| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.1952, | |
| "eval_steps": 500, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.032, | |
| "grad_norm": 4.924402713775635, | |
| "learning_rate": 5.69620253164557e-07, | |
| "loss": 1.003, | |
| "mean_token_accuracy": 0.7565429553389549, | |
| "num_tokens": 108351.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.064, | |
| "grad_norm": 5.024679183959961, | |
| "learning_rate": 1.2025316455696204e-06, | |
| "loss": 0.9631, | |
| "mean_token_accuracy": 0.7678055994212627, | |
| "num_tokens": 220746.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.096, | |
| "grad_norm": 3.090874671936035, | |
| "learning_rate": 1.8354430379746838e-06, | |
| "loss": 0.884, | |
| "mean_token_accuracy": 0.779016162455082, | |
| "num_tokens": 333736.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.128, | |
| "grad_norm": 2.65234637260437, | |
| "learning_rate": 2.4683544303797473e-06, | |
| "loss": 0.7815, | |
| "mean_token_accuracy": 0.8002516604959965, | |
| "num_tokens": 449637.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 2.4638168811798096, | |
| "learning_rate": 3.10126582278481e-06, | |
| "loss": 0.75, | |
| "mean_token_accuracy": 0.8030893772840499, | |
| "num_tokens": 562989.0, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 2.374603509902954, | |
| "learning_rate": 3.7341772151898737e-06, | |
| "loss": 0.7152, | |
| "mean_token_accuracy": 0.8119513258337975, | |
| "num_tokens": 673839.0, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.224, | |
| "grad_norm": 2.330502510070801, | |
| "learning_rate": 4.367088607594937e-06, | |
| "loss": 0.6941, | |
| "mean_token_accuracy": 0.815937215834856, | |
| "num_tokens": 781061.0, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.256, | |
| "grad_norm": 2.267749309539795, | |
| "learning_rate": 5e-06, | |
| "loss": 0.696, | |
| "mean_token_accuracy": 0.8140616729855538, | |
| "num_tokens": 891312.0, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 2.236624240875244, | |
| "learning_rate": 4.9994413292115925e-06, | |
| "loss": 0.6568, | |
| "mean_token_accuracy": 0.8207014128565788, | |
| "num_tokens": 1005632.0, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 2.1301517486572266, | |
| "learning_rate": 4.997765566536809e-06, | |
| "loss": 0.6286, | |
| "mean_token_accuracy": 0.8304322719573974, | |
| "num_tokens": 1119621.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.352, | |
| "grad_norm": 2.2088050842285156, | |
| "learning_rate": 4.9949734609353746e-06, | |
| "loss": 0.6173, | |
| "mean_token_accuracy": 0.8298865057528019, | |
| "num_tokens": 1228735.0, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 2.109389305114746, | |
| "learning_rate": 4.991066260301558e-06, | |
| "loss": 0.6392, | |
| "mean_token_accuracy": 0.8268058508634567, | |
| "num_tokens": 1339064.0, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.416, | |
| "grad_norm": 1.9944440126419067, | |
| "learning_rate": 4.986045710906447e-06, | |
| "loss": 0.5951, | |
| "mean_token_accuracy": 0.8359164826571941, | |
| "num_tokens": 1452954.0, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.448, | |
| "grad_norm": 2.0711472034454346, | |
| "learning_rate": 4.979914056617472e-06, | |
| "loss": 0.6136, | |
| "mean_token_accuracy": 0.8322145372629166, | |
| "num_tokens": 1564718.0, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 2.3020684719085693, | |
| "learning_rate": 4.972674037895542e-06, | |
| "loss": 0.593, | |
| "mean_token_accuracy": 0.8376348964869976, | |
| "num_tokens": 1673953.0, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.512, | |
| "grad_norm": 2.002972364425659, | |
| "learning_rate": 4.964328890570231e-06, | |
| "loss": 0.6062, | |
| "mean_token_accuracy": 0.8321727059781552, | |
| "num_tokens": 1780894.0, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.544, | |
| "grad_norm": 2.345299482345581, | |
| "learning_rate": 4.954882344393566e-06, | |
| "loss": 0.6018, | |
| "mean_token_accuracy": 0.8306759320199489, | |
| "num_tokens": 1895107.0, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 2.180422306060791, | |
| "learning_rate": 4.94433862137307e-06, | |
| "loss": 0.5945, | |
| "mean_token_accuracy": 0.834496195614338, | |
| "num_tokens": 2004705.0, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.608, | |
| "grad_norm": 2.1533453464508057, | |
| "learning_rate": 4.9327024338847836e-06, | |
| "loss": 0.5989, | |
| "mean_token_accuracy": 0.8349006526172161, | |
| "num_tokens": 2117436.0, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 2.128485918045044, | |
| "learning_rate": 4.919978982567138e-06, | |
| "loss": 0.5582, | |
| "mean_token_accuracy": 0.8449515514075756, | |
| "num_tokens": 2229604.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 2.323518753051758, | |
| "learning_rate": 4.906173953996596e-06, | |
| "loss": 0.5596, | |
| "mean_token_accuracy": 0.8436533316969872, | |
| "num_tokens": 2343563.0, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.704, | |
| "grad_norm": 2.2156243324279785, | |
| "learning_rate": 4.891293518146112e-06, | |
| "loss": 0.5855, | |
| "mean_token_accuracy": 0.8377246461808682, | |
| "num_tokens": 2453154.0, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.736, | |
| "grad_norm": 2.2261300086975098, | |
| "learning_rate": 4.875344325627551e-06, | |
| "loss": 0.5716, | |
| "mean_token_accuracy": 0.8415978990495205, | |
| "num_tokens": 2566369.0, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 2.0974161624908447, | |
| "learning_rate": 4.85833350471928e-06, | |
| "loss": 0.5628, | |
| "mean_token_accuracy": 0.8448635570704937, | |
| "num_tokens": 2674001.0, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 2.181368112564087, | |
| "learning_rate": 4.840268658180281e-06, | |
| "loss": 0.5575, | |
| "mean_token_accuracy": 0.8437174156308174, | |
| "num_tokens": 2785131.0, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.832, | |
| "grad_norm": 1.9916293621063232, | |
| "learning_rate": 4.821157859852201e-06, | |
| "loss": 0.5496, | |
| "mean_token_accuracy": 0.8463803254067898, | |
| "num_tokens": 2898708.0, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 1.9616676568984985, | |
| "learning_rate": 4.801009651050856e-06, | |
| "loss": 0.5599, | |
| "mean_token_accuracy": 0.8417543187737465, | |
| "num_tokens": 3008577.0, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.896, | |
| "grad_norm": 2.2142021656036377, | |
| "learning_rate": 4.779833036748801e-06, | |
| "loss": 0.5268, | |
| "mean_token_accuracy": 0.8514743112027645, | |
| "num_tokens": 3120102.0, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.928, | |
| "grad_norm": 1.8409771919250488, | |
| "learning_rate": 4.757637481550683e-06, | |
| "loss": 0.5443, | |
| "mean_token_accuracy": 0.8460176154971123, | |
| "num_tokens": 3234456.0, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 1.9530022144317627, | |
| "learning_rate": 4.73443290546316e-06, | |
| "loss": 0.5487, | |
| "mean_token_accuracy": 0.8462878614664078, | |
| "num_tokens": 3348821.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.992, | |
| "grad_norm": 1.9719116687774658, | |
| "learning_rate": 4.7102296794612865e-06, | |
| "loss": 0.5229, | |
| "mean_token_accuracy": 0.8531442761421204, | |
| "num_tokens": 3455469.0, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.0224, | |
| "grad_norm": 1.9529277086257935, | |
| "learning_rate": 4.6850386208533424e-06, | |
| "loss": 0.4849, | |
| "mean_token_accuracy": 0.8625573249239671, | |
| "num_tokens": 3561864.0, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.0544, | |
| "grad_norm": 2.105419635772705, | |
| "learning_rate": 4.658870988446189e-06, | |
| "loss": 0.4819, | |
| "mean_token_accuracy": 0.864888047426939, | |
| "num_tokens": 3672405.0, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.0864, | |
| "grad_norm": 2.035449504852295, | |
| "learning_rate": 4.631738477513286e-06, | |
| "loss": 0.4735, | |
| "mean_token_accuracy": 0.8661263197660446, | |
| "num_tokens": 3780404.0, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.1184, | |
| "grad_norm": 2.1748783588409424, | |
| "learning_rate": 4.603653214567654e-06, | |
| "loss": 0.4638, | |
| "mean_token_accuracy": 0.8661996729671955, | |
| "num_tokens": 3894359.0, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.1504, | |
| "grad_norm": 2.0794739723205566, | |
| "learning_rate": 4.574627751942087e-06, | |
| "loss": 0.4683, | |
| "mean_token_accuracy": 0.8664901047945023, | |
| "num_tokens": 4007235.0, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.1824, | |
| "grad_norm": 2.2591140270233154, | |
| "learning_rate": 4.544675062179056e-06, | |
| "loss": 0.4616, | |
| "mean_token_accuracy": 0.8676725938916207, | |
| "num_tokens": 4116127.0, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.2144, | |
| "grad_norm": 2.111358165740967, | |
| "learning_rate": 4.513808532232804e-06, | |
| "loss": 0.4742, | |
| "mean_token_accuracy": 0.8660650610923767, | |
| "num_tokens": 4228679.0, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.2464, | |
| "grad_norm": 2.115079402923584, | |
| "learning_rate": 4.482041957486229e-06, | |
| "loss": 0.4679, | |
| "mean_token_accuracy": 0.8653517335653305, | |
| "num_tokens": 4341814.0, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.2784, | |
| "grad_norm": 2.15097713470459, | |
| "learning_rate": 4.4493895355852176e-06, | |
| "loss": 0.448, | |
| "mean_token_accuracy": 0.8716798067092896, | |
| "num_tokens": 4451355.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.3104, | |
| "grad_norm": 2.124791383743286, | |
| "learning_rate": 4.415865860093199e-06, | |
| "loss": 0.4576, | |
| "mean_token_accuracy": 0.8684222124516964, | |
| "num_tokens": 4566924.0, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.3424, | |
| "grad_norm": 2.033034563064575, | |
| "learning_rate": 4.381485913968747e-06, | |
| "loss": 0.4558, | |
| "mean_token_accuracy": 0.8672670938074589, | |
| "num_tokens": 4677684.0, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.3744, | |
| "grad_norm": 2.179928779602051, | |
| "learning_rate": 4.346265062869147e-06, | |
| "loss": 0.468, | |
| "mean_token_accuracy": 0.8666120313107968, | |
| "num_tokens": 4787468.0, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.4064, | |
| "grad_norm": 2.1226675510406494, | |
| "learning_rate": 4.310219048282921e-06, | |
| "loss": 0.4514, | |
| "mean_token_accuracy": 0.8693858951330184, | |
| "num_tokens": 4899692.0, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.4384000000000001, | |
| "grad_norm": 2.226494550704956, | |
| "learning_rate": 4.273363980494379e-06, | |
| "loss": 0.4761, | |
| "mean_token_accuracy": 0.8673883900046349, | |
| "num_tokens": 5008859.0, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.4704, | |
| "grad_norm": 2.039381980895996, | |
| "learning_rate": 4.235716331383343e-06, | |
| "loss": 0.4652, | |
| "mean_token_accuracy": 0.8683492943644524, | |
| "num_tokens": 5117719.0, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.5024, | |
| "grad_norm": 2.1596784591674805, | |
| "learning_rate": 4.197292927063263e-06, | |
| "loss": 0.4701, | |
| "mean_token_accuracy": 0.863686279207468, | |
| "num_tokens": 5227041.0, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.5344, | |
| "grad_norm": 2.153256893157959, | |
| "learning_rate": 4.158110940361007e-06, | |
| "loss": 0.4559, | |
| "mean_token_accuracy": 0.8705666847527027, | |
| "num_tokens": 5334777.0, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.5664, | |
| "grad_norm": 2.5348920822143555, | |
| "learning_rate": 4.118187883141694e-06, | |
| "loss": 0.461, | |
| "mean_token_accuracy": 0.8692857407033443, | |
| "num_tokens": 5441237.0, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.5984, | |
| "grad_norm": 2.2534334659576416, | |
| "learning_rate": 4.077541598482009e-06, | |
| "loss": 0.4748, | |
| "mean_token_accuracy": 0.8672291107475758, | |
| "num_tokens": 5553620.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.6303999999999998, | |
| "grad_norm": 2.1640467643737793, | |
| "learning_rate": 4.036190252695467e-06, | |
| "loss": 0.4874, | |
| "mean_token_accuracy": 0.8607077576220036, | |
| "num_tokens": 5663373.0, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.6623999999999999, | |
| "grad_norm": 2.1866941452026367, | |
| "learning_rate": 3.994152327213232e-06, | |
| "loss": 0.4545, | |
| "mean_token_accuracy": 0.8709075525403023, | |
| "num_tokens": 5776290.0, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.6944, | |
| "grad_norm": 2.4210782051086426, | |
| "learning_rate": 3.95144661032408e-06, | |
| "loss": 0.4316, | |
| "mean_token_accuracy": 0.8738773807883262, | |
| "num_tokens": 5882282.0, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.7264, | |
| "grad_norm": 2.2260217666625977, | |
| "learning_rate": 3.9080921887772314e-06, | |
| "loss": 0.4497, | |
| "mean_token_accuracy": 0.8696616031229496, | |
| "num_tokens": 5999645.0, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.7584, | |
| "grad_norm": 2.1372995376586914, | |
| "learning_rate": 3.864108439251779e-06, | |
| "loss": 0.4466, | |
| "mean_token_accuracy": 0.8693803034722805, | |
| "num_tokens": 6108677.0, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.7904, | |
| "grad_norm": 2.0875401496887207, | |
| "learning_rate": 3.8195150196965414e-06, | |
| "loss": 0.4618, | |
| "mean_token_accuracy": 0.866703423857689, | |
| "num_tokens": 6222393.0, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.8224, | |
| "grad_norm": 2.06887149810791, | |
| "learning_rate": 3.7743318605442093e-06, | |
| "loss": 0.4593, | |
| "mean_token_accuracy": 0.8678097896277904, | |
| "num_tokens": 6337550.0, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.8544, | |
| "grad_norm": 1.9830808639526367, | |
| "learning_rate": 3.728579155803697e-06, | |
| "loss": 0.4577, | |
| "mean_token_accuracy": 0.8675460621714592, | |
| "num_tokens": 6449551.0, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.8864, | |
| "grad_norm": 2.165144443511963, | |
| "learning_rate": 3.6822773540347107e-06, | |
| "loss": 0.439, | |
| "mean_token_accuracy": 0.8739401362836361, | |
| "num_tokens": 6562113.0, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.9184, | |
| "grad_norm": 2.1738622188568115, | |
| "learning_rate": 3.635447149208528e-06, | |
| "loss": 0.4412, | |
| "mean_token_accuracy": 0.8708230577409267, | |
| "num_tokens": 6674860.0, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.9504000000000001, | |
| "grad_norm": 2.4379212856292725, | |
| "learning_rate": 3.5881094714591102e-06, | |
| "loss": 0.4522, | |
| "mean_token_accuracy": 0.8691440485417843, | |
| "num_tokens": 6789358.0, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.9824000000000002, | |
| "grad_norm": 2.1988134384155273, | |
| "learning_rate": 3.540285477728658e-06, | |
| "loss": 0.4351, | |
| "mean_token_accuracy": 0.8745189763605594, | |
| "num_tokens": 6899897.0, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 2.0128, | |
| "grad_norm": 2.226576089859009, | |
| "learning_rate": 3.491996542311795e-06, | |
| "loss": 0.4196, | |
| "mean_token_accuracy": 0.8793611534331974, | |
| "num_tokens": 7005635.0, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 2.0448, | |
| "grad_norm": 2.284040927886963, | |
| "learning_rate": 3.44326424730262e-06, | |
| "loss": 0.3729, | |
| "mean_token_accuracy": 0.8900357685983181, | |
| "num_tokens": 7118882.0, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.0768, | |
| "grad_norm": 2.3776021003723145, | |
| "learning_rate": 3.394110372948871e-06, | |
| "loss": 0.3784, | |
| "mean_token_accuracy": 0.8912972524762154, | |
| "num_tokens": 7229223.0, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.1088, | |
| "grad_norm": 2.3752377033233643, | |
| "learning_rate": 3.344556887917539e-06, | |
| "loss": 0.3754, | |
| "mean_token_accuracy": 0.8933634988963604, | |
| "num_tokens": 7344352.0, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.1408, | |
| "grad_norm": 2.2204787731170654, | |
| "learning_rate": 3.2946259394762645e-06, | |
| "loss": 0.3603, | |
| "mean_token_accuracy": 0.8949612870812416, | |
| "num_tokens": 7452978.0, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.1728, | |
| "grad_norm": 2.3392176628112793, | |
| "learning_rate": 3.244339843594913e-06, | |
| "loss": 0.3633, | |
| "mean_token_accuracy": 0.8934120312333107, | |
| "num_tokens": 7562987.0, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.2048, | |
| "grad_norm": 2.548868417739868, | |
| "learning_rate": 3.19372107497175e-06, | |
| "loss": 0.3688, | |
| "mean_token_accuracy": 0.8925295993685722, | |
| "num_tokens": 7672018.0, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.2368, | |
| "grad_norm": 2.524747371673584, | |
| "learning_rate": 3.1427922569886756e-06, | |
| "loss": 0.3535, | |
| "mean_token_accuracy": 0.8959250211715698, | |
| "num_tokens": 7778694.0, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.2688, | |
| "grad_norm": 2.2159647941589355, | |
| "learning_rate": 3.0915761516000053e-06, | |
| "loss": 0.3645, | |
| "mean_token_accuracy": 0.8916135348379612, | |
| "num_tokens": 7889175.0, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.3008, | |
| "grad_norm": 2.6435368061065674, | |
| "learning_rate": 3.0400956491593215e-06, | |
| "loss": 0.3742, | |
| "mean_token_accuracy": 0.8912425316870213, | |
| "num_tokens": 7997644.0, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.3327999999999998, | |
| "grad_norm": 2.5125110149383545, | |
| "learning_rate": 2.9883737581889315e-06, | |
| "loss": 0.3767, | |
| "mean_token_accuracy": 0.8903645075857639, | |
| "num_tokens": 8109860.0, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.3648, | |
| "grad_norm": 2.482232093811035, | |
| "learning_rate": 2.9364335950965227e-06, | |
| "loss": 0.3531, | |
| "mean_token_accuracy": 0.8950154200196266, | |
| "num_tokens": 8219541.0, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.3968, | |
| "grad_norm": 2.360308885574341, | |
| "learning_rate": 2.8842983738435855e-06, | |
| "loss": 0.3602, | |
| "mean_token_accuracy": 0.892794082313776, | |
| "num_tokens": 8329765.0, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.4288, | |
| "grad_norm": 2.5826432704925537, | |
| "learning_rate": 2.831991395570249e-06, | |
| "loss": 0.364, | |
| "mean_token_accuracy": 0.8930977649986744, | |
| "num_tokens": 8443917.0, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.4608, | |
| "grad_norm": 2.588911771774292, | |
| "learning_rate": 2.779536038181146e-06, | |
| "loss": 0.3771, | |
| "mean_token_accuracy": 0.8915269665420056, | |
| "num_tokens": 8550550.0, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.4928, | |
| "grad_norm": 2.4379560947418213, | |
| "learning_rate": 2.726955745896972e-06, | |
| "loss": 0.3726, | |
| "mean_token_accuracy": 0.8931849867105484, | |
| "num_tokens": 8659752.0, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.5248, | |
| "grad_norm": 2.285378932952881, | |
| "learning_rate": 2.6742740187764026e-06, | |
| "loss": 0.3644, | |
| "mean_token_accuracy": 0.8937488354742527, | |
| "num_tokens": 8775392.0, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.5568, | |
| "grad_norm": 2.6339035034179688, | |
| "learning_rate": 2.621514402213059e-06, | |
| "loss": 0.3528, | |
| "mean_token_accuracy": 0.8967724144458771, | |
| "num_tokens": 8888499.0, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.5888, | |
| "grad_norm": 2.2561960220336914, | |
| "learning_rate": 2.568700476412204e-06, | |
| "loss": 0.3846, | |
| "mean_token_accuracy": 0.8879048563539982, | |
| "num_tokens": 9001912.0, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 2.6208, | |
| "grad_norm": 2.3487298488616943, | |
| "learning_rate": 2.515855845851894e-06, | |
| "loss": 0.3602, | |
| "mean_token_accuracy": 0.8923349373042584, | |
| "num_tokens": 9113468.0, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.6528, | |
| "grad_norm": 2.4302189350128174, | |
| "learning_rate": 2.46300412873326e-06, | |
| "loss": 0.3613, | |
| "mean_token_accuracy": 0.8928801603615284, | |
| "num_tokens": 9229335.0, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 2.6848, | |
| "grad_norm": 2.48608136177063, | |
| "learning_rate": 2.4101689464246816e-06, | |
| "loss": 0.3642, | |
| "mean_token_accuracy": 0.8935355007648468, | |
| "num_tokens": 9344595.0, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 2.7168, | |
| "grad_norm": 2.357041120529175, | |
| "learning_rate": 2.3573739129045227e-06, | |
| "loss": 0.3579, | |
| "mean_token_accuracy": 0.8953744530677795, | |
| "num_tokens": 9455015.0, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.7488, | |
| "grad_norm": 2.649960517883301, | |
| "learning_rate": 2.3046426242071834e-06, | |
| "loss": 0.3689, | |
| "mean_token_accuracy": 0.8912502139806747, | |
| "num_tokens": 9560806.0, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 2.7808, | |
| "grad_norm": 2.3866405487060547, | |
| "learning_rate": 2.251998647877169e-06, | |
| "loss": 0.364, | |
| "mean_token_accuracy": 0.8922963574528694, | |
| "num_tokens": 9673310.0, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 2.8128, | |
| "grad_norm": 2.3306539058685303, | |
| "learning_rate": 2.1994655124358875e-06, | |
| "loss": 0.347, | |
| "mean_token_accuracy": 0.8968791022896767, | |
| "num_tokens": 9787134.0, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 2.8448, | |
| "grad_norm": 2.3708155155181885, | |
| "learning_rate": 2.1470666968658956e-06, | |
| "loss": 0.3576, | |
| "mean_token_accuracy": 0.8956275820732117, | |
| "num_tokens": 9901311.0, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 2.8768000000000002, | |
| "grad_norm": 2.466033935546875, | |
| "learning_rate": 2.0948256201172772e-06, | |
| "loss": 0.3495, | |
| "mean_token_accuracy": 0.8957146309316159, | |
| "num_tokens": 10017025.0, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.9088000000000003, | |
| "grad_norm": 2.630037784576416, | |
| "learning_rate": 2.0427656306408594e-06, | |
| "loss": 0.3643, | |
| "mean_token_accuracy": 0.892961747199297, | |
| "num_tokens": 10131105.0, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 2.9408, | |
| "grad_norm": 2.4972527027130127, | |
| "learning_rate": 1.990909995952936e-06, | |
| "loss": 0.3542, | |
| "mean_token_accuracy": 0.8959409989416599, | |
| "num_tokens": 10240555.0, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 2.9728, | |
| "grad_norm": 2.722083330154419, | |
| "learning_rate": 1.939281892236157e-06, | |
| "loss": 0.3531, | |
| "mean_token_accuracy": 0.8945782586932183, | |
| "num_tokens": 10347889.0, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 3.0032, | |
| "grad_norm": 2.2683236598968506, | |
| "learning_rate": 1.8879043939812503e-06, | |
| "loss": 0.3579, | |
| "mean_token_accuracy": 0.8970388119157992, | |
| "num_tokens": 10453648.0, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 3.0352, | |
| "grad_norm": 2.5510222911834717, | |
| "learning_rate": 1.8368004636741794e-06, | |
| "loss": 0.3062, | |
| "mean_token_accuracy": 0.9120431691408157, | |
| "num_tokens": 10565156.0, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 3.0672, | |
| "grad_norm": 2.730621337890625, | |
| "learning_rate": 1.7859929415333725e-06, | |
| "loss": 0.2912, | |
| "mean_token_accuracy": 0.9148132555186749, | |
| "num_tokens": 10676677.0, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 3.0992, | |
| "grad_norm": 2.5102102756500244, | |
| "learning_rate": 1.735504535301592e-06, | |
| "loss": 0.2882, | |
| "mean_token_accuracy": 0.9159041397273541, | |
| "num_tokens": 10787468.0, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 3.1312, | |
| "grad_norm": 2.3301403522491455, | |
| "learning_rate": 1.6853578100970086e-06, | |
| "loss": 0.3016, | |
| "mean_token_accuracy": 0.911161607503891, | |
| "num_tokens": 10897279.0, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 3.1632, | |
| "grad_norm": 2.4964194297790527, | |
| "learning_rate": 1.6355751783280283e-06, | |
| "loss": 0.2927, | |
| "mean_token_accuracy": 0.9142641559243202, | |
| "num_tokens": 11008731.0, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 3.1952, | |
| "grad_norm": 2.547372341156006, | |
| "learning_rate": 1.5861788896763625e-06, | |
| "loss": 0.2991, | |
| "mean_token_accuracy": 0.9131025090813637, | |
| "num_tokens": 11123820.0, | |
| "step": 1000 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1565, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.0242157166877082e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |