| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.5, | |
| "global_step": 21362, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 1.6355140186915887e-06, | |
| "loss": 10.7824, | |
| "theoretical_loss": 20.81281780154715, | |
| "tokens_seen": 65536 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 8.177570093457944e-05, | |
| "loss": 8.6889, | |
| "theoretical_loss": 8.563482664611069, | |
| "tokens_seen": 3276800 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 0.0001635514018691589, | |
| "loss": 6.6636, | |
| "theoretical_loss": 7.4777587180480305, | |
| "tokens_seen": 6553600 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 0.0002453271028037383, | |
| "loss": 5.9745, | |
| "theoretical_loss": 6.9337544888949, | |
| "tokens_seen": 9830400 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 0.0003271028037383178, | |
| "loss": 5.5783, | |
| "theoretical_loss": 6.583566228426414, | |
| "tokens_seen": 13107200 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.0004088785046728972, | |
| "loss": 5.3806, | |
| "theoretical_loss": 6.330713565116083, | |
| "tokens_seen": 16384000 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.0004906542056074766, | |
| "loss": 5.2075, | |
| "theoretical_loss": 6.135529231940326, | |
| "tokens_seen": 19660800 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.0005724299065420561, | |
| "loss": 5.0385, | |
| "theoretical_loss": 5.978101583869607, | |
| "tokens_seen": 22937600 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.0006542056074766356, | |
| "loss": 4.946, | |
| "theoretical_loss": 5.8471173262659235, | |
| "tokens_seen": 26214400 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.0006996358993758275, | |
| "loss": 4.7943, | |
| "theoretical_loss": 5.7355768158821245, | |
| "tokens_seen": 29491200 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.0006988083979572536, | |
| "loss": 4.7314, | |
| "theoretical_loss": 5.638870144071353, | |
| "tokens_seen": 32768000 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.0006979808965386797, | |
| "loss": 4.6052, | |
| "theoretical_loss": 5.553812381844907, | |
| "tokens_seen": 36044800 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.000697153395120106, | |
| "loss": 4.5231, | |
| "theoretical_loss": 5.478118080556438, | |
| "tokens_seen": 39321600 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0006963258937015321, | |
| "loss": 4.4354, | |
| "theoretical_loss": 5.410095959579362, | |
| "tokens_seen": 42598400 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0006954983922829582, | |
| "loss": 4.4109, | |
| "theoretical_loss": 5.348462083735834, | |
| "tokens_seen": 45875200 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0006946708908643843, | |
| "loss": 4.258, | |
| "theoretical_loss": 5.292220566937567, | |
| "tokens_seen": 49152000 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0006938433894458105, | |
| "loss": 4.2162, | |
| "theoretical_loss": 5.240584625769978, | |
| "tokens_seen": 52428800 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0006930158880272367, | |
| "loss": 4.1529, | |
| "theoretical_loss": 5.192922724525789, | |
| "tokens_seen": 55705600 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0006921883866086628, | |
| "loss": 4.0864, | |
| "theoretical_loss": 5.1487208633564405, | |
| "tokens_seen": 58982400 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0006913608851900889, | |
| "loss": 3.9637, | |
| "theoretical_loss": 5.107555562405102, | |
| "tokens_seen": 62259200 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.000690533383771515, | |
| "loss": 3.8525, | |
| "theoretical_loss": 5.069074117143246, | |
| "tokens_seen": 65536000 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0006897058823529412, | |
| "loss": 3.8185, | |
| "theoretical_loss": 5.032979909838007, | |
| "tokens_seen": 68812800 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.0006888783809343674, | |
| "loss": 3.759, | |
| "theoretical_loss": 4.999021308224664, | |
| "tokens_seen": 72089600 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.0006880508795157935, | |
| "loss": 3.703, | |
| "theoretical_loss": 4.966983155351962, | |
| "tokens_seen": 75366400 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.0006872233780972196, | |
| "loss": 3.6079, | |
| "theoretical_loss": 4.9366801616251355, | |
| "tokens_seen": 78643200 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.0006863958766786457, | |
| "loss": 3.5951, | |
| "theoretical_loss": 4.907951713830082, | |
| "tokens_seen": 81920000 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.0006855683752600718, | |
| "loss": 3.533, | |
| "theoretical_loss": 4.880657753812926, | |
| "tokens_seen": 85196800 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.000684740873841498, | |
| "loss": 3.5388, | |
| "theoretical_loss": 4.854675474481779, | |
| "tokens_seen": 88473600 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.0006839133724229242, | |
| "loss": 3.4465, | |
| "theoretical_loss": 4.8298966473088125, | |
| "tokens_seen": 91750400 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.0006830858710043503, | |
| "loss": 3.4702, | |
| "theoretical_loss": 4.8062254427779205, | |
| "tokens_seen": 95027200 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.0006822583695857764, | |
| "loss": 3.4191, | |
| "theoretical_loss": 4.783576639276257, | |
| "tokens_seen": 98304000 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.0006814308681672025, | |
| "loss": 3.4562, | |
| "theoretical_loss": 4.761874140772408, | |
| "tokens_seen": 101580800 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.0006806033667486286, | |
| "loss": 3.4558, | |
| "theoretical_loss": 4.741049741962473, | |
| "tokens_seen": 104857600 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.0006797758653300548, | |
| "loss": 3.4434, | |
| "theoretical_loss": 4.721042093249051, | |
| "tokens_seen": 108134400 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.000678948363911481, | |
| "loss": 3.4108, | |
| "theoretical_loss": 4.701795828231866, | |
| "tokens_seen": 111411200 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.0006781208624929071, | |
| "loss": 3.4049, | |
| "theoretical_loss": 4.68326082423593, | |
| "tokens_seen": 114688000 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.0006772933610743332, | |
| "loss": 3.3773, | |
| "theoretical_loss": 4.665391572426282, | |
| "tokens_seen": 117964800 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.0006764658596557593, | |
| "loss": 3.3413, | |
| "theoretical_loss": 4.648146638719739, | |
| "tokens_seen": 121241600 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.0006756383582371856, | |
| "loss": 3.356, | |
| "theoretical_loss": 4.631488200339643, | |
| "tokens_seen": 124518400 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.0006748108568186117, | |
| "loss": 3.3753, | |
| "theoretical_loss": 4.615381645715717, | |
| "tokens_seen": 127795200 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.0006739833554000378, | |
| "loss": 3.3066, | |
| "theoretical_loss": 4.599795227690505, | |
| "tokens_seen": 131072000 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.000673155853981464, | |
| "loss": 3.3297, | |
| "theoretical_loss": 4.584699761792674, | |
| "tokens_seen": 134348800 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.0006723283525628902, | |
| "loss": 3.284, | |
| "theoretical_loss": 4.570068362778516, | |
| "tokens_seen": 137625600 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.0006715008511443163, | |
| "loss": 3.3003, | |
| "theoretical_loss": 4.555876213804037, | |
| "tokens_seen": 140902400 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.0006706733497257424, | |
| "loss": 3.2875, | |
| "theoretical_loss": 4.542100363530799, | |
| "tokens_seen": 144179200 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.0006698458483071685, | |
| "loss": 3.2669, | |
| "theoretical_loss": 4.528719547234816, | |
| "tokens_seen": 147456000 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.0006690183468885946, | |
| "loss": 3.2242, | |
| "theoretical_loss": 4.515714028614996, | |
| "tokens_seen": 150732800 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.0006681908454700209, | |
| "loss": 3.2343, | |
| "theoretical_loss": 4.503065459513339, | |
| "tokens_seen": 154009600 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.000667363344051447, | |
| "loss": 3.2442, | |
| "theoretical_loss": 4.4907567551852665, | |
| "tokens_seen": 157286400 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.0006665358426328731, | |
| "loss": 3.2286, | |
| "theoretical_loss": 4.478771983111967, | |
| "tokens_seen": 160563200 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "objective/train/avg_token_score": 0.027871694415807724, | |
| "objective/train/avg_weight": 0.9746770858764648, | |
| "objective/train/docs_used": 104000, | |
| "objective/train/instantaneous_batch_size": 32, | |
| "objective/train/instantaneous_microbatch_size": 32768, | |
| "objective/train/original_loss": 3.3261609077453613, | |
| "objective/train/std_weight": 0.07171463221311569, | |
| "objective/train/theoretical_loss": 4.467096263641219, | |
| "objective/train/tokens_used": 184300000, | |
| "theoretical_loss": 4.467096263641219, | |
| "tokens_seen": 163840000 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.0006657083412142992, | |
| "loss": 3.2241, | |
| "theoretical_loss": 4.467096263641219, | |
| "tokens_seen": 163840000 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.0006648808397957253, | |
| "loss": 3.2466, | |
| "theoretical_loss": 4.455715680989545, | |
| "tokens_seen": 167116800 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.0006640533383771514, | |
| "loss": 3.1999, | |
| "theoretical_loss": 4.44461720334543, | |
| "tokens_seen": 170393600 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.0006632258369585777, | |
| "loss": 3.1426, | |
| "theoretical_loss": 4.433788610987646, | |
| "tokens_seen": 173670400 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.0006623983355400038, | |
| "loss": 3.1918, | |
| "theoretical_loss": 4.42321843148016, | |
| "tokens_seen": 176947200 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.0006615708341214299, | |
| "loss": 3.1379, | |
| "theoretical_loss": 4.412895881130142, | |
| "tokens_seen": 180224000 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.000660743332702856, | |
| "loss": 3.1736, | |
| "theoretical_loss": 4.4028108120020795, | |
| "tokens_seen": 183500800 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0006599158312842821, | |
| "loss": 3.1695, | |
| "theoretical_loss": 4.392953663871862, | |
| "tokens_seen": 186777600 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0006590883298657083, | |
| "loss": 3.1562, | |
| "theoretical_loss": 4.383315420582533, | |
| "tokens_seen": 190054400 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0006582608284471345, | |
| "loss": 3.1895, | |
| "theoretical_loss": 4.373887570330275, | |
| "tokens_seen": 193331200 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0006574333270285606, | |
| "loss": 3.1465, | |
| "theoretical_loss": 4.364662069466704, | |
| "tokens_seen": 196608000 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0006566058256099867, | |
| "loss": 3.146, | |
| "theoretical_loss": 4.355631309453283, | |
| "tokens_seen": 199884800 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0006557783241914128, | |
| "loss": 3.0991, | |
| "theoretical_loss": 4.346788086646671, | |
| "tokens_seen": 203161600 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0006549508227728391, | |
| "loss": 3.1286, | |
| "theoretical_loss": 4.33812557463116, | |
| "tokens_seen": 206438400 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0006541233213542652, | |
| "loss": 3.1357, | |
| "theoretical_loss": 4.329637298846812, | |
| "tokens_seen": 209715200 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.0006532958199356913, | |
| "loss": 3.0929, | |
| "theoretical_loss": 4.321317113290252, | |
| "tokens_seen": 212992000 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.0006524683185171174, | |
| "loss": 3.0686, | |
| "theoretical_loss": 4.3131591790897925, | |
| "tokens_seen": 216268800 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.0006516408170985437, | |
| "loss": 3.1167, | |
| "theoretical_loss": 4.305157944778228, | |
| "tokens_seen": 219545600 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.0006508133156799698, | |
| "loss": 3.1412, | |
| "theoretical_loss": 4.297308128105687, | |
| "tokens_seen": 222822400 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.0006499858142613959, | |
| "loss": 3.1237, | |
| "theoretical_loss": 4.2896046992515995, | |
| "tokens_seen": 226099200 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.000649158312842822, | |
| "loss": 3.1605, | |
| "theoretical_loss": 4.282042865309616, | |
| "tokens_seen": 229376000 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.0006483308114242481, | |
| "loss": 3.1312, | |
| "theoretical_loss": 4.274618055932298, | |
| "tokens_seen": 232652800 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.0006475033100056744, | |
| "loss": 3.0994, | |
| "theoretical_loss": 4.267325910033897, | |
| "tokens_seen": 235929600 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.0006466758085871005, | |
| "loss": 3.049, | |
| "theoretical_loss": 4.260162263459744, | |
| "tokens_seen": 239206400 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.0006458483071685266, | |
| "loss": 3.0873, | |
| "theoretical_loss": 4.253123137539814, | |
| "tokens_seen": 242483200 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.0006450208057499527, | |
| "loss": 3.1018, | |
| "theoretical_loss": 4.246204728452055, | |
| "tokens_seen": 245760000 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.0006441933043313788, | |
| "loss": 3.116, | |
| "theoretical_loss": 4.239403397328261, | |
| "tokens_seen": 249036800 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.0006433658029128049, | |
| "loss": 3.0758, | |
| "theoretical_loss": 4.232715661041632, | |
| "tokens_seen": 252313600 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.0006425383014942312, | |
| "loss": 3.0471, | |
| "theoretical_loss": 4.226138183620867, | |
| "tokens_seen": 255590400 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.0006417108000756573, | |
| "loss": 3.0065, | |
| "theoretical_loss": 4.219667768240775, | |
| "tokens_seen": 258867200 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.0006408832986570834, | |
| "loss": 3.0265, | |
| "theoretical_loss": 4.213301349743924, | |
| "tokens_seen": 262144000 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.0006400557972385095, | |
| "loss": 2.978, | |
| "theoretical_loss": 4.20703598765197, | |
| "tokens_seen": 265420800 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.0006392282958199356, | |
| "loss": 2.9473, | |
| "theoretical_loss": 4.2008688596290025, | |
| "tokens_seen": 268697600 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.0006384007944013618, | |
| "loss": 2.9058, | |
| "theoretical_loss": 4.194797255362549, | |
| "tokens_seen": 271974400 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.000637573292982788, | |
| "loss": 2.9845, | |
| "theoretical_loss": 4.188818570830883, | |
| "tokens_seen": 275251200 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.0006367457915642141, | |
| "loss": 3.0175, | |
| "theoretical_loss": 4.182930302927963, | |
| "tokens_seen": 278528000 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.0006359182901456402, | |
| "loss": 2.9712, | |
| "theoretical_loss": 4.17713004441978, | |
| "tokens_seen": 281804800 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.0006350907887270663, | |
| "loss": 3.0091, | |
| "theoretical_loss": 4.1714154792080915, | |
| "tokens_seen": 285081600 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.0006342632873084925, | |
| "loss": 3.0192, | |
| "theoretical_loss": 4.165784377879517, | |
| "tokens_seen": 288358400 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.0006334357858899187, | |
| "loss": 2.9853, | |
| "theoretical_loss": 4.160234593519768, | |
| "tokens_seen": 291635200 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.0006326082844713448, | |
| "loss": 3.0073, | |
| "theoretical_loss": 4.15476405777444, | |
| "tokens_seen": 294912000 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.0006317807830527709, | |
| "loss": 2.9819, | |
| "theoretical_loss": 4.149370777139286, | |
| "tokens_seen": 298188800 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.0006309532816341972, | |
| "loss": 2.9752, | |
| "theoretical_loss": 4.144052829464249, | |
| "tokens_seen": 301465600 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.0006301257802156233, | |
| "loss": 2.9458, | |
| "theoretical_loss": 4.138808360656742, | |
| "tokens_seen": 304742400 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.0006292982787970494, | |
| "loss": 2.945, | |
| "theoretical_loss": 4.133635581570836, | |
| "tokens_seen": 308019200 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.0006284707773784755, | |
| "loss": 2.9216, | |
| "theoretical_loss": 4.128532765070004, | |
| "tokens_seen": 311296000 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.0006276432759599016, | |
| "loss": 2.9457, | |
| "theoretical_loss": 4.123498243252032, | |
| "tokens_seen": 314572800 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.0006268157745413279, | |
| "loss": 2.9304, | |
| "theoretical_loss": 4.118530404825556, | |
| "tokens_seen": 317849600 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.000625988273122754, | |
| "loss": 2.9513, | |
| "theoretical_loss": 4.113627692628464, | |
| "tokens_seen": 321126400 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.0006251607717041801, | |
| "loss": 2.9564, | |
| "theoretical_loss": 4.108788601279149, | |
| "tokens_seen": 324403200 | |
| }, | |
| { | |
| "debugging/Self-BLEU-5": 0.5365128506817183, | |
| "debugging/distinct-1-grams": 0.7612814402327299, | |
| "debugging/distinct-2-grams": 0.9694583753853511, | |
| "debugging/entropy-1-grams": 6.003629944255698, | |
| "debugging/entropy-2-grams": 7.054987089269872, | |
| "debugging/length": 495.25, | |
| "debugging/num_segments": 16, | |
| "epoch": 0.12, | |
| "objective/train/avg_token_score": 0.04385810345411301, | |
| "objective/train/avg_weight": 0.9601507186889648, | |
| "objective/train/docs_used": 197327, | |
| "objective/train/instantaneous_batch_size": 32, | |
| "objective/train/instantaneous_microbatch_size": 32768, | |
| "objective/train/original_loss": 2.9074478149414062, | |
| "objective/train/std_weight": 0.14257079362869263, | |
| "objective/train/theoretical_loss": 4.10401167495222, | |
| "objective/train/tokens_used": 348140000, | |
| "theoretical_loss": 4.10401167495222, | |
| "tokens_seen": 327680000 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.0006243332702856062, | |
| "loss": 2.986, | |
| "theoretical_loss": 4.10401167495222, | |
| "tokens_seen": 327680000 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.0006235057688670323, | |
| "loss": 2.9852, | |
| "theoretical_loss": 4.099295505270921, | |
| "tokens_seen": 330956800 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.0006226782674484584, | |
| "loss": 2.9316, | |
| "theoretical_loss": 4.094638729309031, | |
| "tokens_seen": 334233600 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.0006218507660298847, | |
| "loss": 2.9442, | |
| "theoretical_loss": 4.090040027695556, | |
| "tokens_seen": 337510400 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.0006210232646113108, | |
| "loss": 2.9001, | |
| "theoretical_loss": 4.085498122815992, | |
| "tokens_seen": 340787200 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.0006201957631927369, | |
| "loss": 2.9256, | |
| "theoretical_loss": 4.081011777104333, | |
| "tokens_seen": 344064000 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.000619368261774163, | |
| "loss": 2.9297, | |
| "theoretical_loss": 4.076579791420469, | |
| "tokens_seen": 347340800 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.0006185407603555891, | |
| "loss": 2.9439, | |
| "theoretical_loss": 4.0722010035079155, | |
| "tokens_seen": 350617600 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.0006177132589370153, | |
| "loss": 2.9351, | |
| "theoretical_loss": 4.067874286527197, | |
| "tokens_seen": 353894400 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.0006168857575184414, | |
| "loss": 2.9326, | |
| "theoretical_loss": 4.063598547660519, | |
| "tokens_seen": 357171200 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.0006160582560998676, | |
| "loss": 2.9614, | |
| "theoretical_loss": 4.05937272678363, | |
| "tokens_seen": 360448000 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.0006152307546812937, | |
| "loss": 2.9476, | |
| "theoretical_loss": 4.055195795201069, | |
| "tokens_seen": 363724800 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.0006144032532627198, | |
| "loss": 2.9058, | |
| "theoretical_loss": 4.051066754441235, | |
| "tokens_seen": 367001600 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.000613575751844146, | |
| "loss": 2.9081, | |
| "theoretical_loss": 4.04698463510794, | |
| "tokens_seen": 370278400 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.0006127482504255721, | |
| "loss": 2.9117, | |
| "theoretical_loss": 4.042948495785312, | |
| "tokens_seen": 373555200 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.0006119207490069983, | |
| "loss": 2.8717, | |
| "theoretical_loss": 4.038957421993153, | |
| "tokens_seen": 376832000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.0006110932475884244, | |
| "loss": 2.881, | |
| "theoretical_loss": 4.035010525189982, | |
| "tokens_seen": 380108800 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.0006102657461698505, | |
| "loss": 2.9145, | |
| "theoretical_loss": 4.031106941821218, | |
| "tokens_seen": 383385600 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.0006094382447512768, | |
| "loss": 2.877, | |
| "theoretical_loss": 4.027245832410079, | |
| "tokens_seen": 386662400 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.0006086107433327029, | |
| "loss": 2.8523, | |
| "theoretical_loss": 4.023426380688943, | |
| "tokens_seen": 389939200 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.000607783241914129, | |
| "loss": 2.894, | |
| "theoretical_loss": 4.019647792769048, | |
| "tokens_seen": 393216000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.0006069557404955551, | |
| "loss": 2.9554, | |
| "theoretical_loss": 4.015909296346521, | |
| "tokens_seen": 396492800 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.0006061282390769814, | |
| "loss": 2.8873, | |
| "theoretical_loss": 4.012210139942894, | |
| "tokens_seen": 399769600 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.0006053007376584075, | |
| "loss": 2.9151, | |
| "theoretical_loss": 4.008549592178291, | |
| "tokens_seen": 403046400 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0006044732362398336, | |
| "loss": 2.8973, | |
| "theoretical_loss": 4.004926941075674, | |
| "tokens_seen": 406323200 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0006036457348212597, | |
| "loss": 2.9519, | |
| "theoretical_loss": 4.001341493394558, | |
| "tokens_seen": 409600000 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0006028182334026858, | |
| "loss": 2.9595, | |
| "theoretical_loss": 3.997792573992726, | |
| "tokens_seen": 412876800 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0006019907319841119, | |
| "loss": 2.9127, | |
| "theoretical_loss": 3.994279525214554, | |
| "tokens_seen": 416153600 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0006011632305655381, | |
| "loss": 2.9624, | |
| "theoretical_loss": 3.990801706304647, | |
| "tokens_seen": 419430400 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0006003357291469643, | |
| "loss": 2.9186, | |
| "theoretical_loss": 3.987358492845532, | |
| "tokens_seen": 422707200 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0005995082277283904, | |
| "loss": 2.9497, | |
| "theoretical_loss": 3.9839492762182647, | |
| "tokens_seen": 425984000 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0005986807263098165, | |
| "loss": 2.9201, | |
| "theoretical_loss": 3.9805734630848306, | |
| "tokens_seen": 429260800 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0005978532248912426, | |
| "loss": 2.9038, | |
| "theoretical_loss": 3.9772304748913054, | |
| "tokens_seen": 432537600 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.0005970257234726688, | |
| "loss": 2.8974, | |
| "theoretical_loss": 3.973919747390801, | |
| "tokens_seen": 435814400 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.0005961982220540949, | |
| "loss": 2.9079, | |
| "theoretical_loss": 3.9706407301852487, | |
| "tokens_seen": 439091200 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.0005953707206355211, | |
| "loss": 2.8871, | |
| "theoretical_loss": 3.9673928862851655, | |
| "tokens_seen": 442368000 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.0005945432192169472, | |
| "loss": 2.8781, | |
| "theoretical_loss": 3.9641756916865463, | |
| "tokens_seen": 445644800 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.0005937157177983733, | |
| "loss": 2.8865, | |
| "theoretical_loss": 3.960988634964113, | |
| "tokens_seen": 448921600 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.0005928882163797995, | |
| "loss": 2.8596, | |
| "theoretical_loss": 3.9578312168801597, | |
| "tokens_seen": 452198400 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.0005920607149612256, | |
| "loss": 2.8591, | |
| "theoretical_loss": 3.954702950008308, | |
| "tokens_seen": 455475200 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.0005912332135426518, | |
| "loss": 2.8435, | |
| "theoretical_loss": 3.9516033583714734, | |
| "tokens_seen": 458752000 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.0005904057121240779, | |
| "loss": 2.8688, | |
| "theoretical_loss": 3.9485319770934355, | |
| "tokens_seen": 462028800 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.000589578210705504, | |
| "loss": 2.8546, | |
| "theoretical_loss": 3.945488352063391, | |
| "tokens_seen": 465305600 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.0005887507092869302, | |
| "loss": 2.8741, | |
| "theoretical_loss": 3.942472039612926, | |
| "tokens_seen": 468582400 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.0005879232078683564, | |
| "loss": 2.8452, | |
| "theoretical_loss": 3.939482606204863, | |
| "tokens_seen": 471859200 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.0005870957064497825, | |
| "loss": 2.8528, | |
| "theoretical_loss": 3.936519628133466, | |
| "tokens_seen": 475136000 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.0005862682050312086, | |
| "loss": 2.8866, | |
| "theoretical_loss": 3.9335826912355114, | |
| "tokens_seen": 478412800 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.0005854407036126347, | |
| "loss": 2.9264, | |
| "theoretical_loss": 3.93067139061177, | |
| "tokens_seen": 481689600 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.000584613202194061, | |
| "loss": 2.8535, | |
| "theoretical_loss": 3.927785330358441, | |
| "tokens_seen": 484966400 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.0005837857007754871, | |
| "loss": 2.8422, | |
| "theoretical_loss": 3.9249241233081333, | |
| "tokens_seen": 488243200 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "objective/train/avg_token_score": 0.004644907079637051, | |
| "objective/train/avg_weight": 0.9957804679870605, | |
| "objective/train/docs_used": 287192, | |
| "objective/train/instantaneous_batch_size": 32, | |
| "objective/train/instantaneous_microbatch_size": 32768, | |
| "objective/train/original_loss": 3.0019359588623047, | |
| "objective/train/std_weight": 0.02099473774433136, | |
| "objective/train/theoretical_loss": 3.92208739077998, | |
| "objective/train/tokens_used": 511980000, | |
| "theoretical_loss": 3.92208739077998, | |
| "tokens_seen": 491520000 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.0005829581993569132, | |
| "loss": 2.8177, | |
| "theoretical_loss": 3.92208739077998, | |
| "tokens_seen": 491520000 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.0005821306979383393, | |
| "loss": 2.829, | |
| "theoretical_loss": 3.919274762338519, | |
| "tokens_seen": 494796800 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.0005813031965197654, | |
| "loss": 2.8542, | |
| "theoretical_loss": 3.9164858755609613, | |
| "tokens_seen": 498073600 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.0005804756951011916, | |
| "loss": 2.8375, | |
| "theoretical_loss": 3.9137203758125176, | |
| "tokens_seen": 501350400 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.0005796481936826178, | |
| "loss": 2.7995, | |
| "theoretical_loss": 3.910977916029439, | |
| "tokens_seen": 504627200 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.0005788206922640439, | |
| "loss": 2.8108, | |
| "theoretical_loss": 3.908258156509472, | |
| "tokens_seen": 507904000 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.00057799319084547, | |
| "loss": 2.8003, | |
| "theoretical_loss": 3.905560764709417, | |
| "tokens_seen": 511180800 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.0005771656894268961, | |
| "loss": 2.7796, | |
| "theoretical_loss": 3.9028854150495143, | |
| "tokens_seen": 514457600 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.0005763381880083223, | |
| "loss": 2.8048, | |
| "theoretical_loss": 3.9002317887243834, | |
| "tokens_seen": 517734400 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.0005755106865897484, | |
| "loss": 2.803, | |
| "theoretical_loss": 3.897599573520247, | |
| "tokens_seen": 521011200 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.0005746831851711746, | |
| "loss": 2.834, | |
| "theoretical_loss": 3.8949884636382106, | |
| "tokens_seen": 524288000 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.0005738556837526007, | |
| "loss": 2.8202, | |
| "theoretical_loss": 3.892398159523345, | |
| "tokens_seen": 527564800 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.0005730281823340268, | |
| "loss": 2.8716, | |
| "theoretical_loss": 3.889828367699349, | |
| "tokens_seen": 530841600 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.000572200680915453, | |
| "loss": 2.8106, | |
| "theoretical_loss": 3.8872788006085894, | |
| "tokens_seen": 534118400 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.0005713731794968791, | |
| "loss": 2.8356, | |
| "theoretical_loss": 3.8847491764572926, | |
| "tokens_seen": 537395200 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.0005705456780783052, | |
| "loss": 2.8189, | |
| "theoretical_loss": 3.882239219065708, | |
| "tokens_seen": 540672000 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.0005697181766597314, | |
| "loss": 2.8447, | |
| "theoretical_loss": 3.879748657723039, | |
| "tokens_seen": 543948800 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0005688906752411575, | |
| "loss": 2.8729, | |
| "theoretical_loss": 3.8772772270469824, | |
| "tokens_seen": 547225600 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0005680631738225837, | |
| "loss": 2.8352, | |
| "theoretical_loss": 3.8748246668476827, | |
| "tokens_seen": 550502400 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0005672356724040099, | |
| "loss": 2.8532, | |
| "theoretical_loss": 3.8723907219959486, | |
| "tokens_seen": 553779200 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.000566408170985436, | |
| "loss": 2.8698, | |
| "theoretical_loss": 3.869975142295573, | |
| "tokens_seen": 557056000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0005655806695668621, | |
| "loss": 2.9208, | |
| "theoretical_loss": 3.8675776823595998, | |
| "tokens_seen": 560332800 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0005647531681482882, | |
| "loss": 2.889, | |
| "theoretical_loss": 3.8651981014904027, | |
| "tokens_seen": 563609600 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0005639422167580859, | |
| "loss": 2.8663, | |
| "theoretical_loss": 3.8628361635634265, | |
| "tokens_seen": 566886400 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.000563114715339512, | |
| "loss": 2.9028, | |
| "theoretical_loss": 3.8604916369144666, | |
| "tokens_seen": 570163200 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0005622872139209381, | |
| "loss": 2.8744, | |
| "theoretical_loss": 3.858164294230354, | |
| "tokens_seen": 573440000 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.0005614597125023642, | |
| "loss": 2.8465, | |
| "theoretical_loss": 3.85585391244293, | |
| "tokens_seen": 576716800 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.0005606322110837905, | |
| "loss": 2.8498, | |
| "theoretical_loss": 3.8535602726261864, | |
| "tokens_seen": 579993600 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.0005598047096652166, | |
| "loss": 2.8343, | |
| "theoretical_loss": 3.851283159896468, | |
| "tokens_seen": 583270400 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.0005589772082466427, | |
| "loss": 2.8178, | |
| "theoretical_loss": 3.8490223633156173, | |
| "tokens_seen": 586547200 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.0005581497068280688, | |
| "loss": 2.8002, | |
| "theoretical_loss": 3.846777675796974, | |
| "tokens_seen": 589824000 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.000557322205409495, | |
| "loss": 2.801, | |
| "theoretical_loss": 3.844548894014116, | |
| "tokens_seen": 593100800 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.0005564947039909212, | |
| "loss": 2.8109, | |
| "theoretical_loss": 3.8423358183122582, | |
| "tokens_seen": 596377600 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.0005556672025723473, | |
| "loss": 2.7858, | |
| "theoretical_loss": 3.840138252622208, | |
| "tokens_seen": 599654400 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0005548397011537734, | |
| "loss": 2.8172, | |
| "theoretical_loss": 3.837956004376799, | |
| "tokens_seen": 602931200 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0005540121997351996, | |
| "loss": 2.7818, | |
| "theoretical_loss": 3.8357888844297094, | |
| "tokens_seen": 606208000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0005531846983166257, | |
| "loss": 2.8023, | |
| "theoretical_loss": 3.8336367069765958, | |
| "tokens_seen": 609484800 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0005523571968980519, | |
| "loss": 2.755, | |
| "theoretical_loss": 3.8314992894784536, | |
| "tokens_seen": 612761600 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.000551529695479478, | |
| "loss": 2.7988, | |
| "theoretical_loss": 3.829376452587134, | |
| "tokens_seen": 616038400 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0005507021940609041, | |
| "loss": 2.8135, | |
| "theoretical_loss": 3.827268020072948, | |
| "tokens_seen": 619315200 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0005498746926423302, | |
| "loss": 2.8462, | |
| "theoretical_loss": 3.8251738187542843, | |
| "tokens_seen": 622592000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0005490471912237563, | |
| "loss": 2.8284, | |
| "theoretical_loss": 3.8230936784291787, | |
| "tokens_seen": 625868800 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0005482196898051826, | |
| "loss": 2.8606, | |
| "theoretical_loss": 3.8210274318087656, | |
| "tokens_seen": 629145600 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0005473921883866087, | |
| "loss": 2.8068, | |
| "theoretical_loss": 3.818974914452557, | |
| "tokens_seen": 632422400 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0005465646869680348, | |
| "loss": 2.8042, | |
| "theoretical_loss": 3.8169359647054835, | |
| "tokens_seen": 635699200 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0005457371855494609, | |
| "loss": 2.7892, | |
| "theoretical_loss": 3.8149104236366433, | |
| "tokens_seen": 638976000 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.000544909684130887, | |
| "loss": 2.7712, | |
| "theoretical_loss": 3.8128981349797098, | |
| "tokens_seen": 642252800 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0005440821827123131, | |
| "loss": 2.7953, | |
| "theoretical_loss": 3.8108989450749293, | |
| "tokens_seen": 645529600 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0005432546812937394, | |
| "loss": 2.7863, | |
| "theoretical_loss": 3.8089127028126764, | |
| "tokens_seen": 648806400 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0005424271798751655, | |
| "loss": 2.8319, | |
| "theoretical_loss": 3.8069392595785083, | |
| "tokens_seen": 652083200 | |
| }, | |
| { | |
| "debugging/Self-BLEU-5": 0.5265375629586004, | |
| "debugging/distinct-1-grams": 0.7435820408094715, | |
| "debugging/distinct-2-grams": 0.9558103821233092, | |
| "debugging/entropy-1-grams": 5.931434510687563, | |
| "debugging/entropy-2-grams": 6.886416755326388, | |
| "debugging/length": 521.9230769230769, | |
| "debugging/num_segments": 13, | |
| "epoch": 0.23, | |
| "objective/train/avg_token_score": 0.022742915898561478, | |
| "objective/train/avg_weight": 0.9793195724487305, | |
| "objective/train/docs_used": 379091, | |
| "objective/train/instantaneous_batch_size": 32, | |
| "objective/train/instantaneous_microbatch_size": 32768, | |
| "objective/train/original_loss": 3.077260732650757, | |
| "objective/train/std_weight": 0.07130531966686249, | |
| "objective/train/theoretical_loss": 3.804978469199669, | |
| "objective/train/tokens_used": 675820000, | |
| "theoretical_loss": 3.804978469199669, | |
| "tokens_seen": 655360000 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0005416162284849631, | |
| "loss": 2.8518, | |
| "theoretical_loss": 3.804978469199669, | |
| "tokens_seen": 655360000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.0005407887270663892, | |
| "loss": 2.8406, | |
| "theoretical_loss": 3.803030187893005, | |
| "tokens_seen": 658636800 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.0005399612256478154, | |
| "loss": 2.8343, | |
| "theoretical_loss": 3.8010942742142415, | |
| "tokens_seen": 661913600 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.0005391337242292415, | |
| "loss": 2.8597, | |
| "theoretical_loss": 3.799170589008585, | |
| "tokens_seen": 665190400 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.0005383062228106677, | |
| "loss": 2.8447, | |
| "theoretical_loss": 3.7972589953626006, | |
| "tokens_seen": 668467200 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.0005374787213920938, | |
| "loss": 2.8528, | |
| "theoretical_loss": 3.795359358557337, | |
| "tokens_seen": 671744000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.0005366512199735199, | |
| "loss": 2.7855, | |
| "theoretical_loss": 3.79347154602265, | |
| "tokens_seen": 675020800 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.0005358237185549461, | |
| "loss": 2.8356, | |
| "theoretical_loss": 3.7915954272926955, | |
| "tokens_seen": 678297600 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.0005349962171363722, | |
| "loss": 2.7915, | |
| "theoretical_loss": 3.789730873962557, | |
| "tokens_seen": 681574400 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.0005341687157177983, | |
| "loss": 2.7291, | |
| "theoretical_loss": 3.787877759645963, | |
| "tokens_seen": 684851200 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.0005333412142992244, | |
| "loss": 2.7625, | |
| "theoretical_loss": 3.7860359599340776, | |
| "tokens_seen": 688128000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.0005325137128806506, | |
| "loss": 2.8011, | |
| "theoretical_loss": 3.784205352355321, | |
| "tokens_seen": 691404800 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.0005316862114620768, | |
| "loss": 2.83, | |
| "theoretical_loss": 3.782385816336189, | |
| "tokens_seen": 694681600 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.0005308587100435029, | |
| "loss": 2.7875, | |
| "theoretical_loss": 3.7805772331630516, | |
| "tokens_seen": 697958400 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.0005300312086249291, | |
| "loss": 2.7623, | |
| "theoretical_loss": 3.7787794859448898, | |
| "tokens_seen": 701235200 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.0005292037072063552, | |
| "loss": 2.7454, | |
| "theoretical_loss": 3.7769924595769546, | |
| "tokens_seen": 704512000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.0005283762057877813, | |
| "loss": 2.7341, | |
| "theoretical_loss": 3.7752160407053115, | |
| "tokens_seen": 707788800 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.0005275487043692075, | |
| "loss": 2.7184, | |
| "theoretical_loss": 3.7734501176922493, | |
| "tokens_seen": 711065600 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.0005267212029506337, | |
| "loss": 2.7617, | |
| "theoretical_loss": 3.7716945805825337, | |
| "tokens_seen": 714342400 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.0005258937015320598, | |
| "loss": 2.799, | |
| "theoretical_loss": 3.7699493210704667, | |
| "tokens_seen": 717619200 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.0005250662001134859, | |
| "loss": 2.775, | |
| "theoretical_loss": 3.7682142324677455, | |
| "tokens_seen": 720896000 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.000524238698694912, | |
| "loss": 2.8333, | |
| "theoretical_loss": 3.7664892096720886, | |
| "tokens_seen": 724172800 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.0005234111972763382, | |
| "loss": 2.7697, | |
| "theoretical_loss": 3.7647741491366067, | |
| "tokens_seen": 727449600 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.0005225836958577644, | |
| "loss": 2.7844, | |
| "theoretical_loss": 3.7630689488399027, | |
| "tokens_seen": 730726400 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.0005217561944391905, | |
| "loss": 2.7504, | |
| "theoretical_loss": 3.7613735082568764, | |
| "tokens_seen": 734003200 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.0005209286930206166, | |
| "loss": 2.7869, | |
| "theoretical_loss": 3.759687728330217, | |
| "tokens_seen": 737280000 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.0005201011916020427, | |
| "loss": 2.7799, | |
| "theoretical_loss": 3.75801151144256, | |
| "tokens_seen": 740556800 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.0005192736901834689, | |
| "loss": 2.7636, | |
| "theoretical_loss": 3.756344761389295, | |
| "tokens_seen": 743833600 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.000518446188764895, | |
| "loss": 2.7358, | |
| "theoretical_loss": 3.754687383352003, | |
| "tokens_seen": 747110400 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.0005176186873463211, | |
| "loss": 2.7389, | |
| "theoretical_loss": 3.7530392838725097, | |
| "tokens_seen": 750387200 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.0005167911859277473, | |
| "loss": 2.7568, | |
| "theoretical_loss": 3.751400370827529, | |
| "tokens_seen": 753664000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.0005159636845091734, | |
| "loss": 2.7797, | |
| "theoretical_loss": 3.749770553403895, | |
| "tokens_seen": 756940800 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.0005151361830905996, | |
| "loss": 2.7431, | |
| "theoretical_loss": 3.748149742074355, | |
| "tokens_seen": 760217600 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.0005143086816720257, | |
| "loss": 2.7718, | |
| "theoretical_loss": 3.746537848573908, | |
| "tokens_seen": 763494400 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.0005134811802534518, | |
| "loss": 2.7869, | |
| "theoretical_loss": 3.744934785876686, | |
| "tokens_seen": 766771200 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.0005126536788348779, | |
| "loss": 2.7932, | |
| "theoretical_loss": 3.7433404681733475, | |
| "tokens_seen": 770048000 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.0005118261774163041, | |
| "loss": 2.7999, | |
| "theoretical_loss": 3.7417548108489846, | |
| "tokens_seen": 773324800 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.0005109986759977303, | |
| "loss": 2.7733, | |
| "theoretical_loss": 3.740177730461517, | |
| "tokens_seen": 776601600 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.0005101711745791564, | |
| "loss": 2.822, | |
| "theoretical_loss": 3.73860914472057, | |
| "tokens_seen": 779878400 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.0005093436731605825, | |
| "loss": 2.7869, | |
| "theoretical_loss": 3.7370489724668197, | |
| "tokens_seen": 783155200 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.0005085161717420087, | |
| "loss": 2.7992, | |
| "theoretical_loss": 3.735497133651788, | |
| "tokens_seen": 786432000 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.0005076886703234348, | |
| "loss": 2.7876, | |
| "theoretical_loss": 3.733953549318091, | |
| "tokens_seen": 789708800 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.000506861168904861, | |
| "loss": 2.8014, | |
| "theoretical_loss": 3.7324181415801094, | |
| "tokens_seen": 792985600 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.0005060336674862872, | |
| "loss": 2.7963, | |
| "theoretical_loss": 3.7308908336050814, | |
| "tokens_seen": 796262400 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0005052061660677133, | |
| "loss": 2.7868, | |
| "theoretical_loss": 3.729371549594614, | |
| "tokens_seen": 799539200 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0005043786646491394, | |
| "loss": 2.8003, | |
| "theoretical_loss": 3.7278602147665776, | |
| "tokens_seen": 802816000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0005035511632305655, | |
| "loss": 2.7947, | |
| "theoretical_loss": 3.726356755337407, | |
| "tokens_seen": 806092800 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0005027236618119917, | |
| "loss": 2.7845, | |
| "theoretical_loss": 3.724861098504767, | |
| "tokens_seen": 809369600 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0005018961603934179, | |
| "loss": 2.7496, | |
| "theoretical_loss": 3.7233731724305974, | |
| "tokens_seen": 812646400 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.000501068658974844, | |
| "loss": 2.7717, | |
| "theoretical_loss": 3.7218929062245105, | |
| "tokens_seen": 815923200 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "objective/train/avg_token_score": 0.009068925864994526, | |
| "objective/train/avg_weight": 0.9917506575584412, | |
| "objective/train/docs_used": 471128, | |
| "objective/train/instantaneous_batch_size": 32, | |
| "objective/train/instantaneous_microbatch_size": 32768, | |
| "objective/train/original_loss": 2.670485496520996, | |
| "objective/train/std_weight": 0.05290398746728897, | |
| "objective/train/theoretical_loss": 3.7204202299275475, | |
| "objective/train/tokens_used": 839660000, | |
| "theoretical_loss": 3.7204202299275475, | |
| "tokens_seen": 819200000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0005002411575562701, | |
| "loss": 2.7158, | |
| "theoretical_loss": 3.7204202299275475, | |
| "tokens_seen": 819200000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0004994136561376962, | |
| "loss": 2.7397, | |
| "theoretical_loss": 3.7189550744962707, | |
| "tokens_seen": 822476800 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0004985861547191224, | |
| "loss": 2.7188, | |
| "theoretical_loss": 3.717497371787192, | |
| "tokens_seen": 825753600 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.0004977586533005485, | |
| "loss": 2.7159, | |
| "theoretical_loss": 3.7160470545415274, | |
| "tokens_seen": 829030400 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.0004969311518819746, | |
| "loss": 2.7166, | |
| "theoretical_loss": 3.714604056370267, | |
| "tokens_seen": 832307200 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.0004961036504634008, | |
| "loss": 2.726, | |
| "theoretical_loss": 3.713168311739558, | |
| "tokens_seen": 835584000 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.0004952761490448269, | |
| "loss": 2.7102, | |
| "theoretical_loss": 3.7117397559563843, | |
| "tokens_seen": 838860800 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.0004944486476262531, | |
| "loss": 2.6939, | |
| "theoretical_loss": 3.710318325154545, | |
| "tokens_seen": 842137600 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.0004936211462076792, | |
| "loss": 2.7416, | |
| "theoretical_loss": 3.7089039562809223, | |
| "tokens_seen": 845414400 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.0004927936447891053, | |
| "loss": 2.7152, | |
| "theoretical_loss": 3.7074965870820193, | |
| "tokens_seen": 848691200 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.0004919661433705314, | |
| "loss": 2.7146, | |
| "theoretical_loss": 3.7060961560907857, | |
| "tokens_seen": 851968000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0004911386419519576, | |
| "loss": 2.7725, | |
| "theoretical_loss": 3.7047026026137, | |
| "tokens_seen": 855244800 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0004903111405333838, | |
| "loss": 2.7333, | |
| "theoretical_loss": 3.7033158667181154, | |
| "tokens_seen": 858521600 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0004894836391148099, | |
| "loss": 2.7624, | |
| "theoretical_loss": 3.701935889219863, | |
| "tokens_seen": 861798400 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.000488656137696236, | |
| "loss": 2.7316, | |
| "theoretical_loss": 3.7005626116710966, | |
| "tokens_seen": 865075200 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.00048782863627766217, | |
| "loss": 2.7403, | |
| "theoretical_loss": 3.69919597634839, | |
| "tokens_seen": 868352000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0004870011348590883, | |
| "loss": 2.719, | |
| "theoretical_loss": 3.6978359262410603, | |
| "tokens_seen": 871628800 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0004861736334405145, | |
| "loss": 2.7432, | |
| "theoretical_loss": 3.6964824050397276, | |
| "tokens_seen": 874905600 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.00048534613202194063, | |
| "loss": 2.6861, | |
| "theoretical_loss": 3.6951353571251015, | |
| "tokens_seen": 878182400 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.00048451863060336675, | |
| "loss": 2.6866, | |
| "theoretical_loss": 3.693794727556988, | |
| "tokens_seen": 881459200 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.00048369112918479287, | |
| "loss": 2.6974, | |
| "theoretical_loss": 3.692460462063506, | |
| "tokens_seen": 884736000 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.000482863627766219, | |
| "loss": 2.7507, | |
| "theoretical_loss": 3.691132507030521, | |
| "tokens_seen": 888012800 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.00048203612634764516, | |
| "loss": 2.7619, | |
| "theoretical_loss": 3.6898108094912816, | |
| "tokens_seen": 891289600 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.0004812086249290713, | |
| "loss": 2.719, | |
| "theoretical_loss": 3.6884953171162556, | |
| "tokens_seen": 894566400 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.00048038112351049745, | |
| "loss": 2.7252, | |
| "theoretical_loss": 3.6871859782031624, | |
| "tokens_seen": 897843200 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.0004795536220919236, | |
| "loss": 2.7355, | |
| "theoretical_loss": 3.685882741667202, | |
| "tokens_seen": 901120000 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.00047872612067334974, | |
| "loss": 2.7378, | |
| "theoretical_loss": 3.684585557031461, | |
| "tokens_seen": 904396800 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.00047789861925477586, | |
| "loss": 2.7513, | |
| "theoretical_loss": 3.6832943744175126, | |
| "tokens_seen": 907673600 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.000477071117836202, | |
| "loss": 2.7248, | |
| "theoretical_loss": 3.682009144536188, | |
| "tokens_seen": 910950400 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.0004762436164176281, | |
| "loss": 2.7513, | |
| "theoretical_loss": 3.680729818678526, | |
| "tokens_seen": 914227200 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.0004754161149990543, | |
| "loss": 2.7298, | |
| "theoretical_loss": 3.6794563487068936, | |
| "tokens_seen": 917504000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.00047458861358048044, | |
| "loss": 2.7376, | |
| "theoretical_loss": 3.6781886870462692, | |
| "tokens_seen": 920780800 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.00047376111216190656, | |
| "loss": 2.7544, | |
| "theoretical_loss": 3.676926786675698, | |
| "tokens_seen": 924057600 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.0004729336107433327, | |
| "loss": 2.6955, | |
| "theoretical_loss": 3.6756706011198963, | |
| "tokens_seen": 927334400 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.0004721061093247588, | |
| "loss": 2.6633, | |
| "theoretical_loss": 3.6744200844410217, | |
| "tokens_seen": 930611200 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.00047127860790618497, | |
| "loss": 2.6639, | |
| "theoretical_loss": 3.6731751912305914, | |
| "tokens_seen": 933888000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.00047045110648761114, | |
| "loss": 2.6981, | |
| "theoretical_loss": 3.671935876601547, | |
| "tokens_seen": 937164800 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.00046962360506903726, | |
| "loss": 2.6376, | |
| "theoretical_loss": 3.6707020961804715, | |
| "tokens_seen": 940441600 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.00046879610365046343, | |
| "loss": 2.6905, | |
| "theoretical_loss": 3.6694738060999468, | |
| "tokens_seen": 943718400 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.00046796860223188955, | |
| "loss": 2.7087, | |
| "theoretical_loss": 3.668250962991049, | |
| "tokens_seen": 946995200 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.00046714110081331567, | |
| "loss": 2.7094, | |
| "theoretical_loss": 3.667033523975983, | |
| "tokens_seen": 950272000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.0004663135993947418, | |
| "loss": 2.6905, | |
| "theoretical_loss": 3.66582144666085, | |
| "tokens_seen": 953548800 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.0004654860979761679, | |
| "loss": 2.6958, | |
| "theoretical_loss": 3.664614689128546, | |
| "tokens_seen": 956825600 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.00046465859655759413, | |
| "loss": 2.6934, | |
| "theoretical_loss": 3.6634132099317886, | |
| "tokens_seen": 960102400 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.00046383109513902025, | |
| "loss": 2.7015, | |
| "theoretical_loss": 3.662216968086267, | |
| "tokens_seen": 963379200 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.00046300359372044637, | |
| "loss": 2.731, | |
| "theoretical_loss": 3.6610259230639217, | |
| "tokens_seen": 966656000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.0004621760923018725, | |
| "loss": 2.7641, | |
| "theoretical_loss": 3.659840034786333, | |
| "tokens_seen": 969932800 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.00046134859088329866, | |
| "loss": 2.8244, | |
| "theoretical_loss": 3.6586592636182376, | |
| "tokens_seen": 973209600 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.0004605210894647248, | |
| "loss": 2.8167, | |
| "theoretical_loss": 3.6574835703611566, | |
| "tokens_seen": 976486400 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.00045969358804615095, | |
| "loss": 2.8036, | |
| "theoretical_loss": 3.6563129162471313, | |
| "tokens_seen": 979763200 | |
| }, | |
| { | |
| "debugging/Self-BLEU-5": 0.4286046663919377, | |
| "debugging/distinct-1-grams": 0.8147567798871364, | |
| "debugging/distinct-2-grams": 0.9823269374342457, | |
| "debugging/entropy-1-grams": 6.1671920556004824, | |
| "debugging/entropy-2-grams": 6.947028138756313, | |
| "debugging/length": 477.53333333333336, | |
| "debugging/num_segments": 15, | |
| "epoch": 0.35, | |
| "objective/train/avg_token_score": 0.020611366257071495, | |
| "objective/train/avg_weight": 0.9812498092651367, | |
| "objective/train/docs_used": 560408, | |
| "objective/train/instantaneous_batch_size": 32, | |
| "objective/train/instantaneous_microbatch_size": 32768, | |
| "objective/train/original_loss": 2.92838716506958, | |
| "objective/train/std_weight": 0.0772768035531044, | |
| "objective/train/theoretical_loss": 3.6551472629325787, | |
| "objective/train/tokens_used": 1003500000, | |
| "theoretical_loss": 3.6551472629325787, | |
| "tokens_seen": 983040000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.00045886608662757707, | |
| "loss": 2.8015, | |
| "theoretical_loss": 3.6551472629325787, | |
| "tokens_seen": 983040000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.00045803858520900324, | |
| "loss": 2.8028, | |
| "theoretical_loss": 3.653986572492247, | |
| "tokens_seen": 986316800 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.00045721108379042936, | |
| "loss": 2.771, | |
| "theoretical_loss": 3.65283080741328, | |
| "tokens_seen": 989593600 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.0004563835823718555, | |
| "loss": 2.7847, | |
| "theoretical_loss": 3.6516799305893866, | |
| "tokens_seen": 992870400 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0004555560809532816, | |
| "loss": 2.7772, | |
| "theoretical_loss": 3.6505339053151076, | |
| "tokens_seen": 996147200 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0004547285795347078, | |
| "loss": 2.7861, | |
| "theoretical_loss": 3.649392695280186, | |
| "tokens_seen": 999424000 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.00045390107811613394, | |
| "loss": 2.7509, | |
| "theoretical_loss": 3.6482562645640337, | |
| "tokens_seen": 1002700800 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.00045307357669756006, | |
| "loss": 2.7548, | |
| "theoretical_loss": 3.6471245776302883, | |
| "tokens_seen": 1005977600 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0004522460752789862, | |
| "loss": 2.7905, | |
| "theoretical_loss": 3.6459975993214724, | |
| "tokens_seen": 1009254400 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0004514185738604123, | |
| "loss": 2.7515, | |
| "theoretical_loss": 3.6448752948537377, | |
| "tokens_seen": 1012531200 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.00045059107244183847, | |
| "loss": 2.7866, | |
| "theoretical_loss": 3.6437576298116996, | |
| "tokens_seen": 1015808000 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0004497635710232646, | |
| "loss": 2.7797, | |
| "theoretical_loss": 3.6426445701433607, | |
| "tokens_seen": 1019084800 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.00044893606960469076, | |
| "loss": 2.8016, | |
| "theoretical_loss": 3.6415360821551226, | |
| "tokens_seen": 1022361600 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.0004481085681861169, | |
| "loss": 2.7548, | |
| "theoretical_loss": 3.6404321325068754, | |
| "tokens_seen": 1025638400 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.00044728106676754305, | |
| "loss": 2.7788, | |
| "theoretical_loss": 3.639332688207178, | |
| "tokens_seen": 1028915200 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.00044645356534896917, | |
| "loss": 2.7593, | |
| "theoretical_loss": 3.6382377166085096, | |
| "tokens_seen": 1032192000 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.0004456260639303953, | |
| "loss": 2.7166, | |
| "theoretical_loss": 3.6371471854026147, | |
| "tokens_seen": 1035468800 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.0004447985625118214, | |
| "loss": 2.7249, | |
| "theoretical_loss": 3.6360610626159087, | |
| "tokens_seen": 1038745600 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.00044397106109324763, | |
| "loss": 2.6919, | |
| "theoretical_loss": 3.634979316604973, | |
| "tokens_seen": 1042022400 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.00044314355967467375, | |
| "loss": 2.7069, | |
| "theoretical_loss": 3.6339019160521198, | |
| "tokens_seen": 1045299200 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.00044231605825609987, | |
| "loss": 2.6991, | |
| "theoretical_loss": 3.632828829961029, | |
| "tokens_seen": 1048576000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.000441488556837526, | |
| "loss": 2.7159, | |
| "theoretical_loss": 3.631760027652461, | |
| "tokens_seen": 1051852800 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.0004406610554189521, | |
| "loss": 2.7444, | |
| "theoretical_loss": 3.630695478760034, | |
| "tokens_seen": 1055129600 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.0004398335540003783, | |
| "loss": 2.718, | |
| "theoretical_loss": 3.6296351532260767, | |
| "tokens_seen": 1058406400 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.00043900605258180445, | |
| "loss": 2.7124, | |
| "theoretical_loss": 3.6285790212975435, | |
| "tokens_seen": 1061683200 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.00043817855116323057, | |
| "loss": 2.6883, | |
| "theoretical_loss": 3.6275270535220008, | |
| "tokens_seen": 1064960000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.00043735104974465674, | |
| "loss": 2.7127, | |
| "theoretical_loss": 3.626479220743673, | |
| "tokens_seen": 1068236800 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.00043652354832608286, | |
| "loss": 2.7173, | |
| "theoretical_loss": 3.6254354940995586, | |
| "tokens_seen": 1071513600 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.000435696046907509, | |
| "loss": 2.7093, | |
| "theoretical_loss": 3.624395845015602, | |
| "tokens_seen": 1074790400 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.0004348685454889351, | |
| "loss": 2.6681, | |
| "theoretical_loss": 3.6233602452029348, | |
| "tokens_seen": 1078067200 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.0004340410440703612, | |
| "loss": 2.6861, | |
| "theoretical_loss": 3.6223286666541683, | |
| "tokens_seen": 1081344000 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.00043321354265178744, | |
| "loss": 2.7143, | |
| "theoretical_loss": 3.621301081639753, | |
| "tokens_seen": 1084620800 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.00043238604123321356, | |
| "loss": 2.7513, | |
| "theoretical_loss": 3.6202774627043923, | |
| "tokens_seen": 1087897600 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.0004315585398146397, | |
| "loss": 2.7077, | |
| "theoretical_loss": 3.619257782663513, | |
| "tokens_seen": 1091174400 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.0004307310383960658, | |
| "loss": 2.7183, | |
| "theoretical_loss": 3.618242014599793, | |
| "tokens_seen": 1094451200 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.0004299035369774919, | |
| "loss": 2.7047, | |
| "theoretical_loss": 3.617230131859743, | |
| "tokens_seen": 1097728000 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.0004290760355589181, | |
| "loss": 2.6888, | |
| "theoretical_loss": 3.6162221080503416, | |
| "tokens_seen": 1101004800 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.00042824853414034426, | |
| "loss": 2.6753, | |
| "theoretical_loss": 3.615217917035726, | |
| "tokens_seen": 1104281600 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0004274210327217704, | |
| "loss": 2.7061, | |
| "theoretical_loss": 3.614217532933929, | |
| "tokens_seen": 1107558400 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.00042659353130319655, | |
| "loss": 2.694, | |
| "theoretical_loss": 3.6132209301136715, | |
| "tokens_seen": 1110835200 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.00042576602988462267, | |
| "loss": 2.6781, | |
| "theoretical_loss": 3.612228083191205, | |
| "tokens_seen": 1114112000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0004249385284660488, | |
| "loss": 2.737, | |
| "theoretical_loss": 3.611238967027199, | |
| "tokens_seen": 1117388800 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0004241110270474749, | |
| "loss": 2.7779, | |
| "theoretical_loss": 3.610253556723679, | |
| "tokens_seen": 1120665600 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.00042328352562890113, | |
| "loss": 2.7414, | |
| "theoretical_loss": 3.609271827621014, | |
| "tokens_seen": 1123942400 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.00042245602421032725, | |
| "loss": 2.7457, | |
| "theoretical_loss": 3.6082937552949463, | |
| "tokens_seen": 1127219200 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.00042162852279175337, | |
| "loss": 2.7128, | |
| "theoretical_loss": 3.607319315553669, | |
| "tokens_seen": 1130496000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0004208010213731795, | |
| "loss": 2.6753, | |
| "theoretical_loss": 3.6063484844349456, | |
| "tokens_seen": 1133772800 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.0004199735199546056, | |
| "loss": 2.7007, | |
| "theoretical_loss": 3.605381238203279, | |
| "tokens_seen": 1137049600 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.0004191460185360317, | |
| "loss": 2.7095, | |
| "theoretical_loss": 3.604417553347117, | |
| "tokens_seen": 1140326400 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.0004183185171174579, | |
| "loss": 2.7005, | |
| "theoretical_loss": 3.603457406576106, | |
| "tokens_seen": 1143603200 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "objective/train/avg_token_score": 0.022877871990203857, | |
| "objective/train/avg_weight": 0.9792008399963379, | |
| "objective/train/docs_used": 649861, | |
| "objective/train/instantaneous_batch_size": 32, | |
| "objective/train/instantaneous_microbatch_size": 32768, | |
| "objective/train/original_loss": 2.647454261779785, | |
| "objective/train/std_weight": 0.08679035305976868, | |
| "objective/train/theoretical_loss": 3.602500774818379, | |
| "objective/train/tokens_used": 1167340000, | |
| "theoretical_loss": 3.602500774818379, | |
| "tokens_seen": 1146880000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.00041750756572725553, | |
| "loss": 2.6892, | |
| "theoretical_loss": 3.602500774818379, | |
| "tokens_seen": 1146880000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.0004166800643086817, | |
| "loss": 2.7282, | |
| "theoretical_loss": 3.601547635217892, | |
| "tokens_seen": 1150156800 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.0004158525628901078, | |
| "loss": 2.729, | |
| "theoretical_loss": 3.6005979651317976, | |
| "tokens_seen": 1153433600 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.00041502506147153394, | |
| "loss": 2.7375, | |
| "theoretical_loss": 3.599651742127855, | |
| "tokens_seen": 1156710400 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.00041419756005296005, | |
| "loss": 2.6894, | |
| "theoretical_loss": 3.5987089439818805, | |
| "tokens_seen": 1159987200 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.00041337005863438617, | |
| "loss": 2.6691, | |
| "theoretical_loss": 3.5977695486752426, | |
| "tokens_seen": 1163264000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.0004125425572158124, | |
| "loss": 2.7023, | |
| "theoretical_loss": 3.596833534392379, | |
| "tokens_seen": 1166540800 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.0004117150557972385, | |
| "loss": 2.72, | |
| "theoretical_loss": 3.595900879518368, | |
| "tokens_seen": 1169817600 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.00041088755437866464, | |
| "loss": 2.7353, | |
| "theoretical_loss": 3.594971562636521, | |
| "tokens_seen": 1173094400 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.00041006005296009075, | |
| "loss": 2.6902, | |
| "theoretical_loss": 3.5940455625260226, | |
| "tokens_seen": 1176371200 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.0004092325515415169, | |
| "loss": 2.7018, | |
| "theoretical_loss": 3.5931228581595938, | |
| "tokens_seen": 1179648000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.00040840505012294305, | |
| "loss": 2.7002, | |
| "theoretical_loss": 3.5922034287011995, | |
| "tokens_seen": 1182924800 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.0004075775487043692, | |
| "loss": 2.6656, | |
| "theoretical_loss": 3.5912872535037828, | |
| "tokens_seen": 1186201600 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.0004067500472857954, | |
| "loss": 2.7475, | |
| "theoretical_loss": 3.590374312107035, | |
| "tokens_seen": 1189478400 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.0004059225458672215, | |
| "loss": 2.733, | |
| "theoretical_loss": 3.5894645842351993, | |
| "tokens_seen": 1192755200 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.0004050950444486476, | |
| "loss": 2.7186, | |
| "theoretical_loss": 3.588558049794902, | |
| "tokens_seen": 1196032000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.00040426754303007374, | |
| "loss": 2.698, | |
| "theoretical_loss": 3.5876546888730187, | |
| "tokens_seen": 1199308800 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.00040344004161149986, | |
| "loss": 2.7175, | |
| "theoretical_loss": 3.5867544817345713, | |
| "tokens_seen": 1202585600 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.0004026125401929261, | |
| "loss": 2.7441, | |
| "theoretical_loss": 3.585857408820652, | |
| "tokens_seen": 1205862400 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.0004017850387743522, | |
| "loss": 2.751, | |
| "theoretical_loss": 3.58496345074638, | |
| "tokens_seen": 1209139200 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.0004009575373557783, | |
| "loss": 2.7643, | |
| "theoretical_loss": 3.5840725882988873, | |
| "tokens_seen": 1212416000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.00040013003593720444, | |
| "loss": 2.7308, | |
| "theoretical_loss": 3.5831848024353317, | |
| "tokens_seen": 1215692800 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.00039930253451863056, | |
| "loss": 2.756, | |
| "theoretical_loss": 3.5823000742809374, | |
| "tokens_seen": 1218969600 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.00039847503310005674, | |
| "loss": 2.7264, | |
| "theoretical_loss": 3.5814183851270673, | |
| "tokens_seen": 1222246400 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.00039764753168148285, | |
| "loss": 2.7167, | |
| "theoretical_loss": 3.5805397164293167, | |
| "tokens_seen": 1225523200 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.000396820030262909, | |
| "loss": 2.6978, | |
| "theoretical_loss": 3.5796640498056407, | |
| "tokens_seen": 1228800000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0003959925288443352, | |
| "loss": 2.7057, | |
| "theoretical_loss": 3.5787913670345013, | |
| "tokens_seen": 1232076800 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0003951650274257613, | |
| "loss": 2.6838, | |
| "theoretical_loss": 3.577921650053045, | |
| "tokens_seen": 1235353600 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.00039433752600718743, | |
| "loss": 2.6807, | |
| "theoretical_loss": 3.577054880955303, | |
| "tokens_seen": 1238630400 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.00039351002458861355, | |
| "loss": 2.6645, | |
| "theoretical_loss": 3.5761910419904193, | |
| "tokens_seen": 1241907200 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.00039268252317003967, | |
| "loss": 2.7265, | |
| "theoretical_loss": 3.5753301155609014, | |
| "tokens_seen": 1245184000 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.0003918550217514659, | |
| "loss": 2.6666, | |
| "theoretical_loss": 3.574472084220896, | |
| "tokens_seen": 1248460800 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.000391027520332892, | |
| "loss": 2.656, | |
| "theoretical_loss": 3.5736169306744885, | |
| "tokens_seen": 1251737600 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.00039020001891431813, | |
| "loss": 2.6265, | |
| "theoretical_loss": 3.572764637774024, | |
| "tokens_seen": 1255014400 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.00038937251749574425, | |
| "loss": 2.6361, | |
| "theoretical_loss": 3.571915188518457, | |
| "tokens_seen": 1258291200 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.00038854501607717037, | |
| "loss": 2.6557, | |
| "theoretical_loss": 3.571068566051716, | |
| "tokens_seen": 1261568000 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.00038771751465859654, | |
| "loss": 2.6973, | |
| "theoretical_loss": 3.5702247536610976, | |
| "tokens_seen": 1264844800 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.0003868900132400227, | |
| "loss": 2.6764, | |
| "theoretical_loss": 3.5693837347756783, | |
| "tokens_seen": 1268121600 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.00038606251182144883, | |
| "loss": 2.6491, | |
| "theoretical_loss": 3.5685454929647475, | |
| "tokens_seen": 1271398400 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.000385235010402875, | |
| "loss": 2.6612, | |
| "theoretical_loss": 3.5677100119362675, | |
| "tokens_seen": 1274675200 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.0003844075089843011, | |
| "loss": 2.6389, | |
| "theoretical_loss": 3.566877275535345, | |
| "tokens_seen": 1277952000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.00038358000756572724, | |
| "loss": 2.6385, | |
| "theoretical_loss": 3.566047267742733, | |
| "tokens_seen": 1281228800 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.00038275250614715336, | |
| "loss": 2.6157, | |
| "theoretical_loss": 3.5652199726733453, | |
| "tokens_seen": 1284505600 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.0003819250047285795, | |
| "loss": 2.6116, | |
| "theoretical_loss": 3.564395374574796, | |
| "tokens_seen": 1287782400 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.0003810975033100057, | |
| "loss": 2.6711, | |
| "theoretical_loss": 3.5635734578259557, | |
| "tokens_seen": 1291059200 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.0003802700018914318, | |
| "loss": 2.6847, | |
| "theoretical_loss": 3.5627542069355282, | |
| "tokens_seen": 1294336000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.00037944250047285794, | |
| "loss": 2.6503, | |
| "theoretical_loss": 3.5619376065406474, | |
| "tokens_seen": 1297612800 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.00037861499905428406, | |
| "loss": 2.665, | |
| "theoretical_loss": 3.5611236414054868, | |
| "tokens_seen": 1300889600 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.00037778749763571023, | |
| "loss": 2.67, | |
| "theoretical_loss": 3.560312296419899, | |
| "tokens_seen": 1304166400 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.00037695999621713635, | |
| "loss": 2.6669, | |
| "theoretical_loss": 3.55950355659806, | |
| "tokens_seen": 1307443200 | |
| }, | |
| { | |
| "debugging/Self-BLEU-5": 0.49020908264157476, | |
| "debugging/distinct-1-grams": 0.768901113497886, | |
| "debugging/distinct-2-grams": 0.9428782333551957, | |
| "debugging/entropy-1-grams": 6.085999550681761, | |
| "debugging/entropy-2-grams": 7.0033060167714964, | |
| "debugging/length": 490.2352941176471, | |
| "debugging/num_segments": 17, | |
| "epoch": 0.47, | |
| "objective/train/avg_token_score": 0.02056093141436577, | |
| "objective/train/avg_weight": 0.9812931418418884, | |
| "objective/train/docs_used": 741674, | |
| "objective/train/instantaneous_batch_size": 32, | |
| "objective/train/instantaneous_microbatch_size": 32768, | |
| "objective/train/original_loss": 2.6537561416625977, | |
| "objective/train/std_weight": 0.0998743399977684, | |
| "objective/train/theoretical_loss": 3.558697407077142, | |
| "objective/train/tokens_used": 1331180000, | |
| "theoretical_loss": 3.558697407077142, | |
| "tokens_seen": 1310720000 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.0003761324947985625, | |
| "loss": 2.6587, | |
| "theoretical_loss": 3.558697407077142, | |
| "tokens_seen": 1310720000 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.00037530499337998864, | |
| "loss": 2.6727, | |
| "theoretical_loss": 3.5578938331159975, | |
| "tokens_seen": 1313996800 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.0003744774919614148, | |
| "loss": 2.6464, | |
| "theoretical_loss": 3.557092820093863, | |
| "tokens_seen": 1317273600 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.00037364999054284093, | |
| "loss": 2.6667, | |
| "theoretical_loss": 3.556294353509079, | |
| "tokens_seen": 1320550400 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.00037282248912426705, | |
| "loss": 2.6582, | |
| "theoretical_loss": 3.555498418977828, | |
| "tokens_seen": 1323827200 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.00037199498770569317, | |
| "loss": 2.6524, | |
| "theoretical_loss": 3.5547050022328874, | |
| "tokens_seen": 1327104000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.0003711674862871194, | |
| "loss": 2.6501, | |
| "theoretical_loss": 3.553914089122399, | |
| "tokens_seen": 1330380800 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.0003703399848685455, | |
| "loss": 2.7001, | |
| "theoretical_loss": 3.553125665608655, | |
| "tokens_seen": 1333657600 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00036951248344997163, | |
| "loss": 2.6745, | |
| "theoretical_loss": 3.5523397177669005, | |
| "tokens_seen": 1336934400 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00036868498203139775, | |
| "loss": 2.6856, | |
| "theoretical_loss": 3.551556231784149, | |
| "tokens_seen": 1340211200 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00036785748061282387, | |
| "loss": 2.6404, | |
| "theoretical_loss": 3.5507751939580148, | |
| "tokens_seen": 1343488000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00036702997919425004, | |
| "loss": 2.6402, | |
| "theoretical_loss": 3.5499965906955606, | |
| "tokens_seen": 1346764800 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00036620247777567616, | |
| "loss": 2.6887, | |
| "theoretical_loss": 3.549220408512161, | |
| "tokens_seen": 1350041600 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00036537497635710233, | |
| "loss": 2.665, | |
| "theoretical_loss": 3.5484466340303755, | |
| "tokens_seen": 1353318400 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00036454747493852845, | |
| "loss": 2.65, | |
| "theoretical_loss": 3.547675253978843, | |
| "tokens_seen": 1356595200 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.0003637199735199546, | |
| "loss": 2.6883, | |
| "theoretical_loss": 3.5469062551911854, | |
| "tokens_seen": 1359872000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.00036289247210138074, | |
| "loss": 2.6945, | |
| "theoretical_loss": 3.5461396246049244, | |
| "tokens_seen": 1363148800 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.0003620815207111783, | |
| "loss": 2.6619, | |
| "theoretical_loss": 3.545375349260419, | |
| "tokens_seen": 1366425600 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.00036127056932097596, | |
| "loss": 2.6992, | |
| "theoretical_loss": 3.544613416299808, | |
| "tokens_seen": 1369702400 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.00036044306790240213, | |
| "loss": 2.6756, | |
| "theoretical_loss": 3.5438538129659687, | |
| "tokens_seen": 1372979200 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.00035961556648382825, | |
| "loss": 2.6955, | |
| "theoretical_loss": 3.5430965266014933, | |
| "tokens_seen": 1376256000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.0003587880650652544, | |
| "loss": 2.6993, | |
| "theoretical_loss": 3.5423415446476705, | |
| "tokens_seen": 1379532800 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.00035796056364668054, | |
| "loss": 2.6348, | |
| "theoretical_loss": 3.541588854643487, | |
| "tokens_seen": 1382809600 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.00035713306222810666, | |
| "loss": 2.6618, | |
| "theoretical_loss": 3.5408384442246343, | |
| "tokens_seen": 1386086400 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.0003563055608095328, | |
| "loss": 2.6165, | |
| "theoretical_loss": 3.540090301122535, | |
| "tokens_seen": 1389363200 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.000355478059390959, | |
| "loss": 2.6408, | |
| "theoretical_loss": 3.5393444131633762, | |
| "tokens_seen": 1392640000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.0003546505579723851, | |
| "loss": 2.6885, | |
| "theoretical_loss": 3.5386007682671576, | |
| "tokens_seen": 1395916800 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.00035382305655381124, | |
| "loss": 2.6683, | |
| "theoretical_loss": 3.5378593544467494, | |
| "tokens_seen": 1399193600 | |
| } | |
| ], | |
| "max_steps": 42724, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 7.14460209610752e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |