| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.114616497829233, | |
| "eval_steps": 50, | |
| "global_step": 99, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0011577424023154848, | |
| "grad_norm": 0.2338106632232666, | |
| "learning_rate": 5e-05, | |
| "loss": 1.067, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0011577424023154848, | |
| "eval_loss": 1.1207107305526733, | |
| "eval_runtime": 12.9724, | |
| "eval_samples_per_second": 28.06, | |
| "eval_steps_per_second": 14.03, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0023154848046309695, | |
| "grad_norm": 0.30521467328071594, | |
| "learning_rate": 0.0001, | |
| "loss": 1.209, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0034732272069464545, | |
| "grad_norm": 0.27888786792755127, | |
| "learning_rate": 0.00015, | |
| "loss": 1.317, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.004630969609261939, | |
| "grad_norm": 0.3030368685722351, | |
| "learning_rate": 0.0002, | |
| "loss": 0.9915, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.005788712011577424, | |
| "grad_norm": 0.3091076612472534, | |
| "learning_rate": 0.00025, | |
| "loss": 1.1124, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.006946454413892909, | |
| "grad_norm": 0.49404653906822205, | |
| "learning_rate": 0.0003, | |
| "loss": 1.1443, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.008104196816208394, | |
| "grad_norm": 0.45804426074028015, | |
| "learning_rate": 0.00035, | |
| "loss": 0.848, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.009261939218523878, | |
| "grad_norm": 0.42264845967292786, | |
| "learning_rate": 0.0004, | |
| "loss": 0.8745, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.010419681620839363, | |
| "grad_norm": 0.4398542642593384, | |
| "learning_rate": 0.00045000000000000004, | |
| "loss": 0.7968, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.011577424023154847, | |
| "grad_norm": 0.5333066582679749, | |
| "learning_rate": 0.0005, | |
| "loss": 0.6171, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.012735166425470333, | |
| "grad_norm": 0.6457921862602234, | |
| "learning_rate": 0.0004998442655654946, | |
| "loss": 0.6853, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.013892908827785818, | |
| "grad_norm": 0.6219100952148438, | |
| "learning_rate": 0.0004993772562876909, | |
| "loss": 0.8119, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.015050651230101303, | |
| "grad_norm": 0.4639165997505188, | |
| "learning_rate": 0.0004985995540019955, | |
| "loss": 0.5416, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.016208393632416787, | |
| "grad_norm": 0.4277282655239105, | |
| "learning_rate": 0.0004975121276286136, | |
| "loss": 0.5533, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.017366136034732273, | |
| "grad_norm": 0.5572855472564697, | |
| "learning_rate": 0.0004961163319653958, | |
| "loss": 0.7072, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.018523878437047756, | |
| "grad_norm": 0.4737989604473114, | |
| "learning_rate": 0.0004944139059999286, | |
| "loss": 0.6697, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.019681620839363242, | |
| "grad_norm": 0.4247514009475708, | |
| "learning_rate": 0.000492406970742972, | |
| "loss": 0.5848, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.020839363241678725, | |
| "grad_norm": 0.40353336930274963, | |
| "learning_rate": 0.0004900980265859448, | |
| "loss": 0.6247, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.02199710564399421, | |
| "grad_norm": 0.426436185836792, | |
| "learning_rate": 0.0004874899501857477, | |
| "loss": 0.608, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.023154848046309694, | |
| "grad_norm": 0.49942946434020996, | |
| "learning_rate": 0.00048458599088080736, | |
| "loss": 0.671, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.02431259044862518, | |
| "grad_norm": 0.39419907331466675, | |
| "learning_rate": 0.0004813897666428053, | |
| "loss": 0.5482, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.025470332850940667, | |
| "grad_norm": 0.36672383546829224, | |
| "learning_rate": 0.00047790525956913543, | |
| "loss": 0.5659, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.02662807525325615, | |
| "grad_norm": 0.3333776891231537, | |
| "learning_rate": 0.0004741368109217071, | |
| "loss": 0.5094, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.027785817655571636, | |
| "grad_norm": 0.3671005666255951, | |
| "learning_rate": 0.00047008911571827283, | |
| "loss": 0.5006, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.02894356005788712, | |
| "grad_norm": 0.2863791882991791, | |
| "learning_rate": 0.00046576721688302105, | |
| "loss": 0.4838, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.030101302460202605, | |
| "grad_norm": 0.37180379033088684, | |
| "learning_rate": 0.0004611764989637205, | |
| "loss": 0.447, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.03125904486251809, | |
| "grad_norm": 0.3161357343196869, | |
| "learning_rate": 0.0004563226814232444, | |
| "loss": 0.468, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.032416787264833574, | |
| "grad_norm": 0.42159467935562134, | |
| "learning_rate": 0.0004512118115138315, | |
| "loss": 0.5334, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.03357452966714906, | |
| "grad_norm": 0.3421444892883301, | |
| "learning_rate": 0.0004458502567429631, | |
| "loss": 0.4989, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.03473227206946455, | |
| "grad_norm": 0.37070757150650024, | |
| "learning_rate": 0.00044024469694024196, | |
| "loss": 0.4877, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.03589001447178003, | |
| "grad_norm": 0.3389771580696106, | |
| "learning_rate": 0.00043440211593515554, | |
| "loss": 0.454, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.03704775687409551, | |
| "grad_norm": 0.3566928803920746, | |
| "learning_rate": 0.0004283297928560951, | |
| "loss": 0.6784, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.038205499276410995, | |
| "grad_norm": 0.2971787452697754, | |
| "learning_rate": 0.0004220352930614672, | |
| "loss": 0.5285, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.039363241678726485, | |
| "grad_norm": 0.3165189027786255, | |
| "learning_rate": 0.00041552645871420013, | |
| "loss": 0.4749, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.04052098408104197, | |
| "grad_norm": 0.2689637839794159, | |
| "learning_rate": 0.00040881139901138467, | |
| "loss": 0.441, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.04167872648335745, | |
| "grad_norm": 0.3806246221065521, | |
| "learning_rate": 0.00040189848008122475, | |
| "loss": 0.537, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.04283646888567294, | |
| "grad_norm": 0.2700049579143524, | |
| "learning_rate": 0.00039479631455988334, | |
| "loss": 0.5311, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.04399421128798842, | |
| "grad_norm": 0.3233337998390198, | |
| "learning_rate": 0.0003875137508612103, | |
| "loss": 0.4348, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.045151953690303906, | |
| "grad_norm": 0.2585344612598419, | |
| "learning_rate": 0.00038005986215272055, | |
| "loss": 0.4228, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.04630969609261939, | |
| "grad_norm": 0.2899521589279175, | |
| "learning_rate": 0.0003724439350515571, | |
| "loss": 0.4575, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.04746743849493488, | |
| "grad_norm": 0.5732170343399048, | |
| "learning_rate": 0.0003646754580545226, | |
| "loss": 0.4874, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.04862518089725036, | |
| "grad_norm": 0.2810184955596924, | |
| "learning_rate": 0.000356764109716594, | |
| "loss": 0.4141, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.049782923299565844, | |
| "grad_norm": 0.3357601463794708, | |
| "learning_rate": 0.00034871974659264783, | |
| "loss": 0.5582, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.050940665701881334, | |
| "grad_norm": 0.299430787563324, | |
| "learning_rate": 0.0003405523909574206, | |
| "loss": 0.4751, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.05209840810419682, | |
| "grad_norm": 0.2595735490322113, | |
| "learning_rate": 0.0003322722183190025, | |
| "loss": 0.3575, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0532561505065123, | |
| "grad_norm": 0.32403305172920227, | |
| "learning_rate": 0.0003238895447414211, | |
| "loss": 0.4585, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.05441389290882779, | |
| "grad_norm": 0.3115895390510559, | |
| "learning_rate": 0.0003154148139921102, | |
| "loss": 0.6248, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.05557163531114327, | |
| "grad_norm": 0.26747116446495056, | |
| "learning_rate": 0.00030685858453027663, | |
| "loss": 0.4572, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.056729377713458755, | |
| "grad_norm": 0.24940979480743408, | |
| "learning_rate": 0.0002982315163523742, | |
| "loss": 0.3957, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.05788712011577424, | |
| "grad_norm": 0.31252002716064453, | |
| "learning_rate": 0.000289544357711076, | |
| "loss": 0.3606, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.05788712011577424, | |
| "eval_loss": 0.4854944944381714, | |
| "eval_runtime": 12.8574, | |
| "eval_samples_per_second": 28.311, | |
| "eval_steps_per_second": 14.155, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.05904486251808973, | |
| "grad_norm": 0.26592421531677246, | |
| "learning_rate": 0.0002808079317242896, | |
| "loss": 0.5123, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.06020260492040521, | |
| "grad_norm": 0.3215560019016266, | |
| "learning_rate": 0.0002720331228909005, | |
| "loss": 0.6261, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.06136034732272069, | |
| "grad_norm": 0.2619563043117523, | |
| "learning_rate": 0.00026323086353004075, | |
| "loss": 0.4538, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.06251808972503618, | |
| "grad_norm": 0.25585314631462097, | |
| "learning_rate": 0.0002544121201607822, | |
| "loss": 0.4621, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.06367583212735166, | |
| "grad_norm": 0.2522573471069336, | |
| "learning_rate": 0.00024558787983921783, | |
| "loss": 0.4833, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.06483357452966715, | |
| "grad_norm": 0.27701646089553833, | |
| "learning_rate": 0.0002367691364699592, | |
| "loss": 0.4159, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.06599131693198264, | |
| "grad_norm": 0.33121973276138306, | |
| "learning_rate": 0.00022796687710909964, | |
| "loss": 0.5848, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.06714905933429811, | |
| "grad_norm": 0.27029693126678467, | |
| "learning_rate": 0.00021919206827571036, | |
| "loss": 0.4566, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.0683068017366136, | |
| "grad_norm": 0.3517721891403198, | |
| "learning_rate": 0.00021045564228892402, | |
| "loss": 0.4093, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.0694645441389291, | |
| "grad_norm": 0.3207434117794037, | |
| "learning_rate": 0.00020176848364762578, | |
| "loss": 0.432, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.07062228654124457, | |
| "grad_norm": 0.25897809863090515, | |
| "learning_rate": 0.00019314141546972343, | |
| "loss": 0.4708, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.07178002894356006, | |
| "grad_norm": 0.34539109468460083, | |
| "learning_rate": 0.00018458518600788986, | |
| "loss": 0.4638, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.07293777134587555, | |
| "grad_norm": 0.24760332703590393, | |
| "learning_rate": 0.00017611045525857898, | |
| "loss": 0.3803, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.07409551374819102, | |
| "grad_norm": 0.26825204491615295, | |
| "learning_rate": 0.0001677277816809975, | |
| "loss": 0.5163, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.07525325615050651, | |
| "grad_norm": 0.43493548035621643, | |
| "learning_rate": 0.00015944760904257942, | |
| "loss": 0.5339, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.07641099855282199, | |
| "grad_norm": 0.26649484038352966, | |
| "learning_rate": 0.0001512802534073522, | |
| "loss": 0.4251, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.07756874095513748, | |
| "grad_norm": 0.273527055978775, | |
| "learning_rate": 0.00014323589028340596, | |
| "loss": 0.4912, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.07872648335745297, | |
| "grad_norm": 0.3016795516014099, | |
| "learning_rate": 0.00013532454194547733, | |
| "loss": 0.4472, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.07988422575976845, | |
| "grad_norm": 0.31133025884628296, | |
| "learning_rate": 0.00012755606494844294, | |
| "loss": 0.5907, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.08104196816208394, | |
| "grad_norm": 0.3809708058834076, | |
| "learning_rate": 0.00011994013784727947, | |
| "loss": 0.586, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.08219971056439943, | |
| "grad_norm": 0.402822881937027, | |
| "learning_rate": 0.00011248624913878966, | |
| "loss": 0.6835, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.0833574529667149, | |
| "grad_norm": 0.2550395131111145, | |
| "learning_rate": 0.0001052036854401166, | |
| "loss": 0.2994, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.08451519536903039, | |
| "grad_norm": 0.2937041223049164, | |
| "learning_rate": 9.810151991877531e-05, | |
| "loss": 0.507, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.08567293777134588, | |
| "grad_norm": 0.3934008479118347, | |
| "learning_rate": 9.118860098861537e-05, | |
| "loss": 0.4647, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.08683068017366136, | |
| "grad_norm": 0.31899431347846985, | |
| "learning_rate": 8.44735412857999e-05, | |
| "loss": 0.3375, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.08798842257597685, | |
| "grad_norm": 0.5414046049118042, | |
| "learning_rate": 7.79647069385328e-05, | |
| "loss": 0.4782, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.08914616497829234, | |
| "grad_norm": 0.2682500183582306, | |
| "learning_rate": 7.167020714390501e-05, | |
| "loss": 0.422, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 0.09030390738060781, | |
| "grad_norm": 0.24800242483615875, | |
| "learning_rate": 6.559788406484446e-05, | |
| "loss": 0.4617, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.0914616497829233, | |
| "grad_norm": 0.361648827791214, | |
| "learning_rate": 5.975530305975807e-05, | |
| "loss": 0.6129, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 0.09261939218523878, | |
| "grad_norm": 0.2895083725452423, | |
| "learning_rate": 5.414974325703686e-05, | |
| "loss": 0.5065, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.09377713458755427, | |
| "grad_norm": 0.26388344168663025, | |
| "learning_rate": 4.8788188486168616e-05, | |
| "loss": 0.5139, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.09493487698986976, | |
| "grad_norm": 0.23608814179897308, | |
| "learning_rate": 4.367731857675569e-05, | |
| "loss": 0.4606, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 0.09609261939218523, | |
| "grad_norm": 0.2052265703678131, | |
| "learning_rate": 3.882350103627952e-05, | |
| "loss": 0.3697, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 0.09725036179450072, | |
| "grad_norm": 0.24611227214336395, | |
| "learning_rate": 3.423278311697897e-05, | |
| "loss": 0.5282, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.09840810419681621, | |
| "grad_norm": 0.23457454144954681, | |
| "learning_rate": 2.9910884281727225e-05, | |
| "loss": 0.4415, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.09956584659913169, | |
| "grad_norm": 0.26964303851127625, | |
| "learning_rate": 2.586318907829291e-05, | |
| "loss": 0.3641, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 0.10072358900144718, | |
| "grad_norm": 0.3212110996246338, | |
| "learning_rate": 2.209474043086457e-05, | |
| "loss": 0.4868, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.10188133140376267, | |
| "grad_norm": 0.38655808568000793, | |
| "learning_rate": 1.861023335719475e-05, | |
| "loss": 0.5066, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.10303907380607814, | |
| "grad_norm": 0.25838810205459595, | |
| "learning_rate": 1.5414009119192633e-05, | |
| "loss": 0.372, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 0.10419681620839363, | |
| "grad_norm": 0.3056221604347229, | |
| "learning_rate": 1.25100498142523e-05, | |
| "loss": 0.4282, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.10535455861070912, | |
| "grad_norm": 0.20704999566078186, | |
| "learning_rate": 9.901973414055187e-06, | |
| "loss": 0.3774, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 0.1065123010130246, | |
| "grad_norm": 0.24676023423671722, | |
| "learning_rate": 7.593029257027956e-06, | |
| "loss": 0.3745, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.10767004341534009, | |
| "grad_norm": 0.30260273814201355, | |
| "learning_rate": 5.5860940000714015e-06, | |
| "loss": 0.4059, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.10882778581765558, | |
| "grad_norm": 0.2307083010673523, | |
| "learning_rate": 3.8836680346041594e-06, | |
| "loss": 0.3894, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 0.10998552821997105, | |
| "grad_norm": 0.3418976068496704, | |
| "learning_rate": 2.487872371386424e-06, | |
| "loss": 0.5039, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.11114327062228654, | |
| "grad_norm": 0.25175032019615173, | |
| "learning_rate": 1.4004459980045125e-06, | |
| "loss": 0.4215, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.11230101302460202, | |
| "grad_norm": 0.24384403228759766, | |
| "learning_rate": 6.22743712309054e-07, | |
| "loss": 0.4133, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 0.11345875542691751, | |
| "grad_norm": 0.2955150604248047, | |
| "learning_rate": 1.557344345054501e-07, | |
| "loss": 0.4622, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 0.114616497829233, | |
| "grad_norm": 0.2382400780916214, | |
| "learning_rate": 0.0, | |
| "loss": 0.4178, | |
| "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": 1.970786582868787e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |