| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 1.0, |
| "eval_steps": 500, |
| "global_step": 3092, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.008085381630012937, |
| "grad_norm": 0.3009789288043976, |
| "learning_rate": 0.00013440860215053763, |
| "loss": 1.7048, |
| "mean_token_accuracy": 0.661682380437851, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.016170763260025874, |
| "grad_norm": 0.4460005462169647, |
| "learning_rate": 0.00026881720430107527, |
| "loss": 1.2718, |
| "mean_token_accuracy": 0.713244981765747, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.02425614489003881, |
| "grad_norm": 0.25601527094841003, |
| "learning_rate": 0.0004032258064516129, |
| "loss": 0.9995, |
| "mean_token_accuracy": 0.7553433787822723, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.03234152652005175, |
| "grad_norm": 0.2627596855163574, |
| "learning_rate": 0.0004999932787358948, |
| "loss": 0.9036, |
| "mean_token_accuracy": 0.7764883422851563, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.04042690815006468, |
| "grad_norm": 0.4143337905406952, |
| "learning_rate": 0.0004998595518201121, |
| "loss": 0.8985, |
| "mean_token_accuracy": 0.7727250349521637, |
| "step": 125 |
| }, |
| { |
| "epoch": 0.04851228978007762, |
| "grad_norm": 0.23922297358512878, |
| "learning_rate": 0.00049955446943686, |
| "loss": 0.7909, |
| "mean_token_accuracy": 0.7968144822120666, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.056597671410090554, |
| "grad_norm": 0.2834464907646179, |
| "learning_rate": 0.0004990782408138185, |
| "loss": 0.822, |
| "mean_token_accuracy": 0.7941457486152649, |
| "step": 175 |
| }, |
| { |
| "epoch": 0.0646830530401035, |
| "grad_norm": 0.29430675506591797, |
| "learning_rate": 0.000498431192551983, |
| "loss": 0.8546, |
| "mean_token_accuracy": 0.7836022305488587, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.07276843467011643, |
| "grad_norm": 0.23746679723262787, |
| "learning_rate": 0.0004976137684016788, |
| "loss": 0.7788, |
| "mean_token_accuracy": 0.8025749707221985, |
| "step": 225 |
| }, |
| { |
| "epoch": 0.08085381630012936, |
| "grad_norm": 0.24229322373867035, |
| "learning_rate": 0.0004966265289582338, |
| "loss": 0.8583, |
| "mean_token_accuracy": 0.7800547671318054, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.08893919793014231, |
| "grad_norm": 0.24889522790908813, |
| "learning_rate": 0.0004954701512775184, |
| "loss": 0.685, |
| "mean_token_accuracy": 0.8196275687217712, |
| "step": 275 |
| }, |
| { |
| "epoch": 0.09702457956015524, |
| "grad_norm": 0.22737586498260498, |
| "learning_rate": 0.0004941454284116157, |
| "loss": 0.8068, |
| "mean_token_accuracy": 0.7950991153717041, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.10510996119016817, |
| "grad_norm": 0.2224387228488922, |
| "learning_rate": 0.0004926532688649407, |
| "loss": 0.8054, |
| "mean_token_accuracy": 0.7932506108283996, |
| "step": 325 |
| }, |
| { |
| "epoch": 0.11319534282018111, |
| "grad_norm": 0.2293468415737152, |
| "learning_rate": 0.0004909946959711816, |
| "loss": 0.7816, |
| "mean_token_accuracy": 0.7965369141101837, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.12128072445019406, |
| "grad_norm": 0.2561907470226288, |
| "learning_rate": 0.0004891708471914897, |
| "loss": 0.7487, |
| "mean_token_accuracy": 0.8107043826580047, |
| "step": 375 |
| }, |
| { |
| "epoch": 0.129366106080207, |
| "grad_norm": 0.2583022117614746, |
| "learning_rate": 0.0004871829733344012, |
| "loss": 0.7786, |
| "mean_token_accuracy": 0.7999367988109589, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.13745148771021992, |
| "grad_norm": 0.2550680339336395, |
| "learning_rate": 0.00048503243769802327, |
| "loss": 0.7576, |
| "mean_token_accuracy": 0.8000009500980377, |
| "step": 425 |
| }, |
| { |
| "epoch": 0.14553686934023286, |
| "grad_norm": 0.24982689321041107, |
| "learning_rate": 0.0004827207151350745, |
| "loss": 0.7897, |
| "mean_token_accuracy": 0.7945795893669129, |
| "step": 450 |
| }, |
| { |
| "epoch": 0.1536222509702458, |
| "grad_norm": 0.2784912884235382, |
| "learning_rate": 0.0004802493910414205, |
| "loss": 0.8013, |
| "mean_token_accuracy": 0.792938197851181, |
| "step": 475 |
| }, |
| { |
| "epoch": 0.16170763260025872, |
| "grad_norm": 0.2220473736524582, |
| "learning_rate": 0.00047762016026879807, |
| "loss": 0.711, |
| "mean_token_accuracy": 0.8124627935886383, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.16979301423027165, |
| "grad_norm": 0.2841341197490692, |
| "learning_rate": 0.00047483482596247353, |
| "loss": 0.7382, |
| "mean_token_accuracy": 0.8129073297977447, |
| "step": 525 |
| }, |
| { |
| "epoch": 0.17787839586028462, |
| "grad_norm": 0.23232664167881012, |
| "learning_rate": 0.00047189529832463296, |
| "loss": 0.7588, |
| "mean_token_accuracy": 0.8059421420097351, |
| "step": 550 |
| }, |
| { |
| "epoch": 0.18596377749029755, |
| "grad_norm": 0.22178468108177185, |
| "learning_rate": 0.00046880359330435216, |
| "loss": 0.8146, |
| "mean_token_accuracy": 0.7915572142601013, |
| "step": 575 |
| }, |
| { |
| "epoch": 0.19404915912031048, |
| "grad_norm": 0.23006677627563477, |
| "learning_rate": 0.0004655618312150437, |
| "loss": 0.7822, |
| "mean_token_accuracy": 0.7942324769496918, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.20213454075032342, |
| "grad_norm": 0.32772719860076904, |
| "learning_rate": 0.00046217223528033146, |
| "loss": 0.7632, |
| "mean_token_accuracy": 0.8043809008598327, |
| "step": 625 |
| }, |
| { |
| "epoch": 0.21021992238033635, |
| "grad_norm": 0.18335266411304474, |
| "learning_rate": 0.0004586371301093476, |
| "loss": 0.7218, |
| "mean_token_accuracy": 0.816128705739975, |
| "step": 650 |
| }, |
| { |
| "epoch": 0.21830530401034928, |
| "grad_norm": 0.22064125537872314, |
| "learning_rate": 0.00045495894010249915, |
| "loss": 0.7364, |
| "mean_token_accuracy": 0.8087958478927613, |
| "step": 675 |
| }, |
| { |
| "epoch": 0.22639068564036222, |
| "grad_norm": 0.21018439531326294, |
| "learning_rate": 0.0004511401877887967, |
| "loss": 0.738, |
| "mean_token_accuracy": 0.8069515633583069, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.23447606727037515, |
| "grad_norm": 0.21225683391094208, |
| "learning_rate": 0.0004471834920958864, |
| "loss": 0.7397, |
| "mean_token_accuracy": 0.8051351046562195, |
| "step": 725 |
| }, |
| { |
| "epoch": 0.2425614489003881, |
| "grad_norm": 0.22577480971813202, |
| "learning_rate": 0.00044309156655397003, |
| "loss": 0.7872, |
| "mean_token_accuracy": 0.7951467871665955, |
| "step": 750 |
| }, |
| { |
| "epoch": 0.25064683053040104, |
| "grad_norm": 0.19751884043216705, |
| "learning_rate": 0.000438867217434847, |
| "loss": 0.7147, |
| "mean_token_accuracy": 0.8134795188903808, |
| "step": 775 |
| }, |
| { |
| "epoch": 0.258732212160414, |
| "grad_norm": 0.2011658251285553, |
| "learning_rate": 0.0004345133418273529, |
| "loss": 0.7923, |
| "mean_token_accuracy": 0.7959077858924866, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.2668175937904269, |
| "grad_norm": 0.2174764722585678, |
| "learning_rate": 0.00043003292565051544, |
| "loss": 0.7576, |
| "mean_token_accuracy": 0.8044932758808137, |
| "step": 825 |
| }, |
| { |
| "epoch": 0.27490297542043984, |
| "grad_norm": 0.20990514755249023, |
| "learning_rate": 0.0004254290416057898, |
| "loss": 0.739, |
| "mean_token_accuracy": 0.8073045027256012, |
| "step": 850 |
| }, |
| { |
| "epoch": 0.2829883570504528, |
| "grad_norm": 0.1976221799850464, |
| "learning_rate": 0.0004207048470697777, |
| "loss": 0.6717, |
| "mean_token_accuracy": 0.824974125623703, |
| "step": 875 |
| }, |
| { |
| "epoch": 0.2910737386804657, |
| "grad_norm": 0.2552309036254883, |
| "learning_rate": 0.0004158635819288762, |
| "loss": 0.7311, |
| "mean_token_accuracy": 0.8078971183300019, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.29915912031047864, |
| "grad_norm": 0.26244303584098816, |
| "learning_rate": 0.00041090856635734067, |
| "loss": 0.7264, |
| "mean_token_accuracy": 0.8127052938938141, |
| "step": 925 |
| }, |
| { |
| "epoch": 0.3072445019404916, |
| "grad_norm": 0.24333028495311737, |
| "learning_rate": 0.000405843198540285, |
| "loss": 0.7184, |
| "mean_token_accuracy": 0.8114526355266571, |
| "step": 950 |
| }, |
| { |
| "epoch": 0.3153298835705045, |
| "grad_norm": 0.19133129715919495, |
| "learning_rate": 0.0004006709523431822, |
| "loss": 0.7538, |
| "mean_token_accuracy": 0.8016650295257568, |
| "step": 975 |
| }, |
| { |
| "epoch": 0.32341526520051744, |
| "grad_norm": 0.25047338008880615, |
| "learning_rate": 0.00039539537492946285, |
| "loss": 0.8019, |
| "mean_token_accuracy": 0.7935136258602142, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.3315006468305304, |
| "grad_norm": 0.19724124670028687, |
| "learning_rate": 0.0003900200843278449, |
| "loss": 0.6892, |
| "mean_token_accuracy": 0.8166925406455994, |
| "step": 1025 |
| }, |
| { |
| "epoch": 0.3395860284605433, |
| "grad_norm": 0.21111617982387543, |
| "learning_rate": 0.0003845487669510631, |
| "loss": 0.7281, |
| "mean_token_accuracy": 0.8110716784000397, |
| "step": 1050 |
| }, |
| { |
| "epoch": 0.3476714100905563, |
| "grad_norm": 0.2487187534570694, |
| "learning_rate": 0.00037898517506770196, |
| "loss": 0.7962, |
| "mean_token_accuracy": 0.7921491277217865, |
| "step": 1075 |
| }, |
| { |
| "epoch": 0.35575679172056923, |
| "grad_norm": 0.2757723927497864, |
| "learning_rate": 0.0003733331242288622, |
| "loss": 0.7533, |
| "mean_token_accuracy": 0.8056223785877228, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.36384217335058217, |
| "grad_norm": 0.27630847692489624, |
| "learning_rate": 0.0003675964906514289, |
| "loss": 0.7885, |
| "mean_token_accuracy": 0.7973137283325196, |
| "step": 1125 |
| }, |
| { |
| "epoch": 0.3719275549805951, |
| "grad_norm": 0.2365068644285202, |
| "learning_rate": 0.00036177920855973405, |
| "loss": 0.7275, |
| "mean_token_accuracy": 0.8077067303657531, |
| "step": 1150 |
| }, |
| { |
| "epoch": 0.38001293661060803, |
| "grad_norm": 0.2511255443096161, |
| "learning_rate": 0.00035588526748743754, |
| "loss": 0.8052, |
| "mean_token_accuracy": 0.7931141972541809, |
| "step": 1175 |
| }, |
| { |
| "epoch": 0.38809831824062097, |
| "grad_norm": 0.2489156723022461, |
| "learning_rate": 0.0003499187095414763, |
| "loss": 0.7369, |
| "mean_token_accuracy": 0.8076127851009369, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.3961836998706339, |
| "grad_norm": 0.23067978024482727, |
| "learning_rate": 0.00034388362662995855, |
| "loss": 0.7393, |
| "mean_token_accuracy": 0.8086310243606567, |
| "step": 1225 |
| }, |
| { |
| "epoch": 0.40426908150064683, |
| "grad_norm": 0.2202920764684677, |
| "learning_rate": 0.000337784157655904, |
| "loss": 0.7578, |
| "mean_token_accuracy": 0.8064273929595948, |
| "step": 1250 |
| }, |
| { |
| "epoch": 0.41235446313065977, |
| "grad_norm": 0.21269731223583221, |
| "learning_rate": 0.0003316244856787544, |
| "loss": 0.7937, |
| "mean_token_accuracy": 0.795601452589035, |
| "step": 1275 |
| }, |
| { |
| "epoch": 0.4204398447606727, |
| "grad_norm": 0.24461720883846283, |
| "learning_rate": 0.0003254088350456017, |
| "loss": 0.6656, |
| "mean_token_accuracy": 0.8226878666877746, |
| "step": 1300 |
| }, |
| { |
| "epoch": 0.42852522639068563, |
| "grad_norm": 0.24032443761825562, |
| "learning_rate": 0.0003191414684941003, |
| "loss": 0.7778, |
| "mean_token_accuracy": 0.7960509729385375, |
| "step": 1325 |
| }, |
| { |
| "epoch": 0.43661060802069857, |
| "grad_norm": 0.25362804532051086, |
| "learning_rate": 0.0003128266842290513, |
| "loss": 0.6967, |
| "mean_token_accuracy": 0.8132575881481171, |
| "step": 1350 |
| }, |
| { |
| "epoch": 0.4446959896507115, |
| "grad_norm": 0.2388894110918045, |
| "learning_rate": 0.0003064688129746629, |
| "loss": 0.716, |
| "mean_token_accuracy": 0.8112483811378479, |
| "step": 1375 |
| }, |
| { |
| "epoch": 0.45278137128072443, |
| "grad_norm": 0.2121650129556656, |
| "learning_rate": 0.0003000722150045085, |
| "loss": 0.6942, |
| "mean_token_accuracy": 0.8156200110912323, |
| "step": 1400 |
| }, |
| { |
| "epoch": 0.46086675291073737, |
| "grad_norm": 0.2317550778388977, |
| "learning_rate": 0.0002936412771512206, |
| "loss": 0.7493, |
| "mean_token_accuracy": 0.8051575112342835, |
| "step": 1425 |
| }, |
| { |
| "epoch": 0.4689521345407503, |
| "grad_norm": 0.23476050794124603, |
| "learning_rate": 0.0002871804097979687, |
| "loss": 0.7136, |
| "mean_token_accuracy": 0.8104170382022857, |
| "step": 1450 |
| }, |
| { |
| "epoch": 0.4770375161707633, |
| "grad_norm": 0.2009560465812683, |
| "learning_rate": 0.00028069404385378736, |
| "loss": 0.7117, |
| "mean_token_accuracy": 0.8178813803195953, |
| "step": 1475 |
| }, |
| { |
| "epoch": 0.4851228978007762, |
| "grad_norm": 0.21633079648017883, |
| "learning_rate": 0.0002741866277148276, |
| "loss": 0.7392, |
| "mean_token_accuracy": 0.8080459308624267, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.49320827943078915, |
| "grad_norm": 0.2863864004611969, |
| "learning_rate": 0.00026766262421361407, |
| "loss": 0.7429, |
| "mean_token_accuracy": 0.8051086151599884, |
| "step": 1525 |
| }, |
| { |
| "epoch": 0.5012936610608021, |
| "grad_norm": 0.2746196389198303, |
| "learning_rate": 0.0002611265075584034, |
| "loss": 0.7378, |
| "mean_token_accuracy": 0.8071331679821014, |
| "step": 1550 |
| }, |
| { |
| "epoch": 0.509379042690815, |
| "grad_norm": 0.1745605319738388, |
| "learning_rate": 0.0002545827602647397, |
| "loss": 0.7329, |
| "mean_token_accuracy": 0.8071370398998261, |
| "step": 1575 |
| }, |
| { |
| "epoch": 0.517464424320828, |
| "grad_norm": 0.2692930996417999, |
| "learning_rate": 0.0002480358700813135, |
| "loss": 0.6469, |
| "mean_token_accuracy": 0.8244569575786591, |
| "step": 1600 |
| }, |
| { |
| "epoch": 0.5255498059508409, |
| "grad_norm": 0.2551010549068451, |
| "learning_rate": 0.00024149032691223173, |
| "loss": 0.7093, |
| "mean_token_accuracy": 0.8107299065589905, |
| "step": 1625 |
| }, |
| { |
| "epoch": 0.5336351875808538, |
| "grad_norm": 0.19366180896759033, |
| "learning_rate": 0.0002349506197378092, |
| "loss": 0.7518, |
| "mean_token_accuracy": 0.8024619662761688, |
| "step": 1650 |
| }, |
| { |
| "epoch": 0.5417205692108668, |
| "grad_norm": 0.22391283512115479, |
| "learning_rate": 0.00022842123353599369, |
| "loss": 0.6933, |
| "mean_token_accuracy": 0.8174584257602692, |
| "step": 1675 |
| }, |
| { |
| "epoch": 0.5498059508408797, |
| "grad_norm": 0.24723011255264282, |
| "learning_rate": 0.0002219066462065364, |
| "loss": 0.669, |
| "mean_token_accuracy": 0.822363510131836, |
| "step": 1700 |
| }, |
| { |
| "epoch": 0.5578913324708926, |
| "grad_norm": 0.20229819416999817, |
| "learning_rate": 0.00021541132550001584, |
| "loss": 0.6508, |
| "mean_token_accuracy": 0.8255820453166962, |
| "step": 1725 |
| }, |
| { |
| "epoch": 0.5659767141009056, |
| "grad_norm": 0.2818906605243683, |
| "learning_rate": 0.00020893972595382274, |
| "loss": 0.6417, |
| "mean_token_accuracy": 0.830688863992691, |
| "step": 1750 |
| }, |
| { |
| "epoch": 0.5740620957309185, |
| "grad_norm": 0.24256223440170288, |
| "learning_rate": 0.00020249628583720672, |
| "loss": 0.7353, |
| "mean_token_accuracy": 0.8104202616214752, |
| "step": 1775 |
| }, |
| { |
| "epoch": 0.5821474773609314, |
| "grad_norm": 0.24385611712932587, |
| "learning_rate": 0.00019608542410747888, |
| "loss": 0.6876, |
| "mean_token_accuracy": 0.819042581319809, |
| "step": 1800 |
| }, |
| { |
| "epoch": 0.5902328589909444, |
| "grad_norm": 0.15913242101669312, |
| "learning_rate": 0.00018971153737945968, |
| "loss": 0.646, |
| "mean_token_accuracy": 0.8284247839450836, |
| "step": 1825 |
| }, |
| { |
| "epoch": 0.5983182406209573, |
| "grad_norm": 0.17542122304439545, |
| "learning_rate": 0.00018337899691024914, |
| "loss": 0.6216, |
| "mean_token_accuracy": 0.83616614818573, |
| "step": 1850 |
| }, |
| { |
| "epoch": 0.6064036222509702, |
| "grad_norm": 0.22214658558368683, |
| "learning_rate": 0.0001770921456013872, |
| "loss": 0.6947, |
| "mean_token_accuracy": 0.8170740747451782, |
| "step": 1875 |
| }, |
| { |
| "epoch": 0.6144890038809832, |
| "grad_norm": 0.22132755815982819, |
| "learning_rate": 0.00017085529502046073, |
| "loss": 0.6788, |
| "mean_token_accuracy": 0.820680763721466, |
| "step": 1900 |
| }, |
| { |
| "epoch": 0.6225743855109961, |
| "grad_norm": 0.26897531747817993, |
| "learning_rate": 0.00016467272244420029, |
| "loss": 0.6833, |
| "mean_token_accuracy": 0.822064242362976, |
| "step": 1925 |
| }, |
| { |
| "epoch": 0.630659767141009, |
| "grad_norm": 0.25688934326171875, |
| "learning_rate": 0.0001585486679250922, |
| "loss": 0.6945, |
| "mean_token_accuracy": 0.8143122482299805, |
| "step": 1950 |
| }, |
| { |
| "epoch": 0.638745148771022, |
| "grad_norm": 0.21207553148269653, |
| "learning_rate": 0.0001524873313835208, |
| "loss": 0.6596, |
| "mean_token_accuracy": 0.8273860597610474, |
| "step": 1975 |
| }, |
| { |
| "epoch": 0.6468305304010349, |
| "grad_norm": 0.281393438577652, |
| "learning_rate": 0.00014649286972743319, |
| "loss": 0.6767, |
| "mean_token_accuracy": 0.8178416419029236, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.6549159120310478, |
| "grad_norm": 0.27408191561698914, |
| "learning_rate": 0.00014056939400150143, |
| "loss": 0.6974, |
| "mean_token_accuracy": 0.8189209842681885, |
| "step": 2025 |
| }, |
| { |
| "epoch": 0.6630012936610608, |
| "grad_norm": 0.26886942982673645, |
| "learning_rate": 0.00013472096656773913, |
| "loss": 0.6497, |
| "mean_token_accuracy": 0.8288757252693176, |
| "step": 2050 |
| }, |
| { |
| "epoch": 0.6710866752910737, |
| "grad_norm": 0.21919454634189606, |
| "learning_rate": 0.00012895159831950462, |
| "loss": 0.693, |
| "mean_token_accuracy": 0.8163833570480347, |
| "step": 2075 |
| }, |
| { |
| "epoch": 0.6791720569210866, |
| "grad_norm": 0.21283280849456787, |
| "learning_rate": 0.0001232652459308012, |
| "loss": 0.7117, |
| "mean_token_accuracy": 0.808628898859024, |
| "step": 2100 |
| }, |
| { |
| "epoch": 0.6872574385510997, |
| "grad_norm": 0.229765385389328, |
| "learning_rate": 0.00011766580914276209, |
| "loss": 0.7317, |
| "mean_token_accuracy": 0.8035627353191376, |
| "step": 2125 |
| }, |
| { |
| "epoch": 0.6953428201811126, |
| "grad_norm": 0.21411098539829254, |
| "learning_rate": 0.00011215712808918003, |
| "loss": 0.6469, |
| "mean_token_accuracy": 0.8277445828914642, |
| "step": 2150 |
| }, |
| { |
| "epoch": 0.7034282018111255, |
| "grad_norm": 0.2254790961742401, |
| "learning_rate": 0.00010674298066291601, |
| "loss": 0.6976, |
| "mean_token_accuracy": 0.8171502375602722, |
| "step": 2175 |
| }, |
| { |
| "epoch": 0.7115135834411385, |
| "grad_norm": 0.27148380875587463, |
| "learning_rate": 0.0001014270799249933, |
| "loss": 0.717, |
| "mean_token_accuracy": 0.8086051964759826, |
| "step": 2200 |
| }, |
| { |
| "epoch": 0.7195989650711514, |
| "grad_norm": 0.2047407031059265, |
| "learning_rate": 9.621307155815398e-05, |
| "loss": 0.718, |
| "mean_token_accuracy": 0.8121638822555542, |
| "step": 2225 |
| }, |
| { |
| "epoch": 0.7276843467011643, |
| "grad_norm": 0.22144050896167755, |
| "learning_rate": 9.11045313666231e-05, |
| "loss": 0.6623, |
| "mean_token_accuracy": 0.8254709720611573, |
| "step": 2250 |
| }, |
| { |
| "epoch": 0.7357697283311773, |
| "grad_norm": 0.27873218059539795, |
| "learning_rate": 8.610496282379687e-05, |
| "loss": 0.7073, |
| "mean_token_accuracy": 0.8168034076690673, |
| "step": 2275 |
| }, |
| { |
| "epoch": 0.7438551099611902, |
| "grad_norm": 0.25058674812316895, |
| "learning_rate": 8.121779466953572e-05, |
| "loss": 0.7961, |
| "mean_token_accuracy": 0.7905523943901062, |
| "step": 2300 |
| }, |
| { |
| "epoch": 0.7519404915912031, |
| "grad_norm": 0.2540716826915741, |
| "learning_rate": 7.644637855870959e-05, |
| "loss": 0.7561, |
| "mean_token_accuracy": 0.8036962306499481, |
| "step": 2325 |
| }, |
| { |
| "epoch": 0.7600258732212161, |
| "grad_norm": 0.2069474756717682, |
| "learning_rate": 7.179398676260923e-05, |
| "loss": 0.7163, |
| "mean_token_accuracy": 0.8117474913597107, |
| "step": 2350 |
| }, |
| { |
| "epoch": 0.768111254851229, |
| "grad_norm": 0.23127734661102295, |
| "learning_rate": 6.726380992479941e-05, |
| "loss": 0.6983, |
| "mean_token_accuracy": 0.8151715826988221, |
| "step": 2375 |
| }, |
| { |
| "epoch": 0.7761966364812419, |
| "grad_norm": 0.19075877964496613, |
| "learning_rate": 6.285895487295229e-05, |
| "loss": 0.6644, |
| "mean_token_accuracy": 0.8226857626438141, |
| "step": 2400 |
| }, |
| { |
| "epoch": 0.7842820181112549, |
| "grad_norm": 0.26920101046562195, |
| "learning_rate": 5.858244248816302e-05, |
| "loss": 0.678, |
| "mean_token_accuracy": 0.8184169673919678, |
| "step": 2425 |
| }, |
| { |
| "epoch": 0.7923673997412678, |
| "grad_norm": 0.22957506775856018, |
| "learning_rate": 5.443720563320792e-05, |
| "loss": 0.7125, |
| "mean_token_accuracy": 0.8165527045726776, |
| "step": 2450 |
| }, |
| { |
| "epoch": 0.8004527813712807, |
| "grad_norm": 0.23255349695682526, |
| "learning_rate": 5.042608714116612e-05, |
| "loss": 0.6648, |
| "mean_token_accuracy": 0.8205063927173615, |
| "step": 2475 |
| }, |
| { |
| "epoch": 0.8085381630012937, |
| "grad_norm": 0.21010981500148773, |
| "learning_rate": 4.655183786578426e-05, |
| "loss": 0.6833, |
| "mean_token_accuracy": 0.8196286606788635, |
| "step": 2500 |
| }, |
| { |
| "epoch": 0.8166235446313066, |
| "grad_norm": 0.21943055093288422, |
| "learning_rate": 4.2817114794921677e-05, |
| "loss": 0.6897, |
| "mean_token_accuracy": 0.8164256310462952, |
| "step": 2525 |
| }, |
| { |
| "epoch": 0.8247089262613195, |
| "grad_norm": 0.21263104677200317, |
| "learning_rate": 3.92244792283685e-05, |
| "loss": 0.6553, |
| "mean_token_accuracy": 0.8244921159744263, |
| "step": 2550 |
| }, |
| { |
| "epoch": 0.8327943078913325, |
| "grad_norm": 0.30405542254447937, |
| "learning_rate": 3.577639502128843e-05, |
| "loss": 0.6771, |
| "mean_token_accuracy": 0.8220798122882843, |
| "step": 2575 |
| }, |
| { |
| "epoch": 0.8408796895213454, |
| "grad_norm": 0.2544702887535095, |
| "learning_rate": 3.247522689448923e-05, |
| "loss": 0.6825, |
| "mean_token_accuracy": 0.8175348448753357, |
| "step": 2600 |
| }, |
| { |
| "epoch": 0.8489650711513583, |
| "grad_norm": 0.2168809473514557, |
| "learning_rate": 2.9323238812679982e-05, |
| "loss": 0.7437, |
| "mean_token_accuracy": 0.8062794303894043, |
| "step": 2625 |
| }, |
| { |
| "epoch": 0.8570504527813713, |
| "grad_norm": 0.32856041193008423, |
| "learning_rate": 2.6322592431828136e-05, |
| "loss": 0.6968, |
| "mean_token_accuracy": 0.8184567129611969, |
| "step": 2650 |
| }, |
| { |
| "epoch": 0.8651358344113842, |
| "grad_norm": 0.2495715171098709, |
| "learning_rate": 2.3475345616680327e-05, |
| "loss": 0.6864, |
| "mean_token_accuracy": 0.817262338399887, |
| "step": 2675 |
| }, |
| { |
| "epoch": 0.8732212160413971, |
| "grad_norm": 0.21743454039096832, |
| "learning_rate": 2.0783451029463995e-05, |
| "loss": 0.6554, |
| "mean_token_accuracy": 0.8261248970031738, |
| "step": 2700 |
| }, |
| { |
| "epoch": 0.8813065976714101, |
| "grad_norm": 0.19413378834724426, |
| "learning_rate": 1.8248754790737733e-05, |
| "loss": 0.691, |
| "mean_token_accuracy": 0.8169219958782196, |
| "step": 2725 |
| }, |
| { |
| "epoch": 0.889391979301423, |
| "grad_norm": 0.2755376100540161, |
| "learning_rate": 1.5872995213308566e-05, |
| "loss": 0.6868, |
| "mean_token_accuracy": 0.8156666767597198, |
| "step": 2750 |
| }, |
| { |
| "epoch": 0.8974773609314359, |
| "grad_norm": 0.23579077422618866, |
| "learning_rate": 1.3657801610084563e-05, |
| "loss": 0.669, |
| "mean_token_accuracy": 0.8254081463813782, |
| "step": 2775 |
| }, |
| { |
| "epoch": 0.9055627425614489, |
| "grad_norm": 0.21246632933616638, |
| "learning_rate": 1.1604693176680392e-05, |
| "loss": 0.6519, |
| "mean_token_accuracy": 0.8244655966758728, |
| "step": 2800 |
| }, |
| { |
| "epoch": 0.9136481241914618, |
| "grad_norm": 0.29115888476371765, |
| "learning_rate": 9.715077949542184e-06, |
| "loss": 0.6673, |
| "mean_token_accuracy": 0.825350991487503, |
| "step": 2825 |
| }, |
| { |
| "epoch": 0.9217335058214747, |
| "grad_norm": 0.24020685255527496, |
| "learning_rate": 7.990251840305996e-06, |
| "loss": 0.6349, |
| "mean_token_accuracy": 0.8334643471240998, |
| "step": 2850 |
| }, |
| { |
| "epoch": 0.9298188874514877, |
| "grad_norm": 0.2604895532131195, |
| "learning_rate": 6.431397747052342e-06, |
| "loss": 0.6659, |
| "mean_token_accuracy": 0.8195065236091614, |
| "step": 2875 |
| }, |
| { |
| "epoch": 0.9379042690815006, |
| "grad_norm": 0.19995881617069244, |
| "learning_rate": 5.039584743066344e-06, |
| "loss": 0.706, |
| "mean_token_accuracy": 0.8151924252510071, |
| "step": 2900 |
| }, |
| { |
| "epoch": 0.9459896507115135, |
| "grad_norm": 0.22976571321487427, |
| "learning_rate": 3.815767343659377e-06, |
| "loss": 0.6477, |
| "mean_token_accuracy": 0.8256394731998443, |
| "step": 2925 |
| }, |
| { |
| "epoch": 0.9540750323415266, |
| "grad_norm": 0.22688381373882294, |
| "learning_rate": 2.760784851555953e-06, |
| "loss": 0.672, |
| "mean_token_accuracy": 0.8214276111125947, |
| "step": 2950 |
| }, |
| { |
| "epoch": 0.9621604139715395, |
| "grad_norm": 0.23771828413009644, |
| "learning_rate": 1.875360781293689e-06, |
| "loss": 0.6642, |
| "mean_token_accuracy": 0.8211515319347381, |
| "step": 2975 |
| }, |
| { |
| "epoch": 0.9702457956015524, |
| "grad_norm": 0.22044353187084198, |
| "learning_rate": 1.1601023630319064e-06, |
| "loss": 0.6148, |
| "mean_token_accuracy": 0.838649377822876, |
| "step": 3000 |
| }, |
| { |
| "epoch": 0.9783311772315654, |
| "grad_norm": 0.21916313469409943, |
| "learning_rate": 6.155001261089477e-07, |
| "loss": 0.7242, |
| "mean_token_accuracy": 0.8094534778594971, |
| "step": 3025 |
| }, |
| { |
| "epoch": 0.9864165588615783, |
| "grad_norm": 0.20908524096012115, |
| "learning_rate": 2.4192756263349826e-07, |
| "loss": 0.7293, |
| "mean_token_accuracy": 0.8055576062202454, |
| "step": 3050 |
| }, |
| { |
| "epoch": 0.9945019404915912, |
| "grad_norm": 0.2526475489139557, |
| "learning_rate": 3.9640871341173336e-08, |
| "loss": 0.6961, |
| "mean_token_accuracy": 0.8134376859664917, |
| "step": 3075 |
| } |
| ], |
| "logging_steps": 25, |
| "max_steps": 3092, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 1, |
| "save_steps": 0, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.2584006622747034e+17, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
