| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.1554621848739495, | |
| "eval_steps": 50, | |
| "global_step": 1100, | |
| "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 | |
| } | |
| ], | |
| "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": 1.3173860578020557e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |