| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9992532754056073, | |
| "eval_steps": 500, | |
| "global_step": 920, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.010861448645713123, | |
| "grad_norm": 169.19541931152344, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 3.0856, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.021722897291426246, | |
| "grad_norm": 10.523147583007812, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 3.0568, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03258434593713937, | |
| "grad_norm": 13.033576011657715, | |
| "learning_rate": 3e-06, | |
| "loss": 3.1183, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04344579458285249, | |
| "grad_norm": 14.1365966796875, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 2.9091, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.05430724322856561, | |
| "grad_norm": 36.71495819091797, | |
| "learning_rate": 5e-06, | |
| "loss": 2.8426, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.06516869187427873, | |
| "grad_norm": 14.660510063171387, | |
| "learning_rate": 6e-06, | |
| "loss": 2.7139, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.07603014051999185, | |
| "grad_norm": 33.809818267822266, | |
| "learning_rate": 7.000000000000001e-06, | |
| "loss": 2.5489, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.08689158916570498, | |
| "grad_norm": 17.694528579711914, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 2.5378, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.0977530378114181, | |
| "grad_norm": 23.081829071044922, | |
| "learning_rate": 9e-06, | |
| "loss": 2.4152, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.10861448645713122, | |
| "grad_norm": 13.884807586669922, | |
| "learning_rate": 1e-05, | |
| "loss": 2.364, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.11947593510284434, | |
| "grad_norm": 10.270822525024414, | |
| "learning_rate": 1.1000000000000001e-05, | |
| "loss": 2.5817, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.13033738374855747, | |
| "grad_norm": 8.365082740783691, | |
| "learning_rate": 1.2e-05, | |
| "loss": 2.2702, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.14119883239427058, | |
| "grad_norm": 15.889869689941406, | |
| "learning_rate": 1.3000000000000001e-05, | |
| "loss": 2.0207, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.1520602810399837, | |
| "grad_norm": 7.615819454193115, | |
| "learning_rate": 1.4000000000000001e-05, | |
| "loss": 2.0305, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.16292172968569682, | |
| "grad_norm": 10.326943397521973, | |
| "learning_rate": 1.5e-05, | |
| "loss": 2.0734, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.17378317833140997, | |
| "grad_norm": 7.269619941711426, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 2.0625, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.18464462697712308, | |
| "grad_norm": 4.737296104431152, | |
| "learning_rate": 1.7000000000000003e-05, | |
| "loss": 2.0665, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.1955060756228362, | |
| "grad_norm": 5.883551597595215, | |
| "learning_rate": 1.8e-05, | |
| "loss": 2.0005, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.20636752426854932, | |
| "grad_norm": 7.671362400054932, | |
| "learning_rate": 1.9e-05, | |
| "loss": 1.9652, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.21722897291426244, | |
| "grad_norm": 4.762546539306641, | |
| "learning_rate": 2e-05, | |
| "loss": 1.9471, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.22809042155997555, | |
| "grad_norm": 5.938209056854248, | |
| "learning_rate": 2.1e-05, | |
| "loss": 1.8201, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.23895187020568867, | |
| "grad_norm": 4.879518985748291, | |
| "learning_rate": 2.2000000000000003e-05, | |
| "loss": 1.9098, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.24981331885140182, | |
| "grad_norm": 3.9282209873199463, | |
| "learning_rate": 2.3000000000000003e-05, | |
| "loss": 1.7855, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.26067476749711493, | |
| "grad_norm": 5.627065181732178, | |
| "learning_rate": 2.4e-05, | |
| "loss": 1.9036, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.271536216142828, | |
| "grad_norm": 5.127110481262207, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.9096, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.28239766478854117, | |
| "grad_norm": 4.889794826507568, | |
| "learning_rate": 2.6000000000000002e-05, | |
| "loss": 1.8139, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.2932591134342543, | |
| "grad_norm": 6.027666091918945, | |
| "learning_rate": 2.7000000000000002e-05, | |
| "loss": 1.8053, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.3041205620799674, | |
| "grad_norm": 4.673439979553223, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 1.7594, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.31498201072568055, | |
| "grad_norm": 15.802154541015625, | |
| "learning_rate": 2.9e-05, | |
| "loss": 1.7326, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.32584345937139364, | |
| "grad_norm": 4.887749671936035, | |
| "learning_rate": 3e-05, | |
| "loss": 1.8835, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.3367049080171068, | |
| "grad_norm": 4.747198581695557, | |
| "learning_rate": 3.1e-05, | |
| "loss": 1.7743, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.34756635666281993, | |
| "grad_norm": 3.817439079284668, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 1.8374, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.358427805308533, | |
| "grad_norm": 8.614398002624512, | |
| "learning_rate": 3.3e-05, | |
| "loss": 1.8126, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.36928925395424617, | |
| "grad_norm": 4.607050895690918, | |
| "learning_rate": 3.4000000000000007e-05, | |
| "loss": 1.7341, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.38015070259995926, | |
| "grad_norm": 5.569587707519531, | |
| "learning_rate": 3.5e-05, | |
| "loss": 1.7022, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.3910121512456724, | |
| "grad_norm": 4.454966068267822, | |
| "learning_rate": 3.6e-05, | |
| "loss": 1.697, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.4018735998913855, | |
| "grad_norm": 4.7365031242370605, | |
| "learning_rate": 3.7e-05, | |
| "loss": 1.7528, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.41273504853709864, | |
| "grad_norm": 5.313693523406982, | |
| "learning_rate": 3.8e-05, | |
| "loss": 1.7489, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.4235964971828118, | |
| "grad_norm": 11.473447799682617, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 1.6342, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.43445794582852487, | |
| "grad_norm": 4.205170154571533, | |
| "learning_rate": 4e-05, | |
| "loss": 1.6882, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.445319394474238, | |
| "grad_norm": 4.324029445648193, | |
| "learning_rate": 4.1e-05, | |
| "loss": 1.6918, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.4561808431199511, | |
| "grad_norm": 6.601579666137695, | |
| "learning_rate": 4.2e-05, | |
| "loss": 1.7205, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.46704229176566425, | |
| "grad_norm": 3.448477029800415, | |
| "learning_rate": 4.3e-05, | |
| "loss": 1.5888, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.47790374041137734, | |
| "grad_norm": 19.598926544189453, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 1.6581, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.4887651890570905, | |
| "grad_norm": 4.1286702156066895, | |
| "learning_rate": 4.5e-05, | |
| "loss": 1.669, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.49962663770280363, | |
| "grad_norm": 7.4469404220581055, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 1.7023, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.5104880863485167, | |
| "grad_norm": 3.3093454837799072, | |
| "learning_rate": 4.7e-05, | |
| "loss": 1.6477, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.5213495349942299, | |
| "grad_norm": 7.244236946105957, | |
| "learning_rate": 4.8e-05, | |
| "loss": 1.6486, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.532210983639943, | |
| "grad_norm": 4.145318984985352, | |
| "learning_rate": 4.9e-05, | |
| "loss": 1.6016, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.543072432285656, | |
| "grad_norm": 4.232030868530273, | |
| "learning_rate": 5e-05, | |
| "loss": 1.6154, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.543072432285656, | |
| "eval_loss": 1.4837769269943237, | |
| "eval_runtime": 32.1482, | |
| "eval_samples_per_second": 25.445, | |
| "eval_steps_per_second": 25.445, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.5539338809313692, | |
| "grad_norm": 3.5671701431274414, | |
| "learning_rate": 4.880952380952381e-05, | |
| "loss": 1.6524, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.5647953295770823, | |
| "grad_norm": 5.578204154968262, | |
| "learning_rate": 4.761904761904762e-05, | |
| "loss": 1.6353, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.5756567782227955, | |
| "grad_norm": 7.8000993728637695, | |
| "learning_rate": 4.642857142857143e-05, | |
| "loss": 1.6009, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.5865182268685086, | |
| "grad_norm": 10.637792587280273, | |
| "learning_rate": 4.523809523809524e-05, | |
| "loss": 1.5455, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.5973796755142217, | |
| "grad_norm": 4.490563869476318, | |
| "learning_rate": 4.404761904761905e-05, | |
| "loss": 1.7016, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.6082411241599348, | |
| "grad_norm": 4.277554512023926, | |
| "learning_rate": 4.2857142857142856e-05, | |
| "loss": 1.6836, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.619102572805648, | |
| "grad_norm": 4.989715576171875, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 1.6839, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.6299640214513611, | |
| "grad_norm": 5.333950042724609, | |
| "learning_rate": 4.047619047619048e-05, | |
| "loss": 1.6369, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.6408254700970742, | |
| "grad_norm": 6.111197471618652, | |
| "learning_rate": 3.928571428571429e-05, | |
| "loss": 1.6566, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.6516869187427873, | |
| "grad_norm": 9.309179306030273, | |
| "learning_rate": 3.809523809523809e-05, | |
| "loss": 1.6325, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.6625483673885004, | |
| "grad_norm": 3.6230900287628174, | |
| "learning_rate": 3.690476190476191e-05, | |
| "loss": 1.5397, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.6734098160342136, | |
| "grad_norm": 12.543122291564941, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 1.5934, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.6842712646799267, | |
| "grad_norm": 8.352952003479004, | |
| "learning_rate": 3.4523809523809526e-05, | |
| "loss": 1.6255, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.6951327133256399, | |
| "grad_norm": 3.2874395847320557, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 1.5855, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.7059941619713529, | |
| "grad_norm": 4.011010646820068, | |
| "learning_rate": 3.2142857142857144e-05, | |
| "loss": 1.6257, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.716855610617066, | |
| "grad_norm": 3.508124828338623, | |
| "learning_rate": 3.095238095238095e-05, | |
| "loss": 1.5792, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.7277170592627792, | |
| "grad_norm": 8.244254112243652, | |
| "learning_rate": 2.9761904761904762e-05, | |
| "loss": 1.5836, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.7385785079084923, | |
| "grad_norm": 4.174469947814941, | |
| "learning_rate": 2.857142857142857e-05, | |
| "loss": 1.6274, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.7494399565542054, | |
| "grad_norm": 3.8175296783447266, | |
| "learning_rate": 2.7380952380952383e-05, | |
| "loss": 1.524, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 0.7603014051999185, | |
| "grad_norm": 5.269782543182373, | |
| "learning_rate": 2.6190476190476192e-05, | |
| "loss": 1.6037, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.7711628538456317, | |
| "grad_norm": 6.3031511306762695, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.5312, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 0.7820243024913448, | |
| "grad_norm": 2.747074604034424, | |
| "learning_rate": 2.380952380952381e-05, | |
| "loss": 1.5516, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.792885751137058, | |
| "grad_norm": 3.978896141052246, | |
| "learning_rate": 2.261904761904762e-05, | |
| "loss": 1.6539, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 0.803747199782771, | |
| "grad_norm": 4.52466344833374, | |
| "learning_rate": 2.1428571428571428e-05, | |
| "loss": 1.6088, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.8146086484284841, | |
| "grad_norm": 5.761850357055664, | |
| "learning_rate": 2.023809523809524e-05, | |
| "loss": 1.578, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.8254700970741973, | |
| "grad_norm": 4.5440192222595215, | |
| "learning_rate": 1.9047619047619046e-05, | |
| "loss": 1.5776, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.8363315457199104, | |
| "grad_norm": 3.666665554046631, | |
| "learning_rate": 1.785714285714286e-05, | |
| "loss": 1.5747, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 0.8471929943656236, | |
| "grad_norm": 4.0968403816223145, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 1.5851, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.8580544430113366, | |
| "grad_norm": 4.229625225067139, | |
| "learning_rate": 1.5476190476190476e-05, | |
| "loss": 1.5634, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 0.8689158916570497, | |
| "grad_norm": 30.42665672302246, | |
| "learning_rate": 1.4285714285714285e-05, | |
| "loss": 1.5744, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.8797773403027629, | |
| "grad_norm": 9.380242347717285, | |
| "learning_rate": 1.3095238095238096e-05, | |
| "loss": 1.5207, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 0.890638788948476, | |
| "grad_norm": 3.2494077682495117, | |
| "learning_rate": 1.1904761904761905e-05, | |
| "loss": 1.4783, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 0.9015002375941892, | |
| "grad_norm": 7.472358703613281, | |
| "learning_rate": 1.0714285714285714e-05, | |
| "loss": 1.5537, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 0.9123616862399022, | |
| "grad_norm": 4.2244439125061035, | |
| "learning_rate": 9.523809523809523e-06, | |
| "loss": 1.5969, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.9232231348856154, | |
| "grad_norm": 3.104196548461914, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 1.6525, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.9340845835313285, | |
| "grad_norm": 5.433168411254883, | |
| "learning_rate": 7.142857142857143e-06, | |
| "loss": 1.5486, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 0.9449460321770417, | |
| "grad_norm": 3.924849033355713, | |
| "learning_rate": 5.9523809523809525e-06, | |
| "loss": 1.4945, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 0.9558074808227547, | |
| "grad_norm": 3.4002134799957275, | |
| "learning_rate": 4.7619047619047615e-06, | |
| "loss": 1.5937, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 0.9666689294684678, | |
| "grad_norm": 5.583169460296631, | |
| "learning_rate": 3.5714285714285714e-06, | |
| "loss": 1.6058, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 0.977530378114181, | |
| "grad_norm": 7.243485927581787, | |
| "learning_rate": 2.3809523809523808e-06, | |
| "loss": 1.641, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.9883918267598941, | |
| "grad_norm": 174.63710021972656, | |
| "learning_rate": 1.1904761904761904e-06, | |
| "loss": 1.6228, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 0.9992532754056073, | |
| "grad_norm": 6.230108261108398, | |
| "learning_rate": 0.0, | |
| "loss": 1.4502, | |
| "step": 920 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 920, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 1000000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5528070268502016.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |