| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.5, | |
| "global_step": 21362, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 1.1682242990654204e-06, | |
| "loss": 10.8813, | |
| "theoretical_loss": 20.81281176760504, | |
| "tokens_seen": 65536 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 5.841121495327103e-05, | |
| "loss": 8.9731, | |
| "theoretical_loss": 8.563476630668958, | |
| "tokens_seen": 3276800 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 0.00011682242990654206, | |
| "loss": 6.803, | |
| "theoretical_loss": 7.477752684105921, | |
| "tokens_seen": 6553600 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 0.00017523364485981307, | |
| "loss": 5.89, | |
| "theoretical_loss": 6.9337484549527915, | |
| "tokens_seen": 9830400 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 0.00023364485981308412, | |
| "loss": 5.4842, | |
| "theoretical_loss": 6.5835601944843045, | |
| "tokens_seen": 13107200 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.00029205607476635517, | |
| "loss": 5.2294, | |
| "theoretical_loss": 6.3307075311739744, | |
| "tokens_seen": 16384000 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.00035046728971962614, | |
| "loss": 5.0511, | |
| "theoretical_loss": 6.135523197998216, | |
| "tokens_seen": 19660800 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.0004088785046728972, | |
| "loss": 4.9053, | |
| "theoretical_loss": 5.978095549927499, | |
| "tokens_seen": 22937600 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.00046728971962616824, | |
| "loss": 4.8028, | |
| "theoretical_loss": 5.847111292323815, | |
| "tokens_seen": 26214400 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.000499739928125591, | |
| "loss": 4.7268, | |
| "theoretical_loss": 5.735570781940016, | |
| "tokens_seen": 29491200 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.0004991488556837526, | |
| "loss": 4.5965, | |
| "theoretical_loss": 5.638864110129244, | |
| "tokens_seen": 32768000 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.0004985577832419141, | |
| "loss": 4.4507, | |
| "theoretical_loss": 5.553806347902798, | |
| "tokens_seen": 36044800 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.0004979667108000757, | |
| "loss": 4.3796, | |
| "theoretical_loss": 5.478112046614329, | |
| "tokens_seen": 39321600 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0004973756383582371, | |
| "loss": 4.2971, | |
| "theoretical_loss": 5.410089925637252, | |
| "tokens_seen": 42598400 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0004967845659163987, | |
| "loss": 4.1952, | |
| "theoretical_loss": 5.348456049793725, | |
| "tokens_seen": 45875200 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0004961934934745603, | |
| "loss": 4.1713, | |
| "theoretical_loss": 5.292214532995457, | |
| "tokens_seen": 49152000 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0004956024210327218, | |
| "loss": 4.065, | |
| "theoretical_loss": 5.240578591827869, | |
| "tokens_seen": 52428800 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0004950113485908833, | |
| "loss": 3.9725, | |
| "theoretical_loss": 5.192916690583679, | |
| "tokens_seen": 55705600 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0004944202761490448, | |
| "loss": 3.9811, | |
| "theoretical_loss": 5.148714829414331, | |
| "tokens_seen": 58982400 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0004938292037072064, | |
| "loss": 3.8862, | |
| "theoretical_loss": 5.107549528462992, | |
| "tokens_seen": 62259200 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0004932381312653678, | |
| "loss": 3.7569, | |
| "theoretical_loss": 5.069068083201136, | |
| "tokens_seen": 65536000 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0004926470588235294, | |
| "loss": 3.6979, | |
| "theoretical_loss": 5.032973875895897, | |
| "tokens_seen": 68812800 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.000492055986381691, | |
| "loss": 3.5973, | |
| "theoretical_loss": 4.999015274282555, | |
| "tokens_seen": 72089600 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.0004914649139398525, | |
| "loss": 3.5246, | |
| "theoretical_loss": 4.966977121409853, | |
| "tokens_seen": 75366400 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.000490873841498014, | |
| "loss": 3.5081, | |
| "theoretical_loss": 4.936674127683027, | |
| "tokens_seen": 78643200 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.0004902827690561755, | |
| "loss": 3.4698, | |
| "theoretical_loss": 4.907945679887972, | |
| "tokens_seen": 81920000 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.0004896916966143371, | |
| "loss": 3.4115, | |
| "theoretical_loss": 4.8806517198708175, | |
| "tokens_seen": 85196800 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.0004891006241724985, | |
| "loss": 3.4193, | |
| "theoretical_loss": 4.85466944053967, | |
| "tokens_seen": 88473600 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.0004885095517306601, | |
| "loss": 3.3934, | |
| "theoretical_loss": 4.829890613366704, | |
| "tokens_seen": 91750400 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.0004879184792888217, | |
| "loss": 3.35, | |
| "theoretical_loss": 4.806219408835812, | |
| "tokens_seen": 95027200 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.0004873274068469832, | |
| "loss": 3.2861, | |
| "theoretical_loss": 4.783570605334148, | |
| "tokens_seen": 98304000 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.00048673633440514467, | |
| "loss": 3.2664, | |
| "theoretical_loss": 4.761868106830299, | |
| "tokens_seen": 101580800 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.0004861452619633062, | |
| "loss": 3.3073, | |
| "theoretical_loss": 4.741043708020364, | |
| "tokens_seen": 104857600 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.0004855541895214677, | |
| "loss": 3.2906, | |
| "theoretical_loss": 4.721036059306941, | |
| "tokens_seen": 108134400 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.0004849631170796293, | |
| "loss": 3.2249, | |
| "theoretical_loss": 4.701789794289756, | |
| "tokens_seen": 111411200 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.0004843720446377908, | |
| "loss": 3.2799, | |
| "theoretical_loss": 4.68325479029382, | |
| "tokens_seen": 114688000 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.00048378097219595233, | |
| "loss": 3.229, | |
| "theoretical_loss": 4.6653855384841725, | |
| "tokens_seen": 117964800 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.00048318989975411385, | |
| "loss": 3.2623, | |
| "theoretical_loss": 4.6481406047776295, | |
| "tokens_seen": 121241600 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.00048259882731227537, | |
| "loss": 3.2475, | |
| "theoretical_loss": 4.631482166397534, | |
| "tokens_seen": 124518400 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.0004820077548704369, | |
| "loss": 3.1819, | |
| "theoretical_loss": 4.615375611773608, | |
| "tokens_seen": 127795200 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.00048141668242859847, | |
| "loss": 3.2083, | |
| "theoretical_loss": 4.5997891937483955, | |
| "tokens_seen": 131072000 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.00048082560998676, | |
| "loss": 3.189, | |
| "theoretical_loss": 4.584693727850565, | |
| "tokens_seen": 134348800 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.0004802345375449215, | |
| "loss": 3.1967, | |
| "theoretical_loss": 4.570062328836407, | |
| "tokens_seen": 137625600 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.00047964346510308303, | |
| "loss": 3.1635, | |
| "theoretical_loss": 4.5558701798619285, | |
| "tokens_seen": 140902400 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.00047905239266124455, | |
| "loss": 3.1347, | |
| "theoretical_loss": 4.542094329588689, | |
| "tokens_seen": 144179200 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.00047846132021940607, | |
| "loss": 3.1606, | |
| "theoretical_loss": 4.528713513292708, | |
| "tokens_seen": 147456000 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.0004778702477775676, | |
| "loss": 3.0966, | |
| "theoretical_loss": 4.515707994672887, | |
| "tokens_seen": 150732800 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.00047727917533572917, | |
| "loss": 3.1029, | |
| "theoretical_loss": 4.503059425571229, | |
| "tokens_seen": 154009600 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.0004766881028938907, | |
| "loss": 3.1398, | |
| "theoretical_loss": 4.490750721243157, | |
| "tokens_seen": 157286400 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.0004760970304520522, | |
| "loss": 3.0796, | |
| "theoretical_loss": 4.478765949169858, | |
| "tokens_seen": 160563200 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "objective/train/docs_used": 96518, | |
| "objective/train/instantaneous_batch_size": 32, | |
| "objective/train/instantaneous_microbatch_size": 32768, | |
| "objective/train/original_loss": 3.2286458015441895, | |
| "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": 3.0725, | |
| "theoretical_loss": 4.46709022969911, | |
| "tokens_seen": 163840000 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.00047491488556837525, | |
| "loss": 3.0736, | |
| "theoretical_loss": 4.455709647047437, | |
| "tokens_seen": 167116800 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.00047432381312653677, | |
| "loss": 3.0957, | |
| "theoretical_loss": 4.444611169403321, | |
| "tokens_seen": 170393600 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.00047373274068469835, | |
| "loss": 3.0505, | |
| "theoretical_loss": 4.4337825770455375, | |
| "tokens_seen": 173670400 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.00047314166824285987, | |
| "loss": 3.0771, | |
| "theoretical_loss": 4.423212397538051, | |
| "tokens_seen": 176947200 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.0004725505958010214, | |
| "loss": 3.0805, | |
| "theoretical_loss": 4.4128898471880325, | |
| "tokens_seen": 180224000 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0004719595233591829, | |
| "loss": 3.0562, | |
| "theoretical_loss": 4.40280477805997, | |
| "tokens_seen": 183500800 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.00047136845091734443, | |
| "loss": 3.0368, | |
| "theoretical_loss": 4.392947629929752, | |
| "tokens_seen": 186777600 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.00047077737847550595, | |
| "loss": 3.0456, | |
| "theoretical_loss": 4.383309386640423, | |
| "tokens_seen": 190054400 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0004701863060336675, | |
| "loss": 3.0098, | |
| "theoretical_loss": 4.373881536388167, | |
| "tokens_seen": 193331200 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.00046959523359182905, | |
| "loss": 3.0174, | |
| "theoretical_loss": 4.364656035524595, | |
| "tokens_seen": 196608000 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.00046900416114999057, | |
| "loss": 3.0129, | |
| "theoretical_loss": 4.355625275511174, | |
| "tokens_seen": 199884800 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0004684130887081521, | |
| "loss": 2.9953, | |
| "theoretical_loss": 4.346782052704563, | |
| "tokens_seen": 203161600 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0004678220162663136, | |
| "loss": 2.9867, | |
| "theoretical_loss": 4.338119540689052, | |
| "tokens_seen": 206438400 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.00046723094382447513, | |
| "loss": 3.0165, | |
| "theoretical_loss": 4.329631264904703, | |
| "tokens_seen": 209715200 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.00046663987138263665, | |
| "loss": 3.0017, | |
| "theoretical_loss": 4.321311079348144, | |
| "tokens_seen": 212992000 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.0004660487989407982, | |
| "loss": 2.9951, | |
| "theoretical_loss": 4.313153145147683, | |
| "tokens_seen": 216268800 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.00046545772649895975, | |
| "loss": 2.9689, | |
| "theoretical_loss": 4.305151910836119, | |
| "tokens_seen": 219545600 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.00046486665405712127, | |
| "loss": 2.9826, | |
| "theoretical_loss": 4.2973020941635784, | |
| "tokens_seen": 222822400 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.0004642755816152828, | |
| "loss": 2.9874, | |
| "theoretical_loss": 4.28959866530949, | |
| "tokens_seen": 226099200 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.0004636845091734443, | |
| "loss": 2.9269, | |
| "theoretical_loss": 4.282036831367506, | |
| "tokens_seen": 229376000 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.00046309343673160583, | |
| "loss": 2.9501, | |
| "theoretical_loss": 4.274612021990189, | |
| "tokens_seen": 232652800 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.0004625023642897674, | |
| "loss": 2.9073, | |
| "theoretical_loss": 4.267319876091788, | |
| "tokens_seen": 235929600 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.0004619112918479289, | |
| "loss": 2.9524, | |
| "theoretical_loss": 4.260156229517635, | |
| "tokens_seen": 239206400 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.00046132021940609044, | |
| "loss": 2.9937, | |
| "theoretical_loss": 4.253117103597704, | |
| "tokens_seen": 242483200 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.00046072914696425197, | |
| "loss": 2.9917, | |
| "theoretical_loss": 4.246198694509945, | |
| "tokens_seen": 245760000 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.0004601380745224135, | |
| "loss": 2.967, | |
| "theoretical_loss": 4.239397363386152, | |
| "tokens_seen": 249036800 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.000459547002080575, | |
| "loss": 2.9536, | |
| "theoretical_loss": 4.232709627099522, | |
| "tokens_seen": 252313600 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.0004589559296387366, | |
| "loss": 2.9099, | |
| "theoretical_loss": 4.226132149678757, | |
| "tokens_seen": 255590400 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.0004583648571968981, | |
| "loss": 2.9495, | |
| "theoretical_loss": 4.219661734298666, | |
| "tokens_seen": 258867200 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.0004577737847550596, | |
| "loss": 2.9228, | |
| "theoretical_loss": 4.213295315801815, | |
| "tokens_seen": 262144000 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.0004571827123132211, | |
| "loss": 2.914, | |
| "theoretical_loss": 4.207029953709861, | |
| "tokens_seen": 265420800 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.0004565916398713826, | |
| "loss": 2.9065, | |
| "theoretical_loss": 4.200862825686893, | |
| "tokens_seen": 268697600 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.00045600056742954413, | |
| "loss": 2.9208, | |
| "theoretical_loss": 4.19479122142044, | |
| "tokens_seen": 271974400 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.0004554094949877057, | |
| "loss": 2.8755, | |
| "theoretical_loss": 4.188812536888775, | |
| "tokens_seen": 275251200 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.00045481842254586723, | |
| "loss": 2.911, | |
| "theoretical_loss": 4.182924268985855, | |
| "tokens_seen": 278528000 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.00045422735010402875, | |
| "loss": 2.8592, | |
| "theoretical_loss": 4.177124010477671, | |
| "tokens_seen": 281804800 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.00045363627766219027, | |
| "loss": 2.9063, | |
| "theoretical_loss": 4.171409445265983, | |
| "tokens_seen": 285081600 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.0004530452052203518, | |
| "loss": 2.8809, | |
| "theoretical_loss": 4.165778343937409, | |
| "tokens_seen": 288358400 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.0004524541327785133, | |
| "loss": 2.8394, | |
| "theoretical_loss": 4.160228559577659, | |
| "tokens_seen": 291635200 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.00045186306033667483, | |
| "loss": 2.8306, | |
| "theoretical_loss": 4.15475802383233, | |
| "tokens_seen": 294912000 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.0004512719878948364, | |
| "loss": 2.8102, | |
| "theoretical_loss": 4.149364743197177, | |
| "tokens_seen": 298188800 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.00045068091545299793, | |
| "loss": 2.8491, | |
| "theoretical_loss": 4.14404679552214, | |
| "tokens_seen": 301465600 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.00045008984301115945, | |
| "loss": 2.8152, | |
| "theoretical_loss": 4.138802326714632, | |
| "tokens_seen": 304742400 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.00044949877056932097, | |
| "loss": 2.8902, | |
| "theoretical_loss": 4.133629547628726, | |
| "tokens_seen": 308019200 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.0004489076981274825, | |
| "loss": 2.8874, | |
| "theoretical_loss": 4.128526731127894, | |
| "tokens_seen": 311296000 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.000448316625685644, | |
| "loss": 2.8282, | |
| "theoretical_loss": 4.123492209309923, | |
| "tokens_seen": 314572800 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.0004477255532438056, | |
| "loss": 2.8807, | |
| "theoretical_loss": 4.118524370883447, | |
| "tokens_seen": 317849600 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.0004471344808019671, | |
| "loss": 2.8404, | |
| "theoretical_loss": 4.113621658686355, | |
| "tokens_seen": 321126400 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.00044654340836012863, | |
| "loss": 2.8468, | |
| "theoretical_loss": 4.108782567337039, | |
| "tokens_seen": 324403200 | |
| }, | |
| { | |
| "debugging/Self-BLEU-5": 0.5179236306635556, | |
| "debugging/distinct-1-grams": 0.7868809037747271, | |
| "debugging/distinct-2-grams": 0.95444323584591, | |
| "debugging/entropy-1-grams": 5.810297153205875, | |
| "debugging/entropy-2-grams": 6.5112113932041344, | |
| "debugging/length": 540.5, | |
| "debugging/num_segments": 12, | |
| "epoch": 0.12, | |
| "objective/train/docs_used": 182435, | |
| "objective/train/instantaneous_batch_size": 32, | |
| "objective/train/instantaneous_microbatch_size": 32768, | |
| "objective/train/original_loss": 2.7323758602142334, | |
| "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": 2.8315, | |
| "theoretical_loss": 4.104005641010112, | |
| "tokens_seen": 327680000 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.00044536126347645167, | |
| "loss": 2.7719, | |
| "theoretical_loss": 4.099289471328812, | |
| "tokens_seen": 330956800 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.0004447701910346132, | |
| "loss": 2.8239, | |
| "theoretical_loss": 4.094632695366921, | |
| "tokens_seen": 334233600 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.00044417911859277476, | |
| "loss": 2.7712, | |
| "theoretical_loss": 4.090033993753448, | |
| "tokens_seen": 337510400 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.0004435880461509363, | |
| "loss": 2.7701, | |
| "theoretical_loss": 4.085492088873883, | |
| "tokens_seen": 340787200 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.0004429969737090978, | |
| "loss": 2.7771, | |
| "theoretical_loss": 4.081005743162224, | |
| "tokens_seen": 344064000 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.0004424059012672593, | |
| "loss": 2.7947, | |
| "theoretical_loss": 4.076573757478361, | |
| "tokens_seen": 347340800 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.00044181482882542085, | |
| "loss": 2.7985, | |
| "theoretical_loss": 4.072194969565807, | |
| "tokens_seen": 350617600 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.00044122375638358237, | |
| "loss": 2.8102, | |
| "theoretical_loss": 4.067868252585089, | |
| "tokens_seen": 353894400 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.0004406326839417439, | |
| "loss": 2.8096, | |
| "theoretical_loss": 4.063592513718411, | |
| "tokens_seen": 357171200 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.00044004161149990546, | |
| "loss": 2.7988, | |
| "theoretical_loss": 4.059366692841521, | |
| "tokens_seen": 360448000 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.000439450539058067, | |
| "loss": 2.8074, | |
| "theoretical_loss": 4.055189761258959, | |
| "tokens_seen": 363724800 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.0004388594666162285, | |
| "loss": 2.7982, | |
| "theoretical_loss": 4.051060720499127, | |
| "tokens_seen": 367001600 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.00043826839417439, | |
| "loss": 2.7748, | |
| "theoretical_loss": 4.046978601165831, | |
| "tokens_seen": 370278400 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.00043767732173255155, | |
| "loss": 2.7677, | |
| "theoretical_loss": 4.042942461843204, | |
| "tokens_seen": 373555200 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.00043708624929071307, | |
| "loss": 2.7828, | |
| "theoretical_loss": 4.038951388051044, | |
| "tokens_seen": 376832000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.00043649517684887464, | |
| "loss": 2.7974, | |
| "theoretical_loss": 4.035004491247873, | |
| "tokens_seen": 380108800 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.00043590410440703616, | |
| "loss": 2.7766, | |
| "theoretical_loss": 4.031100907879109, | |
| "tokens_seen": 383385600 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.0004353130319651977, | |
| "loss": 2.8236, | |
| "theoretical_loss": 4.02723979846797, | |
| "tokens_seen": 386662400 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.0004347219595233592, | |
| "loss": 2.829, | |
| "theoretical_loss": 4.023420346746835, | |
| "tokens_seen": 389939200 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.0004341308870815207, | |
| "loss": 2.7938, | |
| "theoretical_loss": 4.019641758826938, | |
| "tokens_seen": 393216000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.00043353981463968225, | |
| "loss": 2.7822, | |
| "theoretical_loss": 4.015903262404413, | |
| "tokens_seen": 396492800 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.0004329487421978438, | |
| "loss": 2.7694, | |
| "theoretical_loss": 4.012204106000786, | |
| "tokens_seen": 399769600 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.00043235766975600534, | |
| "loss": 2.7842, | |
| "theoretical_loss": 4.008543558236181, | |
| "tokens_seen": 403046400 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.00043176659731416686, | |
| "loss": 2.7663, | |
| "theoretical_loss": 4.004920907133565, | |
| "tokens_seen": 406323200 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0004311755248723284, | |
| "loss": 2.7278, | |
| "theoretical_loss": 4.001335459452449, | |
| "tokens_seen": 409600000 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0004305844524304899, | |
| "loss": 2.7589, | |
| "theoretical_loss": 3.997786540050617, | |
| "tokens_seen": 412876800 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0004299933799886514, | |
| "loss": 2.7719, | |
| "theoretical_loss": 3.9942734912724456, | |
| "tokens_seen": 416153600 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.00042940230754681295, | |
| "loss": 2.7241, | |
| "theoretical_loss": 3.9907956723625375, | |
| "tokens_seen": 419430400 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0004288112351049745, | |
| "loss": 2.7718, | |
| "theoretical_loss": 3.9873524589034224, | |
| "tokens_seen": 422707200 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.000428220162663136, | |
| "loss": 2.765, | |
| "theoretical_loss": 3.9839432422761556, | |
| "tokens_seen": 425984000 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0004276290902212975, | |
| "loss": 2.804, | |
| "theoretical_loss": 3.980567429142721, | |
| "tokens_seen": 429260800 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.00042703801777945903, | |
| "loss": 2.7606, | |
| "theoretical_loss": 3.977224440949197, | |
| "tokens_seen": 432537600 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.00042644694533762055, | |
| "loss": 2.772, | |
| "theoretical_loss": 3.9739137134486917, | |
| "tokens_seen": 435814400 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.00042585587289578207, | |
| "loss": 2.8015, | |
| "theoretical_loss": 3.9706346962431396, | |
| "tokens_seen": 439091200 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.00042526480045394365, | |
| "loss": 2.8007, | |
| "theoretical_loss": 3.9673868523430564, | |
| "tokens_seen": 442368000 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.00042467372801210517, | |
| "loss": 2.7835, | |
| "theoretical_loss": 3.9641696577444376, | |
| "tokens_seen": 445644800 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.0004240826555702667, | |
| "loss": 2.7811, | |
| "theoretical_loss": 3.9609826010220033, | |
| "tokens_seen": 448921600 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.0004234915831284282, | |
| "loss": 2.8235, | |
| "theoretical_loss": 3.9578251829380506, | |
| "tokens_seen": 452198400 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.00042290051068658973, | |
| "loss": 2.7819, | |
| "theoretical_loss": 3.954696916066199, | |
| "tokens_seen": 455475200 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.00042230943824475125, | |
| "loss": 2.7632, | |
| "theoretical_loss": 3.9515973244293643, | |
| "tokens_seen": 458752000 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.0004217183658029128, | |
| "loss": 2.7585, | |
| "theoretical_loss": 3.948525943151326, | |
| "tokens_seen": 462028800 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.00042112729336107435, | |
| "loss": 2.7676, | |
| "theoretical_loss": 3.9454823181212815, | |
| "tokens_seen": 465305600 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.00042053622091923587, | |
| "loss": 2.7674, | |
| "theoretical_loss": 3.9424660056708167, | |
| "tokens_seen": 468582400 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.0004199451484773974, | |
| "loss": 2.7598, | |
| "theoretical_loss": 3.939476572262754, | |
| "tokens_seen": 471859200 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.0004193540760355589, | |
| "loss": 2.761, | |
| "theoretical_loss": 3.9365135941913563, | |
| "tokens_seen": 475136000 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.00041876300359372043, | |
| "loss": 2.7423, | |
| "theoretical_loss": 3.9335766572934023, | |
| "tokens_seen": 478412800 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.00041817193115188195, | |
| "loss": 2.7487, | |
| "theoretical_loss": 3.9306653566696603, | |
| "tokens_seen": 481689600 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.0004175808587100435, | |
| "loss": 2.7297, | |
| "theoretical_loss": 3.927779296416332, | |
| "tokens_seen": 484966400 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.00041698978626820505, | |
| "loss": 2.7348, | |
| "theoretical_loss": 3.924918089366024, | |
| "tokens_seen": 488243200 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "objective/train/docs_used": 267104, | |
| "objective/train/instantaneous_batch_size": 32, | |
| "objective/train/instantaneous_microbatch_size": 32768, | |
| "objective/train/original_loss": 2.6938159465789795, | |
| "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": 2.6909, | |
| "theoretical_loss": 3.9220813568378707, | |
| "tokens_seen": 491520000 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.0004158076413845281, | |
| "loss": 2.6957, | |
| "theoretical_loss": 3.9192687283964096, | |
| "tokens_seen": 494796800 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.0004152165689426896, | |
| "loss": 2.7163, | |
| "theoretical_loss": 3.9164798416188527, | |
| "tokens_seen": 498073600 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.00041462549650085113, | |
| "loss": 2.6846, | |
| "theoretical_loss": 3.913714341870409, | |
| "tokens_seen": 501350400 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.0004140344240590127, | |
| "loss": 2.6939, | |
| "theoretical_loss": 3.9109718820873303, | |
| "tokens_seen": 504627200 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.0004134433516171742, | |
| "loss": 2.7024, | |
| "theoretical_loss": 3.9082521225673625, | |
| "tokens_seen": 507904000 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.00041285227917533575, | |
| "loss": 2.6622, | |
| "theoretical_loss": 3.9055547307673075, | |
| "tokens_seen": 511180800 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.00041226120673349727, | |
| "loss": 2.7267, | |
| "theoretical_loss": 3.9028793811074056, | |
| "tokens_seen": 514457600 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.0004116701342916588, | |
| "loss": 2.7517, | |
| "theoretical_loss": 3.900225754782274, | |
| "tokens_seen": 517734400 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.0004110790618498203, | |
| "loss": 2.7289, | |
| "theoretical_loss": 3.897593539578138, | |
| "tokens_seen": 521011200 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.0004104879894079819, | |
| "loss": 2.7181, | |
| "theoretical_loss": 3.8949824296961015, | |
| "tokens_seen": 524288000 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.0004098969169661434, | |
| "loss": 2.7306, | |
| "theoretical_loss": 3.8923921255812353, | |
| "tokens_seen": 527564800 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.0004093058445243049, | |
| "loss": 2.7015, | |
| "theoretical_loss": 3.8898223337572393, | |
| "tokens_seen": 530841600 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.00040871477208246645, | |
| "loss": 2.6978, | |
| "theoretical_loss": 3.88727276666648, | |
| "tokens_seen": 534118400 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.00040812369964062797, | |
| "loss": 2.6723, | |
| "theoretical_loss": 3.884743142515184, | |
| "tokens_seen": 537395200 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.0004075326271987895, | |
| "loss": 2.6872, | |
| "theoretical_loss": 3.8822331851235985, | |
| "tokens_seen": 540672000 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.000406941554756951, | |
| "loss": 2.6913, | |
| "theoretical_loss": 3.87974262378093, | |
| "tokens_seen": 543948800 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0004063504823151126, | |
| "loss": 2.6513, | |
| "theoretical_loss": 3.877271193104873, | |
| "tokens_seen": 547225600 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0004057594098732741, | |
| "loss": 2.6908, | |
| "theoretical_loss": 3.8748186329055736, | |
| "tokens_seen": 550502400 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0004051801588802724, | |
| "loss": 2.7044, | |
| "theoretical_loss": 3.87238468805384, | |
| "tokens_seen": 553779200 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0004045890864384339, | |
| "loss": 2.6751, | |
| "theoretical_loss": 3.8699691083534633, | |
| "tokens_seen": 557056000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0004039980139965954, | |
| "loss": 2.6641, | |
| "theoretical_loss": 3.8675716484174907, | |
| "tokens_seen": 560332800 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.00040340694155475694, | |
| "loss": 2.6713, | |
| "theoretical_loss": 3.8651920675482936, | |
| "tokens_seen": 563609600 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0004028158691129185, | |
| "loss": 2.6941, | |
| "theoretical_loss": 3.862830129621318, | |
| "tokens_seen": 566886400 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.00040222479667108003, | |
| "loss": 2.7425, | |
| "theoretical_loss": 3.8604856029723575, | |
| "tokens_seen": 570163200 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.00040163372422924155, | |
| "loss": 2.6816, | |
| "theoretical_loss": 3.8581582602882447, | |
| "tokens_seen": 573440000 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.0004010426517874031, | |
| "loss": 2.7, | |
| "theoretical_loss": 3.8558478785008203, | |
| "tokens_seen": 576716800 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.0004004515793455646, | |
| "loss": 2.7188, | |
| "theoretical_loss": 3.8535542386840778, | |
| "tokens_seen": 579993600 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.0003998605069037261, | |
| "loss": 2.714, | |
| "theoretical_loss": 3.8512771259543586, | |
| "tokens_seen": 583270400 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.0003992694344618877, | |
| "loss": 2.7144, | |
| "theoretical_loss": 3.8490163293735082, | |
| "tokens_seen": 586547200 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.0003986783620200492, | |
| "loss": 2.7168, | |
| "theoretical_loss": 3.8467716418548648, | |
| "tokens_seen": 589824000 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.00039808728957821073, | |
| "loss": 2.7048, | |
| "theoretical_loss": 3.844542860072007, | |
| "tokens_seen": 593100800 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.00039749621713637225, | |
| "loss": 2.7227, | |
| "theoretical_loss": 3.8423297843701496, | |
| "tokens_seen": 596377600 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.0003969051446945338, | |
| "loss": 2.7137, | |
| "theoretical_loss": 3.8401322186800995, | |
| "tokens_seen": 599654400 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0003963140722526953, | |
| "loss": 2.7505, | |
| "theoretical_loss": 3.83794997043469, | |
| "tokens_seen": 602931200 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.00039572299981085687, | |
| "loss": 2.7533, | |
| "theoretical_loss": 3.8357828504876004, | |
| "tokens_seen": 606208000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0003951319273690184, | |
| "loss": 2.7222, | |
| "theoretical_loss": 3.833630673034487, | |
| "tokens_seen": 609484800 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0003945408549271799, | |
| "loss": 2.7234, | |
| "theoretical_loss": 3.831493255536345, | |
| "tokens_seen": 612761600 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.00039394978248534143, | |
| "loss": 2.7374, | |
| "theoretical_loss": 3.8293704186450253, | |
| "tokens_seen": 616038400 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.00039335871004350295, | |
| "loss": 2.7151, | |
| "theoretical_loss": 3.827261986130839, | |
| "tokens_seen": 619315200 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0003927676376016645, | |
| "loss": 2.6627, | |
| "theoretical_loss": 3.825167784812175, | |
| "tokens_seen": 622592000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.000392176565159826, | |
| "loss": 2.69, | |
| "theoretical_loss": 3.823087644487069, | |
| "tokens_seen": 625868800 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.00039158549271798757, | |
| "loss": 2.6656, | |
| "theoretical_loss": 3.8210213978666565, | |
| "tokens_seen": 629145600 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0003909944202761491, | |
| "loss": 2.6679, | |
| "theoretical_loss": 3.8189688805104476, | |
| "tokens_seen": 632422400 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.00039040334783431056, | |
| "loss": 2.6709, | |
| "theoretical_loss": 3.816929930763374, | |
| "tokens_seen": 635699200 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0003898122753924721, | |
| "loss": 2.6848, | |
| "theoretical_loss": 3.8149043896945347, | |
| "tokens_seen": 638976000 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0003892212029506336, | |
| "loss": 2.6777, | |
| "theoretical_loss": 3.812892101037601, | |
| "tokens_seen": 642252800 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0003886301305087951, | |
| "loss": 2.6847, | |
| "theoretical_loss": 3.81089291113282, | |
| "tokens_seen": 645529600 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0003880390580669567, | |
| "loss": 2.6691, | |
| "theoretical_loss": 3.8089066688705673, | |
| "tokens_seen": 648806400 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0003874479856251182, | |
| "loss": 2.6569, | |
| "theoretical_loss": 3.8069332256363992, | |
| "tokens_seen": 652083200 | |
| }, | |
| { | |
| "debugging/Self-BLEU-5": 0.642861845651988, | |
| "debugging/distinct-1-grams": 0.8145672280225638, | |
| "debugging/distinct-2-grams": 0.980694270616048, | |
| "debugging/entropy-1-grams": 5.8621094281307204, | |
| "debugging/entropy-2-grams": 6.737099920464704, | |
| "debugging/length": 648.6666666666666, | |
| "debugging/num_segments": 12, | |
| "epoch": 0.23, | |
| "objective/train/docs_used": 351924, | |
| "objective/train/instantaneous_batch_size": 32, | |
| "objective/train/instantaneous_microbatch_size": 32768, | |
| "objective/train/original_loss": 2.6956400871276855, | |
| "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": 2.6514, | |
| "theoretical_loss": 3.80497243525756, | |
| "tokens_seen": 655360000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.00038626584074144126, | |
| "loss": 2.6215, | |
| "theoretical_loss": 3.8030241539508958, | |
| "tokens_seen": 658636800 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.0003856747682996028, | |
| "loss": 2.619, | |
| "theoretical_loss": 3.8010882402721324, | |
| "tokens_seen": 661913600 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.0003850836958577643, | |
| "loss": 2.656, | |
| "theoretical_loss": 3.7991645550664757, | |
| "tokens_seen": 665190400 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.0003844926234159259, | |
| "loss": 2.6818, | |
| "theoretical_loss": 3.797252961420492, | |
| "tokens_seen": 668467200 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.0003839015509740874, | |
| "loss": 2.6846, | |
| "theoretical_loss": 3.795353324615228, | |
| "tokens_seen": 671744000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.0003833104785322489, | |
| "loss": 2.7187, | |
| "theoretical_loss": 3.793465512080541, | |
| "tokens_seen": 675020800 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.00038271940609041044, | |
| "loss": 2.7152, | |
| "theoretical_loss": 3.791589393350587, | |
| "tokens_seen": 678297600 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.00038212833364857196, | |
| "loss": 2.6926, | |
| "theoretical_loss": 3.7897248400204475, | |
| "tokens_seen": 681574400 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.0003815372612067335, | |
| "loss": 2.6837, | |
| "theoretical_loss": 3.7878717257038534, | |
| "tokens_seen": 684851200 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.000380946188764895, | |
| "loss": 2.6716, | |
| "theoretical_loss": 3.7860299259919685, | |
| "tokens_seen": 688128000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.0003803551163230566, | |
| "loss": 2.6557, | |
| "theoretical_loss": 3.7841993184132114, | |
| "tokens_seen": 691404800 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.0003797640438812181, | |
| "loss": 2.6455, | |
| "theoretical_loss": 3.78237978239408, | |
| "tokens_seen": 694681600 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.0003791729714393796, | |
| "loss": 2.6635, | |
| "theoretical_loss": 3.780571199220942, | |
| "tokens_seen": 697958400 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.00037858189899754114, | |
| "loss": 2.638, | |
| "theoretical_loss": 3.7787734520027803, | |
| "tokens_seen": 701235200 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.00037799082655570266, | |
| "loss": 2.6455, | |
| "theoretical_loss": 3.7769864256348455, | |
| "tokens_seen": 704512000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.0003773997541138642, | |
| "loss": 2.6937, | |
| "theoretical_loss": 3.775210006763202, | |
| "tokens_seen": 707788800 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.00037680868167202575, | |
| "loss": 2.6785, | |
| "theoretical_loss": 3.7734440837501406, | |
| "tokens_seen": 711065600 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.0003762176092301873, | |
| "loss": 2.7156, | |
| "theoretical_loss": 3.7716885466404246, | |
| "tokens_seen": 714342400 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.0003756265367883488, | |
| "loss": 2.7095, | |
| "theoretical_loss": 3.769943287128357, | |
| "tokens_seen": 717619200 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.0003750354643465103, | |
| "loss": 2.7089, | |
| "theoretical_loss": 3.7682081985256364, | |
| "tokens_seen": 720896000 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.00037444439190467184, | |
| "loss": 2.7286, | |
| "theoretical_loss": 3.7664831757299795, | |
| "tokens_seen": 724172800 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.00037385331946283336, | |
| "loss": 2.6792, | |
| "theoretical_loss": 3.7647681151944976, | |
| "tokens_seen": 727449600 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.00037326224702099493, | |
| "loss": 2.6448, | |
| "theoretical_loss": 3.7630629148977937, | |
| "tokens_seen": 730726400 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.00037267117457915645, | |
| "loss": 2.6615, | |
| "theoretical_loss": 3.761367474314768, | |
| "tokens_seen": 734003200 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.000372080102137318, | |
| "loss": 2.6745, | |
| "theoretical_loss": 3.7596816943881084, | |
| "tokens_seen": 737280000 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.0003714890296954795, | |
| "loss": 2.6448, | |
| "theoretical_loss": 3.758005477500451, | |
| "tokens_seen": 740556800 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.000370897957253641, | |
| "loss": 2.6424, | |
| "theoretical_loss": 3.756338727447186, | |
| "tokens_seen": 743833600 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.00037030688481180254, | |
| "loss": 2.6773, | |
| "theoretical_loss": 3.7546813494098945, | |
| "tokens_seen": 747110400 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.00036971581236996406, | |
| "loss": 2.6502, | |
| "theoretical_loss": 3.7530332499304007, | |
| "tokens_seen": 750387200 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.00036912473992812563, | |
| "loss": 2.673, | |
| "theoretical_loss": 3.7513943368854195, | |
| "tokens_seen": 753664000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.00036853366748628715, | |
| "loss": 2.6322, | |
| "theoretical_loss": 3.7497645194617863, | |
| "tokens_seen": 756940800 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.0003679425950444487, | |
| "loss": 2.602, | |
| "theoretical_loss": 3.748143708132246, | |
| "tokens_seen": 760217600 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.0003673515226026102, | |
| "loss": 2.6187, | |
| "theoretical_loss": 3.7465318146317994, | |
| "tokens_seen": 763494400 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.0003667604501607717, | |
| "loss": 2.587, | |
| "theoretical_loss": 3.7449287519345766, | |
| "tokens_seen": 766771200 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.00036616937771893324, | |
| "loss": 2.5946, | |
| "theoretical_loss": 3.7433344342312385, | |
| "tokens_seen": 770048000 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.0003655783052770948, | |
| "loss": 2.6269, | |
| "theoretical_loss": 3.7417487769068756, | |
| "tokens_seen": 773324800 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.00036498723283525633, | |
| "loss": 2.6403, | |
| "theoretical_loss": 3.7401716965194076, | |
| "tokens_seen": 776601600 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.00036439616039341785, | |
| "loss": 2.6323, | |
| "theoretical_loss": 3.738603110778461, | |
| "tokens_seen": 779878400 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.0003638050879515794, | |
| "loss": 2.6448, | |
| "theoretical_loss": 3.73704293852471, | |
| "tokens_seen": 783155200 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.0003632140155097409, | |
| "loss": 2.6451, | |
| "theoretical_loss": 3.7354910997096793, | |
| "tokens_seen": 786432000 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.0003626229430679024, | |
| "loss": 2.656, | |
| "theoretical_loss": 3.7339475153759825, | |
| "tokens_seen": 789708800 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.000362031870626064, | |
| "loss": 2.611, | |
| "theoretical_loss": 3.732412107638, | |
| "tokens_seen": 792985600 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.00036144079818422546, | |
| "loss": 2.6138, | |
| "theoretical_loss": 3.7308847996629724, | |
| "tokens_seen": 796262400 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.000360849725742387, | |
| "loss": 2.6362, | |
| "theoretical_loss": 3.7293655156525043, | |
| "tokens_seen": 799539200 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0003602586533005485, | |
| "loss": 2.6161, | |
| "theoretical_loss": 3.727854180824469, | |
| "tokens_seen": 802816000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.00035966758085871, | |
| "loss": 2.5601, | |
| "theoretical_loss": 3.7263507213952978, | |
| "tokens_seen": 806092800 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.00035907650841687154, | |
| "loss": 2.584, | |
| "theoretical_loss": 3.724855064562658, | |
| "tokens_seen": 809369600 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.00035848543597503306, | |
| "loss": 2.6023, | |
| "theoretical_loss": 3.723367138488488, | |
| "tokens_seen": 812646400 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.00035789436353319464, | |
| "loss": 2.6156, | |
| "theoretical_loss": 3.7218868722824014, | |
| "tokens_seen": 815923200 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "objective/train/docs_used": 436767, | |
| "objective/train/instantaneous_batch_size": 32, | |
| "objective/train/instantaneous_microbatch_size": 32768, | |
| "objective/train/original_loss": 2.712568759918213, | |
| "objective/train/theoretical_loss": 3.7204141959854384, | |
| "objective/train/tokens_used": 839660000, | |
| "theoretical_loss": 3.7204141959854384, | |
| "tokens_seen": 819200000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.00035730329109135616, | |
| "loss": 2.6219, | |
| "theoretical_loss": 3.7204141959854384, | |
| "tokens_seen": 819200000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0003567122186495177, | |
| "loss": 2.6477, | |
| "theoretical_loss": 3.718949040554162, | |
| "tokens_seen": 822476800 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0003561211462076792, | |
| "loss": 2.6576, | |
| "theoretical_loss": 3.7174913378450833, | |
| "tokens_seen": 825753600 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.0003555300737658407, | |
| "loss": 2.6582, | |
| "theoretical_loss": 3.7160410205994183, | |
| "tokens_seen": 829030400 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.00035493900132400224, | |
| "loss": 2.6759, | |
| "theoretical_loss": 3.7145980224281585, | |
| "tokens_seen": 832307200 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.0003543597503310006, | |
| "loss": 2.6768, | |
| "theoretical_loss": 3.713162277797449, | |
| "tokens_seen": 835584000 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.00035376867788916214, | |
| "loss": 2.6422, | |
| "theoretical_loss": 3.7117337220142748, | |
| "tokens_seen": 838860800 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.0003531894268961604, | |
| "loss": 2.6515, | |
| "theoretical_loss": 3.7103122912124364, | |
| "tokens_seen": 842137600 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.0003526101759031587, | |
| "loss": 2.6762, | |
| "theoretical_loss": 3.7088979223388128, | |
| "tokens_seen": 845414400 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.0003520191034613202, | |
| "loss": 2.6563, | |
| "theoretical_loss": 3.70749055313991, | |
| "tokens_seen": 848691200 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.0003514280310194817, | |
| "loss": 2.6461, | |
| "theoretical_loss": 3.7060901221486766, | |
| "tokens_seen": 851968000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0003508369585776433, | |
| "loss": 2.6557, | |
| "theoretical_loss": 3.704696568671591, | |
| "tokens_seen": 855244800 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0003502458861358048, | |
| "loss": 2.6539, | |
| "theoretical_loss": 3.7033098327760063, | |
| "tokens_seen": 858521600 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.00034965481369396634, | |
| "loss": 2.632, | |
| "theoretical_loss": 3.7019298552777533, | |
| "tokens_seen": 861798400 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.00034906374125212786, | |
| "loss": 2.619, | |
| "theoretical_loss": 3.700556577728988, | |
| "tokens_seen": 865075200 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0003484726688102894, | |
| "loss": 2.629, | |
| "theoretical_loss": 3.6991899424062815, | |
| "tokens_seen": 868352000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0003478815963684509, | |
| "loss": 2.609, | |
| "theoretical_loss": 3.697829892298951, | |
| "tokens_seen": 871628800 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0003472905239266125, | |
| "loss": 2.6334, | |
| "theoretical_loss": 3.696476371097618, | |
| "tokens_seen": 874905600 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.000346699451484774, | |
| "loss": 2.6014, | |
| "theoretical_loss": 3.695129323182993, | |
| "tokens_seen": 878182400 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0003461083790429355, | |
| "loss": 2.5987, | |
| "theoretical_loss": 3.693788693614879, | |
| "tokens_seen": 881459200 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.00034551730660109704, | |
| "loss": 2.5732, | |
| "theoretical_loss": 3.6924544281213967, | |
| "tokens_seen": 884736000 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.00034492623415925856, | |
| "loss": 2.5476, | |
| "theoretical_loss": 3.691126473088412, | |
| "tokens_seen": 888012800 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.0003443351617174201, | |
| "loss": 2.5573, | |
| "theoretical_loss": 3.689804775549173, | |
| "tokens_seen": 891289600 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.0003437440892755816, | |
| "loss": 2.5655, | |
| "theoretical_loss": 3.688489283174146, | |
| "tokens_seen": 894566400 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.0003431530168337432, | |
| "loss": 2.6253, | |
| "theoretical_loss": 3.6871799442610538, | |
| "tokens_seen": 897843200 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.0003425619443919047, | |
| "loss": 2.6272, | |
| "theoretical_loss": 3.685876707725093, | |
| "tokens_seen": 901120000 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.0003419708719500662, | |
| "loss": 2.5663, | |
| "theoretical_loss": 3.6845795230893517, | |
| "tokens_seen": 904396800 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.00034137979950822774, | |
| "loss": 2.5763, | |
| "theoretical_loss": 3.6832883404754035, | |
| "tokens_seen": 907673600 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.00034078872706638926, | |
| "loss": 2.5661, | |
| "theoretical_loss": 3.6820031105940796, | |
| "tokens_seen": 910950400 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.0003401976546245508, | |
| "loss": 2.565, | |
| "theoretical_loss": 3.6807237847364176, | |
| "tokens_seen": 914227200 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.00033960658218271236, | |
| "loss": 2.6169, | |
| "theoretical_loss": 3.6794503147647846, | |
| "tokens_seen": 917504000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.0003390155097408739, | |
| "loss": 2.6209, | |
| "theoretical_loss": 3.67818265310416, | |
| "tokens_seen": 920780800 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.0003384244372990354, | |
| "loss": 2.602, | |
| "theoretical_loss": 3.6769207527335888, | |
| "tokens_seen": 924057600 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.0003378333648571969, | |
| "loss": 2.597, | |
| "theoretical_loss": 3.675664567177787, | |
| "tokens_seen": 927334400 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.00033724229241535844, | |
| "loss": 2.596, | |
| "theoretical_loss": 3.674414050498913, | |
| "tokens_seen": 930611200 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.00033665121997351996, | |
| "loss": 2.5945, | |
| "theoretical_loss": 3.6731691572884824, | |
| "tokens_seen": 933888000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.00033606014753168154, | |
| "loss": 2.5671, | |
| "theoretical_loss": 3.671929842659438, | |
| "tokens_seen": 937164800 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.00033546907508984306, | |
| "loss": 2.5612, | |
| "theoretical_loss": 3.6706960622383624, | |
| "tokens_seen": 940441600 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.0003348780026480046, | |
| "loss": 2.5934, | |
| "theoretical_loss": 3.6694677721578377, | |
| "tokens_seen": 943718400 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.0003342869302061661, | |
| "loss": 2.59, | |
| "theoretical_loss": 3.66824492904894, | |
| "tokens_seen": 946995200 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.00033369585776432757, | |
| "loss": 2.5596, | |
| "theoretical_loss": 3.667027490033874, | |
| "tokens_seen": 950272000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.0003331047853224891, | |
| "loss": 2.5686, | |
| "theoretical_loss": 3.6658154127187412, | |
| "tokens_seen": 953548800 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.0003325137128806506, | |
| "loss": 2.5853, | |
| "theoretical_loss": 3.664608655186437, | |
| "tokens_seen": 956825600 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.0003319226404388122, | |
| "loss": 2.5926, | |
| "theoretical_loss": 3.663407175989679, | |
| "tokens_seen": 960102400 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.0003313315679969737, | |
| "loss": 2.5611, | |
| "theoretical_loss": 3.662210934144158, | |
| "tokens_seen": 963379200 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.0003307404955551352, | |
| "loss": 2.587, | |
| "theoretical_loss": 3.661019889121812, | |
| "tokens_seen": 966656000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.00033014942311329675, | |
| "loss": 2.6005, | |
| "theoretical_loss": 3.6598340008442234, | |
| "tokens_seen": 969932800 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.00032955835067145827, | |
| "loss": 2.5902, | |
| "theoretical_loss": 3.6586532296761285, | |
| "tokens_seen": 973209600 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.0003289672782296198, | |
| "loss": 2.6145, | |
| "theoretical_loss": 3.657477536419047, | |
| "tokens_seen": 976486400 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.00032837620578778136, | |
| "loss": 2.6437, | |
| "theoretical_loss": 3.656306882305022, | |
| "tokens_seen": 979763200 | |
| }, | |
| { | |
| "debugging/Self-BLEU-5": 0.6198785821735879, | |
| "debugging/distinct-1-grams": 0.7707728209964196, | |
| "debugging/distinct-2-grams": 0.9604165709732011, | |
| "debugging/entropy-1-grams": 5.836498792012382, | |
| "debugging/entropy-2-grams": 6.692653651879306, | |
| "debugging/length": 640.8333333333334, | |
| "debugging/num_segments": 12, | |
| "epoch": 0.35, | |
| "objective/train/docs_used": 522194, | |
| "objective/train/instantaneous_batch_size": 32, | |
| "objective/train/instantaneous_microbatch_size": 32768, | |
| "objective/train/original_loss": 2.5617499351501465, | |
| "objective/train/theoretical_loss": 3.6551412289904697, | |
| "objective/train/tokens_used": 1003500000, | |
| "theoretical_loss": 3.6551412289904697, | |
| "tokens_seen": 983040000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.0003277851333459429, | |
| "loss": 2.5923, | |
| "theoretical_loss": 3.6551412289904697, | |
| "tokens_seen": 983040000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.0003271940609041044, | |
| "loss": 2.5912, | |
| "theoretical_loss": 3.6539805385501376, | |
| "tokens_seen": 986316800 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.0003266029884622659, | |
| "loss": 2.6037, | |
| "theoretical_loss": 3.652824773471171, | |
| "tokens_seen": 989593600 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.00032601191602042745, | |
| "loss": 2.5777, | |
| "theoretical_loss": 3.651673896647277, | |
| "tokens_seen": 992870400 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.00032542084357858897, | |
| "loss": 2.5642, | |
| "theoretical_loss": 3.6505278713729985, | |
| "tokens_seen": 996147200 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.00032482977113675054, | |
| "loss": 2.5653, | |
| "theoretical_loss": 3.6493866613380774, | |
| "tokens_seen": 999424000 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.00032423869869491206, | |
| "loss": 2.5584, | |
| "theoretical_loss": 3.648250230621924, | |
| "tokens_seen": 1002700800 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0003236476262530736, | |
| "loss": 2.5586, | |
| "theoretical_loss": 3.647118543688179, | |
| "tokens_seen": 1005977600 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0003230565538112351, | |
| "loss": 2.5333, | |
| "theoretical_loss": 3.6459915653793633, | |
| "tokens_seen": 1009254400 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0003224654813693966, | |
| "loss": 2.5645, | |
| "theoretical_loss": 3.644869260911628, | |
| "tokens_seen": 1012531200 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.00032187440892755814, | |
| "loss": 2.5094, | |
| "theoretical_loss": 3.64375159586959, | |
| "tokens_seen": 1015808000 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0003212833364857197, | |
| "loss": 2.5334, | |
| "theoretical_loss": 3.642638536201252, | |
| "tokens_seen": 1019084800 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.00032069226404388124, | |
| "loss": 2.5607, | |
| "theoretical_loss": 3.6415300482130135, | |
| "tokens_seen": 1022361600 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.00032010119160204276, | |
| "loss": 2.5815, | |
| "theoretical_loss": 3.6404260985647667, | |
| "tokens_seen": 1025638400 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.0003195101191602043, | |
| "loss": 2.531, | |
| "theoretical_loss": 3.6393266542650684, | |
| "tokens_seen": 1028915200 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.0003189190467183658, | |
| "loss": 2.5341, | |
| "theoretical_loss": 3.638231682666401, | |
| "tokens_seen": 1032192000 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.0003183279742765273, | |
| "loss": 2.5545, | |
| "theoretical_loss": 3.637141151460505, | |
| "tokens_seen": 1035468800 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.00031773690183468884, | |
| "loss": 2.5548, | |
| "theoretical_loss": 3.636055028673799, | |
| "tokens_seen": 1038745600 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.0003171458293928504, | |
| "loss": 2.5728, | |
| "theoretical_loss": 3.634973282662864, | |
| "tokens_seen": 1042022400 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.00031655475695101194, | |
| "loss": 2.6221, | |
| "theoretical_loss": 3.6338958821100107, | |
| "tokens_seen": 1045299200 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.00031596368450917346, | |
| "loss": 2.6254, | |
| "theoretical_loss": 3.63282279601892, | |
| "tokens_seen": 1048576000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.000315372612067335, | |
| "loss": 2.6667, | |
| "theoretical_loss": 3.631753993710352, | |
| "tokens_seen": 1051852800 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.0003147815396254965, | |
| "loss": 2.6569, | |
| "theoretical_loss": 3.630689444817925, | |
| "tokens_seen": 1055129600 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.000314190467183658, | |
| "loss": 2.648, | |
| "theoretical_loss": 3.629629119283967, | |
| "tokens_seen": 1058406400 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.0003135993947418196, | |
| "loss": 2.668, | |
| "theoretical_loss": 3.628572987355434, | |
| "tokens_seen": 1061683200 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.0003130083222999811, | |
| "loss": 2.6364, | |
| "theoretical_loss": 3.6275210195798913, | |
| "tokens_seen": 1064960000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.00031241724985814264, | |
| "loss": 2.6346, | |
| "theoretical_loss": 3.626473186801564, | |
| "tokens_seen": 1068236800 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.00031182617741630416, | |
| "loss": 2.6531, | |
| "theoretical_loss": 3.6254294601574495, | |
| "tokens_seen": 1071513600 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.0003112351049744657, | |
| "loss": 2.628, | |
| "theoretical_loss": 3.624389811073493, | |
| "tokens_seen": 1074790400 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.0003106440325326272, | |
| "loss": 2.6233, | |
| "theoretical_loss": 3.6233542112608257, | |
| "tokens_seen": 1078067200 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.0003100529600907888, | |
| "loss": 2.6271, | |
| "theoretical_loss": 3.6223226327120592, | |
| "tokens_seen": 1081344000 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.0003094618876489503, | |
| "loss": 2.6132, | |
| "theoretical_loss": 3.621295047697644, | |
| "tokens_seen": 1084620800 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.0003088708152071118, | |
| "loss": 2.6227, | |
| "theoretical_loss": 3.6202714287622833, | |
| "tokens_seen": 1087897600 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.00030827974276527334, | |
| "loss": 2.6147, | |
| "theoretical_loss": 3.6192517487214038, | |
| "tokens_seen": 1091174400 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.00030768867032343486, | |
| "loss": 2.5954, | |
| "theoretical_loss": 3.6182359806576834, | |
| "tokens_seen": 1094451200 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.0003070975978815964, | |
| "loss": 2.6142, | |
| "theoretical_loss": 3.6172240979176333, | |
| "tokens_seen": 1097728000 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.0003065065254397579, | |
| "loss": 2.5959, | |
| "theoretical_loss": 3.616216074108232, | |
| "tokens_seen": 1101004800 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.0003059154529979195, | |
| "loss": 2.6174, | |
| "theoretical_loss": 3.6152118830936164, | |
| "tokens_seen": 1104281600 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.000305324380556081, | |
| "loss": 2.6236, | |
| "theoretical_loss": 3.6142114989918195, | |
| "tokens_seen": 1107558400 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.00030473330811424246, | |
| "loss": 2.6293, | |
| "theoretical_loss": 3.6132148961715624, | |
| "tokens_seen": 1110835200 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.000304142235672404, | |
| "loss": 2.5852, | |
| "theoretical_loss": 3.6122220492490964, | |
| "tokens_seen": 1114112000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0003035511632305655, | |
| "loss": 2.5782, | |
| "theoretical_loss": 3.6112329330850894, | |
| "tokens_seen": 1117388800 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.000302960090788727, | |
| "loss": 2.607, | |
| "theoretical_loss": 3.61024752278157, | |
| "tokens_seen": 1120665600 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0003023690183468886, | |
| "loss": 2.56, | |
| "theoretical_loss": 3.6092657936789054, | |
| "tokens_seen": 1123942400 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0003017779459050501, | |
| "loss": 2.5852, | |
| "theoretical_loss": 3.6082877213528377, | |
| "tokens_seen": 1127219200 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.00030118687346321164, | |
| "loss": 2.5555, | |
| "theoretical_loss": 3.60731328161156, | |
| "tokens_seen": 1130496000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.00030059580102137316, | |
| "loss": 2.5653, | |
| "theoretical_loss": 3.6063424504928365, | |
| "tokens_seen": 1133772800 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.0003000047285795347, | |
| "loss": 2.5872, | |
| "theoretical_loss": 3.60537520426117, | |
| "tokens_seen": 1137049600 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.0002994136561376962, | |
| "loss": 2.5925, | |
| "theoretical_loss": 3.6044115194050086, | |
| "tokens_seen": 1140326400 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.0002988225836958578, | |
| "loss": 2.5761, | |
| "theoretical_loss": 3.603451372633997, | |
| "tokens_seen": 1143603200 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "objective/train/docs_used": 604425, | |
| "objective/train/instantaneous_batch_size": 32, | |
| "objective/train/instantaneous_microbatch_size": 32768, | |
| "objective/train/original_loss": 2.565331220626831, | |
| "objective/train/theoretical_loss": 3.6024947408762698, | |
| "objective/train/tokens_used": 1167340000, | |
| "theoretical_loss": 3.6024947408762698, | |
| "tokens_seen": 1146880000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.0002982315112540193, | |
| "loss": 2.6043, | |
| "theoretical_loss": 3.6024947408762698, | |
| "tokens_seen": 1146880000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.0002976404388121808, | |
| "loss": 2.5725, | |
| "theoretical_loss": 3.601541601275783, | |
| "tokens_seen": 1150156800 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.00029704936637034234, | |
| "loss": 2.548, | |
| "theoretical_loss": 3.6005919311896886, | |
| "tokens_seen": 1153433600 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.00029647011537734067, | |
| "loss": 2.5699, | |
| "theoretical_loss": 3.5996457081857454, | |
| "tokens_seen": 1156710400 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.00029587904293550214, | |
| "loss": 2.5709, | |
| "theoretical_loss": 3.598702910039772, | |
| "tokens_seen": 1159987200 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.00029528797049366366, | |
| "loss": 2.5726, | |
| "theoretical_loss": 3.597763514733133, | |
| "tokens_seen": 1163264000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.00029469689805182523, | |
| "loss": 2.5542, | |
| "theoretical_loss": 3.59682750045027, | |
| "tokens_seen": 1166540800 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.00029410582560998675, | |
| "loss": 2.5631, | |
| "theoretical_loss": 3.5958948455762583, | |
| "tokens_seen": 1169817600 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.0002935147531681483, | |
| "loss": 2.5865, | |
| "theoretical_loss": 3.594965528694412, | |
| "tokens_seen": 1173094400 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.0002929236807263098, | |
| "loss": 2.5499, | |
| "theoretical_loss": 3.594039528583913, | |
| "tokens_seen": 1176371200 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.0002923326082844713, | |
| "loss": 2.559, | |
| "theoretical_loss": 3.5931168242174847, | |
| "tokens_seen": 1179648000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.00029174153584263284, | |
| "loss": 2.5468, | |
| "theoretical_loss": 3.59219739475909, | |
| "tokens_seen": 1182924800 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.0002911504634007944, | |
| "loss": 2.5555, | |
| "theoretical_loss": 3.5912812195616732, | |
| "tokens_seen": 1186201600 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.00029055939095895593, | |
| "loss": 2.5321, | |
| "theoretical_loss": 3.590368278164926, | |
| "tokens_seen": 1189478400 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.00028996831851711745, | |
| "loss": 2.5672, | |
| "theoretical_loss": 3.5894585502930902, | |
| "tokens_seen": 1192755200 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.000289377246075279, | |
| "loss": 2.5742, | |
| "theoretical_loss": 3.588552015852793, | |
| "tokens_seen": 1196032000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.0002887861736334405, | |
| "loss": 2.5502, | |
| "theoretical_loss": 3.5876486549309097, | |
| "tokens_seen": 1199308800 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.000288195101191602, | |
| "loss": 2.5514, | |
| "theoretical_loss": 3.586748447792462, | |
| "tokens_seen": 1202585600 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.0002876040287497636, | |
| "loss": 2.5578, | |
| "theoretical_loss": 3.5858513748785423, | |
| "tokens_seen": 1205862400 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.0002870129563079251, | |
| "loss": 2.6184, | |
| "theoretical_loss": 3.5849574168042704, | |
| "tokens_seen": 1209139200 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.00028642188386608663, | |
| "loss": 2.5991, | |
| "theoretical_loss": 3.5840665543567782, | |
| "tokens_seen": 1212416000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.00028583081142424815, | |
| "loss": 2.5757, | |
| "theoretical_loss": 3.583178768493222, | |
| "tokens_seen": 1215692800 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.00028523973898240967, | |
| "loss": 2.5659, | |
| "theoretical_loss": 3.5822940403388284, | |
| "tokens_seen": 1218969600 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002846486665405712, | |
| "loss": 2.5729, | |
| "theoretical_loss": 3.581412351184958, | |
| "tokens_seen": 1222246400 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002840575940987327, | |
| "loss": 2.5652, | |
| "theoretical_loss": 3.580533682487208, | |
| "tokens_seen": 1225523200 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002834665216568943, | |
| "loss": 2.5672, | |
| "theoretical_loss": 3.579658015863532, | |
| "tokens_seen": 1228800000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002828754492150558, | |
| "loss": 2.5836, | |
| "theoretical_loss": 3.5787853330923927, | |
| "tokens_seen": 1232076800 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.00028228437677321733, | |
| "loss": 2.5493, | |
| "theoretical_loss": 3.577915616110936, | |
| "tokens_seen": 1235353600 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.00028169330433137885, | |
| "loss": 2.5739, | |
| "theoretical_loss": 3.577048847013194, | |
| "tokens_seen": 1238630400 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.00028110223188954037, | |
| "loss": 2.5689, | |
| "theoretical_loss": 3.57618500804831, | |
| "tokens_seen": 1241907200 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002805111594477019, | |
| "loss": 2.5232, | |
| "theoretical_loss": 3.575324081618793, | |
| "tokens_seen": 1245184000 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.00027992008700586347, | |
| "loss": 2.5686, | |
| "theoretical_loss": 3.5744660502787875, | |
| "tokens_seen": 1248460800 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.000279329014564025, | |
| "loss": 2.5236, | |
| "theoretical_loss": 3.5736108967323794, | |
| "tokens_seen": 1251737600 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.0002787379421221865, | |
| "loss": 2.5925, | |
| "theoretical_loss": 3.5727586038319155, | |
| "tokens_seen": 1255014400 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.00027814686968034803, | |
| "loss": 2.5437, | |
| "theoretical_loss": 3.571909154576348, | |
| "tokens_seen": 1258291200 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.00027755579723850955, | |
| "loss": 2.5461, | |
| "theoretical_loss": 3.5710625321096074, | |
| "tokens_seen": 1261568000 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.00027696472479667107, | |
| "loss": 2.5288, | |
| "theoretical_loss": 3.570218719718989, | |
| "tokens_seen": 1264844800 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.00027637365235483265, | |
| "loss": 2.5546, | |
| "theoretical_loss": 3.569377700833569, | |
| "tokens_seen": 1268121600 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.00027578257991299417, | |
| "loss": 2.5354, | |
| "theoretical_loss": 3.568539459022639, | |
| "tokens_seen": 1271398400 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.0002751915074711557, | |
| "loss": 2.5518, | |
| "theoretical_loss": 3.5677039779941584, | |
| "tokens_seen": 1274675200 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.0002746004350293172, | |
| "loss": 2.5598, | |
| "theoretical_loss": 3.566871241593236, | |
| "tokens_seen": 1277952000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.00027400936258747873, | |
| "loss": 2.5445, | |
| "theoretical_loss": 3.5660412338006235, | |
| "tokens_seen": 1281228800 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.00027341829014564025, | |
| "loss": 2.5946, | |
| "theoretical_loss": 3.565213938731236, | |
| "tokens_seen": 1284505600 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.0002728272177038018, | |
| "loss": 2.5831, | |
| "theoretical_loss": 3.5643893406326868, | |
| "tokens_seen": 1287782400 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.00027223614526196335, | |
| "loss": 2.5568, | |
| "theoretical_loss": 3.5635674238838466, | |
| "tokens_seen": 1291059200 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.00027164507282012487, | |
| "loss": 2.5429, | |
| "theoretical_loss": 3.5627481729934196, | |
| "tokens_seen": 1294336000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.0002710540003782864, | |
| "loss": 2.5785, | |
| "theoretical_loss": 3.561931572598538, | |
| "tokens_seen": 1297612800 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.0002704629279364479, | |
| "loss": 2.5493, | |
| "theoretical_loss": 3.5611176074633777, | |
| "tokens_seen": 1300889600 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.00026987185549460943, | |
| "loss": 2.5602, | |
| "theoretical_loss": 3.5603062624777895, | |
| "tokens_seen": 1304166400 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.00026928078305277095, | |
| "loss": 2.6058, | |
| "theoretical_loss": 3.559497522655951, | |
| "tokens_seen": 1307443200 | |
| }, | |
| { | |
| "debugging/Self-BLEU-5": 0.4840744728399488, | |
| "debugging/distinct-1-grams": 0.7704641680561449, | |
| "debugging/distinct-2-grams": 0.9548886056073045, | |
| "debugging/entropy-1-grams": 5.354809677980487, | |
| "debugging/entropy-2-grams": 6.077984561009924, | |
| "debugging/length": 581.0, | |
| "debugging/num_segments": 7, | |
| "epoch": 0.47, | |
| "objective/train/docs_used": 687328, | |
| "objective/train/instantaneous_batch_size": 32, | |
| "objective/train/instantaneous_microbatch_size": 32768, | |
| "objective/train/original_loss": 2.7656729221343994, | |
| "objective/train/theoretical_loss": 3.5586913731350327, | |
| "objective/train/tokens_used": 1331180000, | |
| "theoretical_loss": 3.5586913731350327, | |
| "tokens_seen": 1310720000 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.0002686897106109325, | |
| "loss": 2.6199, | |
| "theoretical_loss": 3.5586913731350327, | |
| "tokens_seen": 1310720000 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.00026809863816909405, | |
| "loss": 2.6217, | |
| "theoretical_loss": 3.557887799173889, | |
| "tokens_seen": 1313996800 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.00026750756572725557, | |
| "loss": 2.5979, | |
| "theoretical_loss": 3.557086786151754, | |
| "tokens_seen": 1317273600 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.0002669164932854171, | |
| "loss": 2.5856, | |
| "theoretical_loss": 3.5562883195669697, | |
| "tokens_seen": 1320550400 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.00026632542084357855, | |
| "loss": 2.5917, | |
| "theoretical_loss": 3.555492385035719, | |
| "tokens_seen": 1323827200 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.0002657343484017401, | |
| "loss": 2.5804, | |
| "theoretical_loss": 3.5546989682907784, | |
| "tokens_seen": 1327104000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00026514327595990165, | |
| "loss": 2.5405, | |
| "theoretical_loss": 3.5539080551802895, | |
| "tokens_seen": 1330380800 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00026455220351806317, | |
| "loss": 2.5578, | |
| "theoretical_loss": 3.553119631666546, | |
| "tokens_seen": 1333657600 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.0002639611310762247, | |
| "loss": 2.5811, | |
| "theoretical_loss": 3.5523336838247914, | |
| "tokens_seen": 1336934400 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.0002633700586343862, | |
| "loss": 2.5485, | |
| "theoretical_loss": 3.55155019784204, | |
| "tokens_seen": 1340211200 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00026277898619254773, | |
| "loss": 2.5404, | |
| "theoretical_loss": 3.5507691600159053, | |
| "tokens_seen": 1343488000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00026218791375070925, | |
| "loss": 2.5348, | |
| "theoretical_loss": 3.5499905567534515, | |
| "tokens_seen": 1346764800 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00026159684130887083, | |
| "loss": 2.5136, | |
| "theoretical_loss": 3.549214374570052, | |
| "tokens_seen": 1350041600 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00026100576886703235, | |
| "loss": 2.5227, | |
| "theoretical_loss": 3.5484406000882665, | |
| "tokens_seen": 1353318400 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00026041469642519387, | |
| "loss": 2.4971, | |
| "theoretical_loss": 3.5476692200367346, | |
| "tokens_seen": 1356595200 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.0002598236239833554, | |
| "loss": 2.5241, | |
| "theoretical_loss": 3.546900221249076, | |
| "tokens_seen": 1359872000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.0002592325515415169, | |
| "loss": 2.5345, | |
| "theoretical_loss": 3.5461335906628157, | |
| "tokens_seen": 1363148800 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.00025864147909967843, | |
| "loss": 2.5072, | |
| "theoretical_loss": 3.54536931531831, | |
| "tokens_seen": 1366425600 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.00025805040665783995, | |
| "loss": 2.5089, | |
| "theoretical_loss": 3.5446073823576985, | |
| "tokens_seen": 1369702400 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.00025745933421600153, | |
| "loss": 2.5158, | |
| "theoretical_loss": 3.543847779023859, | |
| "tokens_seen": 1372979200 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.00025686826177416305, | |
| "loss": 2.5466, | |
| "theoretical_loss": 3.543090492659384, | |
| "tokens_seen": 1376256000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.00025627718933232457, | |
| "loss": 2.5079, | |
| "theoretical_loss": 3.542335510705562, | |
| "tokens_seen": 1379532800 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.0002556861168904861, | |
| "loss": 2.5099, | |
| "theoretical_loss": 3.541582820701378, | |
| "tokens_seen": 1382809600 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.0002550950444486476, | |
| "loss": 2.4877, | |
| "theoretical_loss": 3.5408324102825253, | |
| "tokens_seen": 1386086400 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.00025450397200680913, | |
| "loss": 2.4898, | |
| "theoretical_loss": 3.5400842671804265, | |
| "tokens_seen": 1389363200 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.0002539128995649707, | |
| "loss": 2.4993, | |
| "theoretical_loss": 3.5393383792212676, | |
| "tokens_seen": 1392640000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.00025332182712313223, | |
| "loss": 2.4967, | |
| "theoretical_loss": 3.5385947343250486, | |
| "tokens_seen": 1395916800 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.00025273075468129375, | |
| "loss": 2.5021, | |
| "theoretical_loss": 3.53785332050464, | |
| "tokens_seen": 1399193600 | |
| } | |
| ], | |
| "max_steps": 42724, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 7.14460209610752e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |