929971989e-05, | |
| "loss": 0.2897, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.26260504201680673, | |
| "eval_loss": 0.3452460467815399, | |
| "eval_runtime": 13.5256, | |
| "eval_samples_per_second": 35.488, | |
| "eval_steps_per_second": 2.218, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.27310924369747897, | |
| "grad_norm": 0.7277679443359375, | |
| "learning_rate": 4.5448179271708687e-05, | |
| "loss": 0.3119, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.28361344537815125, | |
| "grad_norm": 0.6311895251274109, | |
| "learning_rate": 4.527310924369748e-05, | |
| "loss": 0.296, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.29411764705882354, | |
| "grad_norm": 0.5600409507751465, | |
| "learning_rate": 4.5098039215686275e-05, | |
| "loss": 0.3027, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.30462184873949577, | |
| "grad_norm": 0.583111047744751, | |
| "learning_rate": 4.4922969187675076e-05, | |
| "loss": 0.3005, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.31512605042016806, | |
| "grad_norm": 0.6667417287826538, | |
| "learning_rate": 4.474789915966387e-05, | |
| "loss": 0.2924, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.31512605042016806, | |
| "eval_loss": 0.34308406710624695, | |
| "eval_runtime": 13.5267, | |
| "eval_samples_per_second": 35.485, | |
| "eval_steps_per_second": 2.218, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.32563025210084034, | |
| "grad_norm": 0.6434429883956909, | |
| "learning_rate": 4.4572829131652665e-05, | |
| "loss": 0.2893, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.33613445378151263, | |
| "grad_norm": 0.5765148401260376, | |
| "learning_rate": 4.439775910364146e-05, | |
| "loss": 0.2842, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.34663865546218486, | |
| "grad_norm": 0.5374988317489624, | |
| "learning_rate": 4.422268907563025e-05, | |
| "loss": 0.2838, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.35714285714285715, | |
| "grad_norm": 0.5548204779624939, | |
| "learning_rate": 4.404761904761905e-05, | |
| "loss": 0.2708, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.36764705882352944, | |
| "grad_norm": 0.5466386079788208, | |
| "learning_rate": 4.387254901960784e-05, | |
| "loss": 0.2833, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.36764705882352944, | |
| "eval_loss": 0.3345324695110321, | |
| "eval_runtime": 13.5203, | |
| "eval_samples_per_second": 35.502, | |
| "eval_steps_per_second": 2.219, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.37815126050420167, | |
| "grad_norm": 0.655726432800293, | |
| "learning_rate": 4.369747899159664e-05, | |
| "loss": 0.2842, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.38865546218487396, | |
| "grad_norm": 0.6633383631706238, | |
| "learning_rate": 4.352240896358544e-05, | |
| "loss": 0.2752, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.39915966386554624, | |
| "grad_norm": 0.549430787563324, | |
| "learning_rate": 4.334733893557423e-05, | |
| "loss": 0.2918, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.4096638655462185, | |
| "grad_norm": 0.49975866079330444, | |
| "learning_rate": 4.317226890756303e-05, | |
| "loss": 0.2733, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.42016806722689076, | |
| "grad_norm": 0.5723116397857666, | |
| "learning_rate": 4.2997198879551826e-05, | |
| "loss": 0.2852, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.42016806722689076, | |
| "eval_loss": 0.334602415561676, | |
| "eval_runtime": 13.5187, | |
| "eval_samples_per_second": 35.506, | |
| "eval_steps_per_second": 2.219, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.43067226890756305, | |
| "grad_norm": 0.6224997639656067, | |
| "learning_rate": 4.2822128851540614e-05, | |
| "loss": 0.2982, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.4411764705882353, | |
| "grad_norm": 0.548751711845398, | |
| "learning_rate": 4.2647058823529415e-05, | |
| "loss": 0.2867, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.45168067226890757, | |
| "grad_norm": 0.5709792971611023, | |
| "learning_rate": 4.247198879551821e-05, | |
| "loss": 0.2588, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.46218487394957986, | |
| "grad_norm": 0.537171483039856, | |
| "learning_rate": 4.2296918767507e-05, | |
| "loss": 0.2782, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.4726890756302521, | |
| "grad_norm": 0.5505544543266296, | |
| "learning_rate": 4.21218487394958e-05, | |
| "loss": 0.2703, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.4726890756302521, | |
| "eval_loss": 0.3293306231498718, | |
| "eval_runtime": 13.5091, | |
| "eval_samples_per_second": 35.532, | |
| "eval_steps_per_second": 2.221, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.4831932773109244, | |
| "grad_norm": 0.5049700736999512, | |
| "learning_rate": 4.19467787114846e-05, | |
| "loss": 0.2439, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.49369747899159666, | |
| "grad_norm": 0.6546272039413452, | |
| "learning_rate": 4.177170868347339e-05, | |
| "loss": 0.2747, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.5042016806722689, | |
| "grad_norm": 0.4740765690803528, | |
| "learning_rate": 4.159663865546219e-05, | |
| "loss": 0.2692, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.5147058823529411, | |
| "grad_norm": 0.5289779901504517, | |
| "learning_rate": 4.142156862745099e-05, | |
| "loss": 0.29, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.5252100840336135, | |
| "grad_norm": 0.5422542095184326, | |
| "learning_rate": 4.1246498599439776e-05, | |
| "loss": 0.2721, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.5252100840336135, | |
| "eval_loss": 0.32664668560028076, | |
| "eval_runtime": 13.4992, | |
| "eval_samples_per_second": 35.558, | |
| "eval_steps_per_second": 2.222, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.5357142857142857, | |
| "grad_norm": 0.586115837097168, | |
| "learning_rate": 4.107142857142857e-05, | |
| "loss": 0.2686, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.5462184873949579, | |
| "grad_norm": 0.608102560043335, | |
| "learning_rate": 4.089635854341737e-05, | |
| "loss": 0.2712, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.5567226890756303, | |
| "grad_norm": 0.5184093117713928, | |
| "learning_rate": 4.0721288515406165e-05, | |
| "loss": 0.2773, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.5672268907563025, | |
| "grad_norm": 0.524999737739563, | |
| "learning_rate": 4.054621848739496e-05, | |
| "loss": 0.2725, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.5777310924369747, | |
| "grad_norm": 0.6716265082359314, | |
| "learning_rate": 4.0371148459383754e-05, | |
| "loss": 0.2618, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.5777310924369747, | |
| "eval_loss": 0.32582366466522217, | |
| "eval_runtime": 13.4847, | |
| "eval_samples_per_second": 35.596, | |
| "eval_steps_per_second": 2.225, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.5882352941176471, | |
| "grad_norm": 0.5464914441108704, | |
| "learning_rate": 4.0196078431372555e-05, | |
| "loss": 0.2566, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.5987394957983193, | |
| "grad_norm": 0.5436871647834778, | |
| "learning_rate": 4.002100840336135e-05, | |
| "loss": 0.2637, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.6092436974789915, | |
| "grad_norm": 0.5792732238769531, | |
| "learning_rate": 3.984593837535014e-05, | |
| "loss": 0.2678, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.6197478991596639, | |
| "grad_norm": 0.5623672008514404, | |
| "learning_rate": 3.967086834733894e-05, | |
| "loss": 0.2589, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.6302521008403361, | |
| "grad_norm": 0.5475742220878601, | |
| "learning_rate": 3.949579831932773e-05, | |
| "loss": 0.2619, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.6302521008403361, | |
| "eval_loss": 0.3220784366130829, | |
| "eval_runtime": 13.4828, | |
| "eval_samples_per_second": 35.601, | |
| "eval_steps_per_second": 2.225, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.6407563025210085, | |
| "grad_norm": 0.5120112895965576, | |
| "learning_rate": 3.9320728291316526e-05, | |
| "loss": 0.2551, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.6512605042016807, | |
| "grad_norm": 0.5247055292129517, | |
| "learning_rate": 3.914565826330533e-05, | |
| "loss": 0.2578, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.6617647058823529, | |
| "grad_norm": 0.4846402406692505, | |
| "learning_rate": 3.897058823529412e-05, | |
| "loss": 0.2686, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.6722689075630253, | |
| "grad_norm": 0.5671103000640869, | |
| "learning_rate": 3.8795518207282915e-05, | |
| "loss": 0.2773, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.6827731092436975, | |
| "grad_norm": 0.5367168188095093, | |
| "learning_rate": 3.862044817927171e-05, | |
| "loss": 0.2632, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.6827731092436975, | |
| "eval_loss": 0.3196863830089569, | |
| "eval_runtime": 13.4716, | |
| "eval_samples_per_second": 35.631, | |
| "eval_steps_per_second": 2.227, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.6932773109243697, | |
| "grad_norm": 0.6599178910255432, | |
| "learning_rate": 3.844537815126051e-05, | |
| "loss": 0.2603, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.7037815126050421, | |
| "grad_norm": 0.4438059628009796, | |
| "learning_rate": 3.82703081232493e-05, | |
| "loss": 0.2495, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.7142857142857143, | |
| "grad_norm": 0.5657041072845459, | |
| "learning_rate": 3.809523809523809e-05, | |
| "loss": 0.2361, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.7247899159663865, | |
| "grad_norm": 0.5392199754714966, | |
| "learning_rate": 3.792016806722689e-05, | |
| "loss": 0.2541, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 0.7352941176470589, | |
| "grad_norm": 0.5997769832611084, | |
| "learning_rate": 3.774509803921569e-05, | |
| "loss": 0.2643, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.7352941176470589, | |
| "eval_loss": 0.32053324580192566, | |
| "eval_runtime": 13.4863, | |
| "eval_samples_per_second": 35.592, | |
| "eval_steps_per_second": 2.224, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.7457983193277311, | |
| "grad_norm": 0.6396230459213257, | |
| "learning_rate": 3.757002801120448e-05, | |
| "loss": 0.2454, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 0.7563025210084033, | |
| "grad_norm": 0.4835526645183563, | |
| "learning_rate": 3.739495798319328e-05, | |
| "loss": 0.2529, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.7668067226890757, | |
| "grad_norm": 0.5814074277877808, | |
| "learning_rate": 3.721988795518208e-05, | |
| "loss": 0.2445, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 0.7773109243697479, | |
| "grad_norm": 0.4852081537246704, | |
| "learning_rate": 3.704481792717087e-05, | |
| "loss": 0.2522, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.7878151260504201, | |
| "grad_norm": 0.5790079236030579, | |
| "learning_rate": 3.6869747899159665e-05, | |
| "loss": 0.2561, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.7878151260504201, | |
| "eval_loss": 0.32097509503364563, | |
| "eval_runtime": 13.4958, | |
| "eval_samples_per_second": 35.567, | |
| "eval_steps_per_second": 2.223, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.7983193277310925, | |
| "grad_norm": 0.5733464360237122, | |
| "learning_rate": 3.669467787114846e-05, | |
| "loss": 0.248, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.8088235294117647, | |
| "grad_norm": 0.6230357885360718, | |
| "learning_rate": 3.6519607843137254e-05, | |
| "loss": 0.2536, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 0.819327731092437, | |
| "grad_norm": 0.6231834888458252, | |
| "learning_rate": 3.634453781512605e-05, | |
| "loss": 0.258, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.8298319327731093, | |
| "grad_norm": 0.5959641933441162, | |
| "learning_rate": 3.616946778711485e-05, | |
| "loss": 0.2412, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 0.8403361344537815, | |
| "grad_norm": 0.5139572024345398, | |
| "learning_rate": 3.5994397759103643e-05, | |
| "loss": 0.2516, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.8403361344537815, | |
| "eval_loss": 0.3214908242225647, | |
| "eval_runtime": 13.5134, | |
| "eval_samples_per_second": 35.52, | |
| "eval_steps_per_second": 2.22, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.8508403361344538, | |
| "grad_norm": 0.5790511965751648, | |
| "learning_rate": 3.581932773109244e-05, | |
| "loss": 0.2427, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 0.8613445378151261, | |
| "grad_norm": 0.5952211618423462, | |
| "learning_rate": 3.564425770308123e-05, | |
| "loss": 0.2673, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 0.8718487394957983, | |
| "grad_norm": 0.5401202440261841, | |
| "learning_rate": 3.546918767507003e-05, | |
| "loss": 0.2392, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 0.8823529411764706, | |
| "grad_norm": 0.5725980401039124, | |
| "learning_rate": 3.529411764705883e-05, | |
| "loss": 0.2381, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.8928571428571429, | |
| "grad_norm": 0.6116746664047241, | |
| "learning_rate": 3.511904761904762e-05, | |
| "loss": 0.2403, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.8928571428571429, | |
| "eval_loss": 0.3206528425216675, | |
| "eval_runtime": 13.5142, | |
| "eval_samples_per_second": 35.518, | |
| "eval_steps_per_second": 2.22, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.9033613445378151, | |
| "grad_norm": 0.49307090044021606, | |
| "learning_rate": 3.4943977591036416e-05, | |
| "loss": 0.2375, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 0.9138655462184874, | |
| "grad_norm": 0.6631060242652893, | |
| "learning_rate": 3.476890756302521e-05, | |
| "loss": 0.2416, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 0.9243697478991597, | |
| "grad_norm": 0.5449891090393066, | |
| "learning_rate": 3.4593837535014004e-05, | |
| "loss": 0.2614, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 0.9348739495798319, | |
| "grad_norm": 0.5289740562438965, | |
| "learning_rate": 3.4418767507002805e-05, | |
| "loss": 0.2388, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 0.9453781512605042, | |
| "grad_norm": 0.5789199471473694, | |
| "learning_rate": 3.42436974789916e-05, | |
| "loss": 0.2461, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.9453781512605042, | |
| "eval_loss": 0.32027536630630493, | |
| "eval_runtime": 13.5157, | |
| "eval_samples_per_second": 35.514, | |
| "eval_steps_per_second": 2.22, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.9558823529411765, | |
| "grad_norm": 0.5729469060897827, | |
| "learning_rate": 3.4068627450980394e-05, | |
| "loss": 0.2393, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 0.9663865546218487, | |
| "grad_norm": 0.6298367977142334, | |
| "learning_rate": 3.389355742296919e-05, | |
| "loss": 0.2512, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 0.976890756302521, | |
| "grad_norm": 0.6292509436607361, | |
| "learning_rate": 3.371848739495799e-05, | |
| "loss": 0.2374, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 0.9873949579831933, | |
| "grad_norm": 0.6284404397010803, | |
| "learning_rate": 3.3543417366946776e-05, | |
| "loss": 0.2391, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 0.9978991596638656, | |
| "grad_norm": 0.5431928634643555, | |
| "learning_rate": 3.336834733893557e-05, | |
| "loss": 0.2487, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.9978991596638656, | |
| "eval_loss": 0.31819915771484375, | |
| "eval_runtime": 13.5091, | |
| "eval_samples_per_second": 35.532, | |
| "eval_steps_per_second": 2.221, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.0084033613445378, | |
| "grad_norm": 0.47679176926612854, | |
| "learning_rate": 3.319327731092437e-05, | |
| "loss": 0.2011, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 1.01890756302521, | |
| "grad_norm": 0.7469049096107483, | |
| "learning_rate": 3.3018207282913166e-05, | |
| "loss": 0.1979, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 1.0294117647058822, | |
| "grad_norm": 0.5701093077659607, | |
| "learning_rate": 3.284313725490196e-05, | |
| "loss": 0.1986, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 1.0399159663865547, | |
| "grad_norm": 0.5581035017967224, | |
| "learning_rate": 3.266806722689076e-05, | |
| "loss": 0.2039, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 1.050420168067227, | |
| "grad_norm": 0.6445596218109131, | |
| "learning_rate": 3.2492997198879555e-05, | |
| "loss": 0.2183, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.050420168067227, | |
| "eval_loss": 0.3302117586135864, | |
| "eval_runtime": 13.4775, | |
| "eval_samples_per_second": 35.615, | |
| "eval_steps_per_second": 2.226, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0609243697478992, | |
| "grad_norm": 0.48644450306892395, | |
| "learning_rate": 3.231792717086835e-05, | |
| "loss": 0.2044, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 1.0714285714285714, | |
| "grad_norm": 0.555849552154541, | |
| "learning_rate": 3.2142857142857144e-05, | |
| "loss": 0.2029, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 1.0819327731092436, | |
| "grad_norm": 0.657832145690918, | |
| "learning_rate": 3.196778711484594e-05, | |
| "loss": 0.2022, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 1.092436974789916, | |
| "grad_norm": 0.5357531309127808, | |
| "learning_rate": 3.179271708683473e-05, | |
| "loss": 0.2033, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 1.1029411764705883, | |
| "grad_norm": 0.6504517793655396, | |
| "learning_rate": 3.161764705882353e-05, | |
| "loss": 0.2029, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.1029411764705883, | |
| "eval_loss": 0.33139654994010925, | |
| "eval_runtime": 13.4851, | |
| "eval_samples_per_second": 35.595, | |
| "eval_steps_per_second": 2.225, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.1134453781512605, | |
| "grad_norm": 0.6484816670417786, | |
| "learning_rate": 3.144257703081233e-05, | |
| "loss": 0.2, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 1.1239495798319328, | |
| "grad_norm": 0.5106396079063416, | |
| "learning_rate": 3.126750700280112e-05, | |
| "loss": 0.1903, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 1.134453781512605, | |
| "grad_norm": 0.624066948890686, | |
| "learning_rate": 3.1092436974789916e-05, | |
| "loss": 0.201, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 1.1449579831932772, | |
| "grad_norm": 0.6842947602272034, | |
| "learning_rate": 3.091736694677872e-05, | |
| "loss": 0.2057, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 1.1554621848739495, | |
| "grad_norm": 0.5321388244628906, | |
| "learning_rate": 3.074229691876751e-05, | |
| "loss": 0.2037, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.1554621848739495, | |
| "eval_loss": 0.33459416031837463, | |
| "eval_runtime": 13.4814, | |
| "eval_samples_per_second": 35.605, | |
| "eval_steps_per_second": 2.225, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.165966386554622, | |
| "grad_norm": 0.6166321635246277, | |
| "learning_rate": 3.0567226890756306e-05, | |
| "loss": 0.1956, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 1.1764705882352942, | |
| "grad_norm": 0.5974798798561096, | |
| "learning_rate": 3.0392156862745097e-05, | |
| "loss": 0.2099, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 1.1869747899159664, | |
| "grad_norm": 0.6835567951202393, | |
| "learning_rate": 3.0217086834733894e-05, | |
| "loss": 0.2029, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 1.1974789915966386, | |
| "grad_norm": 0.6046004295349121, | |
| "learning_rate": 3.004201680672269e-05, | |
| "loss": 0.1952, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 1.2079831932773109, | |
| "grad_norm": 0.6110156178474426, | |
| "learning_rate": 2.9866946778711486e-05, | |
| "loss": 0.209, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 1.2079831932773109, | |
| "eval_loss": 0.3331316411495209, | |
| "eval_runtime": 13.4871, | |
| "eval_samples_per_second": 35.59, | |
| "eval_steps_per_second": 2.224, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 1.2184873949579833, | |
| "grad_norm": 0.6486967206001282, | |
| "learning_rate": 2.969187675070028e-05, | |
| "loss": 0.2068, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 1.2289915966386555, | |
| "grad_norm": 0.6242414712905884, | |
| "learning_rate": 2.9516806722689078e-05, | |
| "loss": 0.1926, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 1.2394957983193278, | |
| "grad_norm": 0.6824544072151184, | |
| "learning_rate": 2.9341736694677872e-05, | |
| "loss": 0.1985, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 0.6598100662231445, | |
| "learning_rate": 2.916666666666667e-05, | |
| "loss": 0.1968, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 1.2605042016806722, | |
| "grad_norm": 0.5652765035629272, | |
| "learning_rate": 2.8991596638655467e-05, | |
| "loss": 0.2055, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.2605042016806722, | |
| "eval_loss": 0.337155282497406, | |
| "eval_runtime": 13.5062, | |
| "eval_samples_per_second": 35.539, | |
| "eval_steps_per_second": 2.221, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.2710084033613445, | |
| "grad_norm": 0.6030492186546326, | |
| "learning_rate": 2.8816526610644258e-05, | |
| "loss": 0.2006, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 1.2815126050420167, | |
| "grad_norm": 0.6905779242515564, | |
| "learning_rate": 2.8641456582633052e-05, | |
| "loss": 0.1956, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 1.2920168067226891, | |
| "grad_norm": 0.6660848259925842, | |
| "learning_rate": 2.846638655462185e-05, | |
| "loss": 0.2072, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 1.3025210084033614, | |
| "grad_norm": 0.561509370803833, | |
| "learning_rate": 2.8291316526610644e-05, | |
| "loss": 0.1935, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 1.3130252100840336, | |
| "grad_norm": 0.9078445434570312, | |
| "learning_rate": 2.8116246498599442e-05, | |
| "loss": 0.2102, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 1.3130252100840336, | |
| "eval_loss": 0.3378737270832062, | |
| "eval_runtime": 13.5144, | |
| "eval_samples_per_second": 35.518, | |
| "eval_steps_per_second": 2.22, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 1.3235294117647058, | |
| "grad_norm": 0.782431960105896, | |
| "learning_rate": 2.7941176470588236e-05, | |
| "loss": 0.2044, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 1.334033613445378, | |
| "grad_norm": 0.7222577929496765, | |
| "learning_rate": 2.7766106442577034e-05, | |
| "loss": 0.2023, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 1.3445378151260505, | |
| "grad_norm": 0.7043188810348511, | |
| "learning_rate": 2.7591036414565828e-05, | |
| "loss": 0.2002, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 1.3550420168067228, | |
| "grad_norm": 0.5669278502464294, | |
| "learning_rate": 2.7415966386554626e-05, | |
| "loss": 0.1903, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 1.365546218487395, | |
| "grad_norm": 0.6404176950454712, | |
| "learning_rate": 2.7240896358543417e-05, | |
| "loss": 0.1982, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.365546218487395, | |
| "eval_loss": 0.33893150091171265, | |
| "eval_runtime": 13.5229, | |
| "eval_samples_per_second": 35.495, | |
| "eval_steps_per_second": 2.218, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.3760504201680672, | |
| "grad_norm": 0.6714435815811157, | |
| "learning_rate": 2.706582633053221e-05, | |
| "loss": 0.1954, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 1.3865546218487395, | |
| "grad_norm": 0.5384875535964966, | |
| "learning_rate": 2.689075630252101e-05, | |
| "loss": 0.201, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 1.3970588235294117, | |
| "grad_norm": 0.7454679012298584, | |
| "learning_rate": 2.6715686274509806e-05, | |
| "loss": 0.1973, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 1.407563025210084, | |
| "grad_norm": 0.7705392837524414, | |
| "learning_rate": 2.65406162464986e-05, | |
| "loss": 0.2001, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 1.4180672268907564, | |
| "grad_norm": 0.6535789370536804, | |
| "learning_rate": 2.6365546218487398e-05, | |
| "loss": 0.189, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 1.4180672268907564, | |
| "eval_loss": 0.33827126026153564, | |
| "eval_runtime": 13.5137, | |
| "eval_samples_per_second": 35.52, | |
| "eval_steps_per_second": 2.22, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "grad_norm": 0.6305943727493286, | |
| "learning_rate": 2.6190476190476192e-05, | |
| "loss": 0.1894, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 1.4390756302521008, | |
| "grad_norm": 0.7726532816886902, | |
| "learning_rate": 2.601540616246499e-05, | |
| "loss": 0.1971, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 1.449579831932773, | |
| "grad_norm": 0.814151406288147, | |
| "learning_rate": 2.5840336134453784e-05, | |
| "loss": 0.198, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 1.4600840336134453, | |
| "grad_norm": 0.6826948523521423, | |
| "learning_rate": 2.5665266106442575e-05, | |
| "loss": 0.1913, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 1.4705882352941178, | |
| "grad_norm": 0.8195087909698486, | |
| "learning_rate": 2.5490196078431373e-05, | |
| "loss": 0.1814, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.4705882352941178, | |
| "eval_loss": 0.3441857695579529, | |
| "eval_runtime": 13.5128, | |
| "eval_samples_per_second": 35.522, | |
| "eval_steps_per_second": 2.22, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.48109243697479, | |
| "grad_norm": 0.54868084192276, | |
| "learning_rate": 2.5315126050420167e-05, | |
| "loss": 0.1954, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 1.4915966386554622, | |
| "grad_norm": 0.7310366630554199, | |
| "learning_rate": 2.5140056022408964e-05, | |
| "loss": 0.1986, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 1.5021008403361344, | |
| "grad_norm": 0.7019846439361572, | |
| "learning_rate": 2.4964985994397762e-05, | |
| "loss": 0.1946, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 1.5126050420168067, | |
| "grad_norm": 0.839182436466217, | |
| "learning_rate": 2.4789915966386556e-05, | |
| "loss": 0.196, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 1.523109243697479, | |
| "grad_norm": 0.6928364634513855, | |
| "learning_rate": 2.4614845938375354e-05, | |
| "loss": 0.2009, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 1.523109243697479, | |
| "eval_loss": 0.3419071137905121, | |
| "eval_runtime": 13.5014, | |
| "eval_samples_per_second": 35.552, | |
| "eval_steps_per_second": 2.222, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 1.5336134453781511, | |
| "grad_norm": 0.6039723753929138, | |
| "learning_rate": 2.4439775910364145e-05, | |
| "loss": 0.191, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 1.5441176470588234, | |
| "grad_norm": 0.6368713974952698, | |
| "learning_rate": 2.4264705882352942e-05, | |
| "loss": 0.1927, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 1.5546218487394958, | |
| "grad_norm": 0.7114837765693665, | |
| "learning_rate": 2.4089635854341737e-05, | |
| "loss": 0.1884, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 1.565126050420168, | |
| "grad_norm": 0.7814504504203796, | |
| "learning_rate": 2.3914565826330534e-05, | |
| "loss": 0.1974, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 1.5756302521008403, | |
| "grad_norm": 0.738370954990387, | |
| "learning_rate": 2.373949579831933e-05, | |
| "loss": 0.202, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.5756302521008403, | |
| "eval_loss": 0.34229767322540283, | |
| "eval_runtime": 13.5047, | |
| "eval_samples_per_second": 35.543, | |
| "eval_steps_per_second": 2.221, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.5861344537815127, | |
| "grad_norm": 0.7205056548118591, | |
| "learning_rate": 2.3564425770308123e-05, | |
| "loss": 0.1975, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 1.596638655462185, | |
| "grad_norm": 0.8840786218643188, | |
| "learning_rate": 2.338935574229692e-05, | |
| "loss": 0.1876, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 1.6071428571428572, | |
| "grad_norm": 0.7528152465820312, | |
| "learning_rate": 2.3214285714285715e-05, | |
| "loss": 0.193, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 1.6176470588235294, | |
| "grad_norm": 0.6973660588264465, | |
| "learning_rate": 2.303921568627451e-05, | |
| "loss": 0.1768, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 1.6281512605042017, | |
| "grad_norm": 0.8335407376289368, | |
| "learning_rate": 2.2864145658263307e-05, | |
| "loss": 0.1827, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 1.6281512605042017, | |
| "eval_loss": 0.3449813425540924, | |
| "eval_runtime": 13.4784, | |
| "eval_samples_per_second": 35.612, | |
| "eval_steps_per_second": 2.226, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 1.638655462184874, | |
| "grad_norm": 0.834244430065155, | |
| "learning_rate": 2.26890756302521e-05, | |
| "loss": 0.1908, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 1.6491596638655461, | |
| "grad_norm": 0.8551121354103088, | |
| "learning_rate": 2.25140056022409e-05, | |
| "loss": 0.186, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 1.6596638655462184, | |
| "grad_norm": 0.6920718550682068, | |
| "learning_rate": 2.2338935574229693e-05, | |
| "loss": 0.1841, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 1.6701680672268906, | |
| "grad_norm": 0.7398483157157898, | |
| "learning_rate": 2.2163865546218487e-05, | |
| "loss": 0.1897, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 1.680672268907563, | |
| "grad_norm": 0.9140138030052185, | |
| "learning_rate": 2.1988795518207285e-05, | |
| "loss": 0.1922, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.680672268907563, | |
| "eval_loss": 0.3434249460697174, | |
| "eval_runtime": 13.4753, | |
| "eval_samples_per_second": 35.621, | |
| "eval_steps_per_second": 2.226, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.6911764705882353, | |
| "grad_norm": 0.6723321080207825, | |
| "learning_rate": 2.181372549019608e-05, | |
| "loss": 0.1857, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 1.7016806722689075, | |
| "grad_norm": 0.6897566318511963, | |
| "learning_rate": 2.1638655462184876e-05, | |
| "loss": 0.1776, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 1.71218487394958, | |
| "grad_norm": 0.6626412272453308, | |
| "learning_rate": 2.146358543417367e-05, | |
| "loss": 0.1854, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 1.7226890756302522, | |
| "grad_norm": 0.7039874792098999, | |
| "learning_rate": 2.1288515406162465e-05, | |
| "loss": 0.1915, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 1.7331932773109244, | |
| "grad_norm": 0.7689247727394104, | |
| "learning_rate": 2.1113445378151263e-05, | |
| "loss": 0.1753, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 1.7331932773109244, | |
| "eval_loss": 0.3487314283847809, | |
| "eval_runtime": 13.4843, | |
| "eval_samples_per_second": 35.597, | |
| "eval_steps_per_second": 2.225, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 1.7436974789915967, | |
| "grad_norm": 0.6633319854736328, | |
| "learning_rate": 2.0938375350140057e-05, | |
| "loss": 0.187, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 1.754201680672269, | |
| "grad_norm": 0.7929345369338989, | |
| "learning_rate": 2.0763305322128854e-05, | |
| "loss": 0.1796, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 1.7647058823529411, | |
| "grad_norm": 0.6714622378349304, | |
| "learning_rate": 2.058823529411765e-05, | |
| "loss": 0.181, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 1.7752100840336134, | |
| "grad_norm": 0.6277990341186523, | |
| "learning_rate": 2.0413165266106443e-05, | |
| "loss": 0.1852, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 1.7857142857142856, | |
| "grad_norm": 0.6239778995513916, | |
| "learning_rate": 2.023809523809524e-05, | |
| "loss": 0.1929, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.7857142857142856, | |
| "eval_loss": 0.3445141911506653, | |
| "eval_runtime": 13.4945, | |
| "eval_samples_per_second": 35.57, | |
| "eval_steps_per_second": 2.223, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.7962184873949578, | |
| "grad_norm": 0.6844814419746399, | |
| "learning_rate": 2.0063025210084035e-05, | |
| "loss": 0.1821, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 1.8067226890756303, | |
| "grad_norm": 0.663127601146698, | |
| "learning_rate": 1.988795518207283e-05, | |
| "loss": 0.1837, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 1.8172268907563025, | |
| "grad_norm": 0.5808536410331726, | |
| "learning_rate": 1.9712885154061627e-05, | |
| "loss": 0.1775, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 1.8277310924369747, | |
| "grad_norm": 0.7033660411834717, | |
| "learning_rate": 1.953781512605042e-05, | |
| "loss": 0.1791, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 1.8382352941176472, | |
| "grad_norm": 0.8857737183570862, | |
| "learning_rate": 1.936274509803922e-05, | |
| "loss": 0.1842, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 1.8382352941176472, | |
| "eval_loss": 0.35038644075393677, | |
| "eval_runtime": 13.508, | |
| "eval_samples_per_second": 35.534, | |
| "eval_steps_per_second": 2.221, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 1.8487394957983194, | |
| "grad_norm": 0.7379726767539978, | |
| "learning_rate": 1.9187675070028013e-05, | |
| "loss": 0.184, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 1.8592436974789917, | |
| "grad_norm": 0.704282283782959, | |
| "learning_rate": 1.9012605042016807e-05, | |
| "loss": 0.1761, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 1.8697478991596639, | |
| "grad_norm": 0.6162270307540894, | |
| "learning_rate": 1.88375350140056e-05, | |
| "loss": 0.1804, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 1.8802521008403361, | |
| "grad_norm": 0.8076377511024475, | |
| "learning_rate": 1.86624649859944e-05, | |
| "loss": 0.1747, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 1.8907563025210083, | |
| "grad_norm": 0.6141780018806458, | |
| "learning_rate": 1.8487394957983196e-05, | |
| "loss": 0.1717, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.8907563025210083, | |
| "eval_loss": 0.3557831943035126, | |
| "eval_runtime": 13.5096, | |
| "eval_samples_per_second": 35.53, | |
| "eval_steps_per_second": 2.221, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.9012605042016806, | |
| "grad_norm": 0.6518354415893555, | |
| "learning_rate": 1.8312324929971987e-05, | |
| "loss": 0.1762, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 1.9117647058823528, | |
| "grad_norm": 0.8053015470504761, | |
| "learning_rate": 1.8137254901960785e-05, | |
| "loss": 0.1825, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 1.9222689075630253, | |
| "grad_norm": 0.7303296327590942, | |
| "learning_rate": 1.796218487394958e-05, | |
| "loss": 0.1817, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 1.9327731092436975, | |
| "grad_norm": 0.6893795132637024, | |
| "learning_rate": 1.7787114845938377e-05, | |
| "loss": 0.1709, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 1.9432773109243697, | |
| "grad_norm": 0.6915561556816101, | |
| "learning_rate": 1.7612044817927174e-05, | |
| "loss": 0.1738, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 1.9432773109243697, | |
| "eval_loss": 0.35537341237068176, | |
| "eval_runtime": 13.5136, | |
| "eval_samples_per_second": 35.52, | |
| "eval_steps_per_second": 2.22, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 1.9537815126050422, | |
| "grad_norm": 0.6528658270835876, | |
| "learning_rate": 1.7436974789915965e-05, | |
| "loss": 0.1732, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 1.9642857142857144, | |
| "grad_norm": 0.6105799674987793, | |
| "learning_rate": 1.7261904761904763e-05, | |
| "loss": 0.1813, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 1.9747899159663866, | |
| "grad_norm": 0.7751848697662354, | |
| "learning_rate": 1.7086834733893557e-05, | |
| "loss": 0.1738, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 1.9852941176470589, | |
| "grad_norm": 0.7051230072975159, | |
| "learning_rate": 1.6911764705882355e-05, | |
| "loss": 0.1746, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 1.995798319327731, | |
| "grad_norm": 0.6930919885635376, | |
| "learning_rate": 1.673669467787115e-05, | |
| "loss": 0.1702, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.995798319327731, | |
| "eval_loss": 0.36112526059150696, | |
| "eval_runtime": 13.5145, | |
| "eval_samples_per_second": 35.517, | |
| "eval_steps_per_second": 2.22, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.0063025210084033, | |
| "grad_norm": 0.5415343046188354, | |
| "learning_rate": 1.6561624649859943e-05, | |
| "loss": 0.1526, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 2.0168067226890756, | |
| "grad_norm": 0.8496296405792236, | |
| "learning_rate": 1.638655462184874e-05, | |
| "loss": 0.1494, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 2.027310924369748, | |
| "grad_norm": 0.6553441882133484, | |
| "learning_rate": 1.6211484593837535e-05, | |
| "loss": 0.1451, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 2.03781512605042, | |
| "grad_norm": 0.7730965614318848, | |
| "learning_rate": 1.6036414565826333e-05, | |
| "loss": 0.1471, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 2.0483193277310923, | |
| "grad_norm": 0.6922443509101868, | |
| "learning_rate": 1.5861344537815127e-05, | |
| "loss": 0.1492, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 2.0483193277310923, | |
| "eval_loss": 0.3929571807384491, | |
| "eval_runtime": 13.5045, | |
| "eval_samples_per_second": 35.544, | |
| "eval_steps_per_second": 2.221, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 2.0588235294117645, | |
| "grad_norm": 0.7806298136711121, | |
| "learning_rate": 1.568627450980392e-05, | |
| "loss": 0.151, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 2.069327731092437, | |
| "grad_norm": 0.8226228356361389, | |
| "learning_rate": 1.551120448179272e-05, | |
| "loss": 0.1467, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 2.0798319327731094, | |
| "grad_norm": 0.6562792658805847, | |
| "learning_rate": 1.5336134453781513e-05, | |
| "loss": 0.1503, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 2.0903361344537816, | |
| "grad_norm": 0.693579912185669, | |
| "learning_rate": 1.5161064425770307e-05, | |
| "loss": 0.1471, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 2.100840336134454, | |
| "grad_norm": 0.665643036365509, | |
| "learning_rate": 1.4985994397759103e-05, | |
| "loss": 0.1525, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.100840336134454, | |
| "eval_loss": 0.3944668471813202, | |
| "eval_runtime": 13.5099, | |
| "eval_samples_per_second": 35.529, | |
| "eval_steps_per_second": 2.221, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.111344537815126, | |
| "grad_norm": 0.5635168552398682, | |
| "learning_rate": 1.48109243697479e-05, | |
| "loss": 0.1476, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 2.1218487394957983, | |
| "grad_norm": 0.6278943419456482, | |
| "learning_rate": 1.4635854341736697e-05, | |
| "loss": 0.1478, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 2.1323529411764706, | |
| "grad_norm": 0.783109724521637, | |
| "learning_rate": 1.4460784313725493e-05, | |
| "loss": 0.1441, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 2.142857142857143, | |
| "grad_norm": 0.8237671256065369, | |
| "learning_rate": 1.4285714285714285e-05, | |
| "loss": 0.1456, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 2.153361344537815, | |
| "grad_norm": 0.8787814974784851, | |
| "learning_rate": 1.4110644257703081e-05, | |
| "loss": 0.1531, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 2.153361344537815, | |
| "eval_loss": 0.4012806713581085, | |
| "eval_runtime": 13.532, | |
| "eval_samples_per_second": 35.471, | |
| "eval_steps_per_second": 2.217, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 2.1638655462184873, | |
| "grad_norm": 0.6932814121246338, | |
| "learning_rate": 1.3935574229691877e-05, | |
| "loss": 0.1435, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 2.1743697478991595, | |
| "grad_norm": 0.6396945118904114, | |
| "learning_rate": 1.3760504201680673e-05, | |
| "loss": 0.1485, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 2.184873949579832, | |
| "grad_norm": 0.6283547878265381, | |
| "learning_rate": 1.3585434173669467e-05, | |
| "loss": 0.1522, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 2.1953781512605044, | |
| "grad_norm": 0.701211154460907, | |
| "learning_rate": 1.3410364145658263e-05, | |
| "loss": 0.1454, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 2.2058823529411766, | |
| "grad_norm": 0.7193111181259155, | |
| "learning_rate": 1.323529411764706e-05, | |
| "loss": 0.1466, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.2058823529411766, | |
| "eval_loss": 0.40620240569114685, | |
| "eval_runtime": 13.5287, | |
| "eval_samples_per_second": 35.48, | |
| "eval_steps_per_second": 2.218, | |
| "step": 2100 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 2856, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.516910569422848e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |