| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 100, | |
| "global_step": 4591, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0021784119376974187, | |
| "grad_norm": 583.7144775390625, | |
| "learning_rate": 1.956521739130435e-07, | |
| "loss": 8.685, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0043568238753948375, | |
| "grad_norm": 563.7721557617188, | |
| "learning_rate": 4.130434782608696e-07, | |
| "loss": 8.2663, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.006535235813092255, | |
| "grad_norm": 223.9811553955078, | |
| "learning_rate": 6.304347826086957e-07, | |
| "loss": 5.5525, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.008713647750789675, | |
| "grad_norm": 579.8836669921875, | |
| "learning_rate": 8.478260869565217e-07, | |
| "loss": 2.5749, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.010892059688487093, | |
| "grad_norm": 234.20037841796875, | |
| "learning_rate": 1.065217391304348e-06, | |
| "loss": 1.9417, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.01307047162618451, | |
| "grad_norm": 62.84681701660156, | |
| "learning_rate": 1.2826086956521742e-06, | |
| "loss": 1.3996, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.01524888356388193, | |
| "grad_norm": 49.981231689453125, | |
| "learning_rate": 1.5e-06, | |
| "loss": 1.2478, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.01742729550157935, | |
| "grad_norm": 80.07563018798828, | |
| "learning_rate": 1.7173913043478263e-06, | |
| "loss": 1.2043, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.019605707439276768, | |
| "grad_norm": 64.25047302246094, | |
| "learning_rate": 1.934782608695652e-06, | |
| "loss": 1.1828, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.021784119376974186, | |
| "grad_norm": 14.763628005981445, | |
| "learning_rate": 2.1521739130434787e-06, | |
| "loss": 1.1666, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.021784119376974186, | |
| "eval_loss": 1.1604220867156982, | |
| "eval_runtime": 9.6262, | |
| "eval_samples_per_second": 518.585, | |
| "eval_steps_per_second": 2.078, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.023962531314671603, | |
| "grad_norm": 52.79157257080078, | |
| "learning_rate": 2.3695652173913043e-06, | |
| "loss": 1.1818, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.02614094325236902, | |
| "grad_norm": 74.7455825805664, | |
| "learning_rate": 2.586956521739131e-06, | |
| "loss": 1.1769, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.028319355190066443, | |
| "grad_norm": 25.616329193115234, | |
| "learning_rate": 2.804347826086957e-06, | |
| "loss": 1.1554, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.03049776712776386, | |
| "grad_norm": 47.48295211791992, | |
| "learning_rate": 3.0217391304347825e-06, | |
| "loss": 1.1475, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.03267617906546128, | |
| "grad_norm": 11.702336311340332, | |
| "learning_rate": 3.239130434782609e-06, | |
| "loss": 1.1473, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.0348545910031587, | |
| "grad_norm": 66.81829833984375, | |
| "learning_rate": 3.456521739130435e-06, | |
| "loss": 1.2304, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.03703300294085612, | |
| "grad_norm": 33.29442596435547, | |
| "learning_rate": 3.673913043478261e-06, | |
| "loss": 1.1667, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.039211414878553535, | |
| "grad_norm": 15.700839042663574, | |
| "learning_rate": 3.891304347826087e-06, | |
| "loss": 1.1553, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.04138982681625095, | |
| "grad_norm": 5.018191814422607, | |
| "learning_rate": 4.108695652173914e-06, | |
| "loss": 1.1635, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.04356823875394837, | |
| "grad_norm": 16.987112045288086, | |
| "learning_rate": 4.32608695652174e-06, | |
| "loss": 1.1794, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.04356823875394837, | |
| "eval_loss": 1.1022545099258423, | |
| "eval_runtime": 9.5442, | |
| "eval_samples_per_second": 523.038, | |
| "eval_steps_per_second": 2.096, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.04574665069164579, | |
| "grad_norm": 13.58285140991211, | |
| "learning_rate": 4.543478260869566e-06, | |
| "loss": 1.1459, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.04792506262934321, | |
| "grad_norm": 17.155811309814453, | |
| "learning_rate": 4.760869565217392e-06, | |
| "loss": 1.1388, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.050103474567040625, | |
| "grad_norm": 76.64653778076172, | |
| "learning_rate": 4.978260869565218e-06, | |
| "loss": 1.1497, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.05228188650473804, | |
| "grad_norm": 25.240867614746094, | |
| "learning_rate": 5.195652173913044e-06, | |
| "loss": 1.1459, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.05446029844243547, | |
| "grad_norm": 34.7780876159668, | |
| "learning_rate": 5.41304347826087e-06, | |
| "loss": 1.1401, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.056638710380132885, | |
| "grad_norm": 54.864139556884766, | |
| "learning_rate": 5.630434782608696e-06, | |
| "loss": 1.1731, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.0588171223178303, | |
| "grad_norm": 35.154762268066406, | |
| "learning_rate": 5.847826086956521e-06, | |
| "loss": 1.1628, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.06099553425552772, | |
| "grad_norm": 15.862505912780762, | |
| "learning_rate": 6.065217391304349e-06, | |
| "loss": 1.1329, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.06317394619322514, | |
| "grad_norm": 12.98637866973877, | |
| "learning_rate": 6.282608695652174e-06, | |
| "loss": 1.1319, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.06535235813092256, | |
| "grad_norm": 25.900012969970703, | |
| "learning_rate": 6.5000000000000004e-06, | |
| "loss": 1.1542, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.06535235813092256, | |
| "eval_loss": 1.096329927444458, | |
| "eval_runtime": 9.6101, | |
| "eval_samples_per_second": 519.452, | |
| "eval_steps_per_second": 2.081, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.06753077006861997, | |
| "grad_norm": 13.545440673828125, | |
| "learning_rate": 6.7173913043478265e-06, | |
| "loss": 1.1598, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.0697091820063174, | |
| "grad_norm": 29.305356979370117, | |
| "learning_rate": 6.934782608695653e-06, | |
| "loss": 1.1462, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.07188759394401481, | |
| "grad_norm": 31.123804092407227, | |
| "learning_rate": 7.152173913043479e-06, | |
| "loss": 1.143, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.07406600588171224, | |
| "grad_norm": 16.058088302612305, | |
| "learning_rate": 7.369565217391305e-06, | |
| "loss": 1.1237, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.07624441781940965, | |
| "grad_norm": 53.67496109008789, | |
| "learning_rate": 7.586956521739131e-06, | |
| "loss": 1.1447, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.07842282975710707, | |
| "grad_norm": 45.335391998291016, | |
| "learning_rate": 7.804347826086957e-06, | |
| "loss": 1.1486, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.08060124169480448, | |
| "grad_norm": 1.5897856950759888, | |
| "learning_rate": 8.021739130434783e-06, | |
| "loss": 1.1459, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.0827796536325019, | |
| "grad_norm": 21.029211044311523, | |
| "learning_rate": 8.239130434782609e-06, | |
| "loss": 1.1447, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.08495806557019933, | |
| "grad_norm": 19.90410041809082, | |
| "learning_rate": 8.456521739130435e-06, | |
| "loss": 1.1186, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.08713647750789674, | |
| "grad_norm": 11.852821350097656, | |
| "learning_rate": 8.673913043478261e-06, | |
| "loss": 1.1265, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.08713647750789674, | |
| "eval_loss": 1.0634950399398804, | |
| "eval_runtime": 9.6061, | |
| "eval_samples_per_second": 519.668, | |
| "eval_steps_per_second": 2.082, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.08931488944559417, | |
| "grad_norm": 14.51545238494873, | |
| "learning_rate": 8.891304347826087e-06, | |
| "loss": 1.1214, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.09149330138329158, | |
| "grad_norm": 13.543514251708984, | |
| "learning_rate": 9.108695652173913e-06, | |
| "loss": 1.1109, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.093671713320989, | |
| "grad_norm": 12.666316986083984, | |
| "learning_rate": 9.32608695652174e-06, | |
| "loss": 1.101, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.09585012525868641, | |
| "grad_norm": 18.478477478027344, | |
| "learning_rate": 9.543478260869567e-06, | |
| "loss": 1.0728, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.09802853719638384, | |
| "grad_norm": 21.53498077392578, | |
| "learning_rate": 9.760869565217393e-06, | |
| "loss": 1.0367, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.10020694913408125, | |
| "grad_norm": 20.17803382873535, | |
| "learning_rate": 9.978260869565218e-06, | |
| "loss": 1.0203, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.10238536107177867, | |
| "grad_norm": 11.546687126159668, | |
| "learning_rate": 9.999882884955554e-06, | |
| "loss": 0.9821, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.10456377300947609, | |
| "grad_norm": 6.028177738189697, | |
| "learning_rate": 9.999478049870269e-06, | |
| "loss": 0.9369, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.10674218494717351, | |
| "grad_norm": 28.20073699951172, | |
| "learning_rate": 9.998784072287277e-06, | |
| "loss": 0.9232, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.10892059688487093, | |
| "grad_norm": 14.939946174621582, | |
| "learning_rate": 9.997800992342448e-06, | |
| "loss": 0.8963, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.10892059688487093, | |
| "eval_loss": 0.8135343790054321, | |
| "eval_runtime": 9.689, | |
| "eval_samples_per_second": 515.224, | |
| "eval_steps_per_second": 2.064, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.11109900882256835, | |
| "grad_norm": 14.081145286560059, | |
| "learning_rate": 9.99652886689176e-06, | |
| "loss": 0.8828, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.11327742076026577, | |
| "grad_norm": 14.179576873779297, | |
| "learning_rate": 9.994967769507992e-06, | |
| "loss": 0.8611, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.11545583269796318, | |
| "grad_norm": 14.495073318481445, | |
| "learning_rate": 9.993117790476494e-06, | |
| "loss": 0.8478, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.1176342446356606, | |
| "grad_norm": 13.599958419799805, | |
| "learning_rate": 9.990979036789943e-06, | |
| "loss": 0.8425, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.11981265657335802, | |
| "grad_norm": 11.789984703063965, | |
| "learning_rate": 9.988551632142171e-06, | |
| "loss": 0.8296, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.12199106851105544, | |
| "grad_norm": 13.627972602844238, | |
| "learning_rate": 9.985835716921e-06, | |
| "loss": 0.8092, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.12416948044875285, | |
| "grad_norm": 24.143400192260742, | |
| "learning_rate": 9.982831448200127e-06, | |
| "loss": 0.8161, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.12634789238645028, | |
| "grad_norm": 7.413597106933594, | |
| "learning_rate": 9.979538999730047e-06, | |
| "loss": 0.8126, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.1285263043241477, | |
| "grad_norm": 4.810773849487305, | |
| "learning_rate": 9.97595856192799e-06, | |
| "loss": 0.777, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.13070471626184513, | |
| "grad_norm": 13.047112464904785, | |
| "learning_rate": 9.972090341866917e-06, | |
| "loss": 0.7587, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.13070471626184513, | |
| "eval_loss": 0.670462429523468, | |
| "eval_runtime": 9.5938, | |
| "eval_samples_per_second": 520.335, | |
| "eval_steps_per_second": 2.085, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.13288312819954254, | |
| "grad_norm": 7.8898539543151855, | |
| "learning_rate": 9.967934563263545e-06, | |
| "loss": 0.7472, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.13506154013723995, | |
| "grad_norm": 12.270000457763672, | |
| "learning_rate": 9.963491466465405e-06, | |
| "loss": 0.7417, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.13723995207493736, | |
| "grad_norm": 11.972742080688477, | |
| "learning_rate": 9.958761308436941e-06, | |
| "loss": 0.727, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.1394183640126348, | |
| "grad_norm": 8.720413208007812, | |
| "learning_rate": 9.953744362744656e-06, | |
| "loss": 0.7179, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.1415967759503322, | |
| "grad_norm": 10.330451965332031, | |
| "learning_rate": 9.948440919541277e-06, | |
| "loss": 0.7123, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.14377518788802962, | |
| "grad_norm": 8.81712818145752, | |
| "learning_rate": 9.942851285548986e-06, | |
| "loss": 0.7052, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.14595359982572703, | |
| "grad_norm": 12.095762252807617, | |
| "learning_rate": 9.936975784041678e-06, | |
| "loss": 0.6999, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.14813201176342447, | |
| "grad_norm": 9.912598609924316, | |
| "learning_rate": 9.930814754826262e-06, | |
| "loss": 0.6873, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.15031042370112188, | |
| "grad_norm": 4.136904716491699, | |
| "learning_rate": 9.924368554223007e-06, | |
| "loss": 0.6988, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 0.1524888356388193, | |
| "grad_norm": 6.229986667633057, | |
| "learning_rate": 9.917637555044942e-06, | |
| "loss": 0.6947, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.1524888356388193, | |
| "eval_loss": 0.606589138507843, | |
| "eval_runtime": 9.6122, | |
| "eval_samples_per_second": 519.339, | |
| "eval_steps_per_second": 2.081, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.15466724757651673, | |
| "grad_norm": 8.056920051574707, | |
| "learning_rate": 9.910622146576285e-06, | |
| "loss": 0.6722, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 0.15684565951421414, | |
| "grad_norm": 6.351996898651123, | |
| "learning_rate": 9.903322734549936e-06, | |
| "loss": 0.6715, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.15902407145191155, | |
| "grad_norm": 4.359755992889404, | |
| "learning_rate": 9.895739741124012e-06, | |
| "loss": 0.6622, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 0.16120248338960896, | |
| "grad_norm": 5.6291327476501465, | |
| "learning_rate": 9.887873604857424e-06, | |
| "loss": 0.6508, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.1633808953273064, | |
| "grad_norm": 4.779679775238037, | |
| "learning_rate": 9.879724780684518e-06, | |
| "loss": 0.638, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.1655593072650038, | |
| "grad_norm": 8.211064338684082, | |
| "learning_rate": 9.87129373988877e-06, | |
| "loss": 0.638, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.16773771920270122, | |
| "grad_norm": 11.320868492126465, | |
| "learning_rate": 9.86258097007552e-06, | |
| "loss": 0.623, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 0.16991613114039866, | |
| "grad_norm": 3.3580856323242188, | |
| "learning_rate": 9.853586975143772e-06, | |
| "loss": 0.5998, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.17209454307809607, | |
| "grad_norm": 9.876014709472656, | |
| "learning_rate": 9.844312275257059e-06, | |
| "loss": 0.5885, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 0.17427295501579348, | |
| "grad_norm": 7.176331520080566, | |
| "learning_rate": 9.834757406813353e-06, | |
| "loss": 0.5813, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.17427295501579348, | |
| "eval_loss": 0.5262527465820312, | |
| "eval_runtime": 9.6798, | |
| "eval_samples_per_second": 515.712, | |
| "eval_steps_per_second": 2.066, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.1764513669534909, | |
| "grad_norm": 8.124480247497559, | |
| "learning_rate": 9.824922922414048e-06, | |
| "loss": 0.5716, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 0.17862977889118833, | |
| "grad_norm": 4.01912260055542, | |
| "learning_rate": 9.814809390831994e-06, | |
| "loss": 0.5618, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 0.18080819082888575, | |
| "grad_norm": 8.250012397766113, | |
| "learning_rate": 9.804417396978605e-06, | |
| "loss": 0.5795, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 0.18298660276658316, | |
| "grad_norm": 3.1755073070526123, | |
| "learning_rate": 9.793747541870034e-06, | |
| "loss": 0.5608, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.18516501470428057, | |
| "grad_norm": 4.885599136352539, | |
| "learning_rate": 9.782800442592408e-06, | |
| "loss": 0.56, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.187343426641978, | |
| "grad_norm": 5.620588302612305, | |
| "learning_rate": 9.771576732266147e-06, | |
| "loss": 0.5522, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 0.18952183857967542, | |
| "grad_norm": 5.071799278259277, | |
| "learning_rate": 9.760077060009337e-06, | |
| "loss": 0.5444, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 0.19170025051737283, | |
| "grad_norm": 3.8804523944854736, | |
| "learning_rate": 9.7483020909002e-06, | |
| "loss": 0.538, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 0.19387866245507027, | |
| "grad_norm": 5.729625701904297, | |
| "learning_rate": 9.73625250593862e-06, | |
| "loss": 0.5356, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 0.19605707439276768, | |
| "grad_norm": 5.823616027832031, | |
| "learning_rate": 9.723929002006761e-06, | |
| "loss": 0.5319, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.19605707439276768, | |
| "eval_loss": 0.4962804913520813, | |
| "eval_runtime": 9.658, | |
| "eval_samples_per_second": 516.876, | |
| "eval_steps_per_second": 2.071, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.1982354863304651, | |
| "grad_norm": 11.661290168762207, | |
| "learning_rate": 9.71133229182877e-06, | |
| "loss": 0.5237, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 0.2004138982681625, | |
| "grad_norm": 8.839614868164062, | |
| "learning_rate": 9.698463103929542e-06, | |
| "loss": 0.538, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 0.20259231020585994, | |
| "grad_norm": 5.292098522186279, | |
| "learning_rate": 9.685322182592604e-06, | |
| "loss": 0.5202, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 0.20477072214355735, | |
| "grad_norm": 8.042999267578125, | |
| "learning_rate": 9.671910287817052e-06, | |
| "loss": 0.5218, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 0.20694913408125476, | |
| "grad_norm": 5.952273368835449, | |
| "learning_rate": 9.658228195273612e-06, | |
| "loss": 0.5167, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.20912754601895217, | |
| "grad_norm": 19.233489990234375, | |
| "learning_rate": 9.644276696259768e-06, | |
| "loss": 0.5153, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 0.2113059579566496, | |
| "grad_norm": 4.609057426452637, | |
| "learning_rate": 9.630056597654002e-06, | |
| "loss": 0.5138, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 0.21348436989434702, | |
| "grad_norm": 2.951719045639038, | |
| "learning_rate": 9.615568721869139e-06, | |
| "loss": 0.5077, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 0.21566278183204443, | |
| "grad_norm": 6.947376251220703, | |
| "learning_rate": 9.600813906804757e-06, | |
| "loss": 0.5024, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 0.21784119376974187, | |
| "grad_norm": 5.463963508605957, | |
| "learning_rate": 9.585793005798759e-06, | |
| "loss": 0.503, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.21784119376974187, | |
| "eval_loss": 0.49929723143577576, | |
| "eval_runtime": 9.6264, | |
| "eval_samples_per_second": 518.576, | |
| "eval_steps_per_second": 2.078, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.22001960570743928, | |
| "grad_norm": 3.5020666122436523, | |
| "learning_rate": 9.570506887577994e-06, | |
| "loss": 0.5113, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 0.2221980176451367, | |
| "grad_norm": 3.083266019821167, | |
| "learning_rate": 9.554956436208031e-06, | |
| "loss": 0.5043, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 0.2243764295828341, | |
| "grad_norm": 6.1208062171936035, | |
| "learning_rate": 9.539142551042024e-06, | |
| "loss": 0.4984, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 0.22655484152053154, | |
| "grad_norm": 2.8016839027404785, | |
| "learning_rate": 9.5230661466687e-06, | |
| "loss": 0.4957, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 0.22873325345822895, | |
| "grad_norm": 6.235162734985352, | |
| "learning_rate": 9.506728152859457e-06, | |
| "loss": 0.4936, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.23091166539592636, | |
| "grad_norm": 8.94963264465332, | |
| "learning_rate": 9.490129514514604e-06, | |
| "loss": 0.49, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 0.2330900773336238, | |
| "grad_norm": 5.008547306060791, | |
| "learning_rate": 9.473271191608699e-06, | |
| "loss": 0.4958, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 0.2352684892713212, | |
| "grad_norm": 8.403609275817871, | |
| "learning_rate": 9.456154159135043e-06, | |
| "loss": 0.4886, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 0.23744690120901862, | |
| "grad_norm": 12.323060035705566, | |
| "learning_rate": 9.438779407049282e-06, | |
| "loss": 0.4933, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 0.23962531314671603, | |
| "grad_norm": 6.123791217803955, | |
| "learning_rate": 9.421147940212152e-06, | |
| "loss": 0.4843, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.23962531314671603, | |
| "eval_loss": 0.4703376889228821, | |
| "eval_runtime": 9.6059, | |
| "eval_samples_per_second": 519.681, | |
| "eval_steps_per_second": 2.082, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.24180372508441347, | |
| "grad_norm": 3.6410789489746094, | |
| "learning_rate": 9.40326077833138e-06, | |
| "loss": 0.4797, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 0.24398213702211088, | |
| "grad_norm": 4.062999248504639, | |
| "learning_rate": 9.385118955902683e-06, | |
| "loss": 0.4844, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 0.2461605489598083, | |
| "grad_norm": 5.992898464202881, | |
| "learning_rate": 9.366723522149965e-06, | |
| "loss": 0.4801, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 0.2483389608975057, | |
| "grad_norm": 3.5720529556274414, | |
| "learning_rate": 9.348075540964611e-06, | |
| "loss": 0.4725, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 0.25051737283520314, | |
| "grad_norm": 3.079658269882202, | |
| "learning_rate": 9.329176090843985e-06, | |
| "loss": 0.4705, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.25269578477290056, | |
| "grad_norm": 4.961737632751465, | |
| "learning_rate": 9.310026264829027e-06, | |
| "loss": 0.4746, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 0.25487419671059797, | |
| "grad_norm": 4.140722274780273, | |
| "learning_rate": 9.290627170441055e-06, | |
| "loss": 0.465, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 0.2570526086482954, | |
| "grad_norm": 2.8426339626312256, | |
| "learning_rate": 9.270979929617712e-06, | |
| "loss": 0.4687, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 0.2592310205859928, | |
| "grad_norm": 6.142527103424072, | |
| "learning_rate": 9.251085678648072e-06, | |
| "loss": 0.4665, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 0.26140943252369025, | |
| "grad_norm": 6.779698371887207, | |
| "learning_rate": 9.230945568106925e-06, | |
| "loss": 0.4691, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.26140943252369025, | |
| "eval_loss": 0.4688015878200531, | |
| "eval_runtime": 9.5403, | |
| "eval_samples_per_second": 523.252, | |
| "eval_steps_per_second": 2.096, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.26358784446138767, | |
| "grad_norm": 4.152688980102539, | |
| "learning_rate": 9.210560762788238e-06, | |
| "loss": 0.4664, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 0.2657662563990851, | |
| "grad_norm": 6.0830817222595215, | |
| "learning_rate": 9.189932441637787e-06, | |
| "loss": 0.466, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 0.2679446683367825, | |
| "grad_norm": 3.1995880603790283, | |
| "learning_rate": 9.169061797684972e-06, | |
| "loss": 0.4615, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 0.2701230802744799, | |
| "grad_norm": 4.960101127624512, | |
| "learning_rate": 9.147950037973821e-06, | |
| "loss": 0.4645, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 0.2723014922121773, | |
| "grad_norm": 2.4431183338165283, | |
| "learning_rate": 9.126598383493179e-06, | |
| "loss": 0.4602, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.2744799041498747, | |
| "grad_norm": 3.671884298324585, | |
| "learning_rate": 9.105008069106093e-06, | |
| "loss": 0.457, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 0.2766583160875722, | |
| "grad_norm": 5.00780725479126, | |
| "learning_rate": 9.083180343478401e-06, | |
| "loss": 0.4622, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 0.2788367280252696, | |
| "grad_norm": 9.24307918548584, | |
| "learning_rate": 9.061116469006504e-06, | |
| "loss": 0.4601, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 0.281015139962967, | |
| "grad_norm": 3.016066789627075, | |
| "learning_rate": 9.038817721744362e-06, | |
| "loss": 0.4608, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 0.2831935519006644, | |
| "grad_norm": 4.565346717834473, | |
| "learning_rate": 9.0162853913297e-06, | |
| "loss": 0.4602, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.2831935519006644, | |
| "eval_loss": 0.45925456285476685, | |
| "eval_runtime": 9.5436, | |
| "eval_samples_per_second": 523.073, | |
| "eval_steps_per_second": 2.096, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.28537196383836183, | |
| "grad_norm": 2.7883694171905518, | |
| "learning_rate": 8.99352078090941e-06, | |
| "loss": 0.4573, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 0.28755037577605924, | |
| "grad_norm": 2.8337409496307373, | |
| "learning_rate": 8.970525207064194e-06, | |
| "loss": 0.4548, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 0.28972878771375665, | |
| "grad_norm": 7.179544448852539, | |
| "learning_rate": 8.947299999732415e-06, | |
| "loss": 0.4565, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 0.29190719965145406, | |
| "grad_norm": 5.800296783447266, | |
| "learning_rate": 8.923846502133186e-06, | |
| "loss": 0.4588, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 0.29408561158915153, | |
| "grad_norm": 5.762913227081299, | |
| "learning_rate": 8.900166070688676e-06, | |
| "loss": 0.4519, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.29626402352684894, | |
| "grad_norm": 10.822574615478516, | |
| "learning_rate": 8.87626007494567e-06, | |
| "loss": 0.456, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 0.29844243546454635, | |
| "grad_norm": 6.762709140777588, | |
| "learning_rate": 8.852129897496367e-06, | |
| "loss": 0.4638, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 0.30062084740224376, | |
| "grad_norm": 3.9109952449798584, | |
| "learning_rate": 8.827776933898397e-06, | |
| "loss": 0.4512, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 0.3027992593399412, | |
| "grad_norm": 4.887868881225586, | |
| "learning_rate": 8.803202592594136e-06, | |
| "loss": 0.4474, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 0.3049776712776386, | |
| "grad_norm": 8.851065635681152, | |
| "learning_rate": 8.778408294829232e-06, | |
| "loss": 0.4495, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.3049776712776386, | |
| "eval_loss": 0.45423299074172974, | |
| "eval_runtime": 9.5738, | |
| "eval_samples_per_second": 521.421, | |
| "eval_steps_per_second": 2.089, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.307156083215336, | |
| "grad_norm": 6.868368148803711, | |
| "learning_rate": 8.753395474570413e-06, | |
| "loss": 0.4507, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 0.30933449515303346, | |
| "grad_norm": 6.118240833282471, | |
| "learning_rate": 8.728165578422558e-06, | |
| "loss": 0.4476, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 0.3115129070907309, | |
| "grad_norm": 4.283236026763916, | |
| "learning_rate": 8.702720065545024e-06, | |
| "loss": 0.4516, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 0.3136913190284283, | |
| "grad_norm": 7.236689567565918, | |
| "learning_rate": 8.677060407567264e-06, | |
| "loss": 0.447, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 0.3158697309661257, | |
| "grad_norm": 1.7403960227966309, | |
| "learning_rate": 8.651188088503713e-06, | |
| "loss": 0.4424, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.3180481429038231, | |
| "grad_norm": 3.7462501525878906, | |
| "learning_rate": 8.625104604667965e-06, | |
| "loss": 0.4436, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 0.3202265548415205, | |
| "grad_norm": 2.083803653717041, | |
| "learning_rate": 8.598811464586225e-06, | |
| "loss": 0.4404, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 0.3224049667792179, | |
| "grad_norm": 3.62467360496521, | |
| "learning_rate": 8.572310188910075e-06, | |
| "loss": 0.4414, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 0.3245833787169154, | |
| "grad_norm": 3.4055850505828857, | |
| "learning_rate": 8.545602310328522e-06, | |
| "loss": 0.4489, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 0.3267617906546128, | |
| "grad_norm": 3.6146199703216553, | |
| "learning_rate": 8.518689373479354e-06, | |
| "loss": 0.4435, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.3267617906546128, | |
| "eval_loss": 0.4498658776283264, | |
| "eval_runtime": 9.6498, | |
| "eval_samples_per_second": 517.318, | |
| "eval_steps_per_second": 2.073, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.3289402025923102, | |
| "grad_norm": 4.079049587249756, | |
| "learning_rate": 8.491572934859808e-06, | |
| "loss": 0.4423, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 0.3311186145300076, | |
| "grad_norm": 4.84357213973999, | |
| "learning_rate": 8.464254562736556e-06, | |
| "loss": 0.4451, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 0.33329702646770504, | |
| "grad_norm": 2.70137619972229, | |
| "learning_rate": 8.436735837054998e-06, | |
| "loss": 0.4429, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 0.33547543840540245, | |
| "grad_norm": 3.2318954467773438, | |
| "learning_rate": 8.409018349347888e-06, | |
| "loss": 0.4394, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 0.33765385034309986, | |
| "grad_norm": 3.6141505241394043, | |
| "learning_rate": 8.381103702643295e-06, | |
| "loss": 0.4406, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 0.3398322622807973, | |
| "grad_norm": 4.4054484367370605, | |
| "learning_rate": 8.352993511371878e-06, | |
| "loss": 0.4397, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 0.34201067421849474, | |
| "grad_norm": 3.0935497283935547, | |
| "learning_rate": 8.324689401273538e-06, | |
| "loss": 0.4349, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 0.34418908615619215, | |
| "grad_norm": 2.5806384086608887, | |
| "learning_rate": 8.296193009303369e-06, | |
| "loss": 0.4366, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 0.34636749809388956, | |
| "grad_norm": 3.518735647201538, | |
| "learning_rate": 8.267505983537003e-06, | |
| "loss": 0.4356, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 0.34854591003158697, | |
| "grad_norm": 4.760187149047852, | |
| "learning_rate": 8.238629983075296e-06, | |
| "loss": 0.4351, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.34854591003158697, | |
| "eval_loss": 0.44457298517227173, | |
| "eval_runtime": 9.6018, | |
| "eval_samples_per_second": 519.901, | |
| "eval_steps_per_second": 2.083, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.3507243219692844, | |
| "grad_norm": 4.355133533477783, | |
| "learning_rate": 8.209566677948358e-06, | |
| "loss": 0.4352, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 0.3529027339069818, | |
| "grad_norm": 3.173840045928955, | |
| "learning_rate": 8.18031774901898e-06, | |
| "loss": 0.4359, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 0.3550811458446792, | |
| "grad_norm": 1.9210759401321411, | |
| "learning_rate": 8.150884887885424e-06, | |
| "loss": 0.4341, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 0.35725955778237667, | |
| "grad_norm": 2.729641914367676, | |
| "learning_rate": 8.121269796783585e-06, | |
| "loss": 0.4326, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 0.3594379697200741, | |
| "grad_norm": 2.5791547298431396, | |
| "learning_rate": 8.091474188488535e-06, | |
| "loss": 0.4313, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 0.3616163816577715, | |
| "grad_norm": 1.8370323181152344, | |
| "learning_rate": 8.061499786215488e-06, | |
| "loss": 0.4328, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 0.3637947935954689, | |
| "grad_norm": 3.280975818634033, | |
| "learning_rate": 8.031348323520113e-06, | |
| "loss": 0.4333, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 0.3659732055331663, | |
| "grad_norm": 3.277691125869751, | |
| "learning_rate": 8.001021544198295e-06, | |
| "loss": 0.4365, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 0.3681516174708637, | |
| "grad_norm": 1.8916176557540894, | |
| "learning_rate": 7.970521202185269e-06, | |
| "loss": 0.4311, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 0.37033002940856113, | |
| "grad_norm": 5.67519998550415, | |
| "learning_rate": 7.939849061454192e-06, | |
| "loss": 0.4335, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.37033002940856113, | |
| "eval_loss": 0.44085103273391724, | |
| "eval_runtime": 9.5393, | |
| "eval_samples_per_second": 523.31, | |
| "eval_steps_per_second": 2.097, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.3725084413462586, | |
| "grad_norm": 4.427841663360596, | |
| "learning_rate": 7.909006895914119e-06, | |
| "loss": 0.4328, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 0.374686853283956, | |
| "grad_norm": 2.4348649978637695, | |
| "learning_rate": 7.877996489307409e-06, | |
| "loss": 0.4273, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 0.3768652652216534, | |
| "grad_norm": 3.5188827514648438, | |
| "learning_rate": 7.846819635106569e-06, | |
| "loss": 0.4291, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 0.37904367715935083, | |
| "grad_norm": 4.911815643310547, | |
| "learning_rate": 7.81547813641052e-06, | |
| "loss": 0.4292, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 0.38122208909704824, | |
| "grad_norm": 5.043581962585449, | |
| "learning_rate": 7.783973805840329e-06, | |
| "loss": 0.4317, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.38340050103474566, | |
| "grad_norm": 5.273864269256592, | |
| "learning_rate": 7.75230846543436e-06, | |
| "loss": 0.4299, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 0.38557891297244307, | |
| "grad_norm": 2.7299153804779053, | |
| "learning_rate": 7.720483946542913e-06, | |
| "loss": 0.4255, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 0.38775732491014053, | |
| "grad_norm": 1.7716805934906006, | |
| "learning_rate": 7.688502089722302e-06, | |
| "loss": 0.4275, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 0.38993573684783794, | |
| "grad_norm": 3.8951661586761475, | |
| "learning_rate": 7.656364744628402e-06, | |
| "loss": 0.4247, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 0.39211414878553535, | |
| "grad_norm": 4.414074420928955, | |
| "learning_rate": 7.624073769909684e-06, | |
| "loss": 0.4259, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.39211414878553535, | |
| "eval_loss": 0.438446044921875, | |
| "eval_runtime": 9.5359, | |
| "eval_samples_per_second": 523.494, | |
| "eval_steps_per_second": 2.097, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.39429256072323277, | |
| "grad_norm": 2.8621246814727783, | |
| "learning_rate": 7.5916310330997175e-06, | |
| "loss": 0.426, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 0.3964709726609302, | |
| "grad_norm": 2.847648859024048, | |
| "learning_rate": 7.559038410509161e-06, | |
| "loss": 0.4263, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 0.3986493845986276, | |
| "grad_norm": 3.055957317352295, | |
| "learning_rate": 7.526297787117247e-06, | |
| "loss": 0.4249, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 0.400827796536325, | |
| "grad_norm": 5.001469612121582, | |
| "learning_rate": 7.493411056462768e-06, | |
| "loss": 0.4236, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 0.40300620847402246, | |
| "grad_norm": 1.1893895864486694, | |
| "learning_rate": 7.460380120534563e-06, | |
| "loss": 0.4238, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 0.4051846204117199, | |
| "grad_norm": 10.078166007995605, | |
| "learning_rate": 7.427206889661511e-06, | |
| "loss": 0.429, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 0.4073630323494173, | |
| "grad_norm": 2.957148551940918, | |
| "learning_rate": 7.393893282402057e-06, | |
| "loss": 0.4287, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 0.4095414442871147, | |
| "grad_norm": 2.5597410202026367, | |
| "learning_rate": 7.360441225433252e-06, | |
| "loss": 0.4264, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 0.4117198562248121, | |
| "grad_norm": 3.9550392627716064, | |
| "learning_rate": 7.326852653439317e-06, | |
| "loss": 0.4231, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 0.4138982681625095, | |
| "grad_norm": 3.9910600185394287, | |
| "learning_rate": 7.293129508999758e-06, | |
| "loss": 0.4254, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.4138982681625095, | |
| "eval_loss": 0.4347061216831207, | |
| "eval_runtime": 9.6662, | |
| "eval_samples_per_second": 516.439, | |
| "eval_steps_per_second": 2.069, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.41607668010020693, | |
| "grad_norm": 1.821865439414978, | |
| "learning_rate": 7.259273742477017e-06, | |
| "loss": 0.4201, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 0.41825509203790434, | |
| "grad_norm": 2.225336790084839, | |
| "learning_rate": 7.2252873119036684e-06, | |
| "loss": 0.4254, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 0.4204335039756018, | |
| "grad_norm": 3.4500694274902344, | |
| "learning_rate": 7.1911721828691904e-06, | |
| "loss": 0.4219, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 0.4226119159132992, | |
| "grad_norm": 2.226417064666748, | |
| "learning_rate": 7.156930328406268e-06, | |
| "loss": 0.4235, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 0.42479032785099663, | |
| "grad_norm": 4.276403427124023, | |
| "learning_rate": 7.122563728876698e-06, | |
| "loss": 0.4216, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 0.42696873978869404, | |
| "grad_norm": 4.915441513061523, | |
| "learning_rate": 7.088074371856847e-06, | |
| "loss": 0.4286, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 0.42914715172639145, | |
| "grad_norm": 4.776780605316162, | |
| "learning_rate": 7.053464252022708e-06, | |
| "loss": 0.4246, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 0.43132556366408886, | |
| "grad_norm": 2.5758256912231445, | |
| "learning_rate": 7.018735371034531e-06, | |
| "loss": 0.42, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 0.4335039756017863, | |
| "grad_norm": 2.303910255432129, | |
| "learning_rate": 6.9838897374210665e-06, | |
| "loss": 0.4183, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 0.43568238753948374, | |
| "grad_norm": 9.267624855041504, | |
| "learning_rate": 6.948929366463397e-06, | |
| "loss": 0.4193, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.43568238753948374, | |
| "eval_loss": 0.4357667863368988, | |
| "eval_runtime": 9.586, | |
| "eval_samples_per_second": 520.759, | |
| "eval_steps_per_second": 2.086, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.43786079947718115, | |
| "grad_norm": 2.5708439350128174, | |
| "learning_rate": 6.913856280078387e-06, | |
| "loss": 0.4255, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 0.44003921141487856, | |
| "grad_norm": 2.246631622314453, | |
| "learning_rate": 6.878672506701742e-06, | |
| "loss": 0.4179, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 0.442217623352576, | |
| "grad_norm": 2.604588270187378, | |
| "learning_rate": 6.843380081170704e-06, | |
| "loss": 0.4193, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 0.4443960352902734, | |
| "grad_norm": 2.7791225910186768, | |
| "learning_rate": 6.807981044606359e-06, | |
| "loss": 0.4199, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 0.4465744472279708, | |
| "grad_norm": 4.1223344802856445, | |
| "learning_rate": 6.772477444295592e-06, | |
| "loss": 0.4198, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 0.4487528591656682, | |
| "grad_norm": 1.6104360818862915, | |
| "learning_rate": 6.736871333572686e-06, | |
| "loss": 0.4166, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 0.45093127110336567, | |
| "grad_norm": 6.032946586608887, | |
| "learning_rate": 6.701164771700564e-06, | |
| "loss": 0.4196, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 0.4531096830410631, | |
| "grad_norm": 3.254844903945923, | |
| "learning_rate": 6.665359823751696e-06, | |
| "loss": 0.4205, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 0.4552880949787605, | |
| "grad_norm": 2.6492910385131836, | |
| "learning_rate": 6.629458560488664e-06, | |
| "loss": 0.4185, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 0.4574665069164579, | |
| "grad_norm": 7.052666664123535, | |
| "learning_rate": 6.593463058244407e-06, | |
| "loss": 0.4164, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.4574665069164579, | |
| "eval_loss": 0.4329228401184082, | |
| "eval_runtime": 9.6522, | |
| "eval_samples_per_second": 517.188, | |
| "eval_steps_per_second": 2.072, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.4596449188541553, | |
| "grad_norm": 3.9242568016052246, | |
| "learning_rate": 6.557375398802124e-06, | |
| "loss": 0.4188, | |
| "step": 2110 | |
| }, | |
| { | |
| "epoch": 0.4618233307918527, | |
| "grad_norm": 2.833991289138794, | |
| "learning_rate": 6.521197669274885e-06, | |
| "loss": 0.4166, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 0.46400174272955014, | |
| "grad_norm": 2.6651554107666016, | |
| "learning_rate": 6.484931961984927e-06, | |
| "loss": 0.4164, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 0.4661801546672476, | |
| "grad_norm": 1.856614112854004, | |
| "learning_rate": 6.448580374342632e-06, | |
| "loss": 0.4159, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 0.468358566604945, | |
| "grad_norm": 2.084019660949707, | |
| "learning_rate": 6.412145008725239e-06, | |
| "loss": 0.4155, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 0.4705369785426424, | |
| "grad_norm": 2.50307297706604, | |
| "learning_rate": 6.375627972355245e-06, | |
| "loss": 0.4155, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 0.47271539048033984, | |
| "grad_norm": 3.0096487998962402, | |
| "learning_rate": 6.339031377178536e-06, | |
| "loss": 0.4231, | |
| "step": 2170 | |
| }, | |
| { | |
| "epoch": 0.47489380241803725, | |
| "grad_norm": 2.756303071975708, | |
| "learning_rate": 6.302357339742245e-06, | |
| "loss": 0.418, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 0.47707221435573466, | |
| "grad_norm": 3.530158042907715, | |
| "learning_rate": 6.265607981072346e-06, | |
| "loss": 0.4165, | |
| "step": 2190 | |
| }, | |
| { | |
| "epoch": 0.47925062629343207, | |
| "grad_norm": 4.0776519775390625, | |
| "learning_rate": 6.228785426550976e-06, | |
| "loss": 0.4142, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.47925062629343207, | |
| "eval_loss": 0.4327417314052582, | |
| "eval_runtime": 9.6888, | |
| "eval_samples_per_second": 515.234, | |
| "eval_steps_per_second": 2.064, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.4814290382311295, | |
| "grad_norm": 2.6378114223480225, | |
| "learning_rate": 6.191891805793523e-06, | |
| "loss": 0.4127, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 0.48360745016882695, | |
| "grad_norm": 2.1185641288757324, | |
| "learning_rate": 6.1549292525254585e-06, | |
| "loss": 0.4145, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 0.48578586210652436, | |
| "grad_norm": 5.306119918823242, | |
| "learning_rate": 6.117899904458935e-06, | |
| "loss": 0.4153, | |
| "step": 2230 | |
| }, | |
| { | |
| "epoch": 0.48796427404422177, | |
| "grad_norm": 2.197924852371216, | |
| "learning_rate": 6.080805903169145e-06, | |
| "loss": 0.4156, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 0.4901426859819192, | |
| "grad_norm": 2.9446542263031006, | |
| "learning_rate": 6.043649393970478e-06, | |
| "loss": 0.414, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 0.4923210979196166, | |
| "grad_norm": 7.410131931304932, | |
| "learning_rate": 6.006432525792435e-06, | |
| "loss": 0.4185, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 0.494499509857314, | |
| "grad_norm": 3.202277898788452, | |
| "learning_rate": 5.9691574510553505e-06, | |
| "loss": 0.4155, | |
| "step": 2270 | |
| }, | |
| { | |
| "epoch": 0.4966779217950114, | |
| "grad_norm": 2.3827662467956543, | |
| "learning_rate": 5.931826325545912e-06, | |
| "loss": 0.412, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 0.4988563337327089, | |
| "grad_norm": 3.133537769317627, | |
| "learning_rate": 5.894441308292471e-06, | |
| "loss": 0.413, | |
| "step": 2290 | |
| }, | |
| { | |
| "epoch": 0.5010347456704063, | |
| "grad_norm": 3.3716533184051514, | |
| "learning_rate": 5.85700456144019e-06, | |
| "loss": 0.4119, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.5010347456704063, | |
| "eval_loss": 0.42866024374961853, | |
| "eval_runtime": 9.6451, | |
| "eval_samples_per_second": 517.568, | |
| "eval_steps_per_second": 2.074, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.5032131576081037, | |
| "grad_norm": 3.1281096935272217, | |
| "learning_rate": 5.819518250125986e-06, | |
| "loss": 0.4141, | |
| "step": 2310 | |
| }, | |
| { | |
| "epoch": 0.5053915695458011, | |
| "grad_norm": 2.40556263923645, | |
| "learning_rate": 5.781984542353313e-06, | |
| "loss": 0.412, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 0.5075699814834985, | |
| "grad_norm": 2.3655827045440674, | |
| "learning_rate": 5.744405608866775e-06, | |
| "loss": 0.4112, | |
| "step": 2330 | |
| }, | |
| { | |
| "epoch": 0.5097483934211959, | |
| "grad_norm": 3.645313024520874, | |
| "learning_rate": 5.7067836230265915e-06, | |
| "loss": 0.4102, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 0.5119268053588933, | |
| "grad_norm": 1.9526983499526978, | |
| "learning_rate": 5.669120760682888e-06, | |
| "loss": 0.414, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 0.5141052172965908, | |
| "grad_norm": 3.076853036880493, | |
| "learning_rate": 5.631419200049867e-06, | |
| "loss": 0.4122, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 0.5162836292342882, | |
| "grad_norm": 2.6010568141937256, | |
| "learning_rate": 5.593681121579828e-06, | |
| "loss": 0.4118, | |
| "step": 2370 | |
| }, | |
| { | |
| "epoch": 0.5184620411719856, | |
| "grad_norm": 1.341263771057129, | |
| "learning_rate": 5.555908707837061e-06, | |
| "loss": 0.4108, | |
| "step": 2380 | |
| }, | |
| { | |
| "epoch": 0.520640453109683, | |
| "grad_norm": 2.0360755920410156, | |
| "learning_rate": 5.518104143371623e-06, | |
| "loss": 0.4132, | |
| "step": 2390 | |
| }, | |
| { | |
| "epoch": 0.5228188650473805, | |
| "grad_norm": 1.9817912578582764, | |
| "learning_rate": 5.4802696145929925e-06, | |
| "loss": 0.4109, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.5228188650473805, | |
| "eval_loss": 0.42882996797561646, | |
| "eval_runtime": 9.7082, | |
| "eval_samples_per_second": 514.203, | |
| "eval_steps_per_second": 2.06, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.5249972769850779, | |
| "grad_norm": 1.8024429082870483, | |
| "learning_rate": 5.44240730964362e-06, | |
| "loss": 0.4126, | |
| "step": 2410 | |
| }, | |
| { | |
| "epoch": 0.5271756889227753, | |
| "grad_norm": 1.9342011213302612, | |
| "learning_rate": 5.404519418272376e-06, | |
| "loss": 0.4135, | |
| "step": 2420 | |
| }, | |
| { | |
| "epoch": 0.5293541008604727, | |
| "grad_norm": 1.8404808044433594, | |
| "learning_rate": 5.366608131707913e-06, | |
| "loss": 0.4089, | |
| "step": 2430 | |
| }, | |
| { | |
| "epoch": 0.5315325127981702, | |
| "grad_norm": 1.6226404905319214, | |
| "learning_rate": 5.328675642531934e-06, | |
| "loss": 0.4097, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 0.5337109247358676, | |
| "grad_norm": 3.6844136714935303, | |
| "learning_rate": 5.290724144552379e-06, | |
| "loss": 0.4098, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 0.535889336673565, | |
| "grad_norm": 4.781176567077637, | |
| "learning_rate": 5.252755832676562e-06, | |
| "loss": 0.4113, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 0.5380677486112624, | |
| "grad_norm": 3.025144338607788, | |
| "learning_rate": 5.214772902784218e-06, | |
| "loss": 0.4084, | |
| "step": 2470 | |
| }, | |
| { | |
| "epoch": 0.5402461605489598, | |
| "grad_norm": 1.8959859609603882, | |
| "learning_rate": 5.176777551600504e-06, | |
| "loss": 0.4085, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 0.5424245724866572, | |
| "grad_norm": 1.4723670482635498, | |
| "learning_rate": 5.13877197656896e-06, | |
| "loss": 0.4089, | |
| "step": 2490 | |
| }, | |
| { | |
| "epoch": 0.5446029844243546, | |
| "grad_norm": 1.6879379749298096, | |
| "learning_rate": 5.100758375724418e-06, | |
| "loss": 0.4117, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.5446029844243546, | |
| "eval_loss": 0.43057578802108765, | |
| "eval_runtime": 9.6671, | |
| "eval_samples_per_second": 516.389, | |
| "eval_steps_per_second": 2.069, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.546781396362052, | |
| "grad_norm": 2.652782440185547, | |
| "learning_rate": 5.062738947565879e-06, | |
| "loss": 0.4075, | |
| "step": 2510 | |
| }, | |
| { | |
| "epoch": 0.5489598082997494, | |
| "grad_norm": 3.549682140350342, | |
| "learning_rate": 5.0247158909293605e-06, | |
| "loss": 0.4061, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 0.5511382202374469, | |
| "grad_norm": 4.6838603019714355, | |
| "learning_rate": 4.986691404860738e-06, | |
| "loss": 0.4106, | |
| "step": 2530 | |
| }, | |
| { | |
| "epoch": 0.5533166321751444, | |
| "grad_norm": 2.4103927612304688, | |
| "learning_rate": 4.948667688488552e-06, | |
| "loss": 0.4093, | |
| "step": 2540 | |
| }, | |
| { | |
| "epoch": 0.5554950441128418, | |
| "grad_norm": 3.744180679321289, | |
| "learning_rate": 4.910646940896828e-06, | |
| "loss": 0.4082, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 0.5576734560505392, | |
| "grad_norm": 3.863417863845825, | |
| "learning_rate": 4.872631360997899e-06, | |
| "loss": 0.4061, | |
| "step": 2560 | |
| }, | |
| { | |
| "epoch": 0.5598518679882366, | |
| "grad_norm": 3.8889987468719482, | |
| "learning_rate": 4.83462314740522e-06, | |
| "loss": 0.4075, | |
| "step": 2570 | |
| }, | |
| { | |
| "epoch": 0.562030279925934, | |
| "grad_norm": 3.1181349754333496, | |
| "learning_rate": 4.796624498306224e-06, | |
| "loss": 0.4113, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 0.5642086918636314, | |
| "grad_norm": 2.6836493015289307, | |
| "learning_rate": 4.758637611335183e-06, | |
| "loss": 0.4088, | |
| "step": 2590 | |
| }, | |
| { | |
| "epoch": 0.5663871038013288, | |
| "grad_norm": 2.120168924331665, | |
| "learning_rate": 4.720664683446111e-06, | |
| "loss": 0.4073, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.5663871038013288, | |
| "eval_loss": 0.4350495934486389, | |
| "eval_runtime": 9.6231, | |
| "eval_samples_per_second": 518.752, | |
| "eval_steps_per_second": 2.078, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.5685655157390263, | |
| "grad_norm": 5.942187786102295, | |
| "learning_rate": 4.682707910785712e-06, | |
| "loss": 0.4149, | |
| "step": 2610 | |
| }, | |
| { | |
| "epoch": 0.5707439276767237, | |
| "grad_norm": 2.514582395553589, | |
| "learning_rate": 4.644769488566351e-06, | |
| "loss": 0.4083, | |
| "step": 2620 | |
| }, | |
| { | |
| "epoch": 0.5729223396144211, | |
| "grad_norm": 3.884115219116211, | |
| "learning_rate": 4.606851610939108e-06, | |
| "loss": 0.4052, | |
| "step": 2630 | |
| }, | |
| { | |
| "epoch": 0.5751007515521185, | |
| "grad_norm": 4.213008403778076, | |
| "learning_rate": 4.5689564708668735e-06, | |
| "loss": 0.4076, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 0.5772791634898159, | |
| "grad_norm": 2.6122772693634033, | |
| "learning_rate": 4.531086259997528e-06, | |
| "loss": 0.4088, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 0.5794575754275133, | |
| "grad_norm": 2.8703975677490234, | |
| "learning_rate": 4.493243168537178e-06, | |
| "loss": 0.4081, | |
| "step": 2660 | |
| }, | |
| { | |
| "epoch": 0.5816359873652107, | |
| "grad_norm": 2.920668840408325, | |
| "learning_rate": 4.455429385123489e-06, | |
| "loss": 0.4074, | |
| "step": 2670 | |
| }, | |
| { | |
| "epoch": 0.5838143993029081, | |
| "grad_norm": 2.7703404426574707, | |
| "learning_rate": 4.417647096699118e-06, | |
| "loss": 0.4037, | |
| "step": 2680 | |
| }, | |
| { | |
| "epoch": 0.5859928112406056, | |
| "grad_norm": 3.3332200050354004, | |
| "learning_rate": 4.379898488385215e-06, | |
| "loss": 0.4063, | |
| "step": 2690 | |
| }, | |
| { | |
| "epoch": 0.5881712231783031, | |
| "grad_norm": 1.8352432250976562, | |
| "learning_rate": 4.342185743355062e-06, | |
| "loss": 0.4062, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.5881712231783031, | |
| "eval_loss": 0.4279891550540924, | |
| "eval_runtime": 9.6691, | |
| "eval_samples_per_second": 516.286, | |
| "eval_steps_per_second": 2.068, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.5903496351160005, | |
| "grad_norm": 1.7296792268753052, | |
| "learning_rate": 4.304511042707802e-06, | |
| "loss": 0.4064, | |
| "step": 2710 | |
| }, | |
| { | |
| "epoch": 0.5925280470536979, | |
| "grad_norm": 3.613008499145508, | |
| "learning_rate": 4.266876565342298e-06, | |
| "loss": 0.4075, | |
| "step": 2720 | |
| }, | |
| { | |
| "epoch": 0.5947064589913953, | |
| "grad_norm": 3.5010597705841064, | |
| "learning_rate": 4.229284487831122e-06, | |
| "loss": 0.4091, | |
| "step": 2730 | |
| }, | |
| { | |
| "epoch": 0.5968848709290927, | |
| "grad_norm": 2.1212668418884277, | |
| "learning_rate": 4.191736984294667e-06, | |
| "loss": 0.4075, | |
| "step": 2740 | |
| }, | |
| { | |
| "epoch": 0.5990632828667901, | |
| "grad_norm": 2.979283571243286, | |
| "learning_rate": 4.154236226275411e-06, | |
| "loss": 0.404, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 0.6012416948044875, | |
| "grad_norm": 1.792733907699585, | |
| "learning_rate": 4.1167843826123265e-06, | |
| "loss": 0.4042, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 0.6034201067421849, | |
| "grad_norm": 3.041388750076294, | |
| "learning_rate": 4.079383619315447e-06, | |
| "loss": 0.4061, | |
| "step": 2770 | |
| }, | |
| { | |
| "epoch": 0.6055985186798823, | |
| "grad_norm": 2.6816020011901855, | |
| "learning_rate": 4.0420360994406e-06, | |
| "loss": 0.4071, | |
| "step": 2780 | |
| }, | |
| { | |
| "epoch": 0.6077769306175798, | |
| "grad_norm": 2.446981430053711, | |
| "learning_rate": 4.004743982964298e-06, | |
| "loss": 0.402, | |
| "step": 2790 | |
| }, | |
| { | |
| "epoch": 0.6099553425552772, | |
| "grad_norm": 2.502835273742676, | |
| "learning_rate": 3.96750942665883e-06, | |
| "loss": 0.4037, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.6099553425552772, | |
| "eval_loss": 0.4277077615261078, | |
| "eval_runtime": 9.743, | |
| "eval_samples_per_second": 512.368, | |
| "eval_steps_per_second": 2.053, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.6121337544929746, | |
| "grad_norm": 1.2970865964889526, | |
| "learning_rate": 3.930334583967514e-06, | |
| "loss": 0.404, | |
| "step": 2810 | |
| }, | |
| { | |
| "epoch": 0.614312166430672, | |
| "grad_norm": 3.3795993328094482, | |
| "learning_rate": 3.8932216048801615e-06, | |
| "loss": 0.4065, | |
| "step": 2820 | |
| }, | |
| { | |
| "epoch": 0.6164905783683695, | |
| "grad_norm": 2.872642755508423, | |
| "learning_rate": 3.856172635808732e-06, | |
| "loss": 0.4059, | |
| "step": 2830 | |
| }, | |
| { | |
| "epoch": 0.6186689903060669, | |
| "grad_norm": 2.930790662765503, | |
| "learning_rate": 3.8191898194631926e-06, | |
| "loss": 0.4053, | |
| "step": 2840 | |
| }, | |
| { | |
| "epoch": 0.6208474022437643, | |
| "grad_norm": 3.7308971881866455, | |
| "learning_rate": 3.782275294727601e-06, | |
| "loss": 0.4055, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 0.6230258141814617, | |
| "grad_norm": 3.462425470352173, | |
| "learning_rate": 3.745431196536398e-06, | |
| "loss": 0.4038, | |
| "step": 2860 | |
| }, | |
| { | |
| "epoch": 0.6252042261191592, | |
| "grad_norm": 2.3263561725616455, | |
| "learning_rate": 3.7086596557509396e-06, | |
| "loss": 0.4045, | |
| "step": 2870 | |
| }, | |
| { | |
| "epoch": 0.6273826380568566, | |
| "grad_norm": 1.485323190689087, | |
| "learning_rate": 3.6719627990362594e-06, | |
| "loss": 0.4025, | |
| "step": 2880 | |
| }, | |
| { | |
| "epoch": 0.629561049994554, | |
| "grad_norm": 3.1288955211639404, | |
| "learning_rate": 3.6353427487380687e-06, | |
| "loss": 0.402, | |
| "step": 2890 | |
| }, | |
| { | |
| "epoch": 0.6317394619322514, | |
| "grad_norm": 2.5085573196411133, | |
| "learning_rate": 3.598801622760021e-06, | |
| "loss": 0.4054, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.6317394619322514, | |
| "eval_loss": 0.42717447876930237, | |
| "eval_runtime": 9.6573, | |
| "eval_samples_per_second": 516.914, | |
| "eval_steps_per_second": 2.071, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.6339178738699488, | |
| "grad_norm": 4.0328049659729, | |
| "learning_rate": 3.5623415344412125e-06, | |
| "loss": 0.4006, | |
| "step": 2910 | |
| }, | |
| { | |
| "epoch": 0.6360962858076462, | |
| "grad_norm": 2.123347759246826, | |
| "learning_rate": 3.525964592433966e-06, | |
| "loss": 0.4046, | |
| "step": 2920 | |
| }, | |
| { | |
| "epoch": 0.6382746977453436, | |
| "grad_norm": 3.348815441131592, | |
| "learning_rate": 3.489672900581876e-06, | |
| "loss": 0.4025, | |
| "step": 2930 | |
| }, | |
| { | |
| "epoch": 0.640453109683041, | |
| "grad_norm": 4.057186603546143, | |
| "learning_rate": 3.453468557798133e-06, | |
| "loss": 0.4037, | |
| "step": 2940 | |
| }, | |
| { | |
| "epoch": 0.6426315216207384, | |
| "grad_norm": 3.547410249710083, | |
| "learning_rate": 3.417353657944137e-06, | |
| "loss": 0.403, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 0.6448099335584359, | |
| "grad_norm": 3.1567604541778564, | |
| "learning_rate": 3.3813302897083955e-06, | |
| "loss": 0.4027, | |
| "step": 2960 | |
| }, | |
| { | |
| "epoch": 0.6469883454961333, | |
| "grad_norm": 4.546987533569336, | |
| "learning_rate": 3.345400536485727e-06, | |
| "loss": 0.4029, | |
| "step": 2970 | |
| }, | |
| { | |
| "epoch": 0.6491667574338308, | |
| "grad_norm": 0.840340793132782, | |
| "learning_rate": 3.3095664762567687e-06, | |
| "loss": 0.4018, | |
| "step": 2980 | |
| }, | |
| { | |
| "epoch": 0.6513451693715282, | |
| "grad_norm": 1.2068535089492798, | |
| "learning_rate": 3.2738301814678015e-06, | |
| "loss": 0.4025, | |
| "step": 2990 | |
| }, | |
| { | |
| "epoch": 0.6535235813092256, | |
| "grad_norm": 2.895763635635376, | |
| "learning_rate": 3.238193718910883e-06, | |
| "loss": 0.4031, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.6535235813092256, | |
| "eval_loss": 0.42841657996177673, | |
| "eval_runtime": 9.6401, | |
| "eval_samples_per_second": 517.835, | |
| "eval_steps_per_second": 2.075, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.655701993246923, | |
| "grad_norm": 4.0693678855896, | |
| "learning_rate": 3.2026591496043226e-06, | |
| "loss": 0.4022, | |
| "step": 3010 | |
| }, | |
| { | |
| "epoch": 0.6578804051846204, | |
| "grad_norm": 3.907064914703369, | |
| "learning_rate": 3.167228528673478e-06, | |
| "loss": 0.4018, | |
| "step": 3020 | |
| }, | |
| { | |
| "epoch": 0.6600588171223178, | |
| "grad_norm": 6.200908660888672, | |
| "learning_rate": 3.1319039052319033e-06, | |
| "loss": 0.4035, | |
| "step": 3030 | |
| }, | |
| { | |
| "epoch": 0.6622372290600153, | |
| "grad_norm": 3.6325767040252686, | |
| "learning_rate": 3.0966873222628368e-06, | |
| "loss": 0.404, | |
| "step": 3040 | |
| }, | |
| { | |
| "epoch": 0.6644156409977127, | |
| "grad_norm": 2.363156795501709, | |
| "learning_rate": 3.0615808165010453e-06, | |
| "loss": 0.4034, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 0.6665940529354101, | |
| "grad_norm": 3.558595895767212, | |
| "learning_rate": 3.0265864183150306e-06, | |
| "loss": 0.4009, | |
| "step": 3060 | |
| }, | |
| { | |
| "epoch": 0.6687724648731075, | |
| "grad_norm": 3.482257604598999, | |
| "learning_rate": 2.991706151589606e-06, | |
| "loss": 0.4024, | |
| "step": 3070 | |
| }, | |
| { | |
| "epoch": 0.6709508768108049, | |
| "grad_norm": 2.492945909500122, | |
| "learning_rate": 2.956942033608843e-06, | |
| "loss": 0.4016, | |
| "step": 3080 | |
| }, | |
| { | |
| "epoch": 0.6731292887485023, | |
| "grad_norm": 2.730677366256714, | |
| "learning_rate": 2.9222960749394073e-06, | |
| "loss": 0.4028, | |
| "step": 3090 | |
| }, | |
| { | |
| "epoch": 0.6753077006861997, | |
| "grad_norm": 1.6951932907104492, | |
| "learning_rate": 2.887770279314273e-06, | |
| "loss": 0.3998, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 0.6753077006861997, | |
| "eval_loss": 0.42821940779685974, | |
| "eval_runtime": 9.5688, | |
| "eval_samples_per_second": 521.694, | |
| "eval_steps_per_second": 2.09, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 0.6774861126238971, | |
| "grad_norm": 3.0309555530548096, | |
| "learning_rate": 2.853366643516841e-06, | |
| "loss": 0.4005, | |
| "step": 3110 | |
| }, | |
| { | |
| "epoch": 0.6796645245615947, | |
| "grad_norm": 3.3345096111297607, | |
| "learning_rate": 2.8190871572654543e-06, | |
| "loss": 0.4013, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 0.6818429364992921, | |
| "grad_norm": 1.4350861310958862, | |
| "learning_rate": 2.784933803098326e-06, | |
| "loss": 0.4007, | |
| "step": 3130 | |
| }, | |
| { | |
| "epoch": 0.6840213484369895, | |
| "grad_norm": 1.698664665222168, | |
| "learning_rate": 2.7509085562588743e-06, | |
| "loss": 0.4002, | |
| "step": 3140 | |
| }, | |
| { | |
| "epoch": 0.6861997603746869, | |
| "grad_norm": 1.3729372024536133, | |
| "learning_rate": 2.717013384581498e-06, | |
| "loss": 0.4017, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 0.6883781723123843, | |
| "grad_norm": 1.2046958208084106, | |
| "learning_rate": 2.683250248377752e-06, | |
| "loss": 0.4025, | |
| "step": 3160 | |
| }, | |
| { | |
| "epoch": 0.6905565842500817, | |
| "grad_norm": 2.475562810897827, | |
| "learning_rate": 2.6496211003229795e-06, | |
| "loss": 0.3992, | |
| "step": 3170 | |
| }, | |
| { | |
| "epoch": 0.6927349961877791, | |
| "grad_norm": 3.6771671772003174, | |
| "learning_rate": 2.6161278853433885e-06, | |
| "loss": 0.402, | |
| "step": 3180 | |
| }, | |
| { | |
| "epoch": 0.6949134081254765, | |
| "grad_norm": 1.5112743377685547, | |
| "learning_rate": 2.582772540503555e-06, | |
| "loss": 0.4014, | |
| "step": 3190 | |
| }, | |
| { | |
| "epoch": 0.6970918200631739, | |
| "grad_norm": 2.9181530475616455, | |
| "learning_rate": 2.5495569948944e-06, | |
| "loss": 0.4003, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.6970918200631739, | |
| "eval_loss": 0.42963868379592896, | |
| "eval_runtime": 9.6232, | |
| "eval_samples_per_second": 518.744, | |
| "eval_steps_per_second": 2.078, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.6992702320008714, | |
| "grad_norm": 3.9118452072143555, | |
| "learning_rate": 2.5164831695216217e-06, | |
| "loss": 0.4006, | |
| "step": 3210 | |
| }, | |
| { | |
| "epoch": 0.7014486439385688, | |
| "grad_norm": 3.3592333793640137, | |
| "learning_rate": 2.4835529771945938e-06, | |
| "loss": 0.4019, | |
| "step": 3220 | |
| }, | |
| { | |
| "epoch": 0.7036270558762662, | |
| "grad_norm": 3.0435197353363037, | |
| "learning_rate": 2.450768322415739e-06, | |
| "loss": 0.3993, | |
| "step": 3230 | |
| }, | |
| { | |
| "epoch": 0.7058054678139636, | |
| "grad_norm": 1.2945152521133423, | |
| "learning_rate": 2.4181311012703833e-06, | |
| "loss": 0.4025, | |
| "step": 3240 | |
| }, | |
| { | |
| "epoch": 0.707983879751661, | |
| "grad_norm": 4.202663421630859, | |
| "learning_rate": 2.3856432013170978e-06, | |
| "loss": 0.4009, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 0.7101622916893584, | |
| "grad_norm": 1.378665566444397, | |
| "learning_rate": 2.35330650147853e-06, | |
| "loss": 0.3978, | |
| "step": 3260 | |
| }, | |
| { | |
| "epoch": 0.7123407036270559, | |
| "grad_norm": 1.7930536270141602, | |
| "learning_rate": 2.321122871932739e-06, | |
| "loss": 0.3982, | |
| "step": 3270 | |
| }, | |
| { | |
| "epoch": 0.7145191155647533, | |
| "grad_norm": 1.7397598028182983, | |
| "learning_rate": 2.289094174005037e-06, | |
| "loss": 0.4005, | |
| "step": 3280 | |
| }, | |
| { | |
| "epoch": 0.7166975275024507, | |
| "grad_norm": 2.92988920211792, | |
| "learning_rate": 2.257222260060335e-06, | |
| "loss": 0.4007, | |
| "step": 3290 | |
| }, | |
| { | |
| "epoch": 0.7188759394401482, | |
| "grad_norm": 4.326840877532959, | |
| "learning_rate": 2.2255089733960162e-06, | |
| "loss": 0.4021, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 0.7188759394401482, | |
| "eval_loss": 0.42818737030029297, | |
| "eval_runtime": 9.6384, | |
| "eval_samples_per_second": 517.929, | |
| "eval_steps_per_second": 2.075, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 0.7210543513778456, | |
| "grad_norm": 1.8383755683898926, | |
| "learning_rate": 2.1939561481353273e-06, | |
| "loss": 0.403, | |
| "step": 3310 | |
| }, | |
| { | |
| "epoch": 0.723232763315543, | |
| "grad_norm": 4.371707439422607, | |
| "learning_rate": 2.1625656091213056e-06, | |
| "loss": 0.4023, | |
| "step": 3320 | |
| }, | |
| { | |
| "epoch": 0.7254111752532404, | |
| "grad_norm": 5.149106502532959, | |
| "learning_rate": 2.131339171811236e-06, | |
| "loss": 0.3982, | |
| "step": 3330 | |
| }, | |
| { | |
| "epoch": 0.7275895871909378, | |
| "grad_norm": 2.2125470638275146, | |
| "learning_rate": 2.100278642171662e-06, | |
| "loss": 0.3994, | |
| "step": 3340 | |
| }, | |
| { | |
| "epoch": 0.7297679991286352, | |
| "grad_norm": 2.824982166290283, | |
| "learning_rate": 2.069385816573928e-06, | |
| "loss": 0.401, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 0.7319464110663326, | |
| "grad_norm": 3.157496213912964, | |
| "learning_rate": 2.0386624816902978e-06, | |
| "loss": 0.3996, | |
| "step": 3360 | |
| }, | |
| { | |
| "epoch": 0.73412482300403, | |
| "grad_norm": 2.6652979850769043, | |
| "learning_rate": 2.008110414390613e-06, | |
| "loss": 0.4014, | |
| "step": 3370 | |
| }, | |
| { | |
| "epoch": 0.7363032349417274, | |
| "grad_norm": 3.388657331466675, | |
| "learning_rate": 1.977731381639541e-06, | |
| "loss": 0.3998, | |
| "step": 3380 | |
| }, | |
| { | |
| "epoch": 0.7384816468794249, | |
| "grad_norm": 3.359178066253662, | |
| "learning_rate": 1.9475271403943686e-06, | |
| "loss": 0.3994, | |
| "step": 3390 | |
| }, | |
| { | |
| "epoch": 0.7406600588171223, | |
| "grad_norm": 1.4487273693084717, | |
| "learning_rate": 1.9174994375034034e-06, | |
| "loss": 0.3982, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 0.7406600588171223, | |
| "eval_loss": 0.429605096578598, | |
| "eval_runtime": 9.6367, | |
| "eval_samples_per_second": 518.021, | |
| "eval_steps_per_second": 2.075, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 0.7428384707548198, | |
| "grad_norm": 3.843029260635376, | |
| "learning_rate": 1.887650009604935e-06, | |
| "loss": 0.4, | |
| "step": 3410 | |
| }, | |
| { | |
| "epoch": 0.7450168826925172, | |
| "grad_norm": 4.428391456604004, | |
| "learning_rate": 1.8579805830268016e-06, | |
| "loss": 0.4005, | |
| "step": 3420 | |
| }, | |
| { | |
| "epoch": 0.7471952946302146, | |
| "grad_norm": 2.102658748626709, | |
| "learning_rate": 1.8284928736865499e-06, | |
| "loss": 0.3997, | |
| "step": 3430 | |
| }, | |
| { | |
| "epoch": 0.749373706567912, | |
| "grad_norm": 2.6355855464935303, | |
| "learning_rate": 1.7991885869921928e-06, | |
| "loss": 0.3983, | |
| "step": 3440 | |
| }, | |
| { | |
| "epoch": 0.7515521185056094, | |
| "grad_norm": 3.8726813793182373, | |
| "learning_rate": 1.7700694177435807e-06, | |
| "loss": 0.4002, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 0.7537305304433068, | |
| "grad_norm": 1.449800729751587, | |
| "learning_rate": 1.7411370500343812e-06, | |
| "loss": 0.3986, | |
| "step": 3460 | |
| }, | |
| { | |
| "epoch": 0.7559089423810043, | |
| "grad_norm": 2.0149781703948975, | |
| "learning_rate": 1.7123931571546826e-06, | |
| "loss": 0.3958, | |
| "step": 3470 | |
| }, | |
| { | |
| "epoch": 0.7580873543187017, | |
| "grad_norm": 1.6387373208999634, | |
| "learning_rate": 1.6838394014942179e-06, | |
| "loss": 0.3997, | |
| "step": 3480 | |
| }, | |
| { | |
| "epoch": 0.7602657662563991, | |
| "grad_norm": 1.898795485496521, | |
| "learning_rate": 1.6554774344462199e-06, | |
| "loss": 0.3986, | |
| "step": 3490 | |
| }, | |
| { | |
| "epoch": 0.7624441781940965, | |
| "grad_norm": 3.0024943351745605, | |
| "learning_rate": 1.6273088963119222e-06, | |
| "loss": 0.3988, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.7624441781940965, | |
| "eval_loss": 0.4297761917114258, | |
| "eval_runtime": 9.6746, | |
| "eval_samples_per_second": 515.992, | |
| "eval_steps_per_second": 2.067, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.7646225901317939, | |
| "grad_norm": 0.9623323082923889, | |
| "learning_rate": 1.5993354162056802e-06, | |
| "loss": 0.3997, | |
| "step": 3510 | |
| }, | |
| { | |
| "epoch": 0.7668010020694913, | |
| "grad_norm": 2.5859196186065674, | |
| "learning_rate": 1.5715586119607617e-06, | |
| "loss": 0.4001, | |
| "step": 3520 | |
| }, | |
| { | |
| "epoch": 0.7689794140071887, | |
| "grad_norm": 2.1984033584594727, | |
| "learning_rate": 1.5439800900357765e-06, | |
| "loss": 0.398, | |
| "step": 3530 | |
| }, | |
| { | |
| "epoch": 0.7711578259448861, | |
| "grad_norm": 1.5113025903701782, | |
| "learning_rate": 1.516601445421768e-06, | |
| "loss": 0.4008, | |
| "step": 3540 | |
| }, | |
| { | |
| "epoch": 0.7733362378825835, | |
| "grad_norm": 1.8336790800094604, | |
| "learning_rate": 1.4894242615499689e-06, | |
| "loss": 0.3986, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 0.7755146498202811, | |
| "grad_norm": 2.773827075958252, | |
| "learning_rate": 1.4624501102002209e-06, | |
| "loss": 0.3984, | |
| "step": 3560 | |
| }, | |
| { | |
| "epoch": 0.7776930617579785, | |
| "grad_norm": 2.334609270095825, | |
| "learning_rate": 1.4356805514100763e-06, | |
| "loss": 0.4002, | |
| "step": 3570 | |
| }, | |
| { | |
| "epoch": 0.7798714736956759, | |
| "grad_norm": 1.920845627784729, | |
| "learning_rate": 1.4091171333845692e-06, | |
| "loss": 0.3957, | |
| "step": 3580 | |
| }, | |
| { | |
| "epoch": 0.7820498856333733, | |
| "grad_norm": 1.4427253007888794, | |
| "learning_rate": 1.3827613924066817e-06, | |
| "loss": 0.396, | |
| "step": 3590 | |
| }, | |
| { | |
| "epoch": 0.7842282975710707, | |
| "grad_norm": 1.5900869369506836, | |
| "learning_rate": 1.3566148527484863e-06, | |
| "loss": 0.3988, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.7842282975710707, | |
| "eval_loss": 0.42993584275245667, | |
| "eval_runtime": 9.6537, | |
| "eval_samples_per_second": 517.108, | |
| "eval_steps_per_second": 2.072, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.7864067095087681, | |
| "grad_norm": 0.8249953389167786, | |
| "learning_rate": 1.330679026582996e-06, | |
| "loss": 0.3996, | |
| "step": 3610 | |
| }, | |
| { | |
| "epoch": 0.7885851214464655, | |
| "grad_norm": 0.8866672515869141, | |
| "learning_rate": 1.3049554138967052e-06, | |
| "loss": 0.3967, | |
| "step": 3620 | |
| }, | |
| { | |
| "epoch": 0.7907635333841629, | |
| "grad_norm": 1.9449257850646973, | |
| "learning_rate": 1.2794455024028396e-06, | |
| "loss": 0.3978, | |
| "step": 3630 | |
| }, | |
| { | |
| "epoch": 0.7929419453218604, | |
| "grad_norm": 2.8876960277557373, | |
| "learning_rate": 1.25415076745532e-06, | |
| "loss": 0.3974, | |
| "step": 3640 | |
| }, | |
| { | |
| "epoch": 0.7951203572595578, | |
| "grad_norm": 1.4642447233200073, | |
| "learning_rate": 1.2290726719634255e-06, | |
| "loss": 0.3963, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 0.7972987691972552, | |
| "grad_norm": 0.9553173184394836, | |
| "learning_rate": 1.2042126663071952e-06, | |
| "loss": 0.3967, | |
| "step": 3660 | |
| }, | |
| { | |
| "epoch": 0.7994771811349526, | |
| "grad_norm": 3.112318515777588, | |
| "learning_rate": 1.1795721882535439e-06, | |
| "loss": 0.3984, | |
| "step": 3670 | |
| }, | |
| { | |
| "epoch": 0.80165559307265, | |
| "grad_norm": 3.6302273273468018, | |
| "learning_rate": 1.1551526628731074e-06, | |
| "loss": 0.3967, | |
| "step": 3680 | |
| }, | |
| { | |
| "epoch": 0.8038340050103474, | |
| "grad_norm": 1.9708794355392456, | |
| "learning_rate": 1.130955502457829e-06, | |
| "loss": 0.3972, | |
| "step": 3690 | |
| }, | |
| { | |
| "epoch": 0.8060124169480449, | |
| "grad_norm": 2.380471706390381, | |
| "learning_rate": 1.1069821064392722e-06, | |
| "loss": 0.3949, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 0.8060124169480449, | |
| "eval_loss": 0.4309428334236145, | |
| "eval_runtime": 9.5957, | |
| "eval_samples_per_second": 520.233, | |
| "eval_steps_per_second": 2.084, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 0.8081908288857423, | |
| "grad_norm": 2.012770652770996, | |
| "learning_rate": 1.083233861307697e-06, | |
| "loss": 0.3977, | |
| "step": 3710 | |
| }, | |
| { | |
| "epoch": 0.8103692408234398, | |
| "grad_norm": 1.0971107482910156, | |
| "learning_rate": 1.059712140531861e-06, | |
| "loss": 0.3984, | |
| "step": 3720 | |
| }, | |
| { | |
| "epoch": 0.8125476527611372, | |
| "grad_norm": 0.9638540744781494, | |
| "learning_rate": 1.0364183044795929e-06, | |
| "loss": 0.3956, | |
| "step": 3730 | |
| }, | |
| { | |
| "epoch": 0.8147260646988346, | |
| "grad_norm": 0.7911339402198792, | |
| "learning_rate": 1.013353700339112e-06, | |
| "loss": 0.397, | |
| "step": 3740 | |
| }, | |
| { | |
| "epoch": 0.816904476636532, | |
| "grad_norm": 0.8662415146827698, | |
| "learning_rate": 9.905196620411178e-07, | |
| "loss": 0.3965, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 0.8190828885742294, | |
| "grad_norm": 1.298143744468689, | |
| "learning_rate": 9.679175101816408e-07, | |
| "loss": 0.395, | |
| "step": 3760 | |
| }, | |
| { | |
| "epoch": 0.8212613005119268, | |
| "grad_norm": 0.6701905131340027, | |
| "learning_rate": 9.455485519456658e-07, | |
| "loss": 0.3947, | |
| "step": 3770 | |
| }, | |
| { | |
| "epoch": 0.8234397124496242, | |
| "grad_norm": 0.8428287506103516, | |
| "learning_rate": 9.234140810315345e-07, | |
| "loss": 0.3962, | |
| "step": 3780 | |
| }, | |
| { | |
| "epoch": 0.8256181243873216, | |
| "grad_norm": 2.1166293621063232, | |
| "learning_rate": 9.015153775761199e-07, | |
| "loss": 0.3946, | |
| "step": 3790 | |
| }, | |
| { | |
| "epoch": 0.827796536325019, | |
| "grad_norm": 3.267383575439453, | |
| "learning_rate": 8.79853708080795e-07, | |
| "loss": 0.3961, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 0.827796536325019, | |
| "eval_loss": 0.4297568202018738, | |
| "eval_runtime": 9.5765, | |
| "eval_samples_per_second": 521.274, | |
| "eval_steps_per_second": 2.088, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 0.8299749482627164, | |
| "grad_norm": 1.0721772909164429, | |
| "learning_rate": 8.584303253381848e-07, | |
| "loss": 0.3958, | |
| "step": 3810 | |
| }, | |
| { | |
| "epoch": 0.8321533602004139, | |
| "grad_norm": 1.8855377435684204, | |
| "learning_rate": 8.372464683597081e-07, | |
| "loss": 0.3953, | |
| "step": 3820 | |
| }, | |
| { | |
| "epoch": 0.8343317721381113, | |
| "grad_norm": 0.7978574633598328, | |
| "learning_rate": 8.16303362303923e-07, | |
| "loss": 0.3931, | |
| "step": 3830 | |
| }, | |
| { | |
| "epoch": 0.8365101840758087, | |
| "grad_norm": 1.466511845588684, | |
| "learning_rate": 7.956022184056689e-07, | |
| "loss": 0.3969, | |
| "step": 3840 | |
| }, | |
| { | |
| "epoch": 0.8386885960135062, | |
| "grad_norm": 1.6769752502441406, | |
| "learning_rate": 7.751442339060161e-07, | |
| "loss": 0.3958, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 0.8408670079512036, | |
| "grad_norm": 1.4472180604934692, | |
| "learning_rate": 7.549305919830241e-07, | |
| "loss": 0.3969, | |
| "step": 3860 | |
| }, | |
| { | |
| "epoch": 0.843045419888901, | |
| "grad_norm": 1.6235806941986084, | |
| "learning_rate": 7.349624616833123e-07, | |
| "loss": 0.3966, | |
| "step": 3870 | |
| }, | |
| { | |
| "epoch": 0.8452238318265984, | |
| "grad_norm": 0.7935024499893188, | |
| "learning_rate": 7.152409978544483e-07, | |
| "loss": 0.3941, | |
| "step": 3880 | |
| }, | |
| { | |
| "epoch": 0.8474022437642958, | |
| "grad_norm": 0.7577910423278809, | |
| "learning_rate": 6.957673410781617e-07, | |
| "loss": 0.3946, | |
| "step": 3890 | |
| }, | |
| { | |
| "epoch": 0.8495806557019933, | |
| "grad_norm": 0.8020604252815247, | |
| "learning_rate": 6.765426176043721e-07, | |
| "loss": 0.3952, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 0.8495806557019933, | |
| "eval_loss": 0.4307285249233246, | |
| "eval_runtime": 9.5556, | |
| "eval_samples_per_second": 522.416, | |
| "eval_steps_per_second": 2.093, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 0.8517590676396907, | |
| "grad_norm": 1.1028465032577515, | |
| "learning_rate": 6.57567939286059e-07, | |
| "loss": 0.396, | |
| "step": 3910 | |
| }, | |
| { | |
| "epoch": 0.8539374795773881, | |
| "grad_norm": 1.3756065368652344, | |
| "learning_rate": 6.388444035149538e-07, | |
| "loss": 0.3975, | |
| "step": 3920 | |
| }, | |
| { | |
| "epoch": 0.8561158915150855, | |
| "grad_norm": 1.218600869178772, | |
| "learning_rate": 6.203730931580798e-07, | |
| "loss": 0.3936, | |
| "step": 3930 | |
| }, | |
| { | |
| "epoch": 0.8582943034527829, | |
| "grad_norm": 0.8649505972862244, | |
| "learning_rate": 6.021550764951157e-07, | |
| "loss": 0.3947, | |
| "step": 3940 | |
| }, | |
| { | |
| "epoch": 0.8604727153904803, | |
| "grad_norm": 3.2568881511688232, | |
| "learning_rate": 5.841914071566179e-07, | |
| "loss": 0.3956, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 0.8626511273281777, | |
| "grad_norm": 1.6868760585784912, | |
| "learning_rate": 5.664831240630841e-07, | |
| "loss": 0.3974, | |
| "step": 3960 | |
| }, | |
| { | |
| "epoch": 0.8648295392658751, | |
| "grad_norm": 0.7339500784873962, | |
| "learning_rate": 5.490312513648666e-07, | |
| "loss": 0.395, | |
| "step": 3970 | |
| }, | |
| { | |
| "epoch": 0.8670079512035725, | |
| "grad_norm": 0.862421989440918, | |
| "learning_rate": 5.318367983829393e-07, | |
| "loss": 0.3947, | |
| "step": 3980 | |
| }, | |
| { | |
| "epoch": 0.8691863631412701, | |
| "grad_norm": 1.4295686483383179, | |
| "learning_rate": 5.149007595505285e-07, | |
| "loss": 0.3943, | |
| "step": 3990 | |
| }, | |
| { | |
| "epoch": 0.8713647750789675, | |
| "grad_norm": 1.2400861978530884, | |
| "learning_rate": 4.982241143555966e-07, | |
| "loss": 0.397, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.8713647750789675, | |
| "eval_loss": 0.43102139234542847, | |
| "eval_runtime": 9.6775, | |
| "eval_samples_per_second": 515.838, | |
| "eval_steps_per_second": 2.067, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.8735431870166649, | |
| "grad_norm": 1.7299386262893677, | |
| "learning_rate": 4.818078272841942e-07, | |
| "loss": 0.3977, | |
| "step": 4010 | |
| }, | |
| { | |
| "epoch": 0.8757215989543623, | |
| "grad_norm": 0.814769446849823, | |
| "learning_rate": 4.6565284776468376e-07, | |
| "loss": 0.3936, | |
| "step": 4020 | |
| }, | |
| { | |
| "epoch": 0.8779000108920597, | |
| "grad_norm": 1.483109712600708, | |
| "learning_rate": 4.4976011011282294e-07, | |
| "loss": 0.398, | |
| "step": 4030 | |
| }, | |
| { | |
| "epoch": 0.8800784228297571, | |
| "grad_norm": 2.0057575702667236, | |
| "learning_rate": 4.3413053347773406e-07, | |
| "loss": 0.3951, | |
| "step": 4040 | |
| }, | |
| { | |
| "epoch": 0.8822568347674545, | |
| "grad_norm": 0.9725803136825562, | |
| "learning_rate": 4.1876502178874403e-07, | |
| "loss": 0.3962, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 0.884435246705152, | |
| "grad_norm": 1.1756798028945923, | |
| "learning_rate": 4.036644637031045e-07, | |
| "loss": 0.3953, | |
| "step": 4060 | |
| }, | |
| { | |
| "epoch": 0.8866136586428494, | |
| "grad_norm": 1.6209012269973755, | |
| "learning_rate": 3.8882973255459975e-07, | |
| "loss": 0.3947, | |
| "step": 4070 | |
| }, | |
| { | |
| "epoch": 0.8887920705805468, | |
| "grad_norm": 1.3361040353775024, | |
| "learning_rate": 3.742616863030352e-07, | |
| "loss": 0.3952, | |
| "step": 4080 | |
| }, | |
| { | |
| "epoch": 0.8909704825182442, | |
| "grad_norm": 0.8442427515983582, | |
| "learning_rate": 3.5996116748461984e-07, | |
| "loss": 0.3943, | |
| "step": 4090 | |
| }, | |
| { | |
| "epoch": 0.8931488944559416, | |
| "grad_norm": 1.1721961498260498, | |
| "learning_rate": 3.459290031632362e-07, | |
| "loss": 0.3935, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 0.8931488944559416, | |
| "eval_loss": 0.4306623339653015, | |
| "eval_runtime": 9.6333, | |
| "eval_samples_per_second": 518.202, | |
| "eval_steps_per_second": 2.076, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 0.895327306393639, | |
| "grad_norm": 0.9018988013267517, | |
| "learning_rate": 3.321660048826092e-07, | |
| "loss": 0.3938, | |
| "step": 4110 | |
| }, | |
| { | |
| "epoch": 0.8975057183313364, | |
| "grad_norm": 0.8927802443504333, | |
| "learning_rate": 3.186729686193718e-07, | |
| "loss": 0.3968, | |
| "step": 4120 | |
| }, | |
| { | |
| "epoch": 0.8996841302690338, | |
| "grad_norm": 1.160204291343689, | |
| "learning_rate": 3.0545067473702617e-07, | |
| "loss": 0.3959, | |
| "step": 4130 | |
| }, | |
| { | |
| "epoch": 0.9018625422067313, | |
| "grad_norm": 0.6945912837982178, | |
| "learning_rate": 2.92499887940817e-07, | |
| "loss": 0.3948, | |
| "step": 4140 | |
| }, | |
| { | |
| "epoch": 0.9040409541444288, | |
| "grad_norm": 0.9483731389045715, | |
| "learning_rate": 2.798213572335001e-07, | |
| "loss": 0.3955, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 0.9062193660821262, | |
| "grad_norm": 0.9849236607551575, | |
| "learning_rate": 2.6741581587202747e-07, | |
| "loss": 0.3946, | |
| "step": 4160 | |
| }, | |
| { | |
| "epoch": 0.9083977780198236, | |
| "grad_norm": 1.0742136240005493, | |
| "learning_rate": 2.55283981325139e-07, | |
| "loss": 0.3954, | |
| "step": 4170 | |
| }, | |
| { | |
| "epoch": 0.910576189957521, | |
| "grad_norm": 0.8884495496749878, | |
| "learning_rate": 2.4342655523186706e-07, | |
| "loss": 0.3922, | |
| "step": 4180 | |
| }, | |
| { | |
| "epoch": 0.9127546018952184, | |
| "grad_norm": 0.7058791518211365, | |
| "learning_rate": 2.3184422336095857e-07, | |
| "loss": 0.3929, | |
| "step": 4190 | |
| }, | |
| { | |
| "epoch": 0.9149330138329158, | |
| "grad_norm": 1.1460115909576416, | |
| "learning_rate": 2.2053765557121298e-07, | |
| "loss": 0.3931, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 0.9149330138329158, | |
| "eval_loss": 0.43217477202415466, | |
| "eval_runtime": 9.5759, | |
| "eval_samples_per_second": 521.307, | |
| "eval_steps_per_second": 2.089, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 0.9171114257706132, | |
| "grad_norm": 1.6075760126113892, | |
| "learning_rate": 2.0950750577274247e-07, | |
| "loss": 0.3943, | |
| "step": 4210 | |
| }, | |
| { | |
| "epoch": 0.9192898377083106, | |
| "grad_norm": 1.4762377738952637, | |
| "learning_rate": 1.9875441188915324e-07, | |
| "loss": 0.395, | |
| "step": 4220 | |
| }, | |
| { | |
| "epoch": 0.921468249646008, | |
| "grad_norm": 0.6985810995101929, | |
| "learning_rate": 1.8827899582064824e-07, | |
| "loss": 0.3941, | |
| "step": 4230 | |
| }, | |
| { | |
| "epoch": 0.9236466615837055, | |
| "grad_norm": 1.9295434951782227, | |
| "learning_rate": 1.7808186340806664e-07, | |
| "loss": 0.3929, | |
| "step": 4240 | |
| }, | |
| { | |
| "epoch": 0.9258250735214029, | |
| "grad_norm": 0.6554338335990906, | |
| "learning_rate": 1.6816360439783797e-07, | |
| "loss": 0.3932, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 0.9280034854591003, | |
| "grad_norm": 0.8134569525718689, | |
| "learning_rate": 1.5852479240787978e-07, | |
| "loss": 0.3929, | |
| "step": 4260 | |
| }, | |
| { | |
| "epoch": 0.9301818973967977, | |
| "grad_norm": 0.6693784594535828, | |
| "learning_rate": 1.4916598489441824e-07, | |
| "loss": 0.3932, | |
| "step": 4270 | |
| }, | |
| { | |
| "epoch": 0.9323603093344952, | |
| "grad_norm": 0.8736308217048645, | |
| "learning_rate": 1.400877231197534e-07, | |
| "loss": 0.3918, | |
| "step": 4280 | |
| }, | |
| { | |
| "epoch": 0.9345387212721926, | |
| "grad_norm": 0.8335554599761963, | |
| "learning_rate": 1.3129053212094956e-07, | |
| "loss": 0.3943, | |
| "step": 4290 | |
| }, | |
| { | |
| "epoch": 0.93671713320989, | |
| "grad_norm": 0.784169614315033, | |
| "learning_rate": 1.2277492067947428e-07, | |
| "loss": 0.3942, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 0.93671713320989, | |
| "eval_loss": 0.4312984347343445, | |
| "eval_runtime": 9.573, | |
| "eval_samples_per_second": 521.466, | |
| "eval_steps_per_second": 2.089, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 0.9388955451475874, | |
| "grad_norm": 0.7941382527351379, | |
| "learning_rate": 1.1454138129177172e-07, | |
| "loss": 0.3953, | |
| "step": 4310 | |
| }, | |
| { | |
| "epoch": 0.9410739570852849, | |
| "grad_norm": 0.8930506110191345, | |
| "learning_rate": 1.0659039014077943e-07, | |
| "loss": 0.3922, | |
| "step": 4320 | |
| }, | |
| { | |
| "epoch": 0.9432523690229823, | |
| "grad_norm": 0.682492733001709, | |
| "learning_rate": 9.892240706838808e-08, | |
| "loss": 0.3958, | |
| "step": 4330 | |
| }, | |
| { | |
| "epoch": 0.9454307809606797, | |
| "grad_norm": 0.6189770698547363, | |
| "learning_rate": 9.15378755488483e-08, | |
| "loss": 0.3945, | |
| "step": 4340 | |
| }, | |
| { | |
| "epoch": 0.9476091928983771, | |
| "grad_norm": 1.1786706447601318, | |
| "learning_rate": 8.443722266312015e-08, | |
| "loss": 0.3943, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 0.9497876048360745, | |
| "grad_norm": 0.7087965607643127, | |
| "learning_rate": 7.762085907417505e-08, | |
| "loss": 0.3931, | |
| "step": 4360 | |
| }, | |
| { | |
| "epoch": 0.9519660167737719, | |
| "grad_norm": 0.8723616600036621, | |
| "learning_rate": 7.108917900324419e-08, | |
| "loss": 0.3934, | |
| "step": 4370 | |
| }, | |
| { | |
| "epoch": 0.9541444287114693, | |
| "grad_norm": 0.8475690484046936, | |
| "learning_rate": 6.484256020701962e-08, | |
| "loss": 0.3937, | |
| "step": 4380 | |
| }, | |
| { | |
| "epoch": 0.9563228406491667, | |
| "grad_norm": 0.9160191416740417, | |
| "learning_rate": 5.8881363955807234e-08, | |
| "loss": 0.3966, | |
| "step": 4390 | |
| }, | |
| { | |
| "epoch": 0.9585012525868641, | |
| "grad_norm": 0.7470030784606934, | |
| "learning_rate": 5.320593501263071e-08, | |
| "loss": 0.3951, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 0.9585012525868641, | |
| "eval_loss": 0.4317244589328766, | |
| "eval_runtime": 9.5949, | |
| "eval_samples_per_second": 520.275, | |
| "eval_steps_per_second": 2.084, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 0.9606796645245615, | |
| "grad_norm": 0.7705516219139099, | |
| "learning_rate": 4.7816601613295263e-08, | |
| "loss": 0.3955, | |
| "step": 4410 | |
| }, | |
| { | |
| "epoch": 0.962858076462259, | |
| "grad_norm": 0.6091457605361938, | |
| "learning_rate": 4.271367544740113e-08, | |
| "loss": 0.3923, | |
| "step": 4420 | |
| }, | |
| { | |
| "epoch": 0.9650364883999565, | |
| "grad_norm": 0.6471231579780579, | |
| "learning_rate": 3.7897451640321326e-08, | |
| "loss": 0.3947, | |
| "step": 4430 | |
| }, | |
| { | |
| "epoch": 0.9672149003376539, | |
| "grad_norm": 0.8809778094291687, | |
| "learning_rate": 3.3368208736128116e-08, | |
| "loss": 0.393, | |
| "step": 4440 | |
| }, | |
| { | |
| "epoch": 0.9693933122753513, | |
| "grad_norm": 0.7096254825592041, | |
| "learning_rate": 2.9126208681488076e-08, | |
| "loss": 0.3916, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 0.9715717242130487, | |
| "grad_norm": 0.8418611288070679, | |
| "learning_rate": 2.5171696810509794e-08, | |
| "loss": 0.3942, | |
| "step": 4460 | |
| }, | |
| { | |
| "epoch": 0.9737501361507461, | |
| "grad_norm": 0.6566638350486755, | |
| "learning_rate": 2.1504901830555203e-08, | |
| "loss": 0.3929, | |
| "step": 4470 | |
| }, | |
| { | |
| "epoch": 0.9759285480884435, | |
| "grad_norm": 0.7813677787780762, | |
| "learning_rate": 1.812603580901462e-08, | |
| "loss": 0.3926, | |
| "step": 4480 | |
| }, | |
| { | |
| "epoch": 0.978106960026141, | |
| "grad_norm": 0.7559650540351868, | |
| "learning_rate": 1.5035294161039882e-08, | |
| "loss": 0.3955, | |
| "step": 4490 | |
| }, | |
| { | |
| "epoch": 0.9802853719638384, | |
| "grad_norm": 0.799744188785553, | |
| "learning_rate": 1.2232855638241726e-08, | |
| "loss": 0.3922, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.9802853719638384, | |
| "eval_loss": 0.43170347809791565, | |
| "eval_runtime": 9.6697, | |
| "eval_samples_per_second": 516.251, | |
| "eval_steps_per_second": 2.068, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.9824637839015358, | |
| "grad_norm": 0.6862944960594177, | |
| "learning_rate": 9.718882318355272e-09, | |
| "loss": 0.395, | |
| "step": 4510 | |
| }, | |
| { | |
| "epoch": 0.9846421958392332, | |
| "grad_norm": 0.7467820048332214, | |
| "learning_rate": 7.49351959586253e-09, | |
| "loss": 0.3943, | |
| "step": 4520 | |
| }, | |
| { | |
| "epoch": 0.9868206077769306, | |
| "grad_norm": 0.789197564125061, | |
| "learning_rate": 5.556896173586346e-09, | |
| "loss": 0.3942, | |
| "step": 4530 | |
| }, | |
| { | |
| "epoch": 0.988999019714628, | |
| "grad_norm": 0.7414279580116272, | |
| "learning_rate": 3.9091240552457945e-09, | |
| "loss": 0.3938, | |
| "step": 4540 | |
| }, | |
| { | |
| "epoch": 0.9911774316523254, | |
| "grad_norm": 0.6362519860267639, | |
| "learning_rate": 2.5502985389774806e-09, | |
| "loss": 0.3957, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 0.9933558435900228, | |
| "grad_norm": 0.7834025621414185, | |
| "learning_rate": 1.4804982118266087e-09, | |
| "loss": 0.3927, | |
| "step": 4560 | |
| }, | |
| { | |
| "epoch": 0.9955342555277203, | |
| "grad_norm": 0.6968185901641846, | |
| "learning_rate": 6.997849451989558e-10, | |
| "loss": 0.3935, | |
| "step": 4570 | |
| }, | |
| { | |
| "epoch": 0.9977126674654178, | |
| "grad_norm": 0.7227827906608582, | |
| "learning_rate": 2.0820389128539674e-10, | |
| "loss": 0.3946, | |
| "step": 4580 | |
| }, | |
| { | |
| "epoch": 0.9998910794031152, | |
| "grad_norm": 0.5763871669769287, | |
| "learning_rate": 5.7834804478851255e-12, | |
| "loss": 0.3973, | |
| "step": 4590 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 4591, | |
| "total_flos": 5.052191529362166e+18, | |
| "train_loss": 0.5649781546881453, | |
| "train_runtime": 40899.9, | |
| "train_samples_per_second": 57.46, | |
| "train_steps_per_second": 0.112 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 4591, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 300, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.052191529362166e+18, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |