| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 2000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.025, | |
| "grad_norm": 37.31816101074219, | |
| "learning_rate": 1.9900000000000003e-05, | |
| "loss": 10.1104, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 35.192100524902344, | |
| "learning_rate": 1.98e-05, | |
| "loss": 7.9082, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.075, | |
| "grad_norm": 25.770776748657227, | |
| "learning_rate": 1.97e-05, | |
| "loss": 7.2319, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 31.321504592895508, | |
| "learning_rate": 1.9600000000000002e-05, | |
| "loss": 6.9716, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.125, | |
| "grad_norm": 34.35246658325195, | |
| "learning_rate": 1.95e-05, | |
| "loss": 6.5322, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 33.44606018066406, | |
| "learning_rate": 1.94e-05, | |
| "loss": 6.1754, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.175, | |
| "grad_norm": 38.797672271728516, | |
| "learning_rate": 1.93e-05, | |
| "loss": 5.7241, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 28.263782501220703, | |
| "learning_rate": 1.9200000000000003e-05, | |
| "loss": 5.305, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.225, | |
| "grad_norm": 41.63703155517578, | |
| "learning_rate": 1.91e-05, | |
| "loss": 4.9392, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 31.407997131347656, | |
| "learning_rate": 1.9e-05, | |
| "loss": 4.4984, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.275, | |
| "grad_norm": 41.87642288208008, | |
| "learning_rate": 1.8900000000000002e-05, | |
| "loss": 4.1325, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 35.91765213012695, | |
| "learning_rate": 1.88e-05, | |
| "loss": 3.7626, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.325, | |
| "grad_norm": 25.258136749267578, | |
| "learning_rate": 1.8700000000000004e-05, | |
| "loss": 3.4081, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 20.95956802368164, | |
| "learning_rate": 1.86e-05, | |
| "loss": 3.0839, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.375, | |
| "grad_norm": 31.322267532348633, | |
| "learning_rate": 1.8500000000000002e-05, | |
| "loss": 2.8921, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 26.40801429748535, | |
| "learning_rate": 1.8400000000000003e-05, | |
| "loss": 2.4794, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.425, | |
| "grad_norm": 14.098700523376465, | |
| "learning_rate": 1.83e-05, | |
| "loss": 2.3116, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 13.986673355102539, | |
| "learning_rate": 1.8200000000000002e-05, | |
| "loss": 2.3836, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.475, | |
| "grad_norm": 9.944358825683594, | |
| "learning_rate": 1.8100000000000003e-05, | |
| "loss": 1.816, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 24.969406127929688, | |
| "learning_rate": 1.8e-05, | |
| "loss": 2.2833, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.525, | |
| "grad_norm": 8.329447746276855, | |
| "learning_rate": 1.79e-05, | |
| "loss": 1.868, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 13.586796760559082, | |
| "learning_rate": 1.7800000000000002e-05, | |
| "loss": 1.8955, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.575, | |
| "grad_norm": 9.99631118774414, | |
| "learning_rate": 1.77e-05, | |
| "loss": 1.7817, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 9.203180313110352, | |
| "learning_rate": 1.76e-05, | |
| "loss": 1.3639, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.625, | |
| "grad_norm": 14.064183235168457, | |
| "learning_rate": 1.7500000000000002e-05, | |
| "loss": 1.7589, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 3.787022590637207, | |
| "learning_rate": 1.7400000000000003e-05, | |
| "loss": 1.5226, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.675, | |
| "grad_norm": 8.696346282958984, | |
| "learning_rate": 1.73e-05, | |
| "loss": 1.6945, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 2.8470327854156494, | |
| "learning_rate": 1.72e-05, | |
| "loss": 1.515, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.725, | |
| "grad_norm": 14.381975173950195, | |
| "learning_rate": 1.7100000000000002e-05, | |
| "loss": 1.5426, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 11.33414077758789, | |
| "learning_rate": 1.7e-05, | |
| "loss": 1.9131, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.775, | |
| "grad_norm": 10.369054794311523, | |
| "learning_rate": 1.69e-05, | |
| "loss": 1.4884, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 10.790966987609863, | |
| "learning_rate": 1.6800000000000002e-05, | |
| "loss": 1.5494, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.825, | |
| "grad_norm": 1.5759797096252441, | |
| "learning_rate": 1.67e-05, | |
| "loss": 1.4403, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 7.573840141296387, | |
| "learning_rate": 1.66e-05, | |
| "loss": 1.6335, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.875, | |
| "grad_norm": 6.928182125091553, | |
| "learning_rate": 1.65e-05, | |
| "loss": 1.6648, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 10.622973442077637, | |
| "learning_rate": 1.64e-05, | |
| "loss": 1.5502, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.925, | |
| "grad_norm": 15.8394136428833, | |
| "learning_rate": 1.63e-05, | |
| "loss": 1.8488, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 8.823399543762207, | |
| "learning_rate": 1.62e-05, | |
| "loss": 1.5358, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.975, | |
| "grad_norm": 13.792437553405762, | |
| "learning_rate": 1.6100000000000002e-05, | |
| "loss": 1.407, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 12.505993843078613, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 1.57, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.6697252988815308, | |
| "eval_runtime": 177.9001, | |
| "eval_samples_per_second": 2.248, | |
| "eval_steps_per_second": 0.281, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.025, | |
| "grad_norm": 2.272650957107544, | |
| "learning_rate": 1.5900000000000004e-05, | |
| "loss": 1.5599, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 8.81256103515625, | |
| "learning_rate": 1.58e-05, | |
| "loss": 1.5504, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.075, | |
| "grad_norm": 9.009818077087402, | |
| "learning_rate": 1.5700000000000002e-05, | |
| "loss": 1.4481, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 10.511366844177246, | |
| "learning_rate": 1.5600000000000003e-05, | |
| "loss": 1.6972, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.125, | |
| "grad_norm": 1.3748583793640137, | |
| "learning_rate": 1.55e-05, | |
| "loss": 1.4778, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 8.614971160888672, | |
| "learning_rate": 1.54e-05, | |
| "loss": 1.9503, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.175, | |
| "grad_norm": 4.834426403045654, | |
| "learning_rate": 1.5300000000000003e-05, | |
| "loss": 1.5342, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 3.6863138675689697, | |
| "learning_rate": 1.5200000000000002e-05, | |
| "loss": 1.6797, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.225, | |
| "grad_norm": 2.9746885299682617, | |
| "learning_rate": 1.5100000000000001e-05, | |
| "loss": 1.6074, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 7.708632469177246, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 1.6568, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.275, | |
| "grad_norm": 4.480324745178223, | |
| "learning_rate": 1.4900000000000001e-05, | |
| "loss": 1.5063, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 5.0911173820495605, | |
| "learning_rate": 1.48e-05, | |
| "loss": 1.59, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.325, | |
| "grad_norm": 1.7006248235702515, | |
| "learning_rate": 1.4700000000000002e-05, | |
| "loss": 1.5901, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "grad_norm": 4.242706775665283, | |
| "learning_rate": 1.46e-05, | |
| "loss": 1.3782, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.375, | |
| "grad_norm": 7.67242956161499, | |
| "learning_rate": 1.45e-05, | |
| "loss": 1.4228, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 6.6367082595825195, | |
| "learning_rate": 1.4400000000000001e-05, | |
| "loss": 1.973, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.425, | |
| "grad_norm": 4.86959981918335, | |
| "learning_rate": 1.43e-05, | |
| "loss": 1.7341, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "grad_norm": 1.4599714279174805, | |
| "learning_rate": 1.4200000000000001e-05, | |
| "loss": 1.9434, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.475, | |
| "grad_norm": 1.5026886463165283, | |
| "learning_rate": 1.41e-05, | |
| "loss": 1.6237, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 1.1954498291015625, | |
| "learning_rate": 1.4e-05, | |
| "loss": 1.399, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.525, | |
| "grad_norm": 6.557270050048828, | |
| "learning_rate": 1.39e-05, | |
| "loss": 1.4375, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "grad_norm": 7.460885524749756, | |
| "learning_rate": 1.38e-05, | |
| "loss": 1.7234, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.575, | |
| "grad_norm": 15.711455345153809, | |
| "learning_rate": 1.3700000000000003e-05, | |
| "loss": 1.4367, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 4.1283392906188965, | |
| "learning_rate": 1.3600000000000002e-05, | |
| "loss": 1.1727, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.625, | |
| "grad_norm": 4.098735332489014, | |
| "learning_rate": 1.3500000000000001e-05, | |
| "loss": 1.4125, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "grad_norm": 7.408526420593262, | |
| "learning_rate": 1.3400000000000002e-05, | |
| "loss": 1.6468, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.675, | |
| "grad_norm": 5.501112937927246, | |
| "learning_rate": 1.3300000000000001e-05, | |
| "loss": 1.8357, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "grad_norm": 4.180479049682617, | |
| "learning_rate": 1.3200000000000002e-05, | |
| "loss": 1.3682, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.725, | |
| "grad_norm": 1.872931718826294, | |
| "learning_rate": 1.3100000000000002e-05, | |
| "loss": 1.5488, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 1.205283522605896, | |
| "learning_rate": 1.3000000000000001e-05, | |
| "loss": 1.6343, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.775, | |
| "grad_norm": 2.79819917678833, | |
| "learning_rate": 1.2900000000000002e-05, | |
| "loss": 1.6792, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 2.925333023071289, | |
| "learning_rate": 1.2800000000000001e-05, | |
| "loss": 1.63, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.825, | |
| "grad_norm": 11.69040298461914, | |
| "learning_rate": 1.27e-05, | |
| "loss": 1.8663, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "grad_norm": 3.8195672035217285, | |
| "learning_rate": 1.2600000000000001e-05, | |
| "loss": 1.4885, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 1.875, | |
| "grad_norm": 6.955050468444824, | |
| "learning_rate": 1.25e-05, | |
| "loss": 1.6633, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 3.9344043731689453, | |
| "learning_rate": 1.2400000000000002e-05, | |
| "loss": 1.6519, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 1.925, | |
| "grad_norm": 3.235872983932495, | |
| "learning_rate": 1.23e-05, | |
| "loss": 1.4941, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "grad_norm": 6.8774518966674805, | |
| "learning_rate": 1.22e-05, | |
| "loss": 1.694, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 1.975, | |
| "grad_norm": 1.166123867034912, | |
| "learning_rate": 1.2100000000000001e-05, | |
| "loss": 1.336, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 13.162934303283691, | |
| "learning_rate": 1.2e-05, | |
| "loss": 1.4602, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.6496708393096924, | |
| "eval_runtime": 179.3668, | |
| "eval_samples_per_second": 2.23, | |
| "eval_steps_per_second": 0.279, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.025, | |
| "grad_norm": 7.529654026031494, | |
| "learning_rate": 1.1900000000000001e-05, | |
| "loss": 1.5174, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "grad_norm": 14.058820724487305, | |
| "learning_rate": 1.18e-05, | |
| "loss": 1.8126, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.075, | |
| "grad_norm": 11.69058895111084, | |
| "learning_rate": 1.17e-05, | |
| "loss": 1.5155, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "grad_norm": 9.208683013916016, | |
| "learning_rate": 1.16e-05, | |
| "loss": 1.5014, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 2.125, | |
| "grad_norm": 10.896733283996582, | |
| "learning_rate": 1.15e-05, | |
| "loss": 1.5513, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "grad_norm": 1.682322382926941, | |
| "learning_rate": 1.14e-05, | |
| "loss": 1.6424, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 2.175, | |
| "grad_norm": 8.488654136657715, | |
| "learning_rate": 1.13e-05, | |
| "loss": 1.3778, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 7.200578689575195, | |
| "learning_rate": 1.1200000000000001e-05, | |
| "loss": 1.5253, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 2.225, | |
| "grad_norm": 2.4229137897491455, | |
| "learning_rate": 1.1100000000000002e-05, | |
| "loss": 1.6402, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "grad_norm": 1.4099198579788208, | |
| "learning_rate": 1.1000000000000001e-05, | |
| "loss": 1.6234, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.275, | |
| "grad_norm": 1.4259026050567627, | |
| "learning_rate": 1.0900000000000002e-05, | |
| "loss": 1.8434, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "grad_norm": 2.157871723175049, | |
| "learning_rate": 1.0800000000000002e-05, | |
| "loss": 1.4639, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 2.325, | |
| "grad_norm": 6.281842231750488, | |
| "learning_rate": 1.0700000000000001e-05, | |
| "loss": 1.5842, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "grad_norm": 12.432153701782227, | |
| "learning_rate": 1.0600000000000002e-05, | |
| "loss": 1.5704, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 2.375, | |
| "grad_norm": 1.2600141763687134, | |
| "learning_rate": 1.0500000000000001e-05, | |
| "loss": 1.5254, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 1.7496198415756226, | |
| "learning_rate": 1.04e-05, | |
| "loss": 1.654, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 2.425, | |
| "grad_norm": 3.237362861633301, | |
| "learning_rate": 1.0300000000000001e-05, | |
| "loss": 1.3958, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "grad_norm": 11.880879402160645, | |
| "learning_rate": 1.02e-05, | |
| "loss": 1.8982, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 2.475, | |
| "grad_norm": 4.163928985595703, | |
| "learning_rate": 1.0100000000000002e-05, | |
| "loss": 1.3325, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 5.584897994995117, | |
| "learning_rate": 1e-05, | |
| "loss": 1.7374, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.525, | |
| "grad_norm": 1.0842101573944092, | |
| "learning_rate": 9.9e-06, | |
| "loss": 1.6294, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "grad_norm": 3.4661624431610107, | |
| "learning_rate": 9.800000000000001e-06, | |
| "loss": 1.8073, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 2.575, | |
| "grad_norm": 2.1857829093933105, | |
| "learning_rate": 9.7e-06, | |
| "loss": 1.5675, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "grad_norm": 1.365864634513855, | |
| "learning_rate": 9.600000000000001e-06, | |
| "loss": 1.7444, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 2.625, | |
| "grad_norm": 6.100329875946045, | |
| "learning_rate": 9.5e-06, | |
| "loss": 1.3475, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "grad_norm": 8.876641273498535, | |
| "learning_rate": 9.4e-06, | |
| "loss": 1.5366, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 2.675, | |
| "grad_norm": 9.202727317810059, | |
| "learning_rate": 9.3e-06, | |
| "loss": 1.7323, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "grad_norm": 8.818689346313477, | |
| "learning_rate": 9.200000000000002e-06, | |
| "loss": 1.3941, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 2.725, | |
| "grad_norm": 1.5399378538131714, | |
| "learning_rate": 9.100000000000001e-06, | |
| "loss": 1.4933, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "grad_norm": 4.687498569488525, | |
| "learning_rate": 9e-06, | |
| "loss": 1.5237, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.775, | |
| "grad_norm": 1.0228453874588013, | |
| "learning_rate": 8.900000000000001e-06, | |
| "loss": 1.3142, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 3.121901273727417, | |
| "learning_rate": 8.8e-06, | |
| "loss": 1.5802, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 2.825, | |
| "grad_norm": 5.920619487762451, | |
| "learning_rate": 8.700000000000001e-06, | |
| "loss": 1.6471, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "grad_norm": 1.708122730255127, | |
| "learning_rate": 8.6e-06, | |
| "loss": 1.5874, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 2.875, | |
| "grad_norm": 6.527494430541992, | |
| "learning_rate": 8.5e-06, | |
| "loss": 1.5082, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "grad_norm": 5.949902057647705, | |
| "learning_rate": 8.400000000000001e-06, | |
| "loss": 1.3408, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 2.925, | |
| "grad_norm": 1.2279417514801025, | |
| "learning_rate": 8.3e-06, | |
| "loss": 1.6026, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "grad_norm": 5.212494850158691, | |
| "learning_rate": 8.2e-06, | |
| "loss": 1.5463, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 2.975, | |
| "grad_norm": 2.9334352016448975, | |
| "learning_rate": 8.1e-06, | |
| "loss": 1.7687, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 4.94165563583374, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 1.5553, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.6061710119247437, | |
| "eval_runtime": 179.0332, | |
| "eval_samples_per_second": 2.234, | |
| "eval_steps_per_second": 0.279, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.025, | |
| "grad_norm": 1.4166760444641113, | |
| "learning_rate": 7.9e-06, | |
| "loss": 1.6249, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "grad_norm": 2.30269455909729, | |
| "learning_rate": 7.800000000000002e-06, | |
| "loss": 1.3924, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 3.075, | |
| "grad_norm": 3.9721744060516357, | |
| "learning_rate": 7.7e-06, | |
| "loss": 1.3955, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "grad_norm": 6.161261558532715, | |
| "learning_rate": 7.600000000000001e-06, | |
| "loss": 1.7152, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 3.125, | |
| "grad_norm": 1.144785761833191, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 1.4025, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "grad_norm": 3.4314322471618652, | |
| "learning_rate": 7.4e-06, | |
| "loss": 1.779, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 3.175, | |
| "grad_norm": 9.453422546386719, | |
| "learning_rate": 7.3e-06, | |
| "loss": 1.523, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 12.822498321533203, | |
| "learning_rate": 7.2000000000000005e-06, | |
| "loss": 1.7858, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 3.225, | |
| "grad_norm": 2.504038095474243, | |
| "learning_rate": 7.100000000000001e-06, | |
| "loss": 1.6753, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "grad_norm": 1.0970689058303833, | |
| "learning_rate": 7e-06, | |
| "loss": 1.6171, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 3.275, | |
| "grad_norm": 2.6205241680145264, | |
| "learning_rate": 6.9e-06, | |
| "loss": 1.2819, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "grad_norm": 1.3209072351455688, | |
| "learning_rate": 6.800000000000001e-06, | |
| "loss": 1.6669, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 3.325, | |
| "grad_norm": 6.18387508392334, | |
| "learning_rate": 6.700000000000001e-06, | |
| "loss": 1.572, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "grad_norm": 7.554809093475342, | |
| "learning_rate": 6.600000000000001e-06, | |
| "loss": 1.621, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 3.375, | |
| "grad_norm": 8.28931713104248, | |
| "learning_rate": 6.5000000000000004e-06, | |
| "loss": 1.3405, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "grad_norm": 7.459342956542969, | |
| "learning_rate": 6.4000000000000006e-06, | |
| "loss": 1.7289, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 3.425, | |
| "grad_norm": 1.0719680786132812, | |
| "learning_rate": 6.300000000000001e-06, | |
| "loss": 1.3691, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "grad_norm": 3.2976412773132324, | |
| "learning_rate": 6.200000000000001e-06, | |
| "loss": 1.6399, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 3.475, | |
| "grad_norm": 6.960519790649414, | |
| "learning_rate": 6.1e-06, | |
| "loss": 1.1903, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "grad_norm": 1.1840627193450928, | |
| "learning_rate": 6e-06, | |
| "loss": 1.6722, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.525, | |
| "grad_norm": 10.626968383789062, | |
| "learning_rate": 5.9e-06, | |
| "loss": 1.4026, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "grad_norm": 2.9099953174591064, | |
| "learning_rate": 5.8e-06, | |
| "loss": 1.333, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 3.575, | |
| "grad_norm": 6.874935150146484, | |
| "learning_rate": 5.7e-06, | |
| "loss": 1.444, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "grad_norm": 1.6039921045303345, | |
| "learning_rate": 5.600000000000001e-06, | |
| "loss": 1.9615, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 3.625, | |
| "grad_norm": 3.3226425647735596, | |
| "learning_rate": 5.500000000000001e-06, | |
| "loss": 1.4471, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "grad_norm": 0.8198592066764832, | |
| "learning_rate": 5.400000000000001e-06, | |
| "loss": 1.111, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 3.675, | |
| "grad_norm": 2.200191020965576, | |
| "learning_rate": 5.300000000000001e-06, | |
| "loss": 1.6108, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "grad_norm": 1.6467258930206299, | |
| "learning_rate": 5.2e-06, | |
| "loss": 1.3662, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 3.725, | |
| "grad_norm": 21.577899932861328, | |
| "learning_rate": 5.1e-06, | |
| "loss": 1.7619, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 10.908743858337402, | |
| "learning_rate": 5e-06, | |
| "loss": 1.5449, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.775, | |
| "grad_norm": 14.157689094543457, | |
| "learning_rate": 4.9000000000000005e-06, | |
| "loss": 1.701, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "grad_norm": 2.318659543991089, | |
| "learning_rate": 4.800000000000001e-06, | |
| "loss": 1.6123, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 3.825, | |
| "grad_norm": 3.1259167194366455, | |
| "learning_rate": 4.7e-06, | |
| "loss": 1.4108, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "grad_norm": 0.9985270500183105, | |
| "learning_rate": 4.600000000000001e-06, | |
| "loss": 1.3939, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 3.875, | |
| "grad_norm": 18.413806915283203, | |
| "learning_rate": 4.5e-06, | |
| "loss": 2.0268, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "grad_norm": 1.4327787160873413, | |
| "learning_rate": 4.4e-06, | |
| "loss": 1.4815, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 3.925, | |
| "grad_norm": 6.25641393661499, | |
| "learning_rate": 4.3e-06, | |
| "loss": 1.5863, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "grad_norm": 7.975721836090088, | |
| "learning_rate": 4.2000000000000004e-06, | |
| "loss": 1.8014, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 3.975, | |
| "grad_norm": 9.99906063079834, | |
| "learning_rate": 4.1e-06, | |
| "loss": 1.6257, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 6.815397262573242, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.7811, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.605211615562439, | |
| "eval_runtime": 179.9082, | |
| "eval_samples_per_second": 2.223, | |
| "eval_steps_per_second": 0.278, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 4.025, | |
| "grad_norm": 8.750836372375488, | |
| "learning_rate": 3.900000000000001e-06, | |
| "loss": 1.4724, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "grad_norm": 10.807592391967773, | |
| "learning_rate": 3.8000000000000005e-06, | |
| "loss": 1.6807, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 4.075, | |
| "grad_norm": 23.482824325561523, | |
| "learning_rate": 3.7e-06, | |
| "loss": 1.5182, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "grad_norm": 10.791790008544922, | |
| "learning_rate": 3.6000000000000003e-06, | |
| "loss": 1.3766, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 4.125, | |
| "grad_norm": 2.6700119972229004, | |
| "learning_rate": 3.5e-06, | |
| "loss": 1.6198, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 4.15, | |
| "grad_norm": 3.3935744762420654, | |
| "learning_rate": 3.4000000000000005e-06, | |
| "loss": 1.5944, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 4.175, | |
| "grad_norm": 4.310108184814453, | |
| "learning_rate": 3.3000000000000006e-06, | |
| "loss": 2.0028, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "grad_norm": 5.834546089172363, | |
| "learning_rate": 3.2000000000000003e-06, | |
| "loss": 1.3597, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 4.225, | |
| "grad_norm": 1.2702422142028809, | |
| "learning_rate": 3.1000000000000004e-06, | |
| "loss": 1.4279, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "grad_norm": 1.1610503196716309, | |
| "learning_rate": 3e-06, | |
| "loss": 1.237, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 4.275, | |
| "grad_norm": 3.733497381210327, | |
| "learning_rate": 2.9e-06, | |
| "loss": 1.4688, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 4.3, | |
| "grad_norm": 1.0653811693191528, | |
| "learning_rate": 2.8000000000000003e-06, | |
| "loss": 1.7002, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 4.325, | |
| "grad_norm": 5.7914934158325195, | |
| "learning_rate": 2.7000000000000004e-06, | |
| "loss": 1.5866, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "grad_norm": 10.097025871276855, | |
| "learning_rate": 2.6e-06, | |
| "loss": 1.3058, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 4.375, | |
| "grad_norm": 2.2957890033721924, | |
| "learning_rate": 2.5e-06, | |
| "loss": 1.5006, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "grad_norm": 1.5741922855377197, | |
| "learning_rate": 2.4000000000000003e-06, | |
| "loss": 1.4366, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 4.425, | |
| "grad_norm": 9.413016319274902, | |
| "learning_rate": 2.3000000000000004e-06, | |
| "loss": 2.1316, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 4.45, | |
| "grad_norm": 11.494050979614258, | |
| "learning_rate": 2.2e-06, | |
| "loss": 1.7012, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 4.475, | |
| "grad_norm": 4.699800968170166, | |
| "learning_rate": 2.1000000000000002e-06, | |
| "loss": 1.4526, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "grad_norm": 3.780972480773926, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 1.4928, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 4.525, | |
| "grad_norm": 9.663272857666016, | |
| "learning_rate": 1.9000000000000002e-06, | |
| "loss": 1.5854, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "grad_norm": 1.1007126569747925, | |
| "learning_rate": 1.8000000000000001e-06, | |
| "loss": 1.5926, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 4.575, | |
| "grad_norm": 3.3589682579040527, | |
| "learning_rate": 1.7000000000000002e-06, | |
| "loss": 1.3675, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 4.6, | |
| "grad_norm": 8.529759407043457, | |
| "learning_rate": 1.6000000000000001e-06, | |
| "loss": 1.5693, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 4.625, | |
| "grad_norm": 4.393673419952393, | |
| "learning_rate": 1.5e-06, | |
| "loss": 1.6012, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "grad_norm": 23.471420288085938, | |
| "learning_rate": 1.4000000000000001e-06, | |
| "loss": 1.5968, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 4.675, | |
| "grad_norm": 10.580934524536133, | |
| "learning_rate": 1.3e-06, | |
| "loss": 1.6844, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 4.7, | |
| "grad_norm": 4.5100579261779785, | |
| "learning_rate": 1.2000000000000002e-06, | |
| "loss": 1.4989, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 4.725, | |
| "grad_norm": 8.684538841247559, | |
| "learning_rate": 1.1e-06, | |
| "loss": 1.7439, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 4.75, | |
| "grad_norm": 10.94477367401123, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 1.7858, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 4.775, | |
| "grad_norm": 9.082775115966797, | |
| "learning_rate": 9.000000000000001e-07, | |
| "loss": 1.2804, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 3.1917006969451904, | |
| "learning_rate": 8.000000000000001e-07, | |
| "loss": 1.3131, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 4.825, | |
| "grad_norm": 1.8910670280456543, | |
| "learning_rate": 7.000000000000001e-07, | |
| "loss": 1.5313, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 4.85, | |
| "grad_norm": 2.0198514461517334, | |
| "learning_rate": 6.000000000000001e-07, | |
| "loss": 1.5972, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 4.875, | |
| "grad_norm": 8.048125267028809, | |
| "learning_rate": 5.000000000000001e-07, | |
| "loss": 1.5084, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "grad_norm": 3.2687554359436035, | |
| "learning_rate": 4.0000000000000003e-07, | |
| "loss": 1.563, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 4.925, | |
| "grad_norm": 3.7933993339538574, | |
| "learning_rate": 3.0000000000000004e-07, | |
| "loss": 1.3889, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "grad_norm": 3.9348926544189453, | |
| "learning_rate": 2.0000000000000002e-07, | |
| "loss": 1.6613, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 4.975, | |
| "grad_norm": 3.605724573135376, | |
| "learning_rate": 1.0000000000000001e-07, | |
| "loss": 1.7075, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.839205265045166, | |
| "learning_rate": 0.0, | |
| "loss": 1.3854, | |
| "step": 2000 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 2000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.747389217800192e+18, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |