| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.6854009595613434, | |
| "eval_steps": 500, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "entropy": 0.9619973443448544, | |
| "epoch": 0.006854009595613434, | |
| "grad_norm": 0.8562721610069275, | |
| "learning_rate": 0.0001995887594242632, | |
| "loss": 0.7973, | |
| "mean_token_accuracy": 0.7519877135753632, | |
| "num_tokens": 15771.0, | |
| "step": 10 | |
| }, | |
| { | |
| "entropy": 0.7034977793693542, | |
| "epoch": 0.013708019191226868, | |
| "grad_norm": 0.5451128482818604, | |
| "learning_rate": 0.0001991318254512223, | |
| "loss": 0.5986, | |
| "mean_token_accuracy": 0.8309322476387024, | |
| "num_tokens": 33062.0, | |
| "step": 20 | |
| }, | |
| { | |
| "entropy": 0.6603402759879827, | |
| "epoch": 0.0205620287868403, | |
| "grad_norm": 0.5171676278114319, | |
| "learning_rate": 0.00019867489147818142, | |
| "loss": 0.633, | |
| "mean_token_accuracy": 0.8433935061097145, | |
| "num_tokens": 48936.0, | |
| "step": 30 | |
| }, | |
| { | |
| "entropy": 0.6830728624016047, | |
| "epoch": 0.027416038382453736, | |
| "grad_norm": 0.4969835877418518, | |
| "learning_rate": 0.0001982179575051405, | |
| "loss": 0.6773, | |
| "mean_token_accuracy": 0.8266744241118431, | |
| "num_tokens": 61449.0, | |
| "step": 40 | |
| }, | |
| { | |
| "entropy": 0.5286078054457903, | |
| "epoch": 0.03427004797806717, | |
| "grad_norm": 0.44698312878608704, | |
| "learning_rate": 0.00019776102353209963, | |
| "loss": 0.5558, | |
| "mean_token_accuracy": 0.8533428102731705, | |
| "num_tokens": 77104.0, | |
| "step": 50 | |
| }, | |
| { | |
| "entropy": 0.5590948283672332, | |
| "epoch": 0.0411240575736806, | |
| "grad_norm": 0.38724300265312195, | |
| "learning_rate": 0.00019730408955905872, | |
| "loss": 0.5771, | |
| "mean_token_accuracy": 0.8514142513275147, | |
| "num_tokens": 91558.0, | |
| "step": 60 | |
| }, | |
| { | |
| "entropy": 0.599293502047658, | |
| "epoch": 0.047978067169294036, | |
| "grad_norm": 0.5922872424125671, | |
| "learning_rate": 0.00019684715558601783, | |
| "loss": 0.5309, | |
| "mean_token_accuracy": 0.851950392127037, | |
| "num_tokens": 105756.0, | |
| "step": 70 | |
| }, | |
| { | |
| "entropy": 0.6024694256484509, | |
| "epoch": 0.05483207676490747, | |
| "grad_norm": 0.5078150629997253, | |
| "learning_rate": 0.00019639022161297692, | |
| "loss": 0.6727, | |
| "mean_token_accuracy": 0.8480887472629547, | |
| "num_tokens": 120163.0, | |
| "step": 80 | |
| }, | |
| { | |
| "entropy": 0.5648054199293255, | |
| "epoch": 0.061686086360520906, | |
| "grad_norm": 0.29077377915382385, | |
| "learning_rate": 0.00019593328763993604, | |
| "loss": 0.5509, | |
| "mean_token_accuracy": 0.8548661589622497, | |
| "num_tokens": 138293.0, | |
| "step": 90 | |
| }, | |
| { | |
| "entropy": 0.5659369576722384, | |
| "epoch": 0.06854009595613433, | |
| "grad_norm": 0.3394547700881958, | |
| "learning_rate": 0.00019547635366689515, | |
| "loss": 0.5729, | |
| "mean_token_accuracy": 0.8460367009043693, | |
| "num_tokens": 157530.0, | |
| "step": 100 | |
| }, | |
| { | |
| "entropy": 0.6021960902959108, | |
| "epoch": 0.07539410555174778, | |
| "grad_norm": 0.42912933230400085, | |
| "learning_rate": 0.00019501941969385424, | |
| "loss": 0.5465, | |
| "mean_token_accuracy": 0.851969163119793, | |
| "num_tokens": 170607.0, | |
| "step": 110 | |
| }, | |
| { | |
| "entropy": 0.610968679189682, | |
| "epoch": 0.0822481151473612, | |
| "grad_norm": 0.3759806752204895, | |
| "learning_rate": 0.00019456248572081335, | |
| "loss": 0.632, | |
| "mean_token_accuracy": 0.83554507791996, | |
| "num_tokens": 185672.0, | |
| "step": 120 | |
| }, | |
| { | |
| "entropy": 0.5211818940937519, | |
| "epoch": 0.08910212474297464, | |
| "grad_norm": 0.503212034702301, | |
| "learning_rate": 0.00019410555174777244, | |
| "loss": 0.547, | |
| "mean_token_accuracy": 0.8610908895730972, | |
| "num_tokens": 200482.0, | |
| "step": 130 | |
| }, | |
| { | |
| "entropy": 0.4536220826208591, | |
| "epoch": 0.09595613433858807, | |
| "grad_norm": 0.7268697619438171, | |
| "learning_rate": 0.00019364861777473156, | |
| "loss": 0.4726, | |
| "mean_token_accuracy": 0.870763523876667, | |
| "num_tokens": 216537.0, | |
| "step": 140 | |
| }, | |
| { | |
| "entropy": 0.5031640276312828, | |
| "epoch": 0.10281014393420151, | |
| "grad_norm": 0.33594396710395813, | |
| "learning_rate": 0.00019319168380169065, | |
| "loss": 0.5923, | |
| "mean_token_accuracy": 0.8628711074590683, | |
| "num_tokens": 232400.0, | |
| "step": 150 | |
| }, | |
| { | |
| "entropy": 0.6555169004946947, | |
| "epoch": 0.10966415352981494, | |
| "grad_norm": 0.5894250869750977, | |
| "learning_rate": 0.00019273474982864976, | |
| "loss": 0.5634, | |
| "mean_token_accuracy": 0.838917362689972, | |
| "num_tokens": 244273.0, | |
| "step": 160 | |
| }, | |
| { | |
| "entropy": 0.6185528110712767, | |
| "epoch": 0.11651816312542837, | |
| "grad_norm": 0.5221670269966125, | |
| "learning_rate": 0.00019227781585560888, | |
| "loss": 0.6818, | |
| "mean_token_accuracy": 0.8385803163051605, | |
| "num_tokens": 262927.0, | |
| "step": 170 | |
| }, | |
| { | |
| "entropy": 0.43512718714773657, | |
| "epoch": 0.12337217272104181, | |
| "grad_norm": 0.4728280007839203, | |
| "learning_rate": 0.00019182088188256796, | |
| "loss": 0.5178, | |
| "mean_token_accuracy": 0.8747259676456451, | |
| "num_tokens": 271716.0, | |
| "step": 180 | |
| }, | |
| { | |
| "entropy": 0.6649946108460426, | |
| "epoch": 0.13022618231665525, | |
| "grad_norm": 0.47320684790611267, | |
| "learning_rate": 0.00019136394790952708, | |
| "loss": 0.6651, | |
| "mean_token_accuracy": 0.8223798260092735, | |
| "num_tokens": 287518.0, | |
| "step": 190 | |
| }, | |
| { | |
| "entropy": 0.44864910580217837, | |
| "epoch": 0.13708019191226867, | |
| "grad_norm": 0.4356485903263092, | |
| "learning_rate": 0.00019090701393648617, | |
| "loss": 0.4631, | |
| "mean_token_accuracy": 0.8835410609841347, | |
| "num_tokens": 306299.0, | |
| "step": 200 | |
| }, | |
| { | |
| "entropy": 0.5033049076795578, | |
| "epoch": 0.1439342015078821, | |
| "grad_norm": 0.48287737369537354, | |
| "learning_rate": 0.00019045007996344528, | |
| "loss": 0.5933, | |
| "mean_token_accuracy": 0.8669374987483025, | |
| "num_tokens": 321955.0, | |
| "step": 210 | |
| }, | |
| { | |
| "entropy": 0.43575111888349055, | |
| "epoch": 0.15078821110349555, | |
| "grad_norm": 0.5973707437515259, | |
| "learning_rate": 0.0001899931459904044, | |
| "loss": 0.4856, | |
| "mean_token_accuracy": 0.8797904253005981, | |
| "num_tokens": 332265.0, | |
| "step": 220 | |
| }, | |
| { | |
| "entropy": 0.6193726476281881, | |
| "epoch": 0.157642220699109, | |
| "grad_norm": 0.28756600618362427, | |
| "learning_rate": 0.0001895362120173635, | |
| "loss": 0.6056, | |
| "mean_token_accuracy": 0.8307039767503739, | |
| "num_tokens": 346377.0, | |
| "step": 230 | |
| }, | |
| { | |
| "entropy": 0.5753613166511059, | |
| "epoch": 0.1644962302947224, | |
| "grad_norm": 0.4320402145385742, | |
| "learning_rate": 0.0001890792780443226, | |
| "loss": 0.5834, | |
| "mean_token_accuracy": 0.8549783885478973, | |
| "num_tokens": 362964.0, | |
| "step": 240 | |
| }, | |
| { | |
| "entropy": 0.5963706407696009, | |
| "epoch": 0.17135023989033585, | |
| "grad_norm": 0.4648321866989136, | |
| "learning_rate": 0.0001886223440712817, | |
| "loss": 0.6272, | |
| "mean_token_accuracy": 0.8493530780076981, | |
| "num_tokens": 375717.0, | |
| "step": 250 | |
| }, | |
| { | |
| "entropy": 0.5467700261622668, | |
| "epoch": 0.1782042494859493, | |
| "grad_norm": 0.3487129211425781, | |
| "learning_rate": 0.00018816541009824083, | |
| "loss": 0.5449, | |
| "mean_token_accuracy": 0.8526464059948922, | |
| "num_tokens": 394586.0, | |
| "step": 260 | |
| }, | |
| { | |
| "entropy": 0.4246529323980212, | |
| "epoch": 0.1850582590815627, | |
| "grad_norm": 0.7286052703857422, | |
| "learning_rate": 0.00018770847612519992, | |
| "loss": 0.45, | |
| "mean_token_accuracy": 0.8814342901110649, | |
| "num_tokens": 411636.0, | |
| "step": 270 | |
| }, | |
| { | |
| "entropy": 0.5389048531651497, | |
| "epoch": 0.19191226867717615, | |
| "grad_norm": 0.3287123441696167, | |
| "learning_rate": 0.00018725154215215904, | |
| "loss": 0.5138, | |
| "mean_token_accuracy": 0.8506704963743686, | |
| "num_tokens": 427077.0, | |
| "step": 280 | |
| }, | |
| { | |
| "entropy": 0.5297756217420101, | |
| "epoch": 0.1987662782727896, | |
| "grad_norm": 0.5151430368423462, | |
| "learning_rate": 0.00018679460817911812, | |
| "loss": 0.5953, | |
| "mean_token_accuracy": 0.8586609676480293, | |
| "num_tokens": 442257.0, | |
| "step": 290 | |
| }, | |
| { | |
| "entropy": 0.5314730744808912, | |
| "epoch": 0.20562028786840303, | |
| "grad_norm": 0.9860548377037048, | |
| "learning_rate": 0.00018633767420607724, | |
| "loss": 0.5143, | |
| "mean_token_accuracy": 0.8650062039494515, | |
| "num_tokens": 458093.0, | |
| "step": 300 | |
| }, | |
| { | |
| "entropy": 0.5666845880448819, | |
| "epoch": 0.21247429746401644, | |
| "grad_norm": 0.8684160113334656, | |
| "learning_rate": 0.00018588074023303635, | |
| "loss": 0.5487, | |
| "mean_token_accuracy": 0.8509581357240676, | |
| "num_tokens": 471306.0, | |
| "step": 310 | |
| }, | |
| { | |
| "entropy": 0.5297997735440731, | |
| "epoch": 0.21932830705962988, | |
| "grad_norm": 0.3815328776836395, | |
| "learning_rate": 0.00018542380625999544, | |
| "loss": 0.6052, | |
| "mean_token_accuracy": 0.8568633005023003, | |
| "num_tokens": 488461.0, | |
| "step": 320 | |
| }, | |
| { | |
| "entropy": 0.5316725082695484, | |
| "epoch": 0.22618231665524333, | |
| "grad_norm": 0.5312303900718689, | |
| "learning_rate": 0.00018496687228695456, | |
| "loss": 0.6035, | |
| "mean_token_accuracy": 0.858753177523613, | |
| "num_tokens": 503665.0, | |
| "step": 330 | |
| }, | |
| { | |
| "entropy": 0.6088610142469406, | |
| "epoch": 0.23303632625085674, | |
| "grad_norm": 0.40660324692726135, | |
| "learning_rate": 0.00018450993831391365, | |
| "loss": 0.6232, | |
| "mean_token_accuracy": 0.8444906592369079, | |
| "num_tokens": 521925.0, | |
| "step": 340 | |
| }, | |
| { | |
| "entropy": 0.6339217025786639, | |
| "epoch": 0.23989033584647018, | |
| "grad_norm": 0.5640454888343811, | |
| "learning_rate": 0.00018405300434087276, | |
| "loss": 0.6188, | |
| "mean_token_accuracy": 0.8382566079497338, | |
| "num_tokens": 535970.0, | |
| "step": 350 | |
| }, | |
| { | |
| "entropy": 0.5411492632701993, | |
| "epoch": 0.24674434544208362, | |
| "grad_norm": 0.42631176114082336, | |
| "learning_rate": 0.00018359607036783185, | |
| "loss": 0.5528, | |
| "mean_token_accuracy": 0.8523587495088577, | |
| "num_tokens": 551676.0, | |
| "step": 360 | |
| }, | |
| { | |
| "entropy": 0.5561750333756208, | |
| "epoch": 0.25359835503769707, | |
| "grad_norm": 0.5579405426979065, | |
| "learning_rate": 0.00018313913639479097, | |
| "loss": 0.5793, | |
| "mean_token_accuracy": 0.8607801914215087, | |
| "num_tokens": 568488.0, | |
| "step": 370 | |
| }, | |
| { | |
| "entropy": 0.5319446712732315, | |
| "epoch": 0.2604523646333105, | |
| "grad_norm": 0.8342606425285339, | |
| "learning_rate": 0.00018268220242175008, | |
| "loss": 0.4994, | |
| "mean_token_accuracy": 0.863979734480381, | |
| "num_tokens": 582963.0, | |
| "step": 380 | |
| }, | |
| { | |
| "entropy": 0.5910112973302603, | |
| "epoch": 0.2673063742289239, | |
| "grad_norm": 0.5433372259140015, | |
| "learning_rate": 0.00018222526844870917, | |
| "loss": 0.6669, | |
| "mean_token_accuracy": 0.8433835208415985, | |
| "num_tokens": 598471.0, | |
| "step": 390 | |
| }, | |
| { | |
| "entropy": 0.46995992250740526, | |
| "epoch": 0.27416038382453733, | |
| "grad_norm": 0.26409879326820374, | |
| "learning_rate": 0.00018176833447566828, | |
| "loss": 0.5199, | |
| "mean_token_accuracy": 0.87328050583601, | |
| "num_tokens": 614036.0, | |
| "step": 400 | |
| }, | |
| { | |
| "entropy": 0.5400116696953774, | |
| "epoch": 0.2810143934201508, | |
| "grad_norm": 0.3498149514198303, | |
| "learning_rate": 0.00018131140050262737, | |
| "loss": 0.5902, | |
| "mean_token_accuracy": 0.8512750566005707, | |
| "num_tokens": 630937.0, | |
| "step": 410 | |
| }, | |
| { | |
| "entropy": 0.45603593066334724, | |
| "epoch": 0.2878684030157642, | |
| "grad_norm": 0.6973631978034973, | |
| "learning_rate": 0.0001808544665295865, | |
| "loss": 0.484, | |
| "mean_token_accuracy": 0.8728810593485832, | |
| "num_tokens": 642492.0, | |
| "step": 420 | |
| }, | |
| { | |
| "entropy": 0.5664497867226601, | |
| "epoch": 0.29472241261137766, | |
| "grad_norm": 0.4047413170337677, | |
| "learning_rate": 0.0001803975325565456, | |
| "loss": 0.5107, | |
| "mean_token_accuracy": 0.8518401965498924, | |
| "num_tokens": 656785.0, | |
| "step": 430 | |
| }, | |
| { | |
| "entropy": 0.5749023761600256, | |
| "epoch": 0.3015764222069911, | |
| "grad_norm": 0.5084949135780334, | |
| "learning_rate": 0.0001799405985835047, | |
| "loss": 0.5558, | |
| "mean_token_accuracy": 0.8492432355880737, | |
| "num_tokens": 671870.0, | |
| "step": 440 | |
| }, | |
| { | |
| "entropy": 0.4889295015484095, | |
| "epoch": 0.30843043180260454, | |
| "grad_norm": 0.42546579241752625, | |
| "learning_rate": 0.0001794836646104638, | |
| "loss": 0.5416, | |
| "mean_token_accuracy": 0.8649413183331489, | |
| "num_tokens": 685980.0, | |
| "step": 450 | |
| }, | |
| { | |
| "entropy": 0.5743775438517332, | |
| "epoch": 0.315284441398218, | |
| "grad_norm": 0.3708641231060028, | |
| "learning_rate": 0.0001790267306374229, | |
| "loss": 0.5976, | |
| "mean_token_accuracy": 0.8467541456222534, | |
| "num_tokens": 699287.0, | |
| "step": 460 | |
| }, | |
| { | |
| "entropy": 0.5913191799074411, | |
| "epoch": 0.32213845099383137, | |
| "grad_norm": 0.37332257628440857, | |
| "learning_rate": 0.000178569796664382, | |
| "loss": 0.5695, | |
| "mean_token_accuracy": 0.8441656738519668, | |
| "num_tokens": 714803.0, | |
| "step": 470 | |
| }, | |
| { | |
| "entropy": 0.45778534524142744, | |
| "epoch": 0.3289924605894448, | |
| "grad_norm": 0.5047005414962769, | |
| "learning_rate": 0.0001781128626913411, | |
| "loss": 0.4778, | |
| "mean_token_accuracy": 0.8752694010734559, | |
| "num_tokens": 732120.0, | |
| "step": 480 | |
| }, | |
| { | |
| "entropy": 0.5643713362514973, | |
| "epoch": 0.33584647018505825, | |
| "grad_norm": 0.4013417065143585, | |
| "learning_rate": 0.0001776559287183002, | |
| "loss": 0.5366, | |
| "mean_token_accuracy": 0.8520827397704125, | |
| "num_tokens": 745974.0, | |
| "step": 490 | |
| }, | |
| { | |
| "entropy": 0.4815288335084915, | |
| "epoch": 0.3427004797806717, | |
| "grad_norm": 0.3859888017177582, | |
| "learning_rate": 0.00017719899474525933, | |
| "loss": 0.5521, | |
| "mean_token_accuracy": 0.8687581121921539, | |
| "num_tokens": 759499.0, | |
| "step": 500 | |
| }, | |
| { | |
| "entropy": 0.5015339620411396, | |
| "epoch": 0.34955448937628514, | |
| "grad_norm": 0.6697199940681458, | |
| "learning_rate": 0.00017674206077221842, | |
| "loss": 0.4885, | |
| "mean_token_accuracy": 0.8649638772010804, | |
| "num_tokens": 773698.0, | |
| "step": 510 | |
| }, | |
| { | |
| "entropy": 0.6096027113497258, | |
| "epoch": 0.3564084989718986, | |
| "grad_norm": 1.021246075630188, | |
| "learning_rate": 0.00017628512679917753, | |
| "loss": 0.6121, | |
| "mean_token_accuracy": 0.8394175350666047, | |
| "num_tokens": 788221.0, | |
| "step": 520 | |
| }, | |
| { | |
| "entropy": 0.5186641301959753, | |
| "epoch": 0.363262508567512, | |
| "grad_norm": 0.8043237924575806, | |
| "learning_rate": 0.00017582819282613662, | |
| "loss": 0.5592, | |
| "mean_token_accuracy": 0.8656348437070847, | |
| "num_tokens": 802330.0, | |
| "step": 530 | |
| }, | |
| { | |
| "entropy": 0.6064855309203268, | |
| "epoch": 0.3701165181631254, | |
| "grad_norm": 0.43781760334968567, | |
| "learning_rate": 0.00017537125885309574, | |
| "loss": 0.5874, | |
| "mean_token_accuracy": 0.8404153436422348, | |
| "num_tokens": 819615.0, | |
| "step": 540 | |
| }, | |
| { | |
| "entropy": 0.5140635691583156, | |
| "epoch": 0.37697052775873885, | |
| "grad_norm": 0.7001516819000244, | |
| "learning_rate": 0.00017491432488005482, | |
| "loss": 0.5547, | |
| "mean_token_accuracy": 0.8677607625722885, | |
| "num_tokens": 835616.0, | |
| "step": 550 | |
| }, | |
| { | |
| "entropy": 0.4637599032372236, | |
| "epoch": 0.3838245373543523, | |
| "grad_norm": 0.38044923543930054, | |
| "learning_rate": 0.00017445739090701394, | |
| "loss": 0.5021, | |
| "mean_token_accuracy": 0.87646614164114, | |
| "num_tokens": 851065.0, | |
| "step": 560 | |
| }, | |
| { | |
| "entropy": 0.5091348428279161, | |
| "epoch": 0.39067854694996573, | |
| "grad_norm": 0.476380318403244, | |
| "learning_rate": 0.00017400045693397305, | |
| "loss": 0.7145, | |
| "mean_token_accuracy": 0.8637179903686046, | |
| "num_tokens": 866992.0, | |
| "step": 570 | |
| }, | |
| { | |
| "entropy": 0.5213964153081179, | |
| "epoch": 0.3975325565455792, | |
| "grad_norm": 0.3205454647541046, | |
| "learning_rate": 0.00017354352296093214, | |
| "loss": 0.5395, | |
| "mean_token_accuracy": 0.8607370749115943, | |
| "num_tokens": 881803.0, | |
| "step": 580 | |
| }, | |
| { | |
| "entropy": 0.619412742741406, | |
| "epoch": 0.4043865661411926, | |
| "grad_norm": 0.6068571209907532, | |
| "learning_rate": 0.00017308658898789126, | |
| "loss": 0.7504, | |
| "mean_token_accuracy": 0.8409741953015327, | |
| "num_tokens": 896283.0, | |
| "step": 590 | |
| }, | |
| { | |
| "entropy": 0.41873827911913397, | |
| "epoch": 0.41124057573680606, | |
| "grad_norm": 0.7212440371513367, | |
| "learning_rate": 0.00017262965501485035, | |
| "loss": 0.4416, | |
| "mean_token_accuracy": 0.8842875167727471, | |
| "num_tokens": 912255.0, | |
| "step": 600 | |
| }, | |
| { | |
| "entropy": 0.5398109834641218, | |
| "epoch": 0.41809458533241944, | |
| "grad_norm": 0.5380451083183289, | |
| "learning_rate": 0.00017217272104180946, | |
| "loss": 0.6092, | |
| "mean_token_accuracy": 0.8535096302628518, | |
| "num_tokens": 924648.0, | |
| "step": 610 | |
| }, | |
| { | |
| "entropy": 0.5522895563393831, | |
| "epoch": 0.4249485949280329, | |
| "grad_norm": 0.3073669373989105, | |
| "learning_rate": 0.00017171578706876858, | |
| "loss": 0.5943, | |
| "mean_token_accuracy": 0.8548912346363068, | |
| "num_tokens": 942485.0, | |
| "step": 620 | |
| }, | |
| { | |
| "entropy": 0.6571722824126482, | |
| "epoch": 0.4318026045236463, | |
| "grad_norm": 0.32408949732780457, | |
| "learning_rate": 0.00017125885309572766, | |
| "loss": 0.6485, | |
| "mean_token_accuracy": 0.8336619213223457, | |
| "num_tokens": 959239.0, | |
| "step": 630 | |
| }, | |
| { | |
| "entropy": 0.44382771104574203, | |
| "epoch": 0.43865661411925977, | |
| "grad_norm": 0.3714044988155365, | |
| "learning_rate": 0.00017080191912268678, | |
| "loss": 0.451, | |
| "mean_token_accuracy": 0.8748382180929184, | |
| "num_tokens": 971292.0, | |
| "step": 640 | |
| }, | |
| { | |
| "entropy": 0.5132732756435872, | |
| "epoch": 0.4455106237148732, | |
| "grad_norm": 0.39163199067115784, | |
| "learning_rate": 0.00017034498514964587, | |
| "loss": 0.5782, | |
| "mean_token_accuracy": 0.8603680655360222, | |
| "num_tokens": 986609.0, | |
| "step": 650 | |
| }, | |
| { | |
| "entropy": 0.6249308317899704, | |
| "epoch": 0.45236463331048665, | |
| "grad_norm": 0.9571526646614075, | |
| "learning_rate": 0.00016988805117660498, | |
| "loss": 0.6102, | |
| "mean_token_accuracy": 0.8412078201770783, | |
| "num_tokens": 1001886.0, | |
| "step": 660 | |
| }, | |
| { | |
| "entropy": 0.5879610646516085, | |
| "epoch": 0.4592186429061001, | |
| "grad_norm": 0.5797366499900818, | |
| "learning_rate": 0.00016943111720356407, | |
| "loss": 0.564, | |
| "mean_token_accuracy": 0.8438061460852623, | |
| "num_tokens": 1016026.0, | |
| "step": 670 | |
| }, | |
| { | |
| "entropy": 0.5628054179251194, | |
| "epoch": 0.4660726525017135, | |
| "grad_norm": 1.196199893951416, | |
| "learning_rate": 0.0001689741832305232, | |
| "loss": 0.5893, | |
| "mean_token_accuracy": 0.8494728982448578, | |
| "num_tokens": 1027357.0, | |
| "step": 680 | |
| }, | |
| { | |
| "entropy": 0.47301769629120827, | |
| "epoch": 0.4729266620973269, | |
| "grad_norm": 0.41828563809394836, | |
| "learning_rate": 0.0001685172492574823, | |
| "loss": 0.5013, | |
| "mean_token_accuracy": 0.8767685040831565, | |
| "num_tokens": 1042347.0, | |
| "step": 690 | |
| }, | |
| { | |
| "entropy": 0.5087582165375352, | |
| "epoch": 0.47978067169294036, | |
| "grad_norm": 0.36651521921157837, | |
| "learning_rate": 0.00016806031528444142, | |
| "loss": 0.5749, | |
| "mean_token_accuracy": 0.8544954568147659, | |
| "num_tokens": 1061719.0, | |
| "step": 700 | |
| }, | |
| { | |
| "entropy": 0.5259541615843772, | |
| "epoch": 0.4866346812885538, | |
| "grad_norm": 0.5560138821601868, | |
| "learning_rate": 0.00016760338131140053, | |
| "loss": 0.5649, | |
| "mean_token_accuracy": 0.860032057762146, | |
| "num_tokens": 1075339.0, | |
| "step": 710 | |
| }, | |
| { | |
| "entropy": 0.4655290380120277, | |
| "epoch": 0.49348869088416725, | |
| "grad_norm": 0.37081801891326904, | |
| "learning_rate": 0.00016714644733835962, | |
| "loss": 0.5218, | |
| "mean_token_accuracy": 0.8754615411162376, | |
| "num_tokens": 1089873.0, | |
| "step": 720 | |
| }, | |
| { | |
| "entropy": 0.5236123736947775, | |
| "epoch": 0.5003427004797807, | |
| "grad_norm": 0.4931930899620056, | |
| "learning_rate": 0.00016668951336531874, | |
| "loss": 0.5333, | |
| "mean_token_accuracy": 0.8595968760550022, | |
| "num_tokens": 1107924.0, | |
| "step": 730 | |
| }, | |
| { | |
| "entropy": 0.5846156869083643, | |
| "epoch": 0.5071967100753941, | |
| "grad_norm": 0.47382351756095886, | |
| "learning_rate": 0.00016623257939227782, | |
| "loss": 0.6035, | |
| "mean_token_accuracy": 0.8439710319042206, | |
| "num_tokens": 1123755.0, | |
| "step": 740 | |
| }, | |
| { | |
| "entropy": 0.5500952435657382, | |
| "epoch": 0.5140507196710076, | |
| "grad_norm": 0.5050795674324036, | |
| "learning_rate": 0.00016577564541923694, | |
| "loss": 0.6073, | |
| "mean_token_accuracy": 0.8468502178788185, | |
| "num_tokens": 1138565.0, | |
| "step": 750 | |
| }, | |
| { | |
| "entropy": 0.5514631005004048, | |
| "epoch": 0.520904729266621, | |
| "grad_norm": 0.5656992793083191, | |
| "learning_rate": 0.00016531871144619603, | |
| "loss": 0.5075, | |
| "mean_token_accuracy": 0.8544194102287292, | |
| "num_tokens": 1157676.0, | |
| "step": 760 | |
| }, | |
| { | |
| "entropy": 0.49982974790036677, | |
| "epoch": 0.5277587388622345, | |
| "grad_norm": 0.4955768585205078, | |
| "learning_rate": 0.00016486177747315514, | |
| "loss": 0.5041, | |
| "mean_token_accuracy": 0.8634250342845917, | |
| "num_tokens": 1174042.0, | |
| "step": 770 | |
| }, | |
| { | |
| "entropy": 0.5862449683248997, | |
| "epoch": 0.5346127484578478, | |
| "grad_norm": 0.5951958298683167, | |
| "learning_rate": 0.00016440484350011426, | |
| "loss": 0.5804, | |
| "mean_token_accuracy": 0.8448848068714142, | |
| "num_tokens": 1190184.0, | |
| "step": 780 | |
| }, | |
| { | |
| "entropy": 0.4969827888533473, | |
| "epoch": 0.5414667580534612, | |
| "grad_norm": 0.4006407856941223, | |
| "learning_rate": 0.00016394790952707335, | |
| "loss": 0.5379, | |
| "mean_token_accuracy": 0.8700526058673859, | |
| "num_tokens": 1205954.0, | |
| "step": 790 | |
| }, | |
| { | |
| "entropy": 0.5849012348800897, | |
| "epoch": 0.5483207676490747, | |
| "grad_norm": 0.624742329120636, | |
| "learning_rate": 0.00016349097555403246, | |
| "loss": 0.6361, | |
| "mean_token_accuracy": 0.8400089010596276, | |
| "num_tokens": 1220942.0, | |
| "step": 800 | |
| }, | |
| { | |
| "entropy": 0.5625248458236456, | |
| "epoch": 0.5551747772446881, | |
| "grad_norm": 1.40684974193573, | |
| "learning_rate": 0.00016303404158099155, | |
| "loss": 0.6314, | |
| "mean_token_accuracy": 0.8519342541694641, | |
| "num_tokens": 1236644.0, | |
| "step": 810 | |
| }, | |
| { | |
| "entropy": 0.49897886253893375, | |
| "epoch": 0.5620287868403016, | |
| "grad_norm": 0.5541409850120544, | |
| "learning_rate": 0.00016257710760795067, | |
| "loss": 0.5443, | |
| "mean_token_accuracy": 0.8650717407464981, | |
| "num_tokens": 1252654.0, | |
| "step": 820 | |
| }, | |
| { | |
| "entropy": 0.4627639502286911, | |
| "epoch": 0.568882796435915, | |
| "grad_norm": 0.48005378246307373, | |
| "learning_rate": 0.00016212017363490975, | |
| "loss": 0.5117, | |
| "mean_token_accuracy": 0.8780093342065811, | |
| "num_tokens": 1267484.0, | |
| "step": 830 | |
| }, | |
| { | |
| "entropy": 0.5190326888114214, | |
| "epoch": 0.5757368060315284, | |
| "grad_norm": 0.6706176400184631, | |
| "learning_rate": 0.00016166323966186887, | |
| "loss": 0.5251, | |
| "mean_token_accuracy": 0.8629909038543702, | |
| "num_tokens": 1282022.0, | |
| "step": 840 | |
| }, | |
| { | |
| "entropy": 0.5338688423857093, | |
| "epoch": 0.5825908156271419, | |
| "grad_norm": 0.5225579738616943, | |
| "learning_rate": 0.00016120630568882798, | |
| "loss": 0.5929, | |
| "mean_token_accuracy": 0.856892392039299, | |
| "num_tokens": 1296087.0, | |
| "step": 850 | |
| }, | |
| { | |
| "entropy": 0.49688454922288655, | |
| "epoch": 0.5894448252227553, | |
| "grad_norm": 0.26188451051712036, | |
| "learning_rate": 0.00016074937171578707, | |
| "loss": 0.6414, | |
| "mean_token_accuracy": 0.8633167922496796, | |
| "num_tokens": 1313509.0, | |
| "step": 860 | |
| }, | |
| { | |
| "entropy": 0.5256685543805361, | |
| "epoch": 0.5962988348183688, | |
| "grad_norm": 0.6064833998680115, | |
| "learning_rate": 0.0001602924377427462, | |
| "loss": 0.4999, | |
| "mean_token_accuracy": 0.8630939826369286, | |
| "num_tokens": 1328689.0, | |
| "step": 870 | |
| }, | |
| { | |
| "entropy": 0.5231131825596094, | |
| "epoch": 0.6031528444139822, | |
| "grad_norm": 0.4005596339702606, | |
| "learning_rate": 0.00015983550376970528, | |
| "loss": 0.5431, | |
| "mean_token_accuracy": 0.8552977308630944, | |
| "num_tokens": 1345534.0, | |
| "step": 880 | |
| }, | |
| { | |
| "entropy": 0.5158163897693158, | |
| "epoch": 0.6100068540095956, | |
| "grad_norm": 0.8037022948265076, | |
| "learning_rate": 0.0001593785697966644, | |
| "loss": 0.6514, | |
| "mean_token_accuracy": 0.8581083044409752, | |
| "num_tokens": 1361497.0, | |
| "step": 890 | |
| }, | |
| { | |
| "entropy": 0.4985586106777191, | |
| "epoch": 0.6168608636052091, | |
| "grad_norm": 0.38318124413490295, | |
| "learning_rate": 0.0001589216358236235, | |
| "loss": 0.5226, | |
| "mean_token_accuracy": 0.8698930114507675, | |
| "num_tokens": 1376582.0, | |
| "step": 900 | |
| }, | |
| { | |
| "entropy": 0.505715012550354, | |
| "epoch": 0.6237148732008225, | |
| "grad_norm": 0.39991825819015503, | |
| "learning_rate": 0.0001584647018505826, | |
| "loss": 0.5799, | |
| "mean_token_accuracy": 0.8601746618747711, | |
| "num_tokens": 1393099.0, | |
| "step": 910 | |
| }, | |
| { | |
| "entropy": 0.5512417580932378, | |
| "epoch": 0.630568882796436, | |
| "grad_norm": 0.3564458191394806, | |
| "learning_rate": 0.0001580077678775417, | |
| "loss": 0.6042, | |
| "mean_token_accuracy": 0.8542352899909019, | |
| "num_tokens": 1405413.0, | |
| "step": 920 | |
| }, | |
| { | |
| "entropy": 0.6023517435416579, | |
| "epoch": 0.6374228923920493, | |
| "grad_norm": 0.5077884197235107, | |
| "learning_rate": 0.0001575508339045008, | |
| "loss": 0.5491, | |
| "mean_token_accuracy": 0.8345923721790314, | |
| "num_tokens": 1420630.0, | |
| "step": 930 | |
| }, | |
| { | |
| "entropy": 0.45138914659619334, | |
| "epoch": 0.6442769019876627, | |
| "grad_norm": 0.526972234249115, | |
| "learning_rate": 0.00015709389993145991, | |
| "loss": 0.4937, | |
| "mean_token_accuracy": 0.8770761311054229, | |
| "num_tokens": 1434843.0, | |
| "step": 940 | |
| }, | |
| { | |
| "entropy": 0.5025592448189855, | |
| "epoch": 0.6511309115832762, | |
| "grad_norm": 0.33170071244239807, | |
| "learning_rate": 0.000156636965958419, | |
| "loss": 0.6154, | |
| "mean_token_accuracy": 0.8634001970291137, | |
| "num_tokens": 1451288.0, | |
| "step": 950 | |
| }, | |
| { | |
| "entropy": 0.4892792083323002, | |
| "epoch": 0.6579849211788896, | |
| "grad_norm": 0.5489900708198547, | |
| "learning_rate": 0.00015618003198537812, | |
| "loss": 0.4766, | |
| "mean_token_accuracy": 0.8682720705866813, | |
| "num_tokens": 1461993.0, | |
| "step": 960 | |
| }, | |
| { | |
| "entropy": 0.45295149497687814, | |
| "epoch": 0.6648389307745031, | |
| "grad_norm": 0.4360350966453552, | |
| "learning_rate": 0.00015572309801233723, | |
| "loss": 0.4526, | |
| "mean_token_accuracy": 0.8701859056949616, | |
| "num_tokens": 1476768.0, | |
| "step": 970 | |
| }, | |
| { | |
| "entropy": 0.5168599784374237, | |
| "epoch": 0.6716929403701165, | |
| "grad_norm": 0.6106163263320923, | |
| "learning_rate": 0.00015526616403929632, | |
| "loss": 0.5304, | |
| "mean_token_accuracy": 0.8633713632822037, | |
| "num_tokens": 1490815.0, | |
| "step": 980 | |
| }, | |
| { | |
| "entropy": 0.5374733801931143, | |
| "epoch": 0.67854694996573, | |
| "grad_norm": 0.6627058982849121, | |
| "learning_rate": 0.00015480923006625544, | |
| "loss": 0.6159, | |
| "mean_token_accuracy": 0.8490797847509384, | |
| "num_tokens": 1509112.0, | |
| "step": 990 | |
| }, | |
| { | |
| "entropy": 0.4429541861638427, | |
| "epoch": 0.6854009595613434, | |
| "grad_norm": 0.4334024488925934, | |
| "learning_rate": 0.00015435229609321452, | |
| "loss": 0.4791, | |
| "mean_token_accuracy": 0.8743316605687141, | |
| "num_tokens": 1525280.0, | |
| "step": 1000 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 4377, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "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": 1.216080158343168e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |