| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 35802, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.013965700240210044, | |
| "grad_norm": 4.268923282623291, | |
| "learning_rate": 0.0004930730126808558, | |
| "loss": 2.5844, | |
| "mean_token_accuracy": 0.5059588371738791, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.027931400480420088, | |
| "grad_norm": 0.6783562302589417, | |
| "learning_rate": 0.0004860901625607508, | |
| "loss": 2.153, | |
| "mean_token_accuracy": 0.5498105692565441, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.04189710072063013, | |
| "grad_norm": 0.49954575300216675, | |
| "learning_rate": 0.0004791073124406458, | |
| "loss": 2.0782, | |
| "mean_token_accuracy": 0.5626471206247806, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.055862800960840175, | |
| "grad_norm": 0.501465916633606, | |
| "learning_rate": 0.0004721244623205407, | |
| "loss": 2.0761, | |
| "mean_token_accuracy": 0.5623862668275833, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.06982850120105022, | |
| "grad_norm": 0.52457594871521, | |
| "learning_rate": 0.0004651416122004357, | |
| "loss": 2.0543, | |
| "mean_token_accuracy": 0.5653762078285217, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.08379420144126026, | |
| "grad_norm": 0.366854727268219, | |
| "learning_rate": 0.0004581587620803307, | |
| "loss": 2.0699, | |
| "mean_token_accuracy": 0.5624443043172359, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.09775990168147031, | |
| "grad_norm": 0.33193984627723694, | |
| "learning_rate": 0.00045118987766046596, | |
| "loss": 2.0706, | |
| "mean_token_accuracy": 0.5632118558287621, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.11172560192168035, | |
| "grad_norm": 0.26177772879600525, | |
| "learning_rate": 0.0004442070275403609, | |
| "loss": 2.0744, | |
| "mean_token_accuracy": 0.564106923341751, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.1256913021618904, | |
| "grad_norm": 0.31001174449920654, | |
| "learning_rate": 0.0004372521088207363, | |
| "loss": 2.0789, | |
| "mean_token_accuracy": 0.5612107511162758, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.13965700240210044, | |
| "grad_norm": 0.26565852761268616, | |
| "learning_rate": 0.0004302692587006313, | |
| "loss": 2.0766, | |
| "mean_token_accuracy": 0.5622789396047592, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.15362270264231048, | |
| "grad_norm": 0.2479698359966278, | |
| "learning_rate": 0.00042331433998100666, | |
| "loss": 2.0658, | |
| "mean_token_accuracy": 0.5642604413330555, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.16758840288252053, | |
| "grad_norm": 0.2320714145898819, | |
| "learning_rate": 0.00041633148986090166, | |
| "loss": 2.0537, | |
| "mean_token_accuracy": 0.5658319992125034, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.18155410312273057, | |
| "grad_norm": 0.20908017456531525, | |
| "learning_rate": 0.00040937657114127705, | |
| "loss": 2.068, | |
| "mean_token_accuracy": 0.5633928556740284, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.19551980336294061, | |
| "grad_norm": 0.21336887776851654, | |
| "learning_rate": 0.000402393721021172, | |
| "loss": 2.056, | |
| "mean_token_accuracy": 0.5659000330269337, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.20948550360315066, | |
| "grad_norm": 0.18956658244132996, | |
| "learning_rate": 0.00039543880230154737, | |
| "loss": 2.0611, | |
| "mean_token_accuracy": 0.5643967908620834, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.2234512038433607, | |
| "grad_norm": 0.19688238203525543, | |
| "learning_rate": 0.00038845595218144237, | |
| "loss": 2.0571, | |
| "mean_token_accuracy": 0.5656557891070843, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.23741690408357075, | |
| "grad_norm": 0.17653051018714905, | |
| "learning_rate": 0.0003815010334618178, | |
| "loss": 2.0642, | |
| "mean_token_accuracy": 0.5636518878340722, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.2513826043237808, | |
| "grad_norm": 0.163355752825737, | |
| "learning_rate": 0.0003745181833417128, | |
| "loss": 2.0687, | |
| "mean_token_accuracy": 0.5622124362289905, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.26534830456399083, | |
| "grad_norm": 0.1576842963695526, | |
| "learning_rate": 0.0003675632646220882, | |
| "loss": 2.0537, | |
| "mean_token_accuracy": 0.5655075218081475, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.2793140048042009, | |
| "grad_norm": 0.1434595286846161, | |
| "learning_rate": 0.00036058041450198313, | |
| "loss": 2.0531, | |
| "mean_token_accuracy": 0.566190888762474, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.2932797050444109, | |
| "grad_norm": 0.15508811175823212, | |
| "learning_rate": 0.0003536254957823585, | |
| "loss": 2.0609, | |
| "mean_token_accuracy": 0.5647544565498829, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.30724540528462096, | |
| "grad_norm": 0.1619788408279419, | |
| "learning_rate": 0.0003466426456622535, | |
| "loss": 2.0572, | |
| "mean_token_accuracy": 0.5651431261301041, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.321211105524831, | |
| "grad_norm": 0.15186958014965057, | |
| "learning_rate": 0.0003396877269426289, | |
| "loss": 2.052, | |
| "mean_token_accuracy": 0.5652778272330761, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.33517680576504105, | |
| "grad_norm": 0.1533740907907486, | |
| "learning_rate": 0.0003327048768225239, | |
| "loss": 2.0507, | |
| "mean_token_accuracy": 0.5665202861726284, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.3491425060052511, | |
| "grad_norm": 0.15380869805812836, | |
| "learning_rate": 0.00032574995810289933, | |
| "loss": 2.0556, | |
| "mean_token_accuracy": 0.5652170441150666, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 0.36310820624546114, | |
| "grad_norm": 0.1550268828868866, | |
| "learning_rate": 0.0003187671079827943, | |
| "loss": 2.0509, | |
| "mean_token_accuracy": 0.565740670889616, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.3770739064856712, | |
| "grad_norm": 0.15138842165470123, | |
| "learning_rate": 0.00031181218926316966, | |
| "loss": 2.0576, | |
| "mean_token_accuracy": 0.5651396196782589, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 0.39103960672588123, | |
| "grad_norm": 0.14943508803844452, | |
| "learning_rate": 0.00030482933914306466, | |
| "loss": 2.0459, | |
| "mean_token_accuracy": 0.5667617973387241, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.40500530696609127, | |
| "grad_norm": 0.15654833614826202, | |
| "learning_rate": 0.00029787442042344004, | |
| "loss": 2.0682, | |
| "mean_token_accuracy": 0.5644841706454754, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 0.4189710072063013, | |
| "grad_norm": 0.1552565097808838, | |
| "learning_rate": 0.00029089157030333504, | |
| "loss": 2.0619, | |
| "mean_token_accuracy": 0.5638643040955067, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.43293670744651136, | |
| "grad_norm": 0.15082110464572906, | |
| "learning_rate": 0.0002839366515837104, | |
| "loss": 2.0546, | |
| "mean_token_accuracy": 0.565163453668356, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 0.4469024076867214, | |
| "grad_norm": 0.16357985138893127, | |
| "learning_rate": 0.00027695380146360536, | |
| "loss": 2.052, | |
| "mean_token_accuracy": 0.5660725966691971, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.46086810792693145, | |
| "grad_norm": 0.1474219560623169, | |
| "learning_rate": 0.00026999888274398075, | |
| "loss": 2.0514, | |
| "mean_token_accuracy": 0.5660765036344528, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 0.4748338081671415, | |
| "grad_norm": 0.15185120701789856, | |
| "learning_rate": 0.00026301603262387575, | |
| "loss": 2.0445, | |
| "mean_token_accuracy": 0.5662275057137013, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.48879950840735154, | |
| "grad_norm": 0.14625252783298492, | |
| "learning_rate": 0.0002560611139042512, | |
| "loss": 2.0506, | |
| "mean_token_accuracy": 0.567681826710701, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 0.5027652086475616, | |
| "grad_norm": 0.1625310778617859, | |
| "learning_rate": 0.0002490782637841461, | |
| "loss": 2.0475, | |
| "mean_token_accuracy": 0.565769705593586, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.5167309088877716, | |
| "grad_norm": 0.1645444631576538, | |
| "learning_rate": 0.00024212334506452154, | |
| "loss": 2.0541, | |
| "mean_token_accuracy": 0.5652133010923862, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 0.5306966091279817, | |
| "grad_norm": 0.15332843363285065, | |
| "learning_rate": 0.00023514049494441654, | |
| "loss": 2.0476, | |
| "mean_token_accuracy": 0.5676583781838417, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.5446623093681917, | |
| "grad_norm": 0.1857738345861435, | |
| "learning_rate": 0.00022818557622479192, | |
| "loss": 2.046, | |
| "mean_token_accuracy": 0.5666445840001106, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 0.5586280096084018, | |
| "grad_norm": 0.1527378261089325, | |
| "learning_rate": 0.0002212027261046869, | |
| "loss": 2.0469, | |
| "mean_token_accuracy": 0.5673894871473313, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.5725937098486118, | |
| "grad_norm": 0.15622954070568085, | |
| "learning_rate": 0.0002142478073850623, | |
| "loss": 2.0451, | |
| "mean_token_accuracy": 0.5668408484160901, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 0.5865594100888218, | |
| "grad_norm": 0.15915007889270782, | |
| "learning_rate": 0.00020726495726495727, | |
| "loss": 2.0423, | |
| "mean_token_accuracy": 0.5684662259817124, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 0.6005251103290319, | |
| "grad_norm": 0.17898404598236084, | |
| "learning_rate": 0.00020031003854533265, | |
| "loss": 2.0485, | |
| "mean_token_accuracy": 0.5655563974380493, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 0.6144908105692419, | |
| "grad_norm": 0.18496471643447876, | |
| "learning_rate": 0.00019332718842522765, | |
| "loss": 2.0495, | |
| "mean_token_accuracy": 0.5659684043526649, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.628456510809452, | |
| "grad_norm": 0.15914028882980347, | |
| "learning_rate": 0.00018637226970560304, | |
| "loss": 2.0587, | |
| "mean_token_accuracy": 0.5645939151346684, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 0.642422211049662, | |
| "grad_norm": 0.1719997376203537, | |
| "learning_rate": 0.000179389419585498, | |
| "loss": 2.0522, | |
| "mean_token_accuracy": 0.5654012953042984, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 0.6563879112898721, | |
| "grad_norm": 0.1447959542274475, | |
| "learning_rate": 0.00017243450086587342, | |
| "loss": 2.0376, | |
| "mean_token_accuracy": 0.5688524658381939, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 0.6703536115300821, | |
| "grad_norm": 0.1650005280971527, | |
| "learning_rate": 0.00016545165074576841, | |
| "loss": 2.0554, | |
| "mean_token_accuracy": 0.5649112530052662, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 0.6843193117702921, | |
| "grad_norm": 0.1643020361661911, | |
| "learning_rate": 0.0001584967320261438, | |
| "loss": 2.0369, | |
| "mean_token_accuracy": 0.5688672766685486, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 0.6982850120105022, | |
| "grad_norm": 0.17811213433742523, | |
| "learning_rate": 0.00015151388190603877, | |
| "loss": 2.0551, | |
| "mean_token_accuracy": 0.5658198243677616, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.7122507122507122, | |
| "grad_norm": 0.18822260200977325, | |
| "learning_rate": 0.00014455896318641418, | |
| "loss": 2.0391, | |
| "mean_token_accuracy": 0.5678052612841129, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 0.7262164124909223, | |
| "grad_norm": 0.1553669422864914, | |
| "learning_rate": 0.00013757611306630915, | |
| "loss": 2.0481, | |
| "mean_token_accuracy": 0.5658735463917255, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 0.7401821127311323, | |
| "grad_norm": 0.1644522100687027, | |
| "learning_rate": 0.00013062119434668453, | |
| "loss": 2.0443, | |
| "mean_token_accuracy": 0.5666062987446785, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 0.7541478129713424, | |
| "grad_norm": 0.18426066637039185, | |
| "learning_rate": 0.00012363834422657953, | |
| "loss": 2.0446, | |
| "mean_token_accuracy": 0.5659443159997464, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 0.7681135132115524, | |
| "grad_norm": 0.19535085558891296, | |
| "learning_rate": 0.00011668342550695491, | |
| "loss": 2.0498, | |
| "mean_token_accuracy": 0.5656883924603462, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 0.7820792134517625, | |
| "grad_norm": 0.1869782954454422, | |
| "learning_rate": 0.0001097005753868499, | |
| "loss": 2.0352, | |
| "mean_token_accuracy": 0.5688391443789005, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 0.7960449136919725, | |
| "grad_norm": 0.17099939286708832, | |
| "learning_rate": 0.0001027456566672253, | |
| "loss": 2.0505, | |
| "mean_token_accuracy": 0.5657015817165375, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 0.8100106139321825, | |
| "grad_norm": 0.17975910007953644, | |
| "learning_rate": 9.576280654712028e-05, | |
| "loss": 2.0318, | |
| "mean_token_accuracy": 0.5694990956783295, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 0.8239763141723926, | |
| "grad_norm": 0.17870086431503296, | |
| "learning_rate": 8.880788782749568e-05, | |
| "loss": 2.0447, | |
| "mean_token_accuracy": 0.5674924266338348, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 0.8379420144126026, | |
| "grad_norm": 0.1745506078004837, | |
| "learning_rate": 8.182503770739065e-05, | |
| "loss": 2.041, | |
| "mean_token_accuracy": 0.5678074882626534, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.8519077146528127, | |
| "grad_norm": 0.18852059543132782, | |
| "learning_rate": 7.487011898776604e-05, | |
| "loss": 2.0407, | |
| "mean_token_accuracy": 0.5682654000520706, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 0.8658734148930227, | |
| "grad_norm": 0.15847346186637878, | |
| "learning_rate": 6.788726886766103e-05, | |
| "loss": 2.0414, | |
| "mean_token_accuracy": 0.5677537434101104, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 0.8798391151332328, | |
| "grad_norm": 0.1863761991262436, | |
| "learning_rate": 6.093235014803642e-05, | |
| "loss": 2.0484, | |
| "mean_token_accuracy": 0.5661664333343506, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 0.8938048153734428, | |
| "grad_norm": 0.18861085176467896, | |
| "learning_rate": 5.39495000279314e-05, | |
| "loss": 2.0404, | |
| "mean_token_accuracy": 0.5677967172861099, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 0.9077705156136529, | |
| "grad_norm": 0.17940880358219147, | |
| "learning_rate": 4.69945813083068e-05, | |
| "loss": 2.0334, | |
| "mean_token_accuracy": 0.5684446477293968, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 0.9217362158538629, | |
| "grad_norm": 0.17555660009384155, | |
| "learning_rate": 4.0011731188201776e-05, | |
| "loss": 2.0472, | |
| "mean_token_accuracy": 0.5666785769164562, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 0.9357019160940729, | |
| "grad_norm": 0.18865884840488434, | |
| "learning_rate": 3.3056812468577174e-05, | |
| "loss": 2.044, | |
| "mean_token_accuracy": 0.5670678113102913, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 0.949667616334283, | |
| "grad_norm": 0.20321083068847656, | |
| "learning_rate": 2.6073962348472154e-05, | |
| "loss": 2.0505, | |
| "mean_token_accuracy": 0.5650117390453816, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 0.963633316574493, | |
| "grad_norm": 0.17523406445980072, | |
| "learning_rate": 1.911904362884755e-05, | |
| "loss": 2.0555, | |
| "mean_token_accuracy": 0.5649900923371315, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 0.9775990168147031, | |
| "grad_norm": 0.1853921115398407, | |
| "learning_rate": 1.213619350874253e-05, | |
| "loss": 2.0464, | |
| "mean_token_accuracy": 0.5669381264448166, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 0.9915647170549131, | |
| "grad_norm": 0.20597019791603088, | |
| "learning_rate": 5.181274789117926e-06, | |
| "loss": 2.0481, | |
| "mean_token_accuracy": 0.5671657482087612, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "mean_token_accuracy": 0.5673895427920171, | |
| "step": 35802, | |
| "total_flos": 6.223165951274123e+18, | |
| "train_loss": 2.0618421919924823, | |
| "train_runtime": 64377.845, | |
| "train_samples_per_second": 17.796, | |
| "train_steps_per_second": 0.556 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 35802, | |
| "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": false, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6.223165951274123e+18, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |