| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.43252595155709345, | |
| "eval_steps": 500, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00865051903114187, | |
| "grad_norm": 0.653178334236145, | |
| "learning_rate": 5.142857142857143e-05, | |
| "loss": 2.462, | |
| "mean_token_accuracy": 0.5231182813644409, | |
| "num_tokens": 20480.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.01730103806228374, | |
| "grad_norm": 0.4405156075954437, | |
| "learning_rate": 0.00010857142857142856, | |
| "loss": 2.0497, | |
| "mean_token_accuracy": 0.5873411521315575, | |
| "num_tokens": 40960.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.025951557093425604, | |
| "grad_norm": 0.6160047054290771, | |
| "learning_rate": 0.00016571428571428575, | |
| "loss": 1.9284, | |
| "mean_token_accuracy": 0.6067448765039444, | |
| "num_tokens": 61440.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.03460207612456748, | |
| "grad_norm": 0.4909583330154419, | |
| "learning_rate": 0.00019999371690018224, | |
| "loss": 1.7739, | |
| "mean_token_accuracy": 0.6309384137392045, | |
| "num_tokens": 81920.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.04325259515570934, | |
| "grad_norm": 0.4441400170326233, | |
| "learning_rate": 0.00019992304109437157, | |
| "loss": 1.9242, | |
| "mean_token_accuracy": 0.6014662817120552, | |
| "num_tokens": 102400.0, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.05190311418685121, | |
| "grad_norm": 0.5243574976921082, | |
| "learning_rate": 0.00019977389129787062, | |
| "loss": 1.8928, | |
| "mean_token_accuracy": 0.6027370542287827, | |
| "num_tokens": 122880.0, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.06055363321799308, | |
| "grad_norm": 0.3737685978412628, | |
| "learning_rate": 0.00019954638464462175, | |
| "loss": 1.8107, | |
| "mean_token_accuracy": 0.6185728281736373, | |
| "num_tokens": 143360.0, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.06920415224913495, | |
| "grad_norm": 0.5457460284233093, | |
| "learning_rate": 0.00019924069980567822, | |
| "loss": 1.7227, | |
| "mean_token_accuracy": 0.6354838699102402, | |
| "num_tokens": 163840.0, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.07785467128027682, | |
| "grad_norm": 0.41756778955459595, | |
| "learning_rate": 0.00019885707684888566, | |
| "loss": 1.7249, | |
| "mean_token_accuracy": 0.6385141760110855, | |
| "num_tokens": 184320.0, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.08650519031141868, | |
| "grad_norm": 0.39207473397254944, | |
| "learning_rate": 0.00019839581705034624, | |
| "loss": 1.6423, | |
| "mean_token_accuracy": 0.6423753708600998, | |
| "num_tokens": 204800.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.09515570934256055, | |
| "grad_norm": 0.46543020009994507, | |
| "learning_rate": 0.00019785728265781324, | |
| "loss": 1.5637, | |
| "mean_token_accuracy": 0.6588465303182602, | |
| "num_tokens": 225280.0, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.10380622837370242, | |
| "grad_norm": 0.4604319930076599, | |
| "learning_rate": 0.0001972418966062018, | |
| "loss": 1.5936, | |
| "mean_token_accuracy": 0.6586021468043327, | |
| "num_tokens": 245760.0, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.11245674740484429, | |
| "grad_norm": 0.34409117698669434, | |
| "learning_rate": 0.0001965501421854394, | |
| "loss": 1.7365, | |
| "mean_token_accuracy": 0.6300097703933716, | |
| "num_tokens": 266240.0, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.12110726643598616, | |
| "grad_norm": 0.46972155570983887, | |
| "learning_rate": 0.0001957825626609169, | |
| "loss": 1.6823, | |
| "mean_token_accuracy": 0.6419354841113091, | |
| "num_tokens": 286720.0, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.12975778546712802, | |
| "grad_norm": 0.45500946044921875, | |
| "learning_rate": 0.00019493976084683813, | |
| "loss": 1.4338, | |
| "mean_token_accuracy": 0.6832355856895447, | |
| "num_tokens": 307200.0, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.1384083044982699, | |
| "grad_norm": 0.4167431890964508, | |
| "learning_rate": 0.0001940223986328032, | |
| "loss": 1.7327, | |
| "mean_token_accuracy": 0.6318181827664375, | |
| "num_tokens": 327680.0, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.14705882352941177, | |
| "grad_norm": 0.43620193004608154, | |
| "learning_rate": 0.00019303119646399728, | |
| "loss": 1.5997, | |
| "mean_token_accuracy": 0.6583577707409859, | |
| "num_tokens": 348160.0, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.15570934256055363, | |
| "grad_norm": 0.48302027583122253, | |
| "learning_rate": 0.00019196693277539307, | |
| "loss": 1.79, | |
| "mean_token_accuracy": 0.617741933465004, | |
| "num_tokens": 368640.0, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.1643598615916955, | |
| "grad_norm": 0.49784767627716064, | |
| "learning_rate": 0.00019083044338041134, | |
| "loss": 1.5624, | |
| "mean_token_accuracy": 0.6568426251411438, | |
| "num_tokens": 389120.0, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.17301038062283736, | |
| "grad_norm": 0.3815184235572815, | |
| "learning_rate": 0.00018962262081451966, | |
| "loss": 1.5723, | |
| "mean_token_accuracy": 0.6639296174049377, | |
| "num_tokens": 409600.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.18166089965397925, | |
| "grad_norm": 0.43787622451782227, | |
| "learning_rate": 0.00018834441363428463, | |
| "loss": 1.7278, | |
| "mean_token_accuracy": 0.6321114301681519, | |
| "num_tokens": 430080.0, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.1903114186851211, | |
| "grad_norm": 0.4128075838088989, | |
| "learning_rate": 0.00018699682567242863, | |
| "loss": 1.6181, | |
| "mean_token_accuracy": 0.6524437993764878, | |
| "num_tokens": 450560.0, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.19896193771626297, | |
| "grad_norm": 0.35929927229881287, | |
| "learning_rate": 0.00018558091524947524, | |
| "loss": 1.6799, | |
| "mean_token_accuracy": 0.6450635358691216, | |
| "num_tokens": 471040.0, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.20761245674740483, | |
| "grad_norm": 0.4328981637954712, | |
| "learning_rate": 0.00018409779434260342, | |
| "loss": 1.2708, | |
| "mean_token_accuracy": 0.7201857253909111, | |
| "num_tokens": 491520.0, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.21626297577854672, | |
| "grad_norm": 0.4529803991317749, | |
| "learning_rate": 0.00018254862771236257, | |
| "loss": 1.718, | |
| "mean_token_accuracy": 0.6376832827925683, | |
| "num_tokens": 512000.0, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.22491349480968859, | |
| "grad_norm": 0.39349377155303955, | |
| "learning_rate": 0.00018093463198793432, | |
| "loss": 1.6715, | |
| "mean_token_accuracy": 0.6447702810168267, | |
| "num_tokens": 532480.0, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.23356401384083045, | |
| "grad_norm": 0.3997224271297455, | |
| "learning_rate": 0.00017925707471165992, | |
| "loss": 1.5151, | |
| "mean_token_accuracy": 0.6776148617267609, | |
| "num_tokens": 552960.0, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.2422145328719723, | |
| "grad_norm": 0.4060993492603302, | |
| "learning_rate": 0.00017751727334358307, | |
| "loss": 1.5702, | |
| "mean_token_accuracy": 0.6609481915831565, | |
| "num_tokens": 573440.0, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.2508650519031142, | |
| "grad_norm": 0.369098424911499, | |
| "learning_rate": 0.00017571659422679003, | |
| "loss": 1.3934, | |
| "mean_token_accuracy": 0.6983382269740105, | |
| "num_tokens": 593920.0, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.25951557093425603, | |
| "grad_norm": 0.4169941842556, | |
| "learning_rate": 0.00017385645151436, | |
| "loss": 1.5608, | |
| "mean_token_accuracy": 0.6624144673347473, | |
| "num_tokens": 614400.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.2681660899653979, | |
| "grad_norm": 0.44405296444892883, | |
| "learning_rate": 0.00017193830605876804, | |
| "loss": 1.6689, | |
| "mean_token_accuracy": 0.6325513184070587, | |
| "num_tokens": 634880.0, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.2768166089965398, | |
| "grad_norm": 0.4252229630947113, | |
| "learning_rate": 0.0001699636642646129, | |
| "loss": 1.4817, | |
| "mean_token_accuracy": 0.6717986300587654, | |
| "num_tokens": 655360.0, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.28546712802768165, | |
| "grad_norm": 0.4406476318836212, | |
| "learning_rate": 0.00016793407690557076, | |
| "loss": 1.6879, | |
| "mean_token_accuracy": 0.640762460231781, | |
| "num_tokens": 675840.0, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.29411764705882354, | |
| "grad_norm": 0.42295652627944946, | |
| "learning_rate": 0.00016585113790650388, | |
| "loss": 1.5637, | |
| "mean_token_accuracy": 0.6610459357500076, | |
| "num_tokens": 696320.0, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.3027681660899654, | |
| "grad_norm": 0.4155745208263397, | |
| "learning_rate": 0.00016371648309168097, | |
| "loss": 1.7478, | |
| "mean_token_accuracy": 0.6266862124204635, | |
| "num_tokens": 716800.0, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.31141868512110726, | |
| "grad_norm": 0.3854524791240692, | |
| "learning_rate": 0.0001615317889000918, | |
| "loss": 1.3819, | |
| "mean_token_accuracy": 0.6973118335008621, | |
| "num_tokens": 737280.0, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.32006920415224915, | |
| "grad_norm": 0.4296940565109253, | |
| "learning_rate": 0.00015929877106886535, | |
| "loss": 1.5729, | |
| "mean_token_accuracy": 0.6544476956129074, | |
| "num_tokens": 757760.0, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.328719723183391, | |
| "grad_norm": 0.5267869830131531, | |
| "learning_rate": 0.0001570191832858256, | |
| "loss": 1.5866, | |
| "mean_token_accuracy": 0.6566959977149963, | |
| "num_tokens": 778240.0, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.3373702422145329, | |
| "grad_norm": 0.4024481177330017, | |
| "learning_rate": 0.00015469481581224272, | |
| "loss": 1.5171, | |
| "mean_token_accuracy": 0.67121212631464, | |
| "num_tokens": 798720.0, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.3460207612456747, | |
| "grad_norm": 0.48981329798698425, | |
| "learning_rate": 0.0001523274940768614, | |
| "loss": 1.6027, | |
| "mean_token_accuracy": 0.6437927693128586, | |
| "num_tokens": 819200.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.3546712802768166, | |
| "grad_norm": 0.39708444476127625, | |
| "learning_rate": 0.00014991907724231122, | |
| "loss": 1.7412, | |
| "mean_token_accuracy": 0.6313782975077629, | |
| "num_tokens": 839680.0, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.3633217993079585, | |
| "grad_norm": 0.3914325535297394, | |
| "learning_rate": 0.0001474714567450234, | |
| "loss": 1.445, | |
| "mean_token_accuracy": 0.6826490700244904, | |
| "num_tokens": 860160.0, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.3719723183391003, | |
| "grad_norm": 0.43571996688842773, | |
| "learning_rate": 0.00014498655480980226, | |
| "loss": 1.4578, | |
| "mean_token_accuracy": 0.6875855386257171, | |
| "num_tokens": 880640.0, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.3806228373702422, | |
| "grad_norm": 0.37934359908103943, | |
| "learning_rate": 0.00014246632294021705, | |
| "loss": 1.332, | |
| "mean_token_accuracy": 0.7121700882911682, | |
| "num_tokens": 901120.0, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.3892733564013841, | |
| "grad_norm": 0.4126007854938507, | |
| "learning_rate": 0.00013991274038599927, | |
| "loss": 1.6188, | |
| "mean_token_accuracy": 0.6448191583156586, | |
| "num_tokens": 921600.0, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.39792387543252594, | |
| "grad_norm": 0.3285645544528961, | |
| "learning_rate": 0.00013732781258865066, | |
| "loss": 1.399, | |
| "mean_token_accuracy": 0.695943309366703, | |
| "num_tokens": 942080.0, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.40657439446366783, | |
| "grad_norm": 0.48543041944503784, | |
| "learning_rate": 0.00013471356960648092, | |
| "loss": 1.693, | |
| "mean_token_accuracy": 0.6381231769919395, | |
| "num_tokens": 962560.0, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.41522491349480967, | |
| "grad_norm": 0.28301766514778137, | |
| "learning_rate": 0.00013207206452031303, | |
| "loss": 1.3507, | |
| "mean_token_accuracy": 0.7064516082406044, | |
| "num_tokens": 983040.0, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.42387543252595156, | |
| "grad_norm": 0.44277122616767883, | |
| "learning_rate": 0.00012940537182110797, | |
| "loss": 1.6078, | |
| "mean_token_accuracy": 0.6580645143985748, | |
| "num_tokens": 1003520.0, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.43252595155709345, | |
| "grad_norm": 0.4079855978488922, | |
| "learning_rate": 0.00012671558578077497, | |
| "loss": 1.7974, | |
| "mean_token_accuracy": 0.6107038110494614, | |
| "num_tokens": 1024000.0, | |
| "step": 500 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1156, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 50, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.7467768307712e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |