| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.10772578890097932, | |
| "eval_steps": 50, | |
| "global_step": 99, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.001088139281828074, | |
| "grad_norm": 0.2615509629249573, | |
| "learning_rate": 5e-05, | |
| "loss": 0.838, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.001088139281828074, | |
| "eval_loss": 0.23379258811473846, | |
| "eval_runtime": 24.7758, | |
| "eval_samples_per_second": 15.62, | |
| "eval_steps_per_second": 7.83, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.002176278563656148, | |
| "grad_norm": 0.27500101923942566, | |
| "learning_rate": 0.0001, | |
| "loss": 1.0311, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.003264417845484222, | |
| "grad_norm": 0.26265749335289, | |
| "learning_rate": 0.00015, | |
| "loss": 0.7598, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.004352557127312296, | |
| "grad_norm": 0.34702393412590027, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7643, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00544069640914037, | |
| "grad_norm": 0.26868095993995667, | |
| "learning_rate": 0.00025, | |
| "loss": 0.8676, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.006528835690968444, | |
| "grad_norm": 0.6326900124549866, | |
| "learning_rate": 0.0003, | |
| "loss": 0.9927, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.007616974972796518, | |
| "grad_norm": 0.6055320501327515, | |
| "learning_rate": 0.00035, | |
| "loss": 0.7319, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.008705114254624592, | |
| "grad_norm": 0.43092861771583557, | |
| "learning_rate": 0.0004, | |
| "loss": 0.7695, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.009793253536452665, | |
| "grad_norm": 0.40037959814071655, | |
| "learning_rate": 0.00045000000000000004, | |
| "loss": 0.7098, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.01088139281828074, | |
| "grad_norm": 0.39341455698013306, | |
| "learning_rate": 0.0005, | |
| "loss": 0.4365, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.011969532100108813, | |
| "grad_norm": 0.8135687112808228, | |
| "learning_rate": 0.0004998442655654946, | |
| "loss": 0.7446, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.013057671381936888, | |
| "grad_norm": 0.8516698479652405, | |
| "learning_rate": 0.0004993772562876909, | |
| "loss": 0.3987, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.014145810663764961, | |
| "grad_norm": 0.3541325628757477, | |
| "learning_rate": 0.0004985995540019955, | |
| "loss": 0.4453, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.015233949945593036, | |
| "grad_norm": 0.5297847986221313, | |
| "learning_rate": 0.0004975121276286136, | |
| "loss": 0.5155, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.01632208922742111, | |
| "grad_norm": 0.442644327878952, | |
| "learning_rate": 0.0004961163319653958, | |
| "loss": 0.4814, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.017410228509249184, | |
| "grad_norm": 0.4812023341655731, | |
| "learning_rate": 0.0004944139059999286, | |
| "loss": 0.5217, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.018498367791077257, | |
| "grad_norm": 0.4284003674983978, | |
| "learning_rate": 0.000492406970742972, | |
| "loss": 0.2965, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.01958650707290533, | |
| "grad_norm": 0.5950977206230164, | |
| "learning_rate": 0.0004900980265859448, | |
| "loss": 0.5757, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.020674646354733407, | |
| "grad_norm": 0.35087430477142334, | |
| "learning_rate": 0.0004874899501857477, | |
| "loss": 0.314, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.02176278563656148, | |
| "grad_norm": 0.2939762771129608, | |
| "learning_rate": 0.00048458599088080736, | |
| "loss": 0.3879, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.022850924918389554, | |
| "grad_norm": 0.3336440324783325, | |
| "learning_rate": 0.0004813897666428053, | |
| "loss": 0.3425, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.023939064200217627, | |
| "grad_norm": 0.5165703296661377, | |
| "learning_rate": 0.00047790525956913543, | |
| "loss": 0.4142, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.025027203482045703, | |
| "grad_norm": 0.41237401962280273, | |
| "learning_rate": 0.0004741368109217071, | |
| "loss": 0.4147, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.026115342763873776, | |
| "grad_norm": 0.5671696066856384, | |
| "learning_rate": 0.00047008911571827283, | |
| "loss": 0.48, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.02720348204570185, | |
| "grad_norm": 0.42801016569137573, | |
| "learning_rate": 0.00046576721688302105, | |
| "loss": 0.3996, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.028291621327529923, | |
| "grad_norm": 0.3613467514514923, | |
| "learning_rate": 0.0004611764989637205, | |
| "loss": 0.3176, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.029379760609358, | |
| "grad_norm": 1.3145064115524292, | |
| "learning_rate": 0.0004563226814232444, | |
| "loss": 0.4326, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.030467899891186073, | |
| "grad_norm": 0.47409093379974365, | |
| "learning_rate": 0.0004512118115138315, | |
| "loss": 0.3246, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.031556039173014146, | |
| "grad_norm": 0.5220752954483032, | |
| "learning_rate": 0.0004458502567429631, | |
| "loss": 0.5013, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.03264417845484222, | |
| "grad_norm": 1.3407182693481445, | |
| "learning_rate": 0.00044024469694024196, | |
| "loss": 0.4893, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.03373231773667029, | |
| "grad_norm": 0.7757295370101929, | |
| "learning_rate": 0.00043440211593515554, | |
| "loss": 0.4163, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.03482045701849837, | |
| "grad_norm": 0.27271905541419983, | |
| "learning_rate": 0.0004283297928560951, | |
| "loss": 0.2256, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.035908596300326445, | |
| "grad_norm": 0.4762435257434845, | |
| "learning_rate": 0.0004220352930614672, | |
| "loss": 0.4584, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.036996735582154515, | |
| "grad_norm": 0.5020000338554382, | |
| "learning_rate": 0.00041552645871420013, | |
| "loss": 0.4403, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.03808487486398259, | |
| "grad_norm": 0.3345811069011688, | |
| "learning_rate": 0.00040881139901138467, | |
| "loss": 0.4192, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.03917301414581066, | |
| "grad_norm": 0.2985716462135315, | |
| "learning_rate": 0.00040189848008122475, | |
| "loss": 0.2805, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.04026115342763874, | |
| "grad_norm": 0.8610369563102722, | |
| "learning_rate": 0.00039479631455988334, | |
| "loss": 0.718, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.041349292709466814, | |
| "grad_norm": 0.4035598635673523, | |
| "learning_rate": 0.0003875137508612103, | |
| "loss": 0.3767, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.042437431991294884, | |
| "grad_norm": 0.3371526002883911, | |
| "learning_rate": 0.00038005986215272055, | |
| "loss": 0.3665, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.04352557127312296, | |
| "grad_norm": 0.9126002788543701, | |
| "learning_rate": 0.0003724439350515571, | |
| "loss": 0.6029, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.04461371055495103, | |
| "grad_norm": 0.34127795696258545, | |
| "learning_rate": 0.0003646754580545226, | |
| "loss": 0.4245, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.04570184983677911, | |
| "grad_norm": 0.4555828869342804, | |
| "learning_rate": 0.000356764109716594, | |
| "loss": 0.6118, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.046789989118607184, | |
| "grad_norm": 0.30817294120788574, | |
| "learning_rate": 0.00034871974659264783, | |
| "loss": 0.3774, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.04787812840043525, | |
| "grad_norm": 0.6372089982032776, | |
| "learning_rate": 0.0003405523909574206, | |
| "loss": 0.4466, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.04896626768226333, | |
| "grad_norm": 0.2218654602766037, | |
| "learning_rate": 0.0003322722183190025, | |
| "loss": 0.2911, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.05005440696409141, | |
| "grad_norm": 0.7268356084823608, | |
| "learning_rate": 0.0003238895447414211, | |
| "loss": 0.5186, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.051142546245919476, | |
| "grad_norm": 0.3474952280521393, | |
| "learning_rate": 0.0003154148139921102, | |
| "loss": 0.3702, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.05223068552774755, | |
| "grad_norm": 0.47476518154144287, | |
| "learning_rate": 0.00030685858453027663, | |
| "loss": 0.3098, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.05331882480957562, | |
| "grad_norm": 0.33052054047584534, | |
| "learning_rate": 0.0002982315163523742, | |
| "loss": 0.3452, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.0544069640914037, | |
| "grad_norm": 0.3936608135700226, | |
| "learning_rate": 0.000289544357711076, | |
| "loss": 0.3504, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.0544069640914037, | |
| "eval_loss": 0.10187384486198425, | |
| "eval_runtime": 24.5923, | |
| "eval_samples_per_second": 15.737, | |
| "eval_steps_per_second": 7.889, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.055495103373231776, | |
| "grad_norm": 0.3658672273159027, | |
| "learning_rate": 0.0002808079317242896, | |
| "loss": 0.3851, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.056583242655059846, | |
| "grad_norm": 0.2899942994117737, | |
| "learning_rate": 0.0002720331228909005, | |
| "loss": 0.2768, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.05767138193688792, | |
| "grad_norm": 0.6674902439117432, | |
| "learning_rate": 0.00026323086353004075, | |
| "loss": 0.6506, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.058759521218716, | |
| "grad_norm": 0.8436214923858643, | |
| "learning_rate": 0.0002544121201607822, | |
| "loss": 0.8227, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.05984766050054407, | |
| "grad_norm": 0.2817295789718628, | |
| "learning_rate": 0.00024558787983921783, | |
| "loss": 0.3449, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.060935799782372145, | |
| "grad_norm": 0.25585779547691345, | |
| "learning_rate": 0.0002367691364699592, | |
| "loss": 0.2408, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.062023939064200215, | |
| "grad_norm": 0.250396192073822, | |
| "learning_rate": 0.00022796687710909964, | |
| "loss": 0.2588, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.06311207834602829, | |
| "grad_norm": 0.6818592548370361, | |
| "learning_rate": 0.00021919206827571036, | |
| "loss": 0.5172, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.06420021762785637, | |
| "grad_norm": 0.8470868468284607, | |
| "learning_rate": 0.00021045564228892402, | |
| "loss": 0.2823, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.06528835690968444, | |
| "grad_norm": 0.48449358344078064, | |
| "learning_rate": 0.00020176848364762578, | |
| "loss": 0.3356, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.06637649619151251, | |
| "grad_norm": 0.25606584548950195, | |
| "learning_rate": 0.00019314141546972343, | |
| "loss": 0.3552, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.06746463547334058, | |
| "grad_norm": 0.21550142765045166, | |
| "learning_rate": 0.00018458518600788986, | |
| "loss": 0.2025, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.06855277475516866, | |
| "grad_norm": 0.33708736300468445, | |
| "learning_rate": 0.00017611045525857898, | |
| "loss": 0.299, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.06964091403699674, | |
| "grad_norm": 0.41520461440086365, | |
| "learning_rate": 0.0001677277816809975, | |
| "loss": 0.3242, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.07072905331882481, | |
| "grad_norm": 0.9485870003700256, | |
| "learning_rate": 0.00015944760904257942, | |
| "loss": 0.4488, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.07181719260065289, | |
| "grad_norm": 0.36165550351142883, | |
| "learning_rate": 0.0001512802534073522, | |
| "loss": 0.5672, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.07290533188248095, | |
| "grad_norm": 0.6473806500434875, | |
| "learning_rate": 0.00014323589028340596, | |
| "loss": 0.6128, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.07399347116430903, | |
| "grad_norm": 0.2864266633987427, | |
| "learning_rate": 0.00013532454194547733, | |
| "loss": 0.3151, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.0750816104461371, | |
| "grad_norm": 0.2809907793998718, | |
| "learning_rate": 0.00012755606494844294, | |
| "loss": 0.1535, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.07616974972796518, | |
| "grad_norm": 0.32490411400794983, | |
| "learning_rate": 0.00011994013784727947, | |
| "loss": 0.3442, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.07725788900979326, | |
| "grad_norm": 0.6548874974250793, | |
| "learning_rate": 0.00011248624913878966, | |
| "loss": 0.5008, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.07834602829162132, | |
| "grad_norm": 0.7361955046653748, | |
| "learning_rate": 0.0001052036854401166, | |
| "loss": 0.6513, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.0794341675734494, | |
| "grad_norm": 0.5869486927986145, | |
| "learning_rate": 9.810151991877531e-05, | |
| "loss": 0.4508, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.08052230685527748, | |
| "grad_norm": 0.3208440840244293, | |
| "learning_rate": 9.118860098861537e-05, | |
| "loss": 0.325, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.08161044613710555, | |
| "grad_norm": 0.35303372144699097, | |
| "learning_rate": 8.44735412857999e-05, | |
| "loss": 0.327, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.08269858541893363, | |
| "grad_norm": 0.22797748446464539, | |
| "learning_rate": 7.79647069385328e-05, | |
| "loss": 0.3183, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.08378672470076169, | |
| "grad_norm": 0.328909307718277, | |
| "learning_rate": 7.167020714390501e-05, | |
| "loss": 0.3248, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 0.08487486398258977, | |
| "grad_norm": 0.34297263622283936, | |
| "learning_rate": 6.559788406484446e-05, | |
| "loss": 0.2397, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.08596300326441784, | |
| "grad_norm": 0.6872113943099976, | |
| "learning_rate": 5.975530305975807e-05, | |
| "loss": 0.5042, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 0.08705114254624592, | |
| "grad_norm": 0.42344143986701965, | |
| "learning_rate": 5.414974325703686e-05, | |
| "loss": 0.3577, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.088139281828074, | |
| "grad_norm": 0.31334200501441956, | |
| "learning_rate": 4.8788188486168616e-05, | |
| "loss": 0.2931, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.08922742110990206, | |
| "grad_norm": 0.394218385219574, | |
| "learning_rate": 4.367731857675569e-05, | |
| "loss": 0.3908, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 0.09031556039173014, | |
| "grad_norm": 0.3325517475605011, | |
| "learning_rate": 3.882350103627952e-05, | |
| "loss": 0.351, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 0.09140369967355821, | |
| "grad_norm": 0.72450190782547, | |
| "learning_rate": 3.423278311697897e-05, | |
| "loss": 0.5302, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.09249183895538629, | |
| "grad_norm": 0.9300814867019653, | |
| "learning_rate": 2.9910884281727225e-05, | |
| "loss": 0.6356, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.09357997823721437, | |
| "grad_norm": 0.36167338490486145, | |
| "learning_rate": 2.586318907829291e-05, | |
| "loss": 0.4763, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 0.09466811751904244, | |
| "grad_norm": 0.3068345785140991, | |
| "learning_rate": 2.209474043086457e-05, | |
| "loss": 0.3455, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.0957562568008705, | |
| "grad_norm": 0.4473964273929596, | |
| "learning_rate": 1.861023335719475e-05, | |
| "loss": 0.5717, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.09684439608269858, | |
| "grad_norm": 0.31159886717796326, | |
| "learning_rate": 1.5414009119192633e-05, | |
| "loss": 0.298, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 0.09793253536452666, | |
| "grad_norm": 0.2886298894882202, | |
| "learning_rate": 1.25100498142523e-05, | |
| "loss": 0.3209, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.09902067464635474, | |
| "grad_norm": 0.27010682225227356, | |
| "learning_rate": 9.901973414055187e-06, | |
| "loss": 0.2501, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 0.10010881392818281, | |
| "grad_norm": 0.3159748315811157, | |
| "learning_rate": 7.593029257027956e-06, | |
| "loss": 0.2336, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.10119695321001088, | |
| "grad_norm": 0.5132943391799927, | |
| "learning_rate": 5.5860940000714015e-06, | |
| "loss": 0.4026, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.10228509249183895, | |
| "grad_norm": 0.3360811769962311, | |
| "learning_rate": 3.8836680346041594e-06, | |
| "loss": 0.3444, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 0.10337323177366703, | |
| "grad_norm": 0.9271918535232544, | |
| "learning_rate": 2.487872371386424e-06, | |
| "loss": 0.7088, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.1044613710554951, | |
| "grad_norm": 0.7888476252555847, | |
| "learning_rate": 1.4004459980045125e-06, | |
| "loss": 0.5126, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.10554951033732318, | |
| "grad_norm": 0.25628480315208435, | |
| "learning_rate": 6.22743712309054e-07, | |
| "loss": 0.2571, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 0.10663764961915125, | |
| "grad_norm": 0.4536930322647095, | |
| "learning_rate": 1.557344345054501e-07, | |
| "loss": 0.5685, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 0.10772578890097932, | |
| "grad_norm": 0.7662109732627869, | |
| "learning_rate": 0.0, | |
| "loss": 0.5405, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 99, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 300, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.988754859143987e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |