| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.01714383679067375, | |
| "eval_steps": 500, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 1.4420362710952759, | |
| "learning_rate": 4e-05, | |
| "loss": 3.3293, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 4e-05, | |
| "loss": 3.8207, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 1.3726439476013184, | |
| "learning_rate": 8e-05, | |
| "loss": 3.4594, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 1.2832139730453491, | |
| "learning_rate": 0.00012, | |
| "loss": 3.297, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 1.2156065702438354, | |
| "learning_rate": 0.00016, | |
| "loss": 3.0036, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 1.196576476097107, | |
| "learning_rate": 0.0002, | |
| "loss": 2.9488, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0002, | |
| "loss": 2.7249, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 1.1029444932937622, | |
| "learning_rate": 0.00019996568291008924, | |
| "loss": 2.8809, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 1.0251176357269287, | |
| "learning_rate": 0.00019993136582017846, | |
| "loss": 2.728, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 1.7124559879302979, | |
| "learning_rate": 0.0001998970487302677, | |
| "loss": 2.6262, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 1.3183832168579102, | |
| "learning_rate": 0.00019986273164035692, | |
| "loss": 2.765, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019986273164035692, | |
| "loss": 2.682, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 1.2592570781707764, | |
| "learning_rate": 0.00019982841455044615, | |
| "loss": 2.7837, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 0.9486250877380371, | |
| "learning_rate": 0.00019979409746053535, | |
| "loss": 2.8407, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 0.8232390284538269, | |
| "learning_rate": 0.00019975978037062458, | |
| "loss": 2.6573, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 1.1916477680206299, | |
| "learning_rate": 0.0001997254632807138, | |
| "loss": 2.8644, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 0.6067786812782288, | |
| "learning_rate": 0.00019969114619080303, | |
| "loss": 2.582, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 0.9011898040771484, | |
| "learning_rate": 0.00019965682910089226, | |
| "loss": 2.7302, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 0.8688616156578064, | |
| "learning_rate": 0.00019962251201098149, | |
| "loss": 2.3353, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 0.7956060767173767, | |
| "learning_rate": 0.0001995881949210707, | |
| "loss": 2.5263, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001995881949210707, | |
| "loss": 2.4818, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 0.8461011052131653, | |
| "learning_rate": 0.00019955387783115994, | |
| "loss": 2.542, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 0.6973548531532288, | |
| "learning_rate": 0.00019951956074124917, | |
| "loss": 2.4787, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 0.6046115756034851, | |
| "learning_rate": 0.0001994852436513384, | |
| "loss": 2.4436, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 0.4711577594280243, | |
| "learning_rate": 0.0001994509265614276, | |
| "loss": 2.3658, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001994509265614276, | |
| "loss": 2.4518, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 0.5012961030006409, | |
| "learning_rate": 0.00019941660947151682, | |
| "loss": 2.3782, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 0.6144160628318787, | |
| "learning_rate": 0.00019938229238160605, | |
| "loss": 2.4487, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019938229238160605, | |
| "loss": 2.3935, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.6648868322372437, | |
| "learning_rate": 0.00019934797529169528, | |
| "loss": 2.5339, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.5968282222747803, | |
| "learning_rate": 0.00019931365820178448, | |
| "loss": 2.4238, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019931365820178448, | |
| "loss": 2.4432, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019931365820178448, | |
| "loss": 2.3153, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.9555236101150513, | |
| "learning_rate": 0.0001992793411118737, | |
| "loss": 2.2495, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.8204368352890015, | |
| "learning_rate": 0.00019924502402196293, | |
| "loss": 2.401, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019924502402196293, | |
| "loss": 2.31, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.8206217885017395, | |
| "learning_rate": 0.00019921070693205216, | |
| "loss": 2.3511, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019921070693205216, | |
| "loss": 2.2848, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.5887894630432129, | |
| "learning_rate": 0.0001991763898421414, | |
| "loss": 2.191, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.5945093035697937, | |
| "learning_rate": 0.00019914207275223062, | |
| "loss": 2.345, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019914207275223062, | |
| "loss": 2.4744, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.7511911988258362, | |
| "learning_rate": 0.00019910775566231984, | |
| "loss": 2.3624, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019910775566231984, | |
| "loss": 2.3179, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.8779334425926208, | |
| "learning_rate": 0.00019907343857240907, | |
| "loss": 2.4634, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.9247270226478577, | |
| "learning_rate": 0.0001990391214824983, | |
| "loss": 2.2699, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.785952627658844, | |
| "learning_rate": 0.0001990048043925875, | |
| "loss": 2.2791, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 1.0079773664474487, | |
| "learning_rate": 0.00019897048730267673, | |
| "loss": 2.4368, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 1.1102200746536255, | |
| "learning_rate": 0.00019893617021276595, | |
| "loss": 2.4482, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 1.1956760883331299, | |
| "learning_rate": 0.00019890185312285518, | |
| "loss": 2.4036, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.7190561294555664, | |
| "learning_rate": 0.0001988675360329444, | |
| "loss": 2.2854, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001988675360329444, | |
| "loss": 2.404, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 1.0126006603240967, | |
| "learning_rate": 0.00019883321894303364, | |
| "loss": 2.1978, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019883321894303364, | |
| "loss": 2.3899, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 1.213659405708313, | |
| "learning_rate": 0.00019879890185312286, | |
| "loss": 2.0958, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 1.3466947078704834, | |
| "learning_rate": 0.0001987645847632121, | |
| "loss": 2.4106, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 1.2530690431594849, | |
| "learning_rate": 0.00019873026767330132, | |
| "loss": 2.5628, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 1.6688309907913208, | |
| "learning_rate": 0.00019869595058339055, | |
| "loss": 2.233, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 2.2483224868774414, | |
| "learning_rate": 0.00019866163349347977, | |
| "loss": 2.2923, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 2.017726182937622, | |
| "learning_rate": 0.000198627316403569, | |
| "loss": 2.354, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.000198627316403569, | |
| "loss": 2.5311, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 1.810524582862854, | |
| "learning_rate": 0.0001985929993136582, | |
| "loss": 2.24, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 2.439291000366211, | |
| "learning_rate": 0.00019855868222374743, | |
| "loss": 2.8512, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019855868222374743, | |
| "loss": 2.5943, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 2.2657182216644287, | |
| "learning_rate": 0.00019852436513383666, | |
| "loss": 2.8789, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 2.1179282665252686, | |
| "learning_rate": 0.00019849004804392589, | |
| "loss": 2.6996, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 2.0068299770355225, | |
| "learning_rate": 0.0001984557309540151, | |
| "loss": 2.3677, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001984557309540151, | |
| "loss": 2.577, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001984557309540151, | |
| "loss": 2.7072, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 4.678086280822754, | |
| "learning_rate": 0.00019842141386410434, | |
| "loss": 2.658, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 4.056044101715088, | |
| "learning_rate": 0.00019838709677419357, | |
| "loss": 2.5174, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 4.784575462341309, | |
| "learning_rate": 0.0001983527796842828, | |
| "loss": 2.6965, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 7.531276226043701, | |
| "learning_rate": 0.00019831846259437202, | |
| "loss": 2.9831, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 4.478590965270996, | |
| "learning_rate": 0.00019828414550446125, | |
| "loss": 3.0535, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019828414550446125, | |
| "loss": 2.7639, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 7.104541778564453, | |
| "learning_rate": 0.00019824982841455048, | |
| "loss": 3.0957, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 5.70428991317749, | |
| "learning_rate": 0.00019821551132463968, | |
| "loss": 3.0107, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 3.2724623680114746, | |
| "learning_rate": 0.0001981811942347289, | |
| "loss": 2.9372, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001981811942347289, | |
| "loss": 3.4289, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001981811942347289, | |
| "loss": 3.4832, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 20.135986328125, | |
| "learning_rate": 0.00019814687714481813, | |
| "loss": 3.2707, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 22.686079025268555, | |
| "learning_rate": 0.00019811256005490736, | |
| "loss": 3.8144, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019811256005490736, | |
| "loss": 4.5631, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 28.4752197265625, | |
| "learning_rate": 0.00019807824296499656, | |
| "loss": 4.8363, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 23.301496505737305, | |
| "learning_rate": 0.0001980439258750858, | |
| "loss": 5.3773, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 19.411945343017578, | |
| "learning_rate": 0.00019800960878517502, | |
| "loss": 5.1105, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019800960878517502, | |
| "loss": 5.8669, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 35.982608795166016, | |
| "learning_rate": 0.00019797529169526424, | |
| "loss": 5.4417, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019797529169526424, | |
| "loss": 5.4101, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 69.57833862304688, | |
| "learning_rate": 0.00019794097460535347, | |
| "loss": 5.2831, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 76.1761703491211, | |
| "learning_rate": 0.0001979066575154427, | |
| "loss": 6.0605, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001979066575154427, | |
| "loss": 6.9964, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 104.58732604980469, | |
| "learning_rate": 0.00019787234042553193, | |
| "loss": 7.346, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 45.56397247314453, | |
| "learning_rate": 0.00019783802333562115, | |
| "loss": 8.6069, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019783802333562115, | |
| "loss": 9.6728, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019783802333562115, | |
| "loss": 8.1841, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019783802333562115, | |
| "loss": 9.4986, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 25.033235549926758, | |
| "learning_rate": 0.00019780370624571035, | |
| "loss": 9.8723, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 74.37015533447266, | |
| "learning_rate": 0.00019776938915579958, | |
| "loss": 9.7078, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 72.31209564208984, | |
| "learning_rate": 0.0001977350720658888, | |
| "loss": 12.0863, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 47.87614059448242, | |
| "learning_rate": 0.00019770075497597804, | |
| "loss": 14.7985, | |
| "step": 100 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 5833, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 100, | |
| "total_flos": 1.2504616383074304e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |