| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "global_step": 42724, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 1.1682242990654204e-06, | |
| "loss": 10.8686, | |
| "theoretical_loss": 20.81281176760504, | |
| "tokens_seen": 65536 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 5.841121495327103e-05, | |
| "loss": 9.0294, | |
| "theoretical_loss": 8.563476630668958, | |
| "tokens_seen": 3276800 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 0.00011682242990654206, | |
| "loss": 6.976, | |
| "theoretical_loss": 7.477752684105921, | |
| "tokens_seen": 6553600 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 0.00017523364485981307, | |
| "loss": 6.1307, | |
| "theoretical_loss": 6.9337484549527915, | |
| "tokens_seen": 9830400 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 0.00023364485981308412, | |
| "loss": 5.7622, | |
| "theoretical_loss": 6.5835601944843045, | |
| "tokens_seen": 13107200 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.00029205607476635517, | |
| "loss": 5.4993, | |
| "theoretical_loss": 6.3307075311739744, | |
| "tokens_seen": 16384000 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.00035046728971962614, | |
| "loss": 5.2788, | |
| "theoretical_loss": 6.135523197998216, | |
| "tokens_seen": 19660800 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.0004088785046728972, | |
| "loss": 5.1139, | |
| "theoretical_loss": 5.978095549927499, | |
| "tokens_seen": 22937600 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.00046728971962616824, | |
| "loss": 5.0316, | |
| "theoretical_loss": 5.847111292323815, | |
| "tokens_seen": 26214400 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.000499739928125591, | |
| "loss": 4.8652, | |
| "theoretical_loss": 5.735570781940016, | |
| "tokens_seen": 29491200 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.0004991488556837526, | |
| "loss": 4.7488, | |
| "theoretical_loss": 5.638864110129244, | |
| "tokens_seen": 32768000 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.0004985577832419141, | |
| "loss": 4.6745, | |
| "theoretical_loss": 5.553806347902798, | |
| "tokens_seen": 36044800 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.0004979667108000757, | |
| "loss": 4.5438, | |
| "theoretical_loss": 5.478112046614329, | |
| "tokens_seen": 39321600 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0004973756383582371, | |
| "loss": 4.4609, | |
| "theoretical_loss": 5.410089925637252, | |
| "tokens_seen": 42598400 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0004967845659163987, | |
| "loss": 4.4206, | |
| "theoretical_loss": 5.348456049793725, | |
| "tokens_seen": 45875200 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0004961934934745603, | |
| "loss": 4.3209, | |
| "theoretical_loss": 5.292214532995457, | |
| "tokens_seen": 49152000 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0004956024210327218, | |
| "loss": 4.2638, | |
| "theoretical_loss": 5.240578591827869, | |
| "tokens_seen": 52428800 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0004950113485908833, | |
| "loss": 4.1812, | |
| "theoretical_loss": 5.192916690583679, | |
| "tokens_seen": 55705600 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0004944202761490448, | |
| "loss": 4.1175, | |
| "theoretical_loss": 5.148714829414331, | |
| "tokens_seen": 58982400 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0004938292037072064, | |
| "loss": 3.9738, | |
| "theoretical_loss": 5.107549528462992, | |
| "tokens_seen": 62259200 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0004932381312653678, | |
| "loss": 3.8582, | |
| "theoretical_loss": 5.069068083201136, | |
| "tokens_seen": 65536000 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0004926470588235294, | |
| "loss": 3.7998, | |
| "theoretical_loss": 5.032973875895897, | |
| "tokens_seen": 68812800 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.000492055986381691, | |
| "loss": 3.7438, | |
| "theoretical_loss": 4.999015274282555, | |
| "tokens_seen": 72089600 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.0004914649139398525, | |
| "loss": 3.7365, | |
| "theoretical_loss": 4.966977121409853, | |
| "tokens_seen": 75366400 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.000490873841498014, | |
| "loss": 3.6564, | |
| "theoretical_loss": 4.936674127683027, | |
| "tokens_seen": 78643200 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.0004902827690561755, | |
| "loss": 3.5971, | |
| "theoretical_loss": 4.907945679887972, | |
| "tokens_seen": 81920000 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.0004896916966143371, | |
| "loss": 3.6012, | |
| "theoretical_loss": 4.8806517198708175, | |
| "tokens_seen": 85196800 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.0004891006241724985, | |
| "loss": 3.5647, | |
| "theoretical_loss": 4.85466944053967, | |
| "tokens_seen": 88473600 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.0004885095517306601, | |
| "loss": 3.5473, | |
| "theoretical_loss": 4.829890613366704, | |
| "tokens_seen": 91750400 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.0004879184792888217, | |
| "loss": 3.4657, | |
| "theoretical_loss": 4.806219408835812, | |
| "tokens_seen": 95027200 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.0004873274068469832, | |
| "loss": 3.4688, | |
| "theoretical_loss": 4.783570605334148, | |
| "tokens_seen": 98304000 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.00048673633440514467, | |
| "loss": 3.4763, | |
| "theoretical_loss": 4.761868106830299, | |
| "tokens_seen": 101580800 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.0004861452619633062, | |
| "loss": 3.4809, | |
| "theoretical_loss": 4.741043708020364, | |
| "tokens_seen": 104857600 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.0004855541895214677, | |
| "loss": 3.4084, | |
| "theoretical_loss": 4.721036059306941, | |
| "tokens_seen": 108134400 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.0004849631170796293, | |
| "loss": 3.4574, | |
| "theoretical_loss": 4.701789794289756, | |
| "tokens_seen": 111411200 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.0004843720446377908, | |
| "loss": 3.4441, | |
| "theoretical_loss": 4.68325479029382, | |
| "tokens_seen": 114688000 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.00048378097219595233, | |
| "loss": 3.4289, | |
| "theoretical_loss": 4.6653855384841725, | |
| "tokens_seen": 117964800 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.00048318989975411385, | |
| "loss": 3.3685, | |
| "theoretical_loss": 4.6481406047776295, | |
| "tokens_seen": 121241600 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.00048259882731227537, | |
| "loss": 3.3637, | |
| "theoretical_loss": 4.631482166397534, | |
| "tokens_seen": 124518400 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.0004820077548704369, | |
| "loss": 3.3737, | |
| "theoretical_loss": 4.615375611773608, | |
| "tokens_seen": 127795200 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.00048141668242859847, | |
| "loss": 3.3587, | |
| "theoretical_loss": 4.5997891937483955, | |
| "tokens_seen": 131072000 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.00048082560998676, | |
| "loss": 3.3354, | |
| "theoretical_loss": 4.584693727850565, | |
| "tokens_seen": 134348800 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.0004802345375449215, | |
| "loss": 3.3025, | |
| "theoretical_loss": 4.570062328836407, | |
| "tokens_seen": 137625600 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.00047964346510308303, | |
| "loss": 3.283, | |
| "theoretical_loss": 4.5558701798619285, | |
| "tokens_seen": 140902400 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.00047905239266124455, | |
| "loss": 3.3052, | |
| "theoretical_loss": 4.542094329588689, | |
| "tokens_seen": 144179200 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.00047846132021940607, | |
| "loss": 3.2876, | |
| "theoretical_loss": 4.528713513292708, | |
| "tokens_seen": 147456000 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.0004778702477775676, | |
| "loss": 3.2959, | |
| "theoretical_loss": 4.515707994672887, | |
| "tokens_seen": 150732800 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.00047727917533572917, | |
| "loss": 3.2606, | |
| "theoretical_loss": 4.503059425571229, | |
| "tokens_seen": 154009600 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.0004766881028938907, | |
| "loss": 3.2412, | |
| "theoretical_loss": 4.490750721243157, | |
| "tokens_seen": 157286400 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.0004760970304520522, | |
| "loss": 3.3205, | |
| "theoretical_loss": 4.478765949169858, | |
| "tokens_seen": 160563200 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "objective/train/docs_used": 101104, | |
| "objective/train/instantaneous_batch_size": 32, | |
| "objective/train/instantaneous_microbatch_size": 32768, | |
| "objective/train/original_loss": 3.560319423675537, | |
| "objective/train/theoretical_loss": 4.46709022969911, | |
| "objective/train/tokens_used": 184300000, | |
| "theoretical_loss": 4.46709022969911, | |
| "tokens_seen": 163840000 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.00047550595801021373, | |
| "loss": 3.2093, | |
| "theoretical_loss": 4.46709022969911, | |
| "tokens_seen": 163840000 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.00047491488556837525, | |
| "loss": 3.2395, | |
| "theoretical_loss": 4.455709647047437, | |
| "tokens_seen": 167116800 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.00047432381312653677, | |
| "loss": 3.1921, | |
| "theoretical_loss": 4.444611169403321, | |
| "tokens_seen": 170393600 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.00047373274068469835, | |
| "loss": 3.1717, | |
| "theoretical_loss": 4.4337825770455375, | |
| "tokens_seen": 173670400 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.00047314166824285987, | |
| "loss": 3.169, | |
| "theoretical_loss": 4.423212397538051, | |
| "tokens_seen": 176947200 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.0004725505958010214, | |
| "loss": 3.1623, | |
| "theoretical_loss": 4.4128898471880325, | |
| "tokens_seen": 180224000 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0004719595233591829, | |
| "loss": 3.1722, | |
| "theoretical_loss": 4.40280477805997, | |
| "tokens_seen": 183500800 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.00047136845091734443, | |
| "loss": 3.1483, | |
| "theoretical_loss": 4.392947629929752, | |
| "tokens_seen": 186777600 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.00047077737847550595, | |
| "loss": 3.176, | |
| "theoretical_loss": 4.383309386640423, | |
| "tokens_seen": 190054400 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0004701863060336675, | |
| "loss": 3.1829, | |
| "theoretical_loss": 4.373881536388167, | |
| "tokens_seen": 193331200 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.00046959523359182905, | |
| "loss": 3.1694, | |
| "theoretical_loss": 4.364656035524595, | |
| "tokens_seen": 196608000 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.00046900416114999057, | |
| "loss": 3.1938, | |
| "theoretical_loss": 4.355625275511174, | |
| "tokens_seen": 199884800 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0004684130887081521, | |
| "loss": 3.1654, | |
| "theoretical_loss": 4.346782052704563, | |
| "tokens_seen": 203161600 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0004678220162663136, | |
| "loss": 3.1277, | |
| "theoretical_loss": 4.338119540689052, | |
| "tokens_seen": 206438400 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.00046723094382447513, | |
| "loss": 3.1532, | |
| "theoretical_loss": 4.329631264904703, | |
| "tokens_seen": 209715200 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.00046663987138263665, | |
| "loss": 3.1625, | |
| "theoretical_loss": 4.321311079348144, | |
| "tokens_seen": 212992000 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.0004660487989407982, | |
| "loss": 3.0982, | |
| "theoretical_loss": 4.313153145147683, | |
| "tokens_seen": 216268800 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.00046545772649895975, | |
| "loss": 3.1152, | |
| "theoretical_loss": 4.305151910836119, | |
| "tokens_seen": 219545600 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.00046486665405712127, | |
| "loss": 3.073, | |
| "theoretical_loss": 4.2973020941635784, | |
| "tokens_seen": 222822400 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.0004642755816152828, | |
| "loss": 3.1169, | |
| "theoretical_loss": 4.28959866530949, | |
| "tokens_seen": 226099200 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.0004636845091734443, | |
| "loss": 3.129, | |
| "theoretical_loss": 4.282036831367506, | |
| "tokens_seen": 229376000 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.00046309343673160583, | |
| "loss": 3.1552, | |
| "theoretical_loss": 4.274612021990189, | |
| "tokens_seen": 232652800 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.0004625023642897674, | |
| "loss": 3.1511, | |
| "theoretical_loss": 4.267319876091788, | |
| "tokens_seen": 235929600 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.0004619112918479289, | |
| "loss": 3.1489, | |
| "theoretical_loss": 4.260156229517635, | |
| "tokens_seen": 239206400 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.00046132021940609044, | |
| "loss": 3.1118, | |
| "theoretical_loss": 4.253117103597704, | |
| "tokens_seen": 242483200 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.00046072914696425197, | |
| "loss": 3.0504, | |
| "theoretical_loss": 4.246198694509945, | |
| "tokens_seen": 245760000 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.0004601380745224135, | |
| "loss": 3.0691, | |
| "theoretical_loss": 4.239397363386152, | |
| "tokens_seen": 249036800 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.000459547002080575, | |
| "loss": 3.0577, | |
| "theoretical_loss": 4.232709627099522, | |
| "tokens_seen": 252313600 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.0004589559296387366, | |
| "loss": 3.0658, | |
| "theoretical_loss": 4.226132149678757, | |
| "tokens_seen": 255590400 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.0004583648571968981, | |
| "loss": 3.1149, | |
| "theoretical_loss": 4.219661734298666, | |
| "tokens_seen": 258867200 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.0004577737847550596, | |
| "loss": 3.0236, | |
| "theoretical_loss": 4.213295315801815, | |
| "tokens_seen": 262144000 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.0004571827123132211, | |
| "loss": 3.0275, | |
| "theoretical_loss": 4.207029953709861, | |
| "tokens_seen": 265420800 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.0004565916398713826, | |
| "loss": 2.9819, | |
| "theoretical_loss": 4.200862825686893, | |
| "tokens_seen": 268697600 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.00045600056742954413, | |
| "loss": 3.0371, | |
| "theoretical_loss": 4.19479122142044, | |
| "tokens_seen": 271974400 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.0004554094949877057, | |
| "loss": 2.9829, | |
| "theoretical_loss": 4.188812536888775, | |
| "tokens_seen": 275251200 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.00045481842254586723, | |
| "loss": 2.9501, | |
| "theoretical_loss": 4.182924268985855, | |
| "tokens_seen": 278528000 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.00045422735010402875, | |
| "loss": 2.9744, | |
| "theoretical_loss": 4.177124010477671, | |
| "tokens_seen": 281804800 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.00045363627766219027, | |
| "loss": 3.0125, | |
| "theoretical_loss": 4.171409445265983, | |
| "tokens_seen": 285081600 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.0004530452052203518, | |
| "loss": 2.9955, | |
| "theoretical_loss": 4.165778343937409, | |
| "tokens_seen": 288358400 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.0004524541327785133, | |
| "loss": 2.9929, | |
| "theoretical_loss": 4.160228559577659, | |
| "tokens_seen": 291635200 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.00045186306033667483, | |
| "loss": 3.0244, | |
| "theoretical_loss": 4.15475802383233, | |
| "tokens_seen": 294912000 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.0004512719878948364, | |
| "loss": 3.0282, | |
| "theoretical_loss": 4.149364743197177, | |
| "tokens_seen": 298188800 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.00045068091545299793, | |
| "loss": 2.9998, | |
| "theoretical_loss": 4.14404679552214, | |
| "tokens_seen": 301465600 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.00045008984301115945, | |
| "loss": 3.0129, | |
| "theoretical_loss": 4.138802326714632, | |
| "tokens_seen": 304742400 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.00044949877056932097, | |
| "loss": 2.9824, | |
| "theoretical_loss": 4.133629547628726, | |
| "tokens_seen": 308019200 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.0004489076981274825, | |
| "loss": 2.9203, | |
| "theoretical_loss": 4.128526731127894, | |
| "tokens_seen": 311296000 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.000448316625685644, | |
| "loss": 2.9514, | |
| "theoretical_loss": 4.123492209309923, | |
| "tokens_seen": 314572800 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.0004477255532438056, | |
| "loss": 2.953, | |
| "theoretical_loss": 4.118524370883447, | |
| "tokens_seen": 317849600 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.0004471344808019671, | |
| "loss": 2.9618, | |
| "theoretical_loss": 4.113621658686355, | |
| "tokens_seen": 321126400 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.00044654340836012863, | |
| "loss": 2.9246, | |
| "theoretical_loss": 4.108782567337039, | |
| "tokens_seen": 324403200 | |
| }, | |
| { | |
| "debugging/Self-BLEU-5": 0.5689206637784514, | |
| "debugging/distinct-1-grams": 0.7604200619411795, | |
| "debugging/distinct-2-grams": 0.9515819938155399, | |
| "debugging/entropy-1-grams": 6.176763024314691, | |
| "debugging/entropy-2-grams": 7.1817278155572515, | |
| "debugging/length": 550.2105263157895, | |
| "debugging/num_segments": 19, | |
| "epoch": 0.12, | |
| "objective/train/docs_used": 191655, | |
| "objective/train/instantaneous_batch_size": 32, | |
| "objective/train/instantaneous_microbatch_size": 32768, | |
| "objective/train/original_loss": 3.2289257049560547, | |
| "objective/train/theoretical_loss": 4.104005641010112, | |
| "objective/train/tokens_used": 348140000, | |
| "theoretical_loss": 4.104005641010112, | |
| "tokens_seen": 327680000 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.00044595233591829015, | |
| "loss": 2.9268, | |
| "theoretical_loss": 4.104005641010112, | |
| "tokens_seen": 327680000 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.00044536126347645167, | |
| "loss": 2.9683, | |
| "theoretical_loss": 4.099289471328812, | |
| "tokens_seen": 330956800 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.0004447701910346132, | |
| "loss": 3.0014, | |
| "theoretical_loss": 4.094632695366921, | |
| "tokens_seen": 334233600 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.00044417911859277476, | |
| "loss": 2.96, | |
| "theoretical_loss": 4.090033993753448, | |
| "tokens_seen": 337510400 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.0004435880461509363, | |
| "loss": 2.9557, | |
| "theoretical_loss": 4.085492088873883, | |
| "tokens_seen": 340787200 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.0004429969737090978, | |
| "loss": 2.9289, | |
| "theoretical_loss": 4.081005743162224, | |
| "tokens_seen": 344064000 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.0004424059012672593, | |
| "loss": 2.9346, | |
| "theoretical_loss": 4.076573757478361, | |
| "tokens_seen": 347340800 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.00044181482882542085, | |
| "loss": 2.9196, | |
| "theoretical_loss": 4.072194969565807, | |
| "tokens_seen": 350617600 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.00044122375638358237, | |
| "loss": 2.9241, | |
| "theoretical_loss": 4.067868252585089, | |
| "tokens_seen": 353894400 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.0004406326839417439, | |
| "loss": 2.9505, | |
| "theoretical_loss": 4.063592513718411, | |
| "tokens_seen": 357171200 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.00044004161149990546, | |
| "loss": 2.9314, | |
| "theoretical_loss": 4.059366692841521, | |
| "tokens_seen": 360448000 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.000439450539058067, | |
| "loss": 2.9642, | |
| "theoretical_loss": 4.055189761258959, | |
| "tokens_seen": 363724800 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.0004388594666162285, | |
| "loss": 2.9659, | |
| "theoretical_loss": 4.051060720499127, | |
| "tokens_seen": 367001600 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.00043826839417439, | |
| "loss": 2.9061, | |
| "theoretical_loss": 4.046978601165831, | |
| "tokens_seen": 370278400 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.00043767732173255155, | |
| "loss": 2.9487, | |
| "theoretical_loss": 4.042942461843204, | |
| "tokens_seen": 373555200 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.00043708624929071307, | |
| "loss": 2.9171, | |
| "theoretical_loss": 4.038951388051044, | |
| "tokens_seen": 376832000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.00043649517684887464, | |
| "loss": 2.892, | |
| "theoretical_loss": 4.035004491247873, | |
| "tokens_seen": 380108800 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.00043590410440703616, | |
| "loss": 2.9348, | |
| "theoretical_loss": 4.031100907879109, | |
| "tokens_seen": 383385600 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.0004353130319651977, | |
| "loss": 2.8516, | |
| "theoretical_loss": 4.02723979846797, | |
| "tokens_seen": 386662400 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.0004347219595233592, | |
| "loss": 2.8972, | |
| "theoretical_loss": 4.023420346746835, | |
| "tokens_seen": 389939200 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.0004341308870815207, | |
| "loss": 2.9333, | |
| "theoretical_loss": 4.019641758826938, | |
| "tokens_seen": 393216000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.00043353981463968225, | |
| "loss": 2.9127, | |
| "theoretical_loss": 4.015903262404413, | |
| "tokens_seen": 396492800 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.0004329487421978438, | |
| "loss": 2.8845, | |
| "theoretical_loss": 4.012204106000786, | |
| "tokens_seen": 399769600 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.00043235766975600534, | |
| "loss": 2.9178, | |
| "theoretical_loss": 4.008543558236181, | |
| "tokens_seen": 403046400 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.00043176659731416686, | |
| "loss": 2.9254, | |
| "theoretical_loss": 4.004920907133565, | |
| "tokens_seen": 406323200 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0004311755248723284, | |
| "loss": 2.9181, | |
| "theoretical_loss": 4.001335459452449, | |
| "tokens_seen": 409600000 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0004305844524304899, | |
| "loss": 2.906, | |
| "theoretical_loss": 3.997786540050617, | |
| "tokens_seen": 412876800 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0004299933799886514, | |
| "loss": 2.9371, | |
| "theoretical_loss": 3.9942734912724456, | |
| "tokens_seen": 416153600 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.00042940230754681295, | |
| "loss": 2.9134, | |
| "theoretical_loss": 3.9907956723625375, | |
| "tokens_seen": 419430400 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0004288112351049745, | |
| "loss": 2.9714, | |
| "theoretical_loss": 3.9873524589034224, | |
| "tokens_seen": 422707200 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.000428220162663136, | |
| "loss": 2.9432, | |
| "theoretical_loss": 3.9839432422761556, | |
| "tokens_seen": 425984000 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0004276290902212975, | |
| "loss": 2.9521, | |
| "theoretical_loss": 3.980567429142721, | |
| "tokens_seen": 429260800 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.00042703801777945903, | |
| "loss": 2.9304, | |
| "theoretical_loss": 3.977224440949197, | |
| "tokens_seen": 432537600 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.00042644694533762055, | |
| "loss": 2.9453, | |
| "theoretical_loss": 3.9739137134486917, | |
| "tokens_seen": 435814400 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.00042585587289578207, | |
| "loss": 2.9322, | |
| "theoretical_loss": 3.9706346962431396, | |
| "tokens_seen": 439091200 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.00042526480045394365, | |
| "loss": 2.9113, | |
| "theoretical_loss": 3.9673868523430564, | |
| "tokens_seen": 442368000 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.00042467372801210517, | |
| "loss": 2.8985, | |
| "theoretical_loss": 3.9641696577444376, | |
| "tokens_seen": 445644800 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.0004240826555702667, | |
| "loss": 2.9016, | |
| "theoretical_loss": 3.9609826010220033, | |
| "tokens_seen": 448921600 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.0004234915831284282, | |
| "loss": 2.9171, | |
| "theoretical_loss": 3.9578251829380506, | |
| "tokens_seen": 452198400 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.00042290051068658973, | |
| "loss": 2.8842, | |
| "theoretical_loss": 3.954696916066199, | |
| "tokens_seen": 455475200 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.00042230943824475125, | |
| "loss": 2.8876, | |
| "theoretical_loss": 3.9515973244293643, | |
| "tokens_seen": 458752000 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.0004217183658029128, | |
| "loss": 2.8417, | |
| "theoretical_loss": 3.948525943151326, | |
| "tokens_seen": 462028800 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.00042112729336107435, | |
| "loss": 2.8284, | |
| "theoretical_loss": 3.9454823181212815, | |
| "tokens_seen": 465305600 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.00042053622091923587, | |
| "loss": 2.8623, | |
| "theoretical_loss": 3.9424660056708167, | |
| "tokens_seen": 468582400 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.0004199451484773974, | |
| "loss": 2.8993, | |
| "theoretical_loss": 3.939476572262754, | |
| "tokens_seen": 471859200 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.0004193540760355589, | |
| "loss": 2.8604, | |
| "theoretical_loss": 3.9365135941913563, | |
| "tokens_seen": 475136000 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.00041876300359372043, | |
| "loss": 2.7975, | |
| "theoretical_loss": 3.9335766572934023, | |
| "tokens_seen": 478412800 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.00041817193115188195, | |
| "loss": 2.8516, | |
| "theoretical_loss": 3.9306653566696603, | |
| "tokens_seen": 481689600 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.0004175808587100435, | |
| "loss": 2.8076, | |
| "theoretical_loss": 3.927779296416332, | |
| "tokens_seen": 484966400 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.00041698978626820505, | |
| "loss": 2.8758, | |
| "theoretical_loss": 3.924918089366024, | |
| "tokens_seen": 488243200 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "objective/train/docs_used": 280313, | |
| "objective/train/instantaneous_batch_size": 32, | |
| "objective/train/instantaneous_microbatch_size": 32768, | |
| "objective/train/original_loss": 2.973356246948242, | |
| "objective/train/theoretical_loss": 3.9220813568378707, | |
| "objective/train/tokens_used": 511980000, | |
| "theoretical_loss": 3.9220813568378707, | |
| "tokens_seen": 491520000 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.00041639871382636657, | |
| "loss": 2.8567, | |
| "theoretical_loss": 3.9220813568378707, | |
| "tokens_seen": 491520000 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.0004158076413845281, | |
| "loss": 2.9071, | |
| "theoretical_loss": 3.9192687283964096, | |
| "tokens_seen": 494796800 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.0004152165689426896, | |
| "loss": 2.878, | |
| "theoretical_loss": 3.9164798416188527, | |
| "tokens_seen": 498073600 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.00041462549650085113, | |
| "loss": 2.8168, | |
| "theoretical_loss": 3.913714341870409, | |
| "tokens_seen": 501350400 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.0004140344240590127, | |
| "loss": 2.8544, | |
| "theoretical_loss": 3.9109718820873303, | |
| "tokens_seen": 504627200 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.0004134433516171742, | |
| "loss": 2.8518, | |
| "theoretical_loss": 3.9082521225673625, | |
| "tokens_seen": 507904000 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.00041285227917533575, | |
| "loss": 2.8566, | |
| "theoretical_loss": 3.9055547307673075, | |
| "tokens_seen": 511180800 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.00041226120673349727, | |
| "loss": 2.8176, | |
| "theoretical_loss": 3.9028793811074056, | |
| "tokens_seen": 514457600 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.0004116701342916588, | |
| "loss": 2.7896, | |
| "theoretical_loss": 3.900225754782274, | |
| "tokens_seen": 517734400 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.0004110790618498203, | |
| "loss": 2.8166, | |
| "theoretical_loss": 3.897593539578138, | |
| "tokens_seen": 521011200 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.0004104879894079819, | |
| "loss": 2.7744, | |
| "theoretical_loss": 3.8949824296961015, | |
| "tokens_seen": 524288000 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.0004098969169661434, | |
| "loss": 2.8054, | |
| "theoretical_loss": 3.8923921255812353, | |
| "tokens_seen": 527564800 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.0004093058445243049, | |
| "loss": 2.795, | |
| "theoretical_loss": 3.8898223337572393, | |
| "tokens_seen": 530841600 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.00040871477208246645, | |
| "loss": 2.8216, | |
| "theoretical_loss": 3.88727276666648, | |
| "tokens_seen": 534118400 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.00040812369964062797, | |
| "loss": 2.8089, | |
| "theoretical_loss": 3.884743142515184, | |
| "tokens_seen": 537395200 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.0004075326271987895, | |
| "loss": 2.8525, | |
| "theoretical_loss": 3.8822331851235985, | |
| "tokens_seen": 540672000 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.000406941554756951, | |
| "loss": 2.8511, | |
| "theoretical_loss": 3.87974262378093, | |
| "tokens_seen": 543948800 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0004063504823151126, | |
| "loss": 2.8256, | |
| "theoretical_loss": 3.877271193104873, | |
| "tokens_seen": 547225600 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0004057594098732741, | |
| "loss": 2.8307, | |
| "theoretical_loss": 3.8748186329055736, | |
| "tokens_seen": 550502400 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0004051683374314356, | |
| "loss": 2.8738, | |
| "theoretical_loss": 3.87238468805384, | |
| "tokens_seen": 553779200 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.00040457726498959715, | |
| "loss": 2.8065, | |
| "theoretical_loss": 3.8699691083534633, | |
| "tokens_seen": 557056000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.00040398619254775867, | |
| "loss": 2.8596, | |
| "theoretical_loss": 3.8675716484174907, | |
| "tokens_seen": 560332800 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0004033951201059202, | |
| "loss": 2.8416, | |
| "theoretical_loss": 3.8651920675482936, | |
| "tokens_seen": 563609600 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.00040280404766408176, | |
| "loss": 2.8393, | |
| "theoretical_loss": 3.862830129621318, | |
| "tokens_seen": 566886400 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0004022129752222433, | |
| "loss": 2.8942, | |
| "theoretical_loss": 3.8604856029723575, | |
| "tokens_seen": 570163200 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0004016219027804048, | |
| "loss": 2.8987, | |
| "theoretical_loss": 3.8581582602882447, | |
| "tokens_seen": 573440000 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.0004010308303385663, | |
| "loss": 2.8711, | |
| "theoretical_loss": 3.8558478785008203, | |
| "tokens_seen": 576716800 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.00040043975789672785, | |
| "loss": 2.869, | |
| "theoretical_loss": 3.8535542386840778, | |
| "tokens_seen": 579993600 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.0003998486854548893, | |
| "loss": 2.8397, | |
| "theoretical_loss": 3.8512771259543586, | |
| "tokens_seen": 583270400 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.0003992694344618877, | |
| "loss": 2.8679, | |
| "theoretical_loss": 3.8490163293735082, | |
| "tokens_seen": 586547200 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.0003986783620200492, | |
| "loss": 2.8821, | |
| "theoretical_loss": 3.8467716418548648, | |
| "tokens_seen": 589824000 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.00039808728957821073, | |
| "loss": 2.8444, | |
| "theoretical_loss": 3.844542860072007, | |
| "tokens_seen": 593100800 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.00039749621713637225, | |
| "loss": 2.8058, | |
| "theoretical_loss": 3.8423297843701496, | |
| "tokens_seen": 596377600 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.0003969051446945338, | |
| "loss": 2.82, | |
| "theoretical_loss": 3.8401322186800995, | |
| "tokens_seen": 599654400 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0003963140722526953, | |
| "loss": 2.8123, | |
| "theoretical_loss": 3.83794997043469, | |
| "tokens_seen": 602931200 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.00039572299981085687, | |
| "loss": 2.8265, | |
| "theoretical_loss": 3.8357828504876004, | |
| "tokens_seen": 606208000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0003951319273690184, | |
| "loss": 2.7866, | |
| "theoretical_loss": 3.833630673034487, | |
| "tokens_seen": 609484800 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0003945408549271799, | |
| "loss": 2.7965, | |
| "theoretical_loss": 3.831493255536345, | |
| "tokens_seen": 612761600 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.00039394978248534143, | |
| "loss": 2.8207, | |
| "theoretical_loss": 3.8293704186450253, | |
| "tokens_seen": 616038400 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.00039335871004350295, | |
| "loss": 2.8041, | |
| "theoretical_loss": 3.827261986130839, | |
| "tokens_seen": 619315200 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0003927676376016645, | |
| "loss": 2.81, | |
| "theoretical_loss": 3.825167784812175, | |
| "tokens_seen": 622592000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.000392176565159826, | |
| "loss": 2.7531, | |
| "theoretical_loss": 3.823087644487069, | |
| "tokens_seen": 625868800 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.00039158549271798757, | |
| "loss": 2.781, | |
| "theoretical_loss": 3.8210213978666565, | |
| "tokens_seen": 629145600 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0003909944202761491, | |
| "loss": 2.8013, | |
| "theoretical_loss": 3.8189688805104476, | |
| "tokens_seen": 632422400 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.00039040334783431056, | |
| "loss": 2.8092, | |
| "theoretical_loss": 3.816929930763374, | |
| "tokens_seen": 635699200 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0003898122753924721, | |
| "loss": 2.8463, | |
| "theoretical_loss": 3.8149043896945347, | |
| "tokens_seen": 638976000 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0003892212029506336, | |
| "loss": 2.8477, | |
| "theoretical_loss": 3.812892101037601, | |
| "tokens_seen": 642252800 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0003886301305087951, | |
| "loss": 2.8368, | |
| "theoretical_loss": 3.81089291113282, | |
| "tokens_seen": 645529600 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0003880390580669567, | |
| "loss": 2.8435, | |
| "theoretical_loss": 3.8089066688705673, | |
| "tokens_seen": 648806400 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0003874479856251182, | |
| "loss": 2.8113, | |
| "theoretical_loss": 3.8069332256363992, | |
| "tokens_seen": 652083200 | |
| }, | |
| { | |
| "debugging/Self-BLEU-5": 0.5486494935750937, | |
| "debugging/distinct-1-grams": 0.8031186825041262, | |
| "debugging/distinct-2-grams": 0.959647362572758, | |
| "debugging/entropy-1-grams": 6.201095616386137, | |
| "debugging/entropy-2-grams": 7.080013788468194, | |
| "debugging/length": 487.95, | |
| "debugging/num_segments": 20, | |
| "epoch": 0.23, | |
| "objective/train/docs_used": 369191, | |
| "objective/train/instantaneous_batch_size": 32, | |
| "objective/train/instantaneous_microbatch_size": 32768, | |
| "objective/train/original_loss": 2.8638339042663574, | |
| "objective/train/theoretical_loss": 3.80497243525756, | |
| "objective/train/tokens_used": 675820000, | |
| "theoretical_loss": 3.80497243525756, | |
| "tokens_seen": 655360000 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.00038685691318327974, | |
| "loss": 2.7935, | |
| "theoretical_loss": 3.80497243525756, | |
| "tokens_seen": 655360000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.00038626584074144126, | |
| "loss": 2.7789, | |
| "theoretical_loss": 3.8030241539508958, | |
| "tokens_seen": 658636800 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.0003856747682996028, | |
| "loss": 2.8171, | |
| "theoretical_loss": 3.8010882402721324, | |
| "tokens_seen": 661913600 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.0003850836958577643, | |
| "loss": 2.8064, | |
| "theoretical_loss": 3.7991645550664757, | |
| "tokens_seen": 665190400 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.0003844926234159259, | |
| "loss": 2.8106, | |
| "theoretical_loss": 3.797252961420492, | |
| "tokens_seen": 668467200 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.0003839015509740874, | |
| "loss": 2.8205, | |
| "theoretical_loss": 3.795353324615228, | |
| "tokens_seen": 671744000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.0003833104785322489, | |
| "loss": 2.833, | |
| "theoretical_loss": 3.793465512080541, | |
| "tokens_seen": 675020800 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.00038271940609041044, | |
| "loss": 2.836, | |
| "theoretical_loss": 3.791589393350587, | |
| "tokens_seen": 678297600 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.00038212833364857196, | |
| "loss": 2.8456, | |
| "theoretical_loss": 3.7897248400204475, | |
| "tokens_seen": 681574400 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.0003815372612067335, | |
| "loss": 2.8388, | |
| "theoretical_loss": 3.7878717257038534, | |
| "tokens_seen": 684851200 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.000380946188764895, | |
| "loss": 2.8519, | |
| "theoretical_loss": 3.7860299259919685, | |
| "tokens_seen": 688128000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.0003803551163230566, | |
| "loss": 2.8263, | |
| "theoretical_loss": 3.7841993184132114, | |
| "tokens_seen": 691404800 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.0003797640438812181, | |
| "loss": 2.8164, | |
| "theoretical_loss": 3.78237978239408, | |
| "tokens_seen": 694681600 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.0003791729714393796, | |
| "loss": 2.8002, | |
| "theoretical_loss": 3.780571199220942, | |
| "tokens_seen": 697958400 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.00037858189899754114, | |
| "loss": 2.7888, | |
| "theoretical_loss": 3.7787734520027803, | |
| "tokens_seen": 701235200 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.00037799082655570266, | |
| "loss": 2.776, | |
| "theoretical_loss": 3.7769864256348455, | |
| "tokens_seen": 704512000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.0003773997541138642, | |
| "loss": 2.7873, | |
| "theoretical_loss": 3.775210006763202, | |
| "tokens_seen": 707788800 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.00037680868167202575, | |
| "loss": 2.7815, | |
| "theoretical_loss": 3.7734440837501406, | |
| "tokens_seen": 711065600 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.0003762176092301873, | |
| "loss": 2.7815, | |
| "theoretical_loss": 3.7716885466404246, | |
| "tokens_seen": 714342400 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.0003756265367883488, | |
| "loss": 2.8186, | |
| "theoretical_loss": 3.769943287128357, | |
| "tokens_seen": 717619200 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.0003750354643465103, | |
| "loss": 2.7491, | |
| "theoretical_loss": 3.7682081985256364, | |
| "tokens_seen": 720896000 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.00037444439190467184, | |
| "loss": 2.7574, | |
| "theoretical_loss": 3.7664831757299795, | |
| "tokens_seen": 724172800 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.00037385331946283336, | |
| "loss": 2.7375, | |
| "theoretical_loss": 3.7647681151944976, | |
| "tokens_seen": 727449600 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.00037326224702099493, | |
| "loss": 2.7197, | |
| "theoretical_loss": 3.7630629148977937, | |
| "tokens_seen": 730726400 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.00037267117457915645, | |
| "loss": 2.7261, | |
| "theoretical_loss": 3.761367474314768, | |
| "tokens_seen": 734003200 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.000372080102137318, | |
| "loss": 2.7753, | |
| "theoretical_loss": 3.7596816943881084, | |
| "tokens_seen": 737280000 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.0003714890296954795, | |
| "loss": 2.7859, | |
| "theoretical_loss": 3.758005477500451, | |
| "tokens_seen": 740556800 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.000370897957253641, | |
| "loss": 2.8045, | |
| "theoretical_loss": 3.756338727447186, | |
| "tokens_seen": 743833600 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.00037030688481180254, | |
| "loss": 2.8171, | |
| "theoretical_loss": 3.7546813494098945, | |
| "tokens_seen": 747110400 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.00036971581236996406, | |
| "loss": 2.7877, | |
| "theoretical_loss": 3.7530332499304007, | |
| "tokens_seen": 750387200 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.00036912473992812563, | |
| "loss": 2.7664, | |
| "theoretical_loss": 3.7513943368854195, | |
| "tokens_seen": 753664000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.00036853366748628715, | |
| "loss": 2.7609, | |
| "theoretical_loss": 3.7497645194617863, | |
| "tokens_seen": 756940800 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.0003679425950444487, | |
| "loss": 2.7662, | |
| "theoretical_loss": 3.748143708132246, | |
| "tokens_seen": 760217600 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.0003673515226026102, | |
| "loss": 2.7662, | |
| "theoretical_loss": 3.7465318146317994, | |
| "tokens_seen": 763494400 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.0003667604501607717, | |
| "loss": 2.6799, | |
| "theoretical_loss": 3.7449287519345766, | |
| "tokens_seen": 766771200 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.00036616937771893324, | |
| "loss": 2.7247, | |
| "theoretical_loss": 3.7433344342312385, | |
| "tokens_seen": 770048000 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.0003655783052770948, | |
| "loss": 2.7386, | |
| "theoretical_loss": 3.7417487769068756, | |
| "tokens_seen": 773324800 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.00036498723283525633, | |
| "loss": 2.7687, | |
| "theoretical_loss": 3.7401716965194076, | |
| "tokens_seen": 776601600 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.00036439616039341785, | |
| "loss": 2.7443, | |
| "theoretical_loss": 3.738603110778461, | |
| "tokens_seen": 779878400 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.0003638050879515794, | |
| "loss": 2.8035, | |
| "theoretical_loss": 3.73704293852471, | |
| "tokens_seen": 783155200 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.0003632140155097409, | |
| "loss": 2.7902, | |
| "theoretical_loss": 3.7354910997096793, | |
| "tokens_seen": 786432000 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.0003626229430679024, | |
| "loss": 2.79, | |
| "theoretical_loss": 3.7339475153759825, | |
| "tokens_seen": 789708800 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.000362031870626064, | |
| "loss": 2.792, | |
| "theoretical_loss": 3.732412107638, | |
| "tokens_seen": 792985600 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.00036144079818422546, | |
| "loss": 2.8192, | |
| "theoretical_loss": 3.7308847996629724, | |
| "tokens_seen": 796262400 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.000360849725742387, | |
| "loss": 2.7965, | |
| "theoretical_loss": 3.7293655156525043, | |
| "tokens_seen": 799539200 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0003602586533005485, | |
| "loss": 2.8091, | |
| "theoretical_loss": 3.727854180824469, | |
| "tokens_seen": 802816000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.00035966758085871, | |
| "loss": 2.7943, | |
| "theoretical_loss": 3.7263507213952978, | |
| "tokens_seen": 806092800 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.00035907650841687154, | |
| "loss": 2.7427, | |
| "theoretical_loss": 3.724855064562658, | |
| "tokens_seen": 809369600 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.00035848543597503306, | |
| "loss": 2.807, | |
| "theoretical_loss": 3.723367138488488, | |
| "tokens_seen": 812646400 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.00035790618498203144, | |
| "loss": 2.7934, | |
| "theoretical_loss": 3.7218868722824014, | |
| "tokens_seen": 815923200 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "objective/train/docs_used": 458194, | |
| "objective/train/instantaneous_batch_size": 32, | |
| "objective/train/instantaneous_microbatch_size": 32768, | |
| "objective/train/original_loss": 2.9236245155334473, | |
| "objective/train/theoretical_loss": 3.7204141959854384, | |
| "objective/train/tokens_used": 839660000, | |
| "theoretical_loss": 3.7204141959854384, | |
| "tokens_seen": 819200000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.00035731511254019296, | |
| "loss": 2.8012, | |
| "theoretical_loss": 3.7204141959854384, | |
| "tokens_seen": 819200000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0003567240400983545, | |
| "loss": 2.7962, | |
| "theoretical_loss": 3.718949040554162, | |
| "tokens_seen": 822476800 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.000356132967656516, | |
| "loss": 2.7889, | |
| "theoretical_loss": 3.7174913378450833, | |
| "tokens_seen": 825753600 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.0003555418952146775, | |
| "loss": 2.7913, | |
| "theoretical_loss": 3.7160410205994183, | |
| "tokens_seen": 829030400 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.00035495082277283904, | |
| "loss": 2.7615, | |
| "theoretical_loss": 3.7145980224281585, | |
| "tokens_seen": 832307200 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.0003543597503310006, | |
| "loss": 2.7366, | |
| "theoretical_loss": 3.713162277797449, | |
| "tokens_seen": 835584000 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.00035376867788916214, | |
| "loss": 2.7191, | |
| "theoretical_loss": 3.7117337220142748, | |
| "tokens_seen": 838860800 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.00035317760544732366, | |
| "loss": 2.7099, | |
| "theoretical_loss": 3.7103122912124364, | |
| "tokens_seen": 842137600 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.0003525865330054852, | |
| "loss": 2.7524, | |
| "theoretical_loss": 3.7088979223388128, | |
| "tokens_seen": 845414400 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.00035199546056364665, | |
| "loss": 2.7272, | |
| "theoretical_loss": 3.70749055313991, | |
| "tokens_seen": 848691200 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.00035140438812180817, | |
| "loss": 2.7231, | |
| "theoretical_loss": 3.7060901221486766, | |
| "tokens_seen": 851968000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.00035081331567996974, | |
| "loss": 2.7398, | |
| "theoretical_loss": 3.704696568671591, | |
| "tokens_seen": 855244800 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.00035022224323813127, | |
| "loss": 2.7242, | |
| "theoretical_loss": 3.7033098327760063, | |
| "tokens_seen": 858521600 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0003496311707962928, | |
| "loss": 2.712, | |
| "theoretical_loss": 3.7019298552777533, | |
| "tokens_seen": 861798400 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0003490400983544543, | |
| "loss": 2.6854, | |
| "theoretical_loss": 3.700556577728988, | |
| "tokens_seen": 865075200 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.00034844902591261583, | |
| "loss": 2.7172, | |
| "theoretical_loss": 3.6991899424062815, | |
| "tokens_seen": 868352000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.00034785795347077735, | |
| "loss": 2.7286, | |
| "theoretical_loss": 3.697829892298951, | |
| "tokens_seen": 871628800 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0003472668810289389, | |
| "loss": 2.7241, | |
| "theoretical_loss": 3.696476371097618, | |
| "tokens_seen": 874905600 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.00034667580858710044, | |
| "loss": 2.721, | |
| "theoretical_loss": 3.695129323182993, | |
| "tokens_seen": 878182400 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.00034608473614526196, | |
| "loss": 2.7104, | |
| "theoretical_loss": 3.693788693614879, | |
| "tokens_seen": 881459200 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.0003454936637034235, | |
| "loss": 2.701, | |
| "theoretical_loss": 3.6924544281213967, | |
| "tokens_seen": 884736000 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.000344902591261585, | |
| "loss": 2.7402, | |
| "theoretical_loss": 3.691126473088412, | |
| "tokens_seen": 888012800 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.00034431151881974653, | |
| "loss": 2.7404, | |
| "theoretical_loss": 3.689804775549173, | |
| "tokens_seen": 891289600 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.00034372044637790805, | |
| "loss": 2.712, | |
| "theoretical_loss": 3.688489283174146, | |
| "tokens_seen": 894566400 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.0003431293739360696, | |
| "loss": 2.7283, | |
| "theoretical_loss": 3.6871799442610538, | |
| "tokens_seen": 897843200 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.00034253830149423114, | |
| "loss": 2.7265, | |
| "theoretical_loss": 3.685876707725093, | |
| "tokens_seen": 901120000 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.00034194722905239266, | |
| "loss": 2.727, | |
| "theoretical_loss": 3.6845795230893517, | |
| "tokens_seen": 904396800 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.0003413561566105542, | |
| "loss": 2.6982, | |
| "theoretical_loss": 3.6832883404754035, | |
| "tokens_seen": 907673600 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.0003407650841687157, | |
| "loss": 2.7179, | |
| "theoretical_loss": 3.6820031105940796, | |
| "tokens_seen": 910950400 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.0003401740117268772, | |
| "loss": 2.7446, | |
| "theoretical_loss": 3.6807237847364176, | |
| "tokens_seen": 914227200 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.0003395829392850388, | |
| "loss": 2.7403, | |
| "theoretical_loss": 3.6794503147647846, | |
| "tokens_seen": 917504000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.0003389918668432003, | |
| "loss": 2.7113, | |
| "theoretical_loss": 3.67818265310416, | |
| "tokens_seen": 920780800 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.00033840079440136184, | |
| "loss": 2.7173, | |
| "theoretical_loss": 3.6769207527335888, | |
| "tokens_seen": 924057600 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.00033780972195952336, | |
| "loss": 2.7436, | |
| "theoretical_loss": 3.675664567177787, | |
| "tokens_seen": 927334400 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.0003372186495176849, | |
| "loss": 2.7563, | |
| "theoretical_loss": 3.674414050498913, | |
| "tokens_seen": 930611200 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.0003366275770758464, | |
| "loss": 2.7517, | |
| "theoretical_loss": 3.6731691572884824, | |
| "tokens_seen": 933888000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.000336036504634008, | |
| "loss": 2.729, | |
| "theoretical_loss": 3.671929842659438, | |
| "tokens_seen": 937164800 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.0003354454321921695, | |
| "loss": 2.7153, | |
| "theoretical_loss": 3.6706960622383624, | |
| "tokens_seen": 940441600 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.000334854359750331, | |
| "loss": 2.7734, | |
| "theoretical_loss": 3.6694677721578377, | |
| "tokens_seen": 943718400 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.00033426328730849254, | |
| "loss": 2.7185, | |
| "theoretical_loss": 3.66824492904894, | |
| "tokens_seen": 946995200 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.00033367221486665406, | |
| "loss": 2.716, | |
| "theoretical_loss": 3.667027490033874, | |
| "tokens_seen": 950272000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.0003330811424248156, | |
| "loss": 2.6816, | |
| "theoretical_loss": 3.6658154127187412, | |
| "tokens_seen": 953548800 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.00033250189143181386, | |
| "loss": 2.6909, | |
| "theoretical_loss": 3.664608655186437, | |
| "tokens_seen": 956825600 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.00033191081898997543, | |
| "loss": 2.6883, | |
| "theoretical_loss": 3.663407175989679, | |
| "tokens_seen": 960102400 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.00033131974654813695, | |
| "loss": 2.6588, | |
| "theoretical_loss": 3.662210934144158, | |
| "tokens_seen": 963379200 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.0003307286741062985, | |
| "loss": 2.6515, | |
| "theoretical_loss": 3.661019889121812, | |
| "tokens_seen": 966656000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.00033013760166446, | |
| "loss": 2.667, | |
| "theoretical_loss": 3.6598340008442234, | |
| "tokens_seen": 969932800 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.0003295465292226215, | |
| "loss": 2.7195, | |
| "theoretical_loss": 3.6586532296761285, | |
| "tokens_seen": 973209600 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.00032895545678078304, | |
| "loss": 2.6838, | |
| "theoretical_loss": 3.657477536419047, | |
| "tokens_seen": 976486400 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.0003283643843389446, | |
| "loss": 2.644, | |
| "theoretical_loss": 3.656306882305022, | |
| "tokens_seen": 979763200 | |
| }, | |
| { | |
| "debugging/Self-BLEU-5": 0.36836749701066873, | |
| "debugging/distinct-1-grams": 0.8145084309773442, | |
| "debugging/distinct-2-grams": 0.9777845741796146, | |
| "debugging/entropy-1-grams": 5.565961728673207, | |
| "debugging/entropy-2-grams": 6.204744460068355, | |
| "debugging/length": 558.2857142857143, | |
| "debugging/num_segments": 7, | |
| "epoch": 0.35, | |
| "objective/train/docs_used": 548756, | |
| "objective/train/instantaneous_batch_size": 32, | |
| "objective/train/instantaneous_microbatch_size": 32768, | |
| "objective/train/original_loss": 2.5849809646606445, | |
| "objective/train/theoretical_loss": 3.6551412289904697, | |
| "objective/train/tokens_used": 1003500000, | |
| "theoretical_loss": 3.6551412289904697, | |
| "tokens_seen": 983040000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.00032777331189710613, | |
| "loss": 2.6889, | |
| "theoretical_loss": 3.6551412289904697, | |
| "tokens_seen": 983040000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.00032718223945526765, | |
| "loss": 2.6755, | |
| "theoretical_loss": 3.6539805385501376, | |
| "tokens_seen": 986316800 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.0003265911670134292, | |
| "loss": 2.6964, | |
| "theoretical_loss": 3.652824773471171, | |
| "tokens_seen": 989593600 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.0003260000945715907, | |
| "loss": 2.7238, | |
| "theoretical_loss": 3.651673896647277, | |
| "tokens_seen": 992870400 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0003254090221297522, | |
| "loss": 2.7876, | |
| "theoretical_loss": 3.6505278713729985, | |
| "tokens_seen": 996147200 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0003248179496879138, | |
| "loss": 2.8354, | |
| "theoretical_loss": 3.6493866613380774, | |
| "tokens_seen": 999424000 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0003242268772460753, | |
| "loss": 2.8495, | |
| "theoretical_loss": 3.648250230621924, | |
| "tokens_seen": 1002700800 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.00032363580480423683, | |
| "loss": 2.7892, | |
| "theoretical_loss": 3.647118543688179, | |
| "tokens_seen": 1005977600 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.00032304473236239835, | |
| "loss": 2.8044, | |
| "theoretical_loss": 3.6459915653793633, | |
| "tokens_seen": 1009254400 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.00032245365992055987, | |
| "loss": 2.7869, | |
| "theoretical_loss": 3.644869260911628, | |
| "tokens_seen": 1012531200 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0003218625874787214, | |
| "loss": 2.8025, | |
| "theoretical_loss": 3.64375159586959, | |
| "tokens_seen": 1015808000 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0003212715150368829, | |
| "loss": 2.7588, | |
| "theoretical_loss": 3.642638536201252, | |
| "tokens_seen": 1019084800 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.0003206804425950445, | |
| "loss": 2.761, | |
| "theoretical_loss": 3.6415300482130135, | |
| "tokens_seen": 1022361600 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.000320089370153206, | |
| "loss": 2.7571, | |
| "theoretical_loss": 3.6404260985647667, | |
| "tokens_seen": 1025638400 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.00031949829771136753, | |
| "loss": 2.7637, | |
| "theoretical_loss": 3.6393266542650684, | |
| "tokens_seen": 1028915200 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.00031890722526952905, | |
| "loss": 2.7481, | |
| "theoretical_loss": 3.638231682666401, | |
| "tokens_seen": 1032192000 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.00031831615282769057, | |
| "loss": 2.7808, | |
| "theoretical_loss": 3.637141151460505, | |
| "tokens_seen": 1035468800 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.0003177250803858521, | |
| "loss": 2.7641, | |
| "theoretical_loss": 3.636055028673799, | |
| "tokens_seen": 1038745600 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.00031713400794401367, | |
| "loss": 2.7917, | |
| "theoretical_loss": 3.634973282662864, | |
| "tokens_seen": 1042022400 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.0003165429355021752, | |
| "loss": 2.7501, | |
| "theoretical_loss": 3.6338958821100107, | |
| "tokens_seen": 1045299200 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.0003159518630603367, | |
| "loss": 2.7534, | |
| "theoretical_loss": 3.63282279601892, | |
| "tokens_seen": 1048576000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.00031536079061849823, | |
| "loss": 2.7475, | |
| "theoretical_loss": 3.631753993710352, | |
| "tokens_seen": 1051852800 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.00031476971817665975, | |
| "loss": 2.7884, | |
| "theoretical_loss": 3.630689444817925, | |
| "tokens_seen": 1055129600 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.0003141786457348212, | |
| "loss": 2.7667, | |
| "theoretical_loss": 3.629629119283967, | |
| "tokens_seen": 1058406400 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.0003135875732929828, | |
| "loss": 2.7601, | |
| "theoretical_loss": 3.628572987355434, | |
| "tokens_seen": 1061683200 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.0003129965008511443, | |
| "loss": 2.7147, | |
| "theoretical_loss": 3.6275210195798913, | |
| "tokens_seen": 1064960000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.00031240542840930583, | |
| "loss": 2.7162, | |
| "theoretical_loss": 3.626473186801564, | |
| "tokens_seen": 1068236800 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.00031181435596746736, | |
| "loss": 2.6936, | |
| "theoretical_loss": 3.6254294601574495, | |
| "tokens_seen": 1071513600 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.0003112232835256289, | |
| "loss": 2.6913, | |
| "theoretical_loss": 3.624389811073493, | |
| "tokens_seen": 1074790400 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.0003106322110837904, | |
| "loss": 2.6906, | |
| "theoretical_loss": 3.6233542112608257, | |
| "tokens_seen": 1078067200 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.0003100411386419519, | |
| "loss": 2.7232, | |
| "theoretical_loss": 3.6223226327120592, | |
| "tokens_seen": 1081344000 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.0003094500662001135, | |
| "loss": 2.6992, | |
| "theoretical_loss": 3.621295047697644, | |
| "tokens_seen": 1084620800 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.000308858993758275, | |
| "loss": 2.6927, | |
| "theoretical_loss": 3.6202714287622833, | |
| "tokens_seen": 1087897600 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.00030826792131643653, | |
| "loss": 2.7535, | |
| "theoretical_loss": 3.6192517487214038, | |
| "tokens_seen": 1091174400 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.00030767684887459806, | |
| "loss": 2.7177, | |
| "theoretical_loss": 3.6182359806576834, | |
| "tokens_seen": 1094451200 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.0003070857764327596, | |
| "loss": 2.7207, | |
| "theoretical_loss": 3.6172240979176333, | |
| "tokens_seen": 1097728000 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.0003064947039909211, | |
| "loss": 2.701, | |
| "theoretical_loss": 3.616216074108232, | |
| "tokens_seen": 1101004800 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.00030590363154908267, | |
| "loss": 2.6874, | |
| "theoretical_loss": 3.6152118830936164, | |
| "tokens_seen": 1104281600 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0003053125591072442, | |
| "loss": 2.7029, | |
| "theoretical_loss": 3.6142114989918195, | |
| "tokens_seen": 1107558400 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0003047214866654057, | |
| "loss": 2.6927, | |
| "theoretical_loss": 3.6132148961715624, | |
| "tokens_seen": 1110835200 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.00030413041422356723, | |
| "loss": 2.7024, | |
| "theoretical_loss": 3.6122220492490964, | |
| "tokens_seen": 1114112000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.00030353934178172875, | |
| "loss": 2.7201, | |
| "theoretical_loss": 3.6112329330850894, | |
| "tokens_seen": 1117388800 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0003029482693398903, | |
| "loss": 2.6931, | |
| "theoretical_loss": 3.61024752278157, | |
| "tokens_seen": 1120665600 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.00030235719689805185, | |
| "loss": 2.6943, | |
| "theoretical_loss": 3.6092657936789054, | |
| "tokens_seen": 1123942400 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.00030176612445621337, | |
| "loss": 2.667, | |
| "theoretical_loss": 3.6082877213528377, | |
| "tokens_seen": 1127219200 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0003011750520143749, | |
| "loss": 2.6584, | |
| "theoretical_loss": 3.60731328161156, | |
| "tokens_seen": 1130496000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0003005839795725364, | |
| "loss": 2.6984, | |
| "theoretical_loss": 3.6063424504928365, | |
| "tokens_seen": 1133772800 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.00029999290713069793, | |
| "loss": 2.6699, | |
| "theoretical_loss": 3.60537520426117, | |
| "tokens_seen": 1137049600 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.00029940183468885945, | |
| "loss": 2.6932, | |
| "theoretical_loss": 3.6044115194050086, | |
| "tokens_seen": 1140326400 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.00029881076224702103, | |
| "loss": 2.685, | |
| "theoretical_loss": 3.603451372633997, | |
| "tokens_seen": 1143603200 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "objective/train/docs_used": 634175, | |
| "objective/train/instantaneous_batch_size": 32, | |
| "objective/train/instantaneous_microbatch_size": 32768, | |
| "objective/train/original_loss": 2.4585118293762207, | |
| "objective/train/theoretical_loss": 3.6024947408762698, | |
| "objective/train/tokens_used": 1167340000, | |
| "theoretical_loss": 3.6024947408762698, | |
| "tokens_seen": 1146880000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.00029821968980518255, | |
| "loss": 2.6539, | |
| "theoretical_loss": 3.6024947408762698, | |
| "tokens_seen": 1146880000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.00029762861736334407, | |
| "loss": 2.756, | |
| "theoretical_loss": 3.601541601275783, | |
| "tokens_seen": 1150156800 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.0002970375449215056, | |
| "loss": 2.7502, | |
| "theoretical_loss": 3.6005919311896886, | |
| "tokens_seen": 1153433600 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.0002964464724796671, | |
| "loss": 2.7149, | |
| "theoretical_loss": 3.5996457081857454, | |
| "tokens_seen": 1156710400 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.00029585540003782863, | |
| "loss": 2.7312, | |
| "theoretical_loss": 3.598702910039772, | |
| "tokens_seen": 1159987200 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.00029526432759599015, | |
| "loss": 2.7239, | |
| "theoretical_loss": 3.597763514733133, | |
| "tokens_seen": 1163264000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.00029467325515415173, | |
| "loss": 2.7269, | |
| "theoretical_loss": 3.59682750045027, | |
| "tokens_seen": 1166540800 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.00029408218271231325, | |
| "loss": 2.685, | |
| "theoretical_loss": 3.5958948455762583, | |
| "tokens_seen": 1169817600 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.00029349111027047477, | |
| "loss": 2.699, | |
| "theoretical_loss": 3.594965528694412, | |
| "tokens_seen": 1173094400 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.0002929000378286363, | |
| "loss": 2.6904, | |
| "theoretical_loss": 3.594039528583913, | |
| "tokens_seen": 1176371200 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.0002923089653867978, | |
| "loss": 2.7125, | |
| "theoretical_loss": 3.5931168242174847, | |
| "tokens_seen": 1179648000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.00029171789294495933, | |
| "loss": 2.6923, | |
| "theoretical_loss": 3.59219739475909, | |
| "tokens_seen": 1182924800 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.0002911268205031209, | |
| "loss": 2.711, | |
| "theoretical_loss": 3.5912812195616732, | |
| "tokens_seen": 1186201600 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.00029053574806128243, | |
| "loss": 2.7153, | |
| "theoretical_loss": 3.590368278164926, | |
| "tokens_seen": 1189478400 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.00028994467561944395, | |
| "loss": 2.6884, | |
| "theoretical_loss": 3.5894585502930902, | |
| "tokens_seen": 1192755200 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.00028935360317760547, | |
| "loss": 2.7023, | |
| "theoretical_loss": 3.588552015852793, | |
| "tokens_seen": 1196032000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.000288762530735767, | |
| "loss": 2.6901, | |
| "theoretical_loss": 3.5876486549309097, | |
| "tokens_seen": 1199308800 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.0002881714582939285, | |
| "loss": 2.7197, | |
| "theoretical_loss": 3.586748447792462, | |
| "tokens_seen": 1202585600 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.0002875803858520901, | |
| "loss": 2.6774, | |
| "theoretical_loss": 3.5858513748785423, | |
| "tokens_seen": 1205862400 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.0002869893134102516, | |
| "loss": 2.6985, | |
| "theoretical_loss": 3.5849574168042704, | |
| "tokens_seen": 1209139200 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.00028639824096841313, | |
| "loss": 2.7111, | |
| "theoretical_loss": 3.5840665543567782, | |
| "tokens_seen": 1212416000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.00028580716852657465, | |
| "loss": 2.7243, | |
| "theoretical_loss": 3.583178768493222, | |
| "tokens_seen": 1215692800 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002852160960847361, | |
| "loss": 2.7112, | |
| "theoretical_loss": 3.5822940403388284, | |
| "tokens_seen": 1218969600 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.00028462502364289764, | |
| "loss": 2.7068, | |
| "theoretical_loss": 3.581412351184958, | |
| "tokens_seen": 1222246400 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.00028403395120105916, | |
| "loss": 2.7265, | |
| "theoretical_loss": 3.580533682487208, | |
| "tokens_seen": 1225523200 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.00028344287875922073, | |
| "loss": 2.7063, | |
| "theoretical_loss": 3.579658015863532, | |
| "tokens_seen": 1228800000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.00028285180631738225, | |
| "loss": 2.6991, | |
| "theoretical_loss": 3.5787853330923927, | |
| "tokens_seen": 1232076800 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002822607338755438, | |
| "loss": 2.7184, | |
| "theoretical_loss": 3.577915616110936, | |
| "tokens_seen": 1235353600 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002816696614337053, | |
| "loss": 2.6728, | |
| "theoretical_loss": 3.577048847013194, | |
| "tokens_seen": 1238630400 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002810785889918668, | |
| "loss": 2.6903, | |
| "theoretical_loss": 3.57618500804831, | |
| "tokens_seen": 1241907200 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.00028048751655002834, | |
| "loss": 2.764, | |
| "theoretical_loss": 3.575324081618793, | |
| "tokens_seen": 1245184000 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.0002798964441081899, | |
| "loss": 2.7676, | |
| "theoretical_loss": 3.5744660502787875, | |
| "tokens_seen": 1248460800 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.00027930537166635143, | |
| "loss": 2.7343, | |
| "theoretical_loss": 3.5736108967323794, | |
| "tokens_seen": 1251737600 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.00027871429922451295, | |
| "loss": 2.6927, | |
| "theoretical_loss": 3.5727586038319155, | |
| "tokens_seen": 1255014400 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.0002781232267826745, | |
| "loss": 2.7064, | |
| "theoretical_loss": 3.571909154576348, | |
| "tokens_seen": 1258291200 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.000277532154340836, | |
| "loss": 2.6997, | |
| "theoretical_loss": 3.5710625321096074, | |
| "tokens_seen": 1261568000 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.0002769410818989975, | |
| "loss": 2.71, | |
| "theoretical_loss": 3.570218719718989, | |
| "tokens_seen": 1264844800 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.0002763500094571591, | |
| "loss": 2.6884, | |
| "theoretical_loss": 3.569377700833569, | |
| "tokens_seen": 1268121600 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.0002757589370153206, | |
| "loss": 2.6788, | |
| "theoretical_loss": 3.568539459022639, | |
| "tokens_seen": 1271398400 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.00027516786457348213, | |
| "loss": 2.659, | |
| "theoretical_loss": 3.5677039779941584, | |
| "tokens_seen": 1274675200 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.00027457679213164365, | |
| "loss": 2.6688, | |
| "theoretical_loss": 3.566871241593236, | |
| "tokens_seen": 1277952000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.0002739857196898052, | |
| "loss": 2.6803, | |
| "theoretical_loss": 3.5660412338006235, | |
| "tokens_seen": 1281228800 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.0002733946472479667, | |
| "loss": 2.6511, | |
| "theoretical_loss": 3.565213938731236, | |
| "tokens_seen": 1284505600 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.0002728035748061282, | |
| "loss": 2.6606, | |
| "theoretical_loss": 3.5643893406326868, | |
| "tokens_seen": 1287782400 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.0002722125023642898, | |
| "loss": 2.6488, | |
| "theoretical_loss": 3.5635674238838466, | |
| "tokens_seen": 1291059200 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.0002716214299224513, | |
| "loss": 2.6633, | |
| "theoretical_loss": 3.5627481729934196, | |
| "tokens_seen": 1294336000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.00027103035748061283, | |
| "loss": 2.7017, | |
| "theoretical_loss": 3.561931572598538, | |
| "tokens_seen": 1297612800 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.00027043928503877435, | |
| "loss": 2.6935, | |
| "theoretical_loss": 3.5611176074633777, | |
| "tokens_seen": 1300889600 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.0002698482125969359, | |
| "loss": 2.6385, | |
| "theoretical_loss": 3.5603062624777895, | |
| "tokens_seen": 1304166400 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.0002692571401550974, | |
| "loss": 2.6251, | |
| "theoretical_loss": 3.559497522655951, | |
| "tokens_seen": 1307443200 | |
| }, | |
| { | |
| "debugging/Self-BLEU-5": 0.5495772280672173, | |
| "debugging/distinct-1-grams": 0.7717677537029302, | |
| "debugging/distinct-2-grams": 0.9421298772004165, | |
| "debugging/entropy-1-grams": 6.12011243732356, | |
| "debugging/entropy-2-grams": 7.053666959871009, | |
| "debugging/length": 531.0, | |
| "debugging/num_segments": 18, | |
| "epoch": 0.47, | |
| "objective/train/docs_used": 722422, | |
| "objective/train/instantaneous_batch_size": 32, | |
| "objective/train/instantaneous_microbatch_size": 32768, | |
| "objective/train/original_loss": 2.7103114128112793, | |
| "objective/train/theoretical_loss": 3.5586913731350327, | |
| "objective/train/tokens_used": 1331180000, | |
| "theoretical_loss": 3.5586913731350327, | |
| "tokens_seen": 1310720000 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.00026866606771325897, | |
| "loss": 2.598, | |
| "theoretical_loss": 3.5586913731350327, | |
| "tokens_seen": 1310720000 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.0002680749952714205, | |
| "loss": 2.6352, | |
| "theoretical_loss": 3.557887799173889, | |
| "tokens_seen": 1313996800 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.000267483922829582, | |
| "loss": 2.6513, | |
| "theoretical_loss": 3.557086786151754, | |
| "tokens_seen": 1317273600 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.00026689285038774353, | |
| "loss": 2.6301, | |
| "theoretical_loss": 3.5562883195669697, | |
| "tokens_seen": 1320550400 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.00026630177794590505, | |
| "loss": 2.6476, | |
| "theoretical_loss": 3.555492385035719, | |
| "tokens_seen": 1323827200 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.0002657107055040666, | |
| "loss": 2.6642, | |
| "theoretical_loss": 3.5546989682907784, | |
| "tokens_seen": 1327104000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00026511963306222815, | |
| "loss": 2.6362, | |
| "theoretical_loss": 3.5539080551802895, | |
| "tokens_seen": 1330380800 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00026452856062038967, | |
| "loss": 2.6501, | |
| "theoretical_loss": 3.553119631666546, | |
| "tokens_seen": 1333657600 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.0002639374881785512, | |
| "loss": 2.6677, | |
| "theoretical_loss": 3.5523336838247914, | |
| "tokens_seen": 1336934400 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.0002633464157367127, | |
| "loss": 2.6693, | |
| "theoretical_loss": 3.55155019784204, | |
| "tokens_seen": 1340211200 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00026275534329487423, | |
| "loss": 2.6824, | |
| "theoretical_loss": 3.5507691600159053, | |
| "tokens_seen": 1343488000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00026216427085303575, | |
| "loss": 2.6875, | |
| "theoretical_loss": 3.5499905567534515, | |
| "tokens_seen": 1346764800 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.0002615731984111973, | |
| "loss": 2.6603, | |
| "theoretical_loss": 3.549214374570052, | |
| "tokens_seen": 1350041600 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.0002609939474181956, | |
| "loss": 2.6334, | |
| "theoretical_loss": 3.5484406000882665, | |
| "tokens_seen": 1353318400 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.0002604028749763571, | |
| "loss": 2.6658, | |
| "theoretical_loss": 3.5476692200367346, | |
| "tokens_seen": 1356595200 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.00025981180253451864, | |
| "loss": 2.6842, | |
| "theoretical_loss": 3.546900221249076, | |
| "tokens_seen": 1359872000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.00025922073009268016, | |
| "loss": 2.6731, | |
| "theoretical_loss": 3.5461335906628157, | |
| "tokens_seen": 1363148800 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.0002586296576508417, | |
| "loss": 2.6841, | |
| "theoretical_loss": 3.54536931531831, | |
| "tokens_seen": 1366425600 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.0002580385852090032, | |
| "loss": 2.6728, | |
| "theoretical_loss": 3.5446073823576985, | |
| "tokens_seen": 1369702400 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.0002574475127671648, | |
| "loss": 2.6405, | |
| "theoretical_loss": 3.543847779023859, | |
| "tokens_seen": 1372979200 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.0002568564403253263, | |
| "loss": 2.6784, | |
| "theoretical_loss": 3.543090492659384, | |
| "tokens_seen": 1376256000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.0002562653678834878, | |
| "loss": 2.6239, | |
| "theoretical_loss": 3.542335510705562, | |
| "tokens_seen": 1379532800 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.00025567429544164934, | |
| "loss": 2.624, | |
| "theoretical_loss": 3.541582820701378, | |
| "tokens_seen": 1382809600 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.00025508322299981086, | |
| "loss": 2.6031, | |
| "theoretical_loss": 3.5408324102825253, | |
| "tokens_seen": 1386086400 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.0002544921505579724, | |
| "loss": 2.6351, | |
| "theoretical_loss": 3.5400842671804265, | |
| "tokens_seen": 1389363200 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.00025390107811613396, | |
| "loss": 2.6747, | |
| "theoretical_loss": 3.5393383792212676, | |
| "tokens_seen": 1392640000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.0002533100056742955, | |
| "loss": 2.6658, | |
| "theoretical_loss": 3.5385947343250486, | |
| "tokens_seen": 1395916800 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.000252718933232457, | |
| "loss": 2.6749, | |
| "theoretical_loss": 3.53785332050464, | |
| "tokens_seen": 1399193600 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.0002521278607906185, | |
| "loss": 2.6646, | |
| "theoretical_loss": 3.537114125864853, | |
| "tokens_seen": 1402470400 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.0002515486097976168, | |
| "loss": 2.6794, | |
| "theoretical_loss": 3.536377138601523, | |
| "tokens_seen": 1405747200 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.0002509575373557783, | |
| "loss": 2.6845, | |
| "theoretical_loss": 3.5356423470006018, | |
| "tokens_seen": 1409024000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.0002503664649139399, | |
| "loss": 2.6747, | |
| "theoretical_loss": 3.5349097394372615, | |
| "tokens_seen": 1412300800 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.00024977539247210135, | |
| "loss": 2.6765, | |
| "theoretical_loss": 3.5341793043750114, | |
| "tokens_seen": 1415577600 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.00024918432003026293, | |
| "loss": 2.6938, | |
| "theoretical_loss": 3.533451030364823, | |
| "tokens_seen": 1418854400 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.00024859324758842445, | |
| "loss": 2.6275, | |
| "theoretical_loss": 3.5327249060442676, | |
| "tokens_seen": 1422131200 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.00024800217514658597, | |
| "loss": 2.6179, | |
| "theoretical_loss": 3.5320009201366616, | |
| "tokens_seen": 1425408000 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0002474111027047475, | |
| "loss": 2.6515, | |
| "theoretical_loss": 3.531279061450226, | |
| "tokens_seen": 1428684800 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.000246820030262909, | |
| "loss": 2.6765, | |
| "theoretical_loss": 3.5305593188772533, | |
| "tokens_seen": 1431961600 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.00024622895782107053, | |
| "loss": 2.6617, | |
| "theoretical_loss": 3.5298416813932865, | |
| "tokens_seen": 1435238400 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0002456378853792321, | |
| "loss": 2.658, | |
| "theoretical_loss": 3.529126138056305, | |
| "tokens_seen": 1438515200 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.00024504681293739363, | |
| "loss": 2.6307, | |
| "theoretical_loss": 3.528412678005922, | |
| "tokens_seen": 1441792000 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.00024445574049555515, | |
| "loss": 2.6269, | |
| "theoretical_loss": 3.527701290462595, | |
| "tokens_seen": 1445068800 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.00024386466805371667, | |
| "loss": 2.5892, | |
| "theoretical_loss": 3.5269919647268377, | |
| "tokens_seen": 1448345600 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.0002432735956118782, | |
| "loss": 2.6232, | |
| "theoretical_loss": 3.526284690178448, | |
| "tokens_seen": 1451622400 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.0002426825231700397, | |
| "loss": 2.6585, | |
| "theoretical_loss": 3.5255794562757443, | |
| "tokens_seen": 1454899200 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.00024209145072820123, | |
| "loss": 2.625, | |
| "theoretical_loss": 3.524876252554808, | |
| "tokens_seen": 1458176000 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.00024150037828636278, | |
| "loss": 2.6278, | |
| "theoretical_loss": 3.524175068628737, | |
| "tokens_seen": 1461452800 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.0002409093058445243, | |
| "loss": 2.6124, | |
| "theoretical_loss": 3.5234758941869075, | |
| "tokens_seen": 1464729600 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.00024031823340268582, | |
| "loss": 2.6236, | |
| "theoretical_loss": 3.522778718994246, | |
| "tokens_seen": 1468006400 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.00023972716096084737, | |
| "loss": 2.5968, | |
| "theoretical_loss": 3.5220835328905062, | |
| "tokens_seen": 1471283200 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "objective/train/docs_used": 814205, | |
| "objective/train/instantaneous_batch_size": 32, | |
| "objective/train/instantaneous_microbatch_size": 32768, | |
| "objective/train/original_loss": 1.9243988990783691, | |
| "objective/train/theoretical_loss": 3.521390325789559, | |
| "objective/train/tokens_used": 1495020000, | |
| "theoretical_loss": 3.521390325789559, | |
| "tokens_seen": 1474560000 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.0002391360885190089, | |
| "loss": 2.6074, | |
| "theoretical_loss": 3.521390325789559, | |
| "tokens_seen": 1474560000 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.0002385450160771704, | |
| "loss": 2.5987, | |
| "theoretical_loss": 3.520699087678688, | |
| "tokens_seen": 1477836800 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.00023795394363533196, | |
| "loss": 2.5673, | |
| "theoretical_loss": 3.5200098086178926, | |
| "tokens_seen": 1481113600 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.00023736287119349348, | |
| "loss": 2.5868, | |
| "theoretical_loss": 3.5193224787392015, | |
| "tokens_seen": 1484390400 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.000236771798751655, | |
| "loss": 2.6053, | |
| "theoretical_loss": 3.518637088245993, | |
| "tokens_seen": 1487667200 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.00023618072630981655, | |
| "loss": 2.5797, | |
| "theoretical_loss": 3.5179536274123233, | |
| "tokens_seen": 1490944000 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.00023558965386797807, | |
| "loss": 2.5724, | |
| "theoretical_loss": 3.5172720865822633, | |
| "tokens_seen": 1494220800 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.0002349985814261396, | |
| "loss": 2.5803, | |
| "theoretical_loss": 3.5165924561692394, | |
| "tokens_seen": 1497497600 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.0002344193304331379, | |
| "loss": 2.6168, | |
| "theoretical_loss": 3.5159147266553887, | |
| "tokens_seen": 1500774400 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.0002338282579912994, | |
| "loss": 2.5949, | |
| "theoretical_loss": 3.515238888590916, | |
| "tokens_seen": 1504051200 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.00023323718554946093, | |
| "loss": 2.6213, | |
| "theoretical_loss": 3.51456493259346, | |
| "tokens_seen": 1507328000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.00023264611310762248, | |
| "loss": 2.6106, | |
| "theoretical_loss": 3.5138928493474664, | |
| "tokens_seen": 1510604800 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.000232055040665784, | |
| "loss": 2.62, | |
| "theoretical_loss": 3.513222629603569, | |
| "tokens_seen": 1513881600 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.00023146396822394552, | |
| "loss": 2.6522, | |
| "theoretical_loss": 3.512554264177979, | |
| "tokens_seen": 1517158400 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.00023087289578210707, | |
| "loss": 2.647, | |
| "theoretical_loss": 3.511887743951874, | |
| "tokens_seen": 1520435200 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.0002302818233402686, | |
| "loss": 2.6618, | |
| "theoretical_loss": 3.511223059870809, | |
| "tokens_seen": 1523712000 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 0.0002296907508984301, | |
| "loss": 2.6569, | |
| "theoretical_loss": 3.5105602029441143, | |
| "tokens_seen": 1526988800 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 0.00022909967845659163, | |
| "loss": 2.6362, | |
| "theoretical_loss": 3.5098991642443185, | |
| "tokens_seen": 1530265600 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 0.00022850860601475318, | |
| "loss": 2.6692, | |
| "theoretical_loss": 3.5092399349065655, | |
| "tokens_seen": 1533542400 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 0.0002279175335729147, | |
| "loss": 2.6821, | |
| "theoretical_loss": 3.5085825061280445, | |
| "tokens_seen": 1536819200 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 0.00022732646113107622, | |
| "loss": 2.6715, | |
| "theoretical_loss": 3.507926869167425, | |
| "tokens_seen": 1540096000 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 0.00022673538868923777, | |
| "loss": 2.6172, | |
| "theoretical_loss": 3.507273015344296, | |
| "tokens_seen": 1543372800 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 0.0002261443162473993, | |
| "loss": 2.6211, | |
| "theoretical_loss": 3.5066209360386154, | |
| "tokens_seen": 1546649600 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 0.0002255532438055608, | |
| "loss": 2.6592, | |
| "theoretical_loss": 3.5059706226901595, | |
| "tokens_seen": 1549926400 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 0.00022496217136372236, | |
| "loss": 2.6639, | |
| "theoretical_loss": 3.505322066797988, | |
| "tokens_seen": 1553203200 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.00022437109892188388, | |
| "loss": 2.6234, | |
| "theoretical_loss": 3.504675259919904, | |
| "tokens_seen": 1556480000 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.0002237800264800454, | |
| "loss": 2.6441, | |
| "theoretical_loss": 3.5040301936719302, | |
| "tokens_seen": 1559756800 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.00022318895403820695, | |
| "loss": 2.6508, | |
| "theoretical_loss": 3.5033868597277813, | |
| "tokens_seen": 1563033600 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.00022259788159636847, | |
| "loss": 2.6046, | |
| "theoretical_loss": 3.502745249818351, | |
| "tokens_seen": 1566310400 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.00022200680915453, | |
| "loss": 2.6174, | |
| "theoretical_loss": 3.5021053557311985, | |
| "tokens_seen": 1569587200 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.00022141573671269154, | |
| "loss": 2.5895, | |
| "theoretical_loss": 3.501467169310044, | |
| "tokens_seen": 1572864000 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.00022082466427085303, | |
| "loss": 2.5947, | |
| "theoretical_loss": 3.500830682454268, | |
| "tokens_seen": 1576140800 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.00022023359182901455, | |
| "loss": 2.6105, | |
| "theoretical_loss": 3.5001958871184167, | |
| "tokens_seen": 1579417600 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.0002196425193871761, | |
| "loss": 2.6151, | |
| "theoretical_loss": 3.4995627753117127, | |
| "tokens_seen": 1582694400 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.00021905144694533762, | |
| "loss": 2.6236, | |
| "theoretical_loss": 3.498931339097571, | |
| "tokens_seen": 1585971200 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.00021846037450349914, | |
| "loss": 2.6656, | |
| "theoretical_loss": 3.4983015705931226, | |
| "tokens_seen": 1589248000 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.00021786930206166066, | |
| "loss": 2.6424, | |
| "theoretical_loss": 3.4976734619687377, | |
| "tokens_seen": 1592524800 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.0002172782296198222, | |
| "loss": 2.6222, | |
| "theoretical_loss": 3.497047005447558, | |
| "tokens_seen": 1595801600 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.00021668715717798373, | |
| "loss": 2.6319, | |
| "theoretical_loss": 3.4964221933050377, | |
| "tokens_seen": 1599078400 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.00021609608473614525, | |
| "loss": 2.5903, | |
| "theoretical_loss": 3.495799017868479, | |
| "tokens_seen": 1602355200 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.0002155050122943068, | |
| "loss": 2.6379, | |
| "theoretical_loss": 3.495177471516584, | |
| "tokens_seen": 1605632000 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.00021491393985246832, | |
| "loss": 2.5803, | |
| "theoretical_loss": 3.4945575466790024, | |
| "tokens_seen": 1608908800 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.00021432286741062984, | |
| "loss": 2.6055, | |
| "theoretical_loss": 3.4939392358358923, | |
| "tokens_seen": 1612185600 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.0002137317949687914, | |
| "loss": 2.6657, | |
| "theoretical_loss": 3.493322531517478, | |
| "tokens_seen": 1615462400 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.0002131407225269529, | |
| "loss": 2.6619, | |
| "theoretical_loss": 3.4927074263036157, | |
| "tokens_seen": 1618739200 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.00021254965008511443, | |
| "loss": 2.6281, | |
| "theoretical_loss": 3.492093912823367, | |
| "tokens_seen": 1622016000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.00021195857764327598, | |
| "loss": 2.6244, | |
| "theoretical_loss": 3.4914819837545705, | |
| "tokens_seen": 1625292800 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.0002113675052014375, | |
| "loss": 2.6007, | |
| "theoretical_loss": 3.490871631823426, | |
| "tokens_seen": 1628569600 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.00021077643275959902, | |
| "loss": 2.5869, | |
| "theoretical_loss": 3.490262849804073, | |
| "tokens_seen": 1631846400 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.00021018536031776057, | |
| "loss": 2.5682, | |
| "theoretical_loss": 3.4896556305181834, | |
| "tokens_seen": 1635123200 | |
| }, | |
| { | |
| "debugging/Self-BLEU-5": 0.3460857686736783, | |
| "debugging/distinct-1-grams": 0.7664962955849348, | |
| "debugging/distinct-2-grams": 0.9583333333333334, | |
| "debugging/entropy-1-grams": 4.7939259983502716, | |
| "debugging/entropy-2-grams": 5.366801704259744, | |
| "debugging/length": 505.0, | |
| "debugging/num_segments": 3, | |
| "epoch": 0.59, | |
| "objective/train/docs_used": 904189, | |
| "objective/train/instantaneous_batch_size": 32, | |
| "objective/train/instantaneous_microbatch_size": 32768, | |
| "objective/train/original_loss": 2.677978277206421, | |
| "objective/train/theoretical_loss": 3.4890499668345516, | |
| "objective/train/tokens_used": 1658860000, | |
| "theoretical_loss": 3.4890499668345516, | |
| "tokens_seen": 1638400000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.0002095942878759221, | |
| "loss": 2.6157, | |
| "theoretical_loss": 3.4890499668345516, | |
| "tokens_seen": 1638400000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.0002090032154340836, | |
| "loss": 2.6266, | |
| "theoretical_loss": 3.4884458516686947, | |
| "tokens_seen": 1641676800 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.00020841214299224516, | |
| "loss": 2.585, | |
| "theoretical_loss": 3.487843277982451, | |
| "tokens_seen": 1644953600 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.00020782107055040668, | |
| "loss": 2.6062, | |
| "theoretical_loss": 3.4872422387835877, | |
| "tokens_seen": 1648230400 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.0002072299981085682, | |
| "loss": 2.6191, | |
| "theoretical_loss": 3.4866427271254077, | |
| "tokens_seen": 1651507200 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.00020663892566672972, | |
| "loss": 2.5949, | |
| "theoretical_loss": 3.4860447361063667, | |
| "tokens_seen": 1654784000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.00020604785322489124, | |
| "loss": 2.6252, | |
| "theoretical_loss": 3.48544825886969, | |
| "tokens_seen": 1658060800 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.00020545678078305276, | |
| "loss": 2.6031, | |
| "theoretical_loss": 3.4848532886029924, | |
| "tokens_seen": 1661337600 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.00020486570834121428, | |
| "loss": 2.625, | |
| "theoretical_loss": 3.4842598185379066, | |
| "tokens_seen": 1664614400 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.00020427463589937583, | |
| "loss": 2.6257, | |
| "theoretical_loss": 3.4836678419497105, | |
| "tokens_seen": 1667891200 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.00020368356345753735, | |
| "loss": 2.5742, | |
| "theoretical_loss": 3.483077352156964, | |
| "tokens_seen": 1671168000 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.00020309249101569887, | |
| "loss": 2.5602, | |
| "theoretical_loss": 3.4824883425211413, | |
| "tokens_seen": 1674444800 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.00020250141857386042, | |
| "loss": 2.577, | |
| "theoretical_loss": 3.4819008064462764, | |
| "tokens_seen": 1677721600 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.00020191034613202194, | |
| "loss": 2.585, | |
| "theoretical_loss": 3.481314737378605, | |
| "tokens_seen": 1680998400 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.00020133109513902024, | |
| "loss": 2.5956, | |
| "theoretical_loss": 3.4807301288062145, | |
| "tokens_seen": 1684275200 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.00020074002269718179, | |
| "loss": 2.5643, | |
| "theoretical_loss": 3.4801469742586955, | |
| "tokens_seen": 1687552000 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.0002001489502553433, | |
| "loss": 2.6222, | |
| "theoretical_loss": 3.4795652673067954, | |
| "tokens_seen": 1690828800 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.00019955787781350483, | |
| "loss": 2.6145, | |
| "theoretical_loss": 3.478985001562082, | |
| "tokens_seen": 1694105600 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.00019896680537166638, | |
| "loss": 2.5763, | |
| "theoretical_loss": 3.4784061706766, | |
| "tokens_seen": 1697382400 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.0001983757329298279, | |
| "loss": 2.5693, | |
| "theoretical_loss": 3.477828768342543, | |
| "tokens_seen": 1700659200 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.00019778466048798942, | |
| "loss": 2.5406, | |
| "theoretical_loss": 3.477252788291917, | |
| "tokens_seen": 1703936000 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.00019719358804615094, | |
| "loss": 2.5817, | |
| "theoretical_loss": 3.4766782242962186, | |
| "tokens_seen": 1707212800 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.00019660251560431246, | |
| "loss": 2.546, | |
| "theoretical_loss": 3.476105070166107, | |
| "tokens_seen": 1710489600 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.00019601144316247398, | |
| "loss": 2.5882, | |
| "theoretical_loss": 3.475533319751085, | |
| "tokens_seen": 1713766400 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.00019542037072063553, | |
| "loss": 2.6089, | |
| "theoretical_loss": 3.4749629669391813, | |
| "tokens_seen": 1717043200 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.00019482929827879705, | |
| "loss": 2.6046, | |
| "theoretical_loss": 3.474394005656636, | |
| "tokens_seen": 1720320000 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.00019423822583695857, | |
| "loss": 2.608, | |
| "theoretical_loss": 3.473826429867589, | |
| "tokens_seen": 1723596800 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.00019364715339512012, | |
| "loss": 2.6189, | |
| "theoretical_loss": 3.4732602335737734, | |
| "tokens_seen": 1726873600 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.00019305608095328164, | |
| "loss": 2.6216, | |
| "theoretical_loss": 3.47269541081421, | |
| "tokens_seen": 1730150400 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.00019246500851144316, | |
| "loss": 2.6166, | |
| "theoretical_loss": 3.4721319556649046, | |
| "tokens_seen": 1733427200 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.00019187393606960468, | |
| "loss": 2.6073, | |
| "theoretical_loss": 3.4715698622385487, | |
| "tokens_seen": 1736704000 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.00019128286362776623, | |
| "loss": 2.6233, | |
| "theoretical_loss": 3.471009124684226, | |
| "tokens_seen": 1739980800 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.00019069179118592775, | |
| "loss": 2.5942, | |
| "theoretical_loss": 3.470449737187118, | |
| "tokens_seen": 1743257600 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.00019010071874408927, | |
| "loss": 2.6462, | |
| "theoretical_loss": 3.4698916939682123, | |
| "tokens_seen": 1746534400 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.00018950964630225082, | |
| "loss": 2.617, | |
| "theoretical_loss": 3.4693349892840186, | |
| "tokens_seen": 1749811200 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.00018891857386041234, | |
| "loss": 2.5931, | |
| "theoretical_loss": 3.4687796174262826, | |
| "tokens_seen": 1753088000 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.00018832750141857386, | |
| "loss": 2.5993, | |
| "theoretical_loss": 3.468225572721704, | |
| "tokens_seen": 1756364800 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.0001877364289767354, | |
| "loss": 2.6055, | |
| "theoretical_loss": 3.4676728495316587, | |
| "tokens_seen": 1759641600 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.00018714535653489693, | |
| "loss": 2.6027, | |
| "theoretical_loss": 3.467121442251922, | |
| "tokens_seen": 1762918400 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.00018655428409305845, | |
| "loss": 2.6185, | |
| "theoretical_loss": 3.4665713453123965, | |
| "tokens_seen": 1766195200 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.00018596321165122, | |
| "loss": 2.6066, | |
| "theoretical_loss": 3.46602255317684, | |
| "tokens_seen": 1769472000 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.00018537213920938152, | |
| "loss": 2.5993, | |
| "theoretical_loss": 3.465475060342598, | |
| "tokens_seen": 1772748800 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.00018478106676754304, | |
| "loss": 2.6411, | |
| "theoretical_loss": 3.4649288613403386, | |
| "tokens_seen": 1776025600 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.00018418999432570459, | |
| "loss": 2.6456, | |
| "theoretical_loss": 3.4643839507337892, | |
| "tokens_seen": 1779302400 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.0001835989218838661, | |
| "loss": 2.6619, | |
| "theoretical_loss": 3.4638403231194763, | |
| "tokens_seen": 1782579200 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.0001830078494420276, | |
| "loss": 2.5926, | |
| "theoretical_loss": 3.463297973126469, | |
| "tokens_seen": 1785856000 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.00018241677700018915, | |
| "loss": 2.6025, | |
| "theoretical_loss": 3.462756895416121, | |
| "tokens_seen": 1789132800 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.00018182570455835067, | |
| "loss": 2.6366, | |
| "theoretical_loss": 3.4622170846818205, | |
| "tokens_seen": 1792409600 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.0001812346321165122, | |
| "loss": 2.6881, | |
| "theoretical_loss": 3.4616785356487387, | |
| "tokens_seen": 1795686400 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.0001806435596746737, | |
| "loss": 2.6392, | |
| "theoretical_loss": 3.461141243073582, | |
| "tokens_seen": 1798963200 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "objective/train/docs_used": 992677, | |
| "objective/train/instantaneous_batch_size": 32, | |
| "objective/train/instantaneous_microbatch_size": 32768, | |
| "objective/train/original_loss": 2.878973960876465, | |
| "objective/train/theoretical_loss": 3.4606052017443467, | |
| "objective/train/tokens_used": 1822700000, | |
| "theoretical_loss": 3.4606052017443467, | |
| "tokens_seen": 1802240000 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.00018005248723283526, | |
| "loss": 2.6433, | |
| "theoretical_loss": 3.4606052017443467, | |
| "tokens_seen": 1802240000 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.00017946141479099678, | |
| "loss": 2.6122, | |
| "theoretical_loss": 3.4600704064800745, | |
| "tokens_seen": 1805516800 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.0001788703423491583, | |
| "loss": 2.6283, | |
| "theoretical_loss": 3.459536852130613, | |
| "tokens_seen": 1808793600 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.00017827926990731985, | |
| "loss": 2.6224, | |
| "theoretical_loss": 3.4590045335763775, | |
| "tokens_seen": 1812070400 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.00017768819746548137, | |
| "loss": 2.6079, | |
| "theoretical_loss": 3.4584734457281128, | |
| "tokens_seen": 1815347200 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.0001770971250236429, | |
| "loss": 2.6239, | |
| "theoretical_loss": 3.457943583526662, | |
| "tokens_seen": 1818624000 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.00017650605258180444, | |
| "loss": 2.6323, | |
| "theoretical_loss": 3.45741494194273, | |
| "tokens_seen": 1821900800 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.00017591498013996596, | |
| "loss": 2.6063, | |
| "theoretical_loss": 3.4568875159766606, | |
| "tokens_seen": 1825177600 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.00017532390769812748, | |
| "loss": 2.6021, | |
| "theoretical_loss": 3.4563613006582017, | |
| "tokens_seen": 1828454400 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.00017473283525628903, | |
| "loss": 2.6058, | |
| "theoretical_loss": 3.4558362910462863, | |
| "tokens_seen": 1831731200 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.00017414176281445055, | |
| "loss": 2.572, | |
| "theoretical_loss": 3.4553124822288055, | |
| "tokens_seen": 1835008000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.00017355069037261207, | |
| "loss": 2.6485, | |
| "theoretical_loss": 3.4547898693223873, | |
| "tokens_seen": 1838284800 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.00017295961793077362, | |
| "loss": 2.5612, | |
| "theoretical_loss": 3.4542684474721805, | |
| "tokens_seen": 1841561600 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.00017236854548893514, | |
| "loss": 2.5701, | |
| "theoretical_loss": 3.453748211851636, | |
| "tokens_seen": 1844838400 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.00017177747304709666, | |
| "loss": 2.5829, | |
| "theoretical_loss": 3.453229157662291, | |
| "tokens_seen": 1848115200 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.0001711864006052582, | |
| "loss": 2.5624, | |
| "theoretical_loss": 3.452711280133558, | |
| "tokens_seen": 1851392000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.00017059532816341973, | |
| "loss": 2.5653, | |
| "theoretical_loss": 3.4521945745225118, | |
| "tokens_seen": 1854668800 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.00017000425572158125, | |
| "loss": 2.5339, | |
| "theoretical_loss": 3.451679036113682, | |
| "tokens_seen": 1857945600 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.00016941318327974277, | |
| "loss": 2.575, | |
| "theoretical_loss": 3.451164660218848, | |
| "tokens_seen": 1861222400 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.00016883393228674107, | |
| "loss": 2.5483, | |
| "theoretical_loss": 3.4506514421768273, | |
| "tokens_seen": 1864499200 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.0001682428598449026, | |
| "loss": 2.5241, | |
| "theoretical_loss": 3.45013937735328, | |
| "tokens_seen": 1867776000 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.0001676517874030641, | |
| "loss": 2.5883, | |
| "theoretical_loss": 3.4496284611405024, | |
| "tokens_seen": 1871052800 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.00016706071496122566, | |
| "loss": 2.5769, | |
| "theoretical_loss": 3.4491186889572303, | |
| "tokens_seen": 1874329600 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.00016646964251938718, | |
| "loss": 2.5537, | |
| "theoretical_loss": 3.4486100562484396, | |
| "tokens_seen": 1877606400 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.0001658785700775487, | |
| "loss": 2.5726, | |
| "theoretical_loss": 3.448102558485152, | |
| "tokens_seen": 1880883200 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.00016528749763571025, | |
| "loss": 2.6023, | |
| "theoretical_loss": 3.447596191164239, | |
| "tokens_seen": 1884160000 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.00016469642519387177, | |
| "loss": 2.5822, | |
| "theoretical_loss": 3.447090949808233, | |
| "tokens_seen": 1887436800 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.0001641053527520333, | |
| "loss": 2.5766, | |
| "theoretical_loss": 3.446586829965132, | |
| "tokens_seen": 1890713600 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.00016351428031019484, | |
| "loss": 2.6042, | |
| "theoretical_loss": 3.4460838272082173, | |
| "tokens_seen": 1893990400 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.00016292320786835636, | |
| "loss": 2.6269, | |
| "theoretical_loss": 3.4455819371358594, | |
| "tokens_seen": 1897267200 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.00016233213542651788, | |
| "loss": 2.6439, | |
| "theoretical_loss": 3.445081155371338, | |
| "tokens_seen": 1900544000 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.00016174106298467942, | |
| "loss": 2.6668, | |
| "theoretical_loss": 3.4445814775626564, | |
| "tokens_seen": 1903820800 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.00016114999054284095, | |
| "loss": 2.6412, | |
| "theoretical_loss": 3.4440828993823605, | |
| "tokens_seen": 1907097600 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.00016055891810100247, | |
| "loss": 2.6627, | |
| "theoretical_loss": 3.4435854165273567, | |
| "tokens_seen": 1910374400 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.00015996784565916401, | |
| "loss": 2.6179, | |
| "theoretical_loss": 3.443089024718735, | |
| "tokens_seen": 1913651200 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.0001593767732173255, | |
| "loss": 2.6268, | |
| "theoretical_loss": 3.4425937197015912, | |
| "tokens_seen": 1916928000 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.00015878570077548703, | |
| "loss": 2.6197, | |
| "theoretical_loss": 3.4420994972448513, | |
| "tokens_seen": 1920204800 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.00015819462833364858, | |
| "loss": 2.576, | |
| "theoretical_loss": 3.4416063531410988, | |
| "tokens_seen": 1923481600 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.0001576035558918101, | |
| "loss": 2.5762, | |
| "theoretical_loss": 3.4411142832064003, | |
| "tokens_seen": 1926758400 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.00015701248344997162, | |
| "loss": 2.512, | |
| "theoretical_loss": 3.4406232832801376, | |
| "tokens_seen": 1930035200 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.00015642141100813314, | |
| "loss": 2.565, | |
| "theoretical_loss": 3.440133349224835, | |
| "tokens_seen": 1933312000 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.00015584216001513147, | |
| "loss": 2.5391, | |
| "theoretical_loss": 3.439644476925995, | |
| "tokens_seen": 1936588800 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.00015525108757329299, | |
| "loss": 2.5388, | |
| "theoretical_loss": 3.4391566622919303, | |
| "tokens_seen": 1939865600 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.0001546600151314545, | |
| "loss": 2.5713, | |
| "theoretical_loss": 3.438669901253598, | |
| "tokens_seen": 1943142400 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.00015406894268961605, | |
| "loss": 2.5858, | |
| "theoretical_loss": 3.4381841897644385, | |
| "tokens_seen": 1946419200 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.00015347787024777758, | |
| "loss": 2.5812, | |
| "theoretical_loss": 3.4376995238002137, | |
| "tokens_seen": 1949696000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.0001528867978059391, | |
| "loss": 2.5584, | |
| "theoretical_loss": 3.4372158993588444, | |
| "tokens_seen": 1952972800 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.00015229572536410064, | |
| "loss": 2.5319, | |
| "theoretical_loss": 3.436733312460253, | |
| "tokens_seen": 1956249600 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.00015170465292226216, | |
| "loss": 2.5283, | |
| "theoretical_loss": 3.436251759146206, | |
| "tokens_seen": 1959526400 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.00015111358048042369, | |
| "loss": 2.5268, | |
| "theoretical_loss": 3.435771235480157, | |
| "tokens_seen": 1962803200 | |
| }, | |
| { | |
| "debugging/Self-BLEU-5": 0.5079118920805717, | |
| "debugging/distinct-1-grams": 0.7766201352507275, | |
| "debugging/distinct-2-grams": 0.9614427024360547, | |
| "debugging/entropy-1-grams": 5.978513760011541, | |
| "debugging/entropy-2-grams": 6.893552160625838, | |
| "debugging/length": 529.4, | |
| "debugging/num_segments": 15, | |
| "epoch": 0.7, | |
| "objective/train/docs_used": 1082414, | |
| "objective/train/instantaneous_batch_size": 32, | |
| "objective/train/instantaneous_microbatch_size": 32768, | |
| "objective/train/original_loss": 2.6117191314697266, | |
| "objective/train/theoretical_loss": 3.435291737547092, | |
| "objective/train/tokens_used": 1986540000, | |
| "theoretical_loss": 3.435291737547092, | |
| "tokens_seen": 1966080000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.00015052250803858523, | |
| "loss": 2.5221, | |
| "theoretical_loss": 3.435291737547092, | |
| "tokens_seen": 1966080000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.00014993143559674673, | |
| "loss": 2.5626, | |
| "theoretical_loss": 3.434813261453378, | |
| "tokens_seen": 1969356800 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.00014934036315490825, | |
| "loss": 2.597, | |
| "theoretical_loss": 3.434335803326607, | |
| "tokens_seen": 1972633600 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.0001487492907130698, | |
| "loss": 2.6264, | |
| "theoretical_loss": 3.433859359315448, | |
| "tokens_seen": 1975910400 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.00014815821827123132, | |
| "loss": 2.634, | |
| "theoretical_loss": 3.433383925589498, | |
| "tokens_seen": 1979187200 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.00014756714582939284, | |
| "loss": 2.6205, | |
| "theoretical_loss": 3.432909498339132, | |
| "tokens_seen": 1982464000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.00014697607338755439, | |
| "loss": 2.5945, | |
| "theoretical_loss": 3.432436073775358, | |
| "tokens_seen": 1985740800 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.0001463850009457159, | |
| "loss": 2.603, | |
| "theoretical_loss": 3.4319636481296687, | |
| "tokens_seen": 1989017600 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.00014579392850387743, | |
| "loss": 2.6237, | |
| "theoretical_loss": 3.4314922176539002, | |
| "tokens_seen": 1992294400 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.00014520285606203897, | |
| "loss": 2.6082, | |
| "theoretical_loss": 3.431021778620088, | |
| "tokens_seen": 1995571200 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.0001446117836202005, | |
| "loss": 2.5486, | |
| "theoretical_loss": 3.4305523273203233, | |
| "tokens_seen": 1998848000 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.00014402071117836202, | |
| "loss": 2.5873, | |
| "theoretical_loss": 3.430083860066617, | |
| "tokens_seen": 2002124800 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.00014342963873652354, | |
| "loss": 2.5733, | |
| "theoretical_loss": 3.429616373190753, | |
| "tokens_seen": 2005401600 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.00014283856629468509, | |
| "loss": 2.5949, | |
| "theoretical_loss": 3.4291498630441586, | |
| "tokens_seen": 2008678400 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.0001422474938528466, | |
| "loss": 2.5914, | |
| "theoretical_loss": 3.4286843259977613, | |
| "tokens_seen": 2011955200 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.00014165642141100813, | |
| "loss": 2.5759, | |
| "theoretical_loss": 3.4282197584418546, | |
| "tokens_seen": 2015232000 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.00014106534896916967, | |
| "loss": 2.5687, | |
| "theoretical_loss": 3.427756156785965, | |
| "tokens_seen": 2018508800 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.0001404742765273312, | |
| "loss": 2.5484, | |
| "theoretical_loss": 3.427293517458717, | |
| "tokens_seen": 2021785600 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.00013988320408549272, | |
| "loss": 2.5743, | |
| "theoretical_loss": 3.4268318369076995, | |
| "tokens_seen": 2025062400 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.00013929213164365426, | |
| "loss": 2.5631, | |
| "theoretical_loss": 3.4263711115993383, | |
| "tokens_seen": 2028339200 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.00013870105920181578, | |
| "loss": 2.5336, | |
| "theoretical_loss": 3.425911338018763, | |
| "tokens_seen": 2031616000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.0001381099867599773, | |
| "loss": 2.543, | |
| "theoretical_loss": 3.425452512669678, | |
| "tokens_seen": 2034892800 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.00013751891431813885, | |
| "loss": 2.5836, | |
| "theoretical_loss": 3.4249946320742364, | |
| "tokens_seen": 2038169600 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.00013692784187630037, | |
| "loss": 2.5969, | |
| "theoretical_loss": 3.4245376927729128, | |
| "tokens_seen": 2041446400 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.0001363367694344619, | |
| "loss": 2.6024, | |
| "theoretical_loss": 3.424081691324375, | |
| "tokens_seen": 2044723200 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.00013574569699262344, | |
| "loss": 2.5964, | |
| "theoretical_loss": 3.4236266243053626, | |
| "tokens_seen": 2048000000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.00013515462455078494, | |
| "loss": 2.6139, | |
| "theoretical_loss": 3.4231724883105628, | |
| "tokens_seen": 2051276800 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.00013456355210894646, | |
| "loss": 2.6353, | |
| "theoretical_loss": 3.422719279952486, | |
| "tokens_seen": 2054553600 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.000133972479667108, | |
| "loss": 2.5992, | |
| "theoretical_loss": 3.422266995861345, | |
| "tokens_seen": 2057830400 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 0.00013338140722526953, | |
| "loss": 2.6145, | |
| "theoretical_loss": 3.421815632684937, | |
| "tokens_seen": 2061107200 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 0.00013279033478343105, | |
| "loss": 2.5986, | |
| "theoretical_loss": 3.4213651870885196, | |
| "tokens_seen": 2064384000 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 0.0001321992623415926, | |
| "loss": 2.6047, | |
| "theoretical_loss": 3.4209156557546954, | |
| "tokens_seen": 2067660800 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 0.00013160818989975412, | |
| "loss": 2.5892, | |
| "theoretical_loss": 3.420467035383294, | |
| "tokens_seen": 2070937600 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 0.00013101711745791564, | |
| "loss": 2.5483, | |
| "theoretical_loss": 3.4200193226912554, | |
| "tokens_seen": 2074214400 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 0.00013042604501607716, | |
| "loss": 2.5657, | |
| "theoretical_loss": 3.419572514412513, | |
| "tokens_seen": 2077491200 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 0.0001298349725742387, | |
| "loss": 2.5535, | |
| "theoretical_loss": 3.4191266072978816, | |
| "tokens_seen": 2080768000 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 0.00012924390013240023, | |
| "loss": 2.5329, | |
| "theoretical_loss": 3.418681598114942, | |
| "tokens_seen": 2084044800 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.00012865282769056175, | |
| "loss": 2.5723, | |
| "theoretical_loss": 3.4182374836479266, | |
| "tokens_seen": 2087321600 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.0001280617552487233, | |
| "loss": 2.5494, | |
| "theoretical_loss": 3.4177942606976126, | |
| "tokens_seen": 2090598400 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.00012747068280688482, | |
| "loss": 2.5239, | |
| "theoretical_loss": 3.4173519260812064, | |
| "tokens_seen": 2093875200 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.00012687961036504634, | |
| "loss": 2.5423, | |
| "theoretical_loss": 3.4169104766322356, | |
| "tokens_seen": 2097152000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.00012628853792320788, | |
| "loss": 2.5346, | |
| "theoretical_loss": 3.4164699092004405, | |
| "tokens_seen": 2100428800 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.0001256974654813694, | |
| "loss": 2.56, | |
| "theoretical_loss": 3.416030220651664, | |
| "tokens_seen": 2103705600 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.00012510639303953093, | |
| "loss": 2.5017, | |
| "theoretical_loss": 3.4155914078677467, | |
| "tokens_seen": 2106982400 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.00012451532059769245, | |
| "loss": 2.5379, | |
| "theoretical_loss": 3.4151534677464195, | |
| "tokens_seen": 2110259200 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.000123924248155854, | |
| "loss": 2.5354, | |
| "theoretical_loss": 3.414716397201196, | |
| "tokens_seen": 2113536000 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.00012333317571401552, | |
| "loss": 2.5039, | |
| "theoretical_loss": 3.4142801931612725, | |
| "tokens_seen": 2116812800 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.00012274210327217704, | |
| "loss": 2.5356, | |
| "theoretical_loss": 3.41384485257142, | |
| "tokens_seen": 2120089600 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.00012215103083033858, | |
| "loss": 2.5503, | |
| "theoretical_loss": 3.4134103723918834, | |
| "tokens_seen": 2123366400 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.00012155995838850009, | |
| "loss": 2.526, | |
| "theoretical_loss": 3.412976749598278, | |
| "tokens_seen": 2126643200 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "objective/train/docs_used": 1175218, | |
| "objective/train/instantaneous_batch_size": 32, | |
| "objective/train/instantaneous_microbatch_size": 32768, | |
| "objective/train/original_loss": 2.4588398933410645, | |
| "objective/train/theoretical_loss": 3.412543981181491, | |
| "objective/train/tokens_used": 2150380000, | |
| "theoretical_loss": 3.412543981181491, | |
| "tokens_seen": 2129920000 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.00012096888594666163, | |
| "loss": 2.5746, | |
| "theoretical_loss": 3.412543981181491, | |
| "tokens_seen": 2129920000 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.00012037781350482316, | |
| "loss": 2.5522, | |
| "theoretical_loss": 3.412112064147577, | |
| "tokens_seen": 2133196800 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.00011978674106298468, | |
| "loss": 2.5646, | |
| "theoretical_loss": 3.4116809955176617, | |
| "tokens_seen": 2136473600 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.00011919566862114621, | |
| "loss": 2.5608, | |
| "theoretical_loss": 3.4112507723278416, | |
| "tokens_seen": 2139750400 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.00011860459617930775, | |
| "loss": 2.5713, | |
| "theoretical_loss": 3.410821391629086, | |
| "tokens_seen": 2143027200 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.00011801352373746927, | |
| "loss": 2.5533, | |
| "theoretical_loss": 3.4103928504871406, | |
| "tokens_seen": 2146304000 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.00011742245129563079, | |
| "loss": 2.5653, | |
| "theoretical_loss": 3.4099651459824294, | |
| "tokens_seen": 2149580800 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.00011683137885379231, | |
| "loss": 2.5656, | |
| "theoretical_loss": 3.409538275209961, | |
| "tokens_seen": 2152857600 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.00011624030641195385, | |
| "loss": 2.6038, | |
| "theoretical_loss": 3.409112235279233, | |
| "tokens_seen": 2156134400 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.00011564923397011538, | |
| "loss": 2.5604, | |
| "theoretical_loss": 3.4086870233141355, | |
| "tokens_seen": 2159411200 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.0001150581615282769, | |
| "loss": 2.5729, | |
| "theoretical_loss": 3.4082626364528625, | |
| "tokens_seen": 2162688000 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.00011446708908643844, | |
| "loss": 2.5849, | |
| "theoretical_loss": 3.4078390718478153, | |
| "tokens_seen": 2165964800 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.00011387601664459997, | |
| "loss": 2.6054, | |
| "theoretical_loss": 3.4074163266655115, | |
| "tokens_seen": 2169241600 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.00011328494420276149, | |
| "loss": 2.5867, | |
| "theoretical_loss": 3.4069943980864945, | |
| "tokens_seen": 2172518400 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.00011269387176092302, | |
| "loss": 2.5776, | |
| "theoretical_loss": 3.4065732833052436, | |
| "tokens_seen": 2175795200 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.00011210279931908456, | |
| "loss": 2.5782, | |
| "theoretical_loss": 3.406152979530081, | |
| "tokens_seen": 2179072000 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.00011151172687724608, | |
| "loss": 2.5612, | |
| "theoretical_loss": 3.405733483983088, | |
| "tokens_seen": 2182348800 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.00011092065443540761, | |
| "loss": 2.5263, | |
| "theoretical_loss": 3.4053147939000117, | |
| "tokens_seen": 2185625600 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.00011032958199356912, | |
| "loss": 2.5404, | |
| "theoretical_loss": 3.4048969065301806, | |
| "tokens_seen": 2188902400 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.00010973850955173066, | |
| "loss": 2.5435, | |
| "theoretical_loss": 3.4044798191364163, | |
| "tokens_seen": 2192179200 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.00010914743710989219, | |
| "loss": 2.5307, | |
| "theoretical_loss": 3.4040635289949472, | |
| "tokens_seen": 2195456000 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 0.00010855636466805371, | |
| "loss": 2.5695, | |
| "theoretical_loss": 3.403648033395324, | |
| "tokens_seen": 2198732800 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 0.00010796529222621525, | |
| "loss": 2.5625, | |
| "theoretical_loss": 3.403233329640333, | |
| "tokens_seen": 2202009600 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 0.00010737421978437678, | |
| "loss": 2.5458, | |
| "theoretical_loss": 3.4028194150459137, | |
| "tokens_seen": 2205286400 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 0.0001067831473425383, | |
| "loss": 2.5627, | |
| "theoretical_loss": 3.4024062869410745, | |
| "tokens_seen": 2208563200 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 0.00010619207490069983, | |
| "loss": 2.5525, | |
| "theoretical_loss": 3.4019939426678096, | |
| "tokens_seen": 2211840000 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 0.00010561282390769813, | |
| "loss": 2.55, | |
| "theoretical_loss": 3.401582379581017, | |
| "tokens_seen": 2215116800 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 0.00010502175146585965, | |
| "loss": 2.544, | |
| "theoretical_loss": 3.401171595048417, | |
| "tokens_seen": 2218393600 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 0.00010444250047285795, | |
| "loss": 2.5597, | |
| "theoretical_loss": 3.40076158645047, | |
| "tokens_seen": 2221670400 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 0.00010385142803101949, | |
| "loss": 2.5823, | |
| "theoretical_loss": 3.400352351180298, | |
| "tokens_seen": 2224947200 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.00010326035558918101, | |
| "loss": 2.5944, | |
| "theoretical_loss": 3.3999438866436047, | |
| "tokens_seen": 2228224000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.00010266928314734254, | |
| "loss": 2.5659, | |
| "theoretical_loss": 3.399536190258594, | |
| "tokens_seen": 2231500800 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.00010207821070550406, | |
| "loss": 2.5591, | |
| "theoretical_loss": 3.399129259455895, | |
| "tokens_seen": 2234777600 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0001014871382636656, | |
| "loss": 2.5303, | |
| "theoretical_loss": 3.3987230916784803, | |
| "tokens_seen": 2238054400 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.00010089606582182713, | |
| "loss": 2.5886, | |
| "theoretical_loss": 3.398317684381593, | |
| "tokens_seen": 2241331200 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.00010030499337998865, | |
| "loss": 2.5194, | |
| "theoretical_loss": 3.397913035032666, | |
| "tokens_seen": 2244608000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 9.971392093815019e-05, | |
| "loss": 2.559, | |
| "theoretical_loss": 3.3975091411112484, | |
| "tokens_seen": 2247884800 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 9.912284849631172e-05, | |
| "loss": 2.5484, | |
| "theoretical_loss": 3.397106000108929, | |
| "tokens_seen": 2251161600 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 9.853177605447323e-05, | |
| "loss": 2.5295, | |
| "theoretical_loss": 3.3967036095292613, | |
| "tokens_seen": 2254438400 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 9.794070361263476e-05, | |
| "loss": 2.5844, | |
| "theoretical_loss": 3.3963019668876897, | |
| "tokens_seen": 2257715200 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 9.73496311707963e-05, | |
| "loss": 2.5484, | |
| "theoretical_loss": 3.395901069711476, | |
| "tokens_seen": 2260992000 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 9.675855872895782e-05, | |
| "loss": 2.6276, | |
| "theoretical_loss": 3.3955009155396247, | |
| "tokens_seen": 2264268800 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 9.616748628711935e-05, | |
| "loss": 2.7715, | |
| "theoretical_loss": 3.3951015019228117, | |
| "tokens_seen": 2267545600 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 9.557641384528087e-05, | |
| "loss": 2.804, | |
| "theoretical_loss": 3.3947028264233112, | |
| "tokens_seen": 2270822400 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 9.498534140344241e-05, | |
| "loss": 2.8085, | |
| "theoretical_loss": 3.3943048866149255, | |
| "tokens_seen": 2274099200 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 9.439426896160394e-05, | |
| "loss": 2.7798, | |
| "theoretical_loss": 3.3939076800829118, | |
| "tokens_seen": 2277376000 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 9.380319651976546e-05, | |
| "loss": 2.7643, | |
| "theoretical_loss": 3.393511204423914, | |
| "tokens_seen": 2280652800 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 9.3212124077927e-05, | |
| "loss": 2.7242, | |
| "theoretical_loss": 3.3931154572458917, | |
| "tokens_seen": 2283929600 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 9.262105163608853e-05, | |
| "loss": 2.6973, | |
| "theoretical_loss": 3.3927204361680503, | |
| "tokens_seen": 2287206400 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 9.202997919425005e-05, | |
| "loss": 2.6357, | |
| "theoretical_loss": 3.3923261388207733, | |
| "tokens_seen": 2290483200 | |
| }, | |
| { | |
| "debugging/Self-BLEU-5": 0.46869240972374726, | |
| "debugging/distinct-1-grams": 0.8121206810717653, | |
| "debugging/distinct-2-grams": 0.9660047807645101, | |
| "debugging/entropy-1-grams": 5.89482428427087, | |
| "debugging/entropy-2-grams": 6.645682303295068, | |
| "debugging/length": 500.8333333333333, | |
| "debugging/num_segments": 12, | |
| "epoch": 0.82, | |
| "objective/train/docs_used": 1259114, | |
| "objective/train/instantaneous_batch_size": 32, | |
| "objective/train/instantaneous_microbatch_size": 32768, | |
| "objective/train/original_loss": 2.5495071411132812, | |
| "objective/train/theoretical_loss": 3.3919325628455534, | |
| "objective/train/tokens_used": 2314220000, | |
| "theoretical_loss": 3.3919325628455534, | |
| "tokens_seen": 2293760000 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 9.143890675241157e-05, | |
| "loss": 2.6191, | |
| "theoretical_loss": 3.3919325628455534, | |
| "tokens_seen": 2293760000 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 9.08478343105731e-05, | |
| "loss": 2.6266, | |
| "theoretical_loss": 3.3915397058949233, | |
| "tokens_seen": 2297036800 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 9.025676186873463e-05, | |
| "loss": 2.575, | |
| "theoretical_loss": 3.3911475656323904, | |
| "tokens_seen": 2300313600 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 8.966568942689616e-05, | |
| "loss": 2.5819, | |
| "theoretical_loss": 3.3907561397323693, | |
| "tokens_seen": 2303590400 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 8.907461698505768e-05, | |
| "loss": 2.5973, | |
| "theoretical_loss": 3.3903654258801152, | |
| "tokens_seen": 2306867200 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 8.848354454321922e-05, | |
| "loss": 2.5693, | |
| "theoretical_loss": 3.389975421771658, | |
| "tokens_seen": 2310144000 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 8.789247210138075e-05, | |
| "loss": 2.5431, | |
| "theoretical_loss": 3.389586125113736, | |
| "tokens_seen": 2313420800 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 8.730139965954227e-05, | |
| "loss": 2.5469, | |
| "theoretical_loss": 3.389197533623734, | |
| "tokens_seen": 2316697600 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 8.671032721770381e-05, | |
| "loss": 2.5209, | |
| "theoretical_loss": 3.3888096450296152, | |
| "tokens_seen": 2319974400 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 8.611925477586534e-05, | |
| "loss": 2.5672, | |
| "theoretical_loss": 3.3884224570698613, | |
| "tokens_seen": 2323251200 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 8.552818233402686e-05, | |
| "loss": 2.5477, | |
| "theoretical_loss": 3.388035967493405, | |
| "tokens_seen": 2326528000 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 8.49371098921884e-05, | |
| "loss": 2.5493, | |
| "theoretical_loss": 3.387650174059569, | |
| "tokens_seen": 2329804800 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 8.43460374503499e-05, | |
| "loss": 2.5458, | |
| "theoretical_loss": 3.3872650745380044, | |
| "tokens_seen": 2333081600 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 8.375496500851144e-05, | |
| "loss": 2.5468, | |
| "theoretical_loss": 3.3868806667086275, | |
| "tokens_seen": 2336358400 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 8.316389256667297e-05, | |
| "loss": 2.523, | |
| "theoretical_loss": 3.3864969483615575, | |
| "tokens_seen": 2339635200 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 8.25728201248345e-05, | |
| "loss": 2.5145, | |
| "theoretical_loss": 3.3861139172970582, | |
| "tokens_seen": 2342912000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 8.198174768299603e-05, | |
| "loss": 2.5311, | |
| "theoretical_loss": 3.385731571325474, | |
| "tokens_seen": 2346188800 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 8.139067524115756e-05, | |
| "loss": 2.526, | |
| "theoretical_loss": 3.385349908267171, | |
| "tokens_seen": 2349465600 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 8.079960279931908e-05, | |
| "loss": 2.5695, | |
| "theoretical_loss": 3.3849689259524784, | |
| "tokens_seen": 2352742400 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 8.020853035748062e-05, | |
| "loss": 2.591, | |
| "theoretical_loss": 3.3845886222216275, | |
| "tokens_seen": 2356019200 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 7.961745791564215e-05, | |
| "loss": 2.6254, | |
| "theoretical_loss": 3.384208994924694, | |
| "tokens_seen": 2359296000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 7.902638547380367e-05, | |
| "loss": 2.7161, | |
| "theoretical_loss": 3.3838300419215384, | |
| "tokens_seen": 2362572800 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 7.843531303196521e-05, | |
| "loss": 2.7479, | |
| "theoretical_loss": 3.383451761081749, | |
| "tokens_seen": 2365849600 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 7.784424059012673e-05, | |
| "loss": 2.6841, | |
| "theoretical_loss": 3.383074150284584, | |
| "tokens_seen": 2369126400 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 7.725316814828825e-05, | |
| "loss": 2.6805, | |
| "theoretical_loss": 3.3826972074189134, | |
| "tokens_seen": 2372403200 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 7.666209570644978e-05, | |
| "loss": 2.6507, | |
| "theoretical_loss": 3.3823209303831634, | |
| "tokens_seen": 2375680000 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 7.60710232646113e-05, | |
| "loss": 2.5914, | |
| "theoretical_loss": 3.3819453170852602, | |
| "tokens_seen": 2378956800 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 7.547995082277284e-05, | |
| "loss": 2.5932, | |
| "theoretical_loss": 3.381570365442573, | |
| "tokens_seen": 2382233600 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 7.488887838093437e-05, | |
| "loss": 2.5743, | |
| "theoretical_loss": 3.3811960733818585, | |
| "tokens_seen": 2385510400 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 7.42978059390959e-05, | |
| "loss": 2.587, | |
| "theoretical_loss": 3.380822438839207, | |
| "tokens_seen": 2388787200 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 7.370673349725743e-05, | |
| "loss": 2.5342, | |
| "theoretical_loss": 3.380449459759986, | |
| "tokens_seen": 2392064000 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 7.311566105541895e-05, | |
| "loss": 2.5706, | |
| "theoretical_loss": 3.380077134098787, | |
| "tokens_seen": 2395340800 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 7.252458861358048e-05, | |
| "loss": 2.5504, | |
| "theoretical_loss": 3.3797054598193723, | |
| "tokens_seen": 2398617600 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 7.193351617174202e-05, | |
| "loss": 2.5328, | |
| "theoretical_loss": 3.379334434894619, | |
| "tokens_seen": 2401894400 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 7.134244372990354e-05, | |
| "loss": 2.5776, | |
| "theoretical_loss": 3.378964057306467, | |
| "tokens_seen": 2405171200 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 7.075137128806507e-05, | |
| "loss": 2.5716, | |
| "theoretical_loss": 3.378594325045867, | |
| "tokens_seen": 2408448000 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 7.01602988462266e-05, | |
| "loss": 2.5901, | |
| "theoretical_loss": 3.378225236112729, | |
| "tokens_seen": 2411724800 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 6.956922640438811e-05, | |
| "loss": 2.5617, | |
| "theoretical_loss": 3.377856788515865, | |
| "tokens_seen": 2415001600 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 6.897815396254965e-05, | |
| "loss": 2.568, | |
| "theoretical_loss": 3.3774889802729446, | |
| "tokens_seen": 2418278400 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 6.838708152071118e-05, | |
| "loss": 2.6481, | |
| "theoretical_loss": 3.377121809410438, | |
| "tokens_seen": 2421555200 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 6.77960090788727e-05, | |
| "loss": 2.6591, | |
| "theoretical_loss": 3.3767552739635684, | |
| "tokens_seen": 2424832000 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 6.720493663703424e-05, | |
| "loss": 2.6231, | |
| "theoretical_loss": 3.37638937197626, | |
| "tokens_seen": 2428108800 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 6.661386419519576e-05, | |
| "loss": 2.5929, | |
| "theoretical_loss": 3.3760241015010872, | |
| "tokens_seen": 2431385600 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 6.602279175335729e-05, | |
| "loss": 2.6067, | |
| "theoretical_loss": 3.375659460599226, | |
| "tokens_seen": 2434662400 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 6.543171931151883e-05, | |
| "loss": 2.5783, | |
| "theoretical_loss": 3.375295447340404, | |
| "tokens_seen": 2437939200 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 6.484064686968035e-05, | |
| "loss": 2.6045, | |
| "theoretical_loss": 3.3749320598028527, | |
| "tokens_seen": 2441216000 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 6.424957442784188e-05, | |
| "loss": 2.5871, | |
| "theoretical_loss": 3.3745692960732545, | |
| "tokens_seen": 2444492800 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 6.365850198600342e-05, | |
| "loss": 2.589, | |
| "theoretical_loss": 3.374207154246699, | |
| "tokens_seen": 2447769600 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 6.306742954416494e-05, | |
| "loss": 2.6142, | |
| "theoretical_loss": 3.373845632426633, | |
| "tokens_seen": 2451046400 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 6.247635710232646e-05, | |
| "loss": 2.5846, | |
| "theoretical_loss": 3.373484728724812, | |
| "tokens_seen": 2454323200 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "objective/train/docs_used": 1345695, | |
| "objective/train/instantaneous_batch_size": 32, | |
| "objective/train/instantaneous_microbatch_size": 32768, | |
| "objective/train/original_loss": 2.3688066005706787, | |
| "objective/train/theoretical_loss": 3.3731244412612527, | |
| "objective/train/tokens_used": 2478060000, | |
| "theoretical_loss": 3.3731244412612527, | |
| "tokens_seen": 2457600000 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 6.188528466048799e-05, | |
| "loss": 2.5939, | |
| "theoretical_loss": 3.3731244412612527, | |
| "tokens_seen": 2457600000 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 6.129421221864951e-05, | |
| "loss": 2.5738, | |
| "theoretical_loss": 3.3727647681641884, | |
| "tokens_seen": 2460876800 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 6.070313977681105e-05, | |
| "loss": 2.5583, | |
| "theoretical_loss": 3.3724057075700182, | |
| "tokens_seen": 2464153600 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 6.0112067334972575e-05, | |
| "loss": 2.6023, | |
| "theoretical_loss": 3.372047257623265, | |
| "tokens_seen": 2467430400 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 5.95209948931341e-05, | |
| "loss": 2.5494, | |
| "theoretical_loss": 3.3716894164765256, | |
| "tokens_seen": 2470707200 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 5.892992245129563e-05, | |
| "loss": 2.5666, | |
| "theoretical_loss": 3.371332182290427, | |
| "tokens_seen": 2473984000 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 5.8338850009457165e-05, | |
| "loss": 2.5024, | |
| "theoretical_loss": 3.37097555323358, | |
| "tokens_seen": 2477260800 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 5.774777756761869e-05, | |
| "loss": 2.5472, | |
| "theoretical_loss": 3.370619527482535, | |
| "tokens_seen": 2480537600 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 5.715670512578021e-05, | |
| "loss": 2.5807, | |
| "theoretical_loss": 3.370264103221736, | |
| "tokens_seen": 2483814400 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 5.656563268394174e-05, | |
| "loss": 2.5391, | |
| "theoretical_loss": 3.369909278643477, | |
| "tokens_seen": 2487091200 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 5.5974560242103275e-05, | |
| "loss": 2.578, | |
| "theoretical_loss": 3.369555051947857, | |
| "tokens_seen": 2490368000 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 5.53834878002648e-05, | |
| "loss": 2.5469, | |
| "theoretical_loss": 3.369201421342737, | |
| "tokens_seen": 2493644800 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 5.479241535842633e-05, | |
| "loss": 2.5474, | |
| "theoretical_loss": 3.3688483850436945, | |
| "tokens_seen": 2496921600 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 5.420134291658786e-05, | |
| "loss": 2.612, | |
| "theoretical_loss": 3.368495941273984, | |
| "tokens_seen": 2500198400 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 5.3610270474749385e-05, | |
| "loss": 2.5129, | |
| "theoretical_loss": 3.3681440882644886, | |
| "tokens_seen": 2503475200 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 5.301919803291091e-05, | |
| "loss": 2.5014, | |
| "theoretical_loss": 3.3677928242536828, | |
| "tokens_seen": 2506752000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 5.242812559107244e-05, | |
| "loss": 2.5047, | |
| "theoretical_loss": 3.3674421474875853, | |
| "tokens_seen": 2510028800 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 5.1837053149233975e-05, | |
| "loss": 2.5355, | |
| "theoretical_loss": 3.367092056219721, | |
| "tokens_seen": 2513305600 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 5.12459807073955e-05, | |
| "loss": 2.5223, | |
| "theoretical_loss": 3.3667425487110756, | |
| "tokens_seen": 2516582400 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 5.065490826555703e-05, | |
| "loss": 2.5242, | |
| "theoretical_loss": 3.366393623230057, | |
| "tokens_seen": 2519859200 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 5.006383582371855e-05, | |
| "loss": 2.5309, | |
| "theoretical_loss": 3.366045278052451, | |
| "tokens_seen": 2523136000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 4.9472763381880085e-05, | |
| "loss": 2.5341, | |
| "theoretical_loss": 3.365697511461384, | |
| "tokens_seen": 2526412800 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 4.888169094004161e-05, | |
| "loss": 2.5344, | |
| "theoretical_loss": 3.365350321747279, | |
| "tokens_seen": 2529689600 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 4.829061849820314e-05, | |
| "loss": 2.5221, | |
| "theoretical_loss": 3.365003707207816, | |
| "tokens_seen": 2532966400 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 4.769954605636467e-05, | |
| "loss": 2.5458, | |
| "theoretical_loss": 3.3646576661478944, | |
| "tokens_seen": 2536243200 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 4.71084736145262e-05, | |
| "loss": 2.5181, | |
| "theoretical_loss": 3.3643121968795895, | |
| "tokens_seen": 2539520000 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 4.651740117268772e-05, | |
| "loss": 2.5356, | |
| "theoretical_loss": 3.3639672977221142, | |
| "tokens_seen": 2542796800 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 4.592632873084925e-05, | |
| "loss": 2.508, | |
| "theoretical_loss": 3.3636229670017808, | |
| "tokens_seen": 2546073600 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 4.533525628901078e-05, | |
| "loss": 2.503, | |
| "theoretical_loss": 3.3632792030519605, | |
| "tokens_seen": 2549350400 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 4.474418384717231e-05, | |
| "loss": 2.5047, | |
| "theoretical_loss": 3.362936004213045, | |
| "tokens_seen": 2552627200 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 4.415311140533384e-05, | |
| "loss": 2.4909, | |
| "theoretical_loss": 3.362593368832409, | |
| "tokens_seen": 2555904000 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 4.356203896349537e-05, | |
| "loss": 2.4579, | |
| "theoretical_loss": 3.36225129526437, | |
| "tokens_seen": 2559180800 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 4.2970966521656895e-05, | |
| "loss": 2.4927, | |
| "theoretical_loss": 3.3619097818701515, | |
| "tokens_seen": 2562457600 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 4.237989407981842e-05, | |
| "loss": 2.4774, | |
| "theoretical_loss": 3.3615688270178454, | |
| "tokens_seen": 2565734400 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 4.178882163797995e-05, | |
| "loss": 2.4833, | |
| "theoretical_loss": 3.3612284290823737, | |
| "tokens_seen": 2569011200 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 4.119774919614148e-05, | |
| "loss": 2.5097, | |
| "theoretical_loss": 3.3608885864454523, | |
| "tokens_seen": 2572288000 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 4.060667675430301e-05, | |
| "loss": 2.5231, | |
| "theoretical_loss": 3.360549297495553, | |
| "tokens_seen": 2575564800 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 4.001560431246454e-05, | |
| "loss": 2.5607, | |
| "theoretical_loss": 3.3602105606278663, | |
| "tokens_seen": 2578841600 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 3.942453187062606e-05, | |
| "loss": 2.5699, | |
| "theoretical_loss": 3.359872374244266, | |
| "tokens_seen": 2582118400 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 3.883345942878759e-05, | |
| "loss": 2.5091, | |
| "theoretical_loss": 3.3595347367532744, | |
| "tokens_seen": 2585395200 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 3.824238698694912e-05, | |
| "loss": 2.494, | |
| "theoretical_loss": 3.359197646570022, | |
| "tokens_seen": 2588672000 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 3.765131454511065e-05, | |
| "loss": 2.4356, | |
| "theoretical_loss": 3.3588611021162156, | |
| "tokens_seen": 2591948800 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 3.706024210327218e-05, | |
| "loss": 2.4527, | |
| "theoretical_loss": 3.3585251018201006, | |
| "tokens_seen": 2595225600 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 3.6469169661433705e-05, | |
| "loss": 2.4592, | |
| "theoretical_loss": 3.358189644116427, | |
| "tokens_seen": 2598502400 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 3.587809721959524e-05, | |
| "loss": 2.4722, | |
| "theoretical_loss": 3.3578547274464134, | |
| "tokens_seen": 2601779200 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 3.528702477775676e-05, | |
| "loss": 2.486, | |
| "theoretical_loss": 3.357520350257713, | |
| "tokens_seen": 2605056000 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 3.469595233591829e-05, | |
| "loss": 2.4389, | |
| "theoretical_loss": 3.357186511004378, | |
| "tokens_seen": 2608332800 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 3.410487989407982e-05, | |
| "loss": 2.4608, | |
| "theoretical_loss": 3.356853208146826, | |
| "tokens_seen": 2611609600 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 3.351380745224135e-05, | |
| "loss": 2.5044, | |
| "theoretical_loss": 3.356520440151805, | |
| "tokens_seen": 2614886400 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 3.292273501040288e-05, | |
| "loss": 2.4754, | |
| "theoretical_loss": 3.3561882054923613, | |
| "tokens_seen": 2618163200 | |
| }, | |
| { | |
| "debugging/Self-BLEU-5": 0.3773267789108808, | |
| "debugging/distinct-1-grams": 0.8169355025323173, | |
| "debugging/distinct-2-grams": 0.9636260389009043, | |
| "debugging/entropy-1-grams": 5.689604031966668, | |
| "debugging/entropy-2-grams": 6.3162845189678904, | |
| "debugging/length": 477.1111111111111, | |
| "debugging/num_segments": 9, | |
| "epoch": 0.94, | |
| "objective/train/docs_used": 1436780, | |
| "objective/train/instantaneous_batch_size": 32, | |
| "objective/train/instantaneous_microbatch_size": 32768, | |
| "objective/train/original_loss": 2.696378231048584, | |
| "objective/train/theoretical_loss": 3.3558565026478027, | |
| "objective/train/tokens_used": 2641900000, | |
| "theoretical_loss": 3.3558565026478027, | |
| "tokens_seen": 2621440000 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 3.2331662568564405e-05, | |
| "loss": 2.4694, | |
| "theoretical_loss": 3.3558565026478027, | |
| "tokens_seen": 2621440000 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 3.174059012672593e-05, | |
| "loss": 2.4595, | |
| "theoretical_loss": 3.3555253301036685, | |
| "tokens_seen": 2624716800 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 3.114951768488746e-05, | |
| "loss": 2.5305, | |
| "theoretical_loss": 3.3551946863516937, | |
| "tokens_seen": 2627993600 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 3.055844524304899e-05, | |
| "loss": 2.5193, | |
| "theoretical_loss": 3.3548645698897777, | |
| "tokens_seen": 2631270400 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 2.996737280121052e-05, | |
| "loss": 2.4535, | |
| "theoretical_loss": 3.3545349792219494, | |
| "tokens_seen": 2634547200 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 2.9376300359372043e-05, | |
| "loss": 2.4608, | |
| "theoretical_loss": 3.354205912858337, | |
| "tokens_seen": 2637824000 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 2.8785227917533574e-05, | |
| "loss": 2.4699, | |
| "theoretical_loss": 3.353877369315133, | |
| "tokens_seen": 2641100800 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 2.81941554756951e-05, | |
| "loss": 2.4735, | |
| "theoretical_loss": 3.353549347114565, | |
| "tokens_seen": 2644377600 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 2.760308303385663e-05, | |
| "loss": 2.5041, | |
| "theoretical_loss": 3.3532218447848616, | |
| "tokens_seen": 2647654400 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 2.7012010592018156e-05, | |
| "loss": 2.4662, | |
| "theoretical_loss": 3.352894860860221, | |
| "tokens_seen": 2650931200 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 2.6420938150179687e-05, | |
| "loss": 2.4468, | |
| "theoretical_loss": 3.352568393880779, | |
| "tokens_seen": 2654208000 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 2.5829865708341215e-05, | |
| "loss": 2.5261, | |
| "theoretical_loss": 3.3522424423925794, | |
| "tokens_seen": 2657484800 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 2.5238793266502742e-05, | |
| "loss": 2.4736, | |
| "theoretical_loss": 3.3519170049475413, | |
| "tokens_seen": 2660761600 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 2.4647720824664273e-05, | |
| "loss": 2.4903, | |
| "theoretical_loss": 3.351592080103429, | |
| "tokens_seen": 2664038400 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 2.4056648382825798e-05, | |
| "loss": 2.523, | |
| "theoretical_loss": 3.3512676664238192, | |
| "tokens_seen": 2667315200 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 2.346557594098733e-05, | |
| "loss": 2.5285, | |
| "theoretical_loss": 3.350943762478075, | |
| "tokens_seen": 2670592000 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 2.2874503499148856e-05, | |
| "loss": 2.5645, | |
| "theoretical_loss": 3.3506203668413095, | |
| "tokens_seen": 2673868800 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 2.2283431057310387e-05, | |
| "loss": 2.5657, | |
| "theoretical_loss": 3.3502974780943613, | |
| "tokens_seen": 2677145600 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 2.169235861547191e-05, | |
| "loss": 2.5185, | |
| "theoretical_loss": 3.34997509482376, | |
| "tokens_seen": 2680422400 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 2.1101286173633442e-05, | |
| "loss": 2.563, | |
| "theoretical_loss": 3.3496532156217005, | |
| "tokens_seen": 2683699200 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 2.051021373179497e-05, | |
| "loss": 2.5605, | |
| "theoretical_loss": 3.34933183908601, | |
| "tokens_seen": 2686976000 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 1.9919141289956497e-05, | |
| "loss": 2.557, | |
| "theoretical_loss": 3.3490109638201195, | |
| "tokens_seen": 2690252800 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 1.9328068848118025e-05, | |
| "loss": 2.5404, | |
| "theoretical_loss": 3.348690588433038, | |
| "tokens_seen": 2693529600 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 1.8736996406279556e-05, | |
| "loss": 2.5472, | |
| "theoretical_loss": 3.3483707115393164, | |
| "tokens_seen": 2696806400 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 1.815774541327785e-05, | |
| "loss": 2.5388, | |
| "theoretical_loss": 3.3480513317590264, | |
| "tokens_seen": 2700083200 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 1.756667297143938e-05, | |
| "loss": 2.534, | |
| "theoretical_loss": 3.347732447717727, | |
| "tokens_seen": 2703360000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 1.697560052960091e-05, | |
| "loss": 2.5488, | |
| "theoretical_loss": 3.347414058046437, | |
| "tokens_seen": 2706636800 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 1.6384528087762434e-05, | |
| "loss": 2.5082, | |
| "theoretical_loss": 3.3470961613816086, | |
| "tokens_seen": 2709913600 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 1.5793455645923965e-05, | |
| "loss": 2.5507, | |
| "theoretical_loss": 3.3467787563650955, | |
| "tokens_seen": 2713190400 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 1.5202383204085493e-05, | |
| "loss": 2.4918, | |
| "theoretical_loss": 3.34646184164413, | |
| "tokens_seen": 2716467200 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 1.4611310762247022e-05, | |
| "loss": 2.5285, | |
| "theoretical_loss": 3.346145415871291, | |
| "tokens_seen": 2719744000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 1.402023832040855e-05, | |
| "loss": 2.5143, | |
| "theoretical_loss": 3.345829477704478, | |
| "tokens_seen": 2723020800 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 1.3429165878570077e-05, | |
| "loss": 2.5145, | |
| "theoretical_loss": 3.345514025806884, | |
| "tokens_seen": 2726297600 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 1.2838093436731606e-05, | |
| "loss": 2.5162, | |
| "theoretical_loss": 3.3451990588469687, | |
| "tokens_seen": 2729574400 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 1.2247020994893134e-05, | |
| "loss": 2.4997, | |
| "theoretical_loss": 3.3448845754984307, | |
| "tokens_seen": 2732851200 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 1.1655948553054663e-05, | |
| "loss": 2.5332, | |
| "theoretical_loss": 3.344570574440179, | |
| "tokens_seen": 2736128000 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 1.106487611121619e-05, | |
| "loss": 2.5341, | |
| "theoretical_loss": 3.3442570543563104, | |
| "tokens_seen": 2739404800 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 1.0473803669377718e-05, | |
| "loss": 2.5641, | |
| "theoretical_loss": 3.34394401393608, | |
| "tokens_seen": 2742681600 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 9.882731227539248e-06, | |
| "loss": 2.5486, | |
| "theoretical_loss": 3.343631451873874, | |
| "tokens_seen": 2745958400 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 9.291658785700775e-06, | |
| "loss": 2.5316, | |
| "theoretical_loss": 3.343319366869187, | |
| "tokens_seen": 2749235200 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 8.700586343862303e-06, | |
| "loss": 2.5689, | |
| "theoretical_loss": 3.343007757626593, | |
| "tokens_seen": 2752512000 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 8.109513902023834e-06, | |
| "loss": 2.5272, | |
| "theoretical_loss": 3.34269662285572, | |
| "tokens_seen": 2755788800 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 7.51844146018536e-06, | |
| "loss": 2.5221, | |
| "theoretical_loss": 3.342385961271226, | |
| "tokens_seen": 2759065600 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 6.927369018346889e-06, | |
| "loss": 2.5087, | |
| "theoretical_loss": 3.3420757715927705, | |
| "tokens_seen": 2762342400 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 6.336296576508416e-06, | |
| "loss": 2.5083, | |
| "theoretical_loss": 3.341766052544993, | |
| "tokens_seen": 2765619200 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 5.745224134669946e-06, | |
| "loss": 2.5091, | |
| "theoretical_loss": 3.341456802857484, | |
| "tokens_seen": 2768896000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 5.154151692831474e-06, | |
| "loss": 2.476, | |
| "theoretical_loss": 3.341148021264762, | |
| "tokens_seen": 2772172800 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 4.563079250993002e-06, | |
| "loss": 2.4923, | |
| "theoretical_loss": 3.3408397065062485, | |
| "tokens_seen": 2775449600 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 3.97200680915453e-06, | |
| "loss": 2.5254, | |
| "theoretical_loss": 3.3405318573262424, | |
| "tokens_seen": 2778726400 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 3.3809343673160584e-06, | |
| "loss": 2.4967, | |
| "theoretical_loss": 3.3402244724738965, | |
| "tokens_seen": 2782003200 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "objective/train/docs_used": 1527817, | |
| "objective/train/instantaneous_batch_size": 32, | |
| "objective/train/instantaneous_microbatch_size": 32768, | |
| "objective/train/original_loss": 2.387871265411377, | |
| "objective/train/theoretical_loss": 3.3399175507031913, | |
| "objective/train/tokens_used": 2805740000, | |
| "theoretical_loss": 3.3399175507031913, | |
| "tokens_seen": 2785280000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 2.789861925477587e-06, | |
| "loss": 2.4951, | |
| "theoretical_loss": 3.3399175507031913, | |
| "tokens_seen": 2785280000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 2.198789483639115e-06, | |
| "loss": 2.5301, | |
| "theoretical_loss": 3.3396110907729133, | |
| "tokens_seen": 2788556800 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1.607717041800643e-06, | |
| "loss": 2.5492, | |
| "theoretical_loss": 3.3393050914466285, | |
| "tokens_seen": 2791833600 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1.0166445999621714e-06, | |
| "loss": 2.5805, | |
| "theoretical_loss": 3.33899955149266, | |
| "tokens_seen": 2795110400 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.255721581236997e-07, | |
| "loss": 2.5805, | |
| "theoretical_loss": 3.338694469684062, | |
| "tokens_seen": 2798387200 | |
| } | |
| ], | |
| "max_steps": 42724, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 1.428920419221504e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |