| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.04278379920136908, | |
| "eval_steps": 9, | |
| "global_step": 75, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0005704506560182544, | |
| "grad_norm": 2.1348867416381836, | |
| "learning_rate": 1e-05, | |
| "loss": 4.4655, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0005704506560182544, | |
| "eval_loss": 4.3506245613098145, | |
| "eval_runtime": 8.1496, | |
| "eval_samples_per_second": 724.452, | |
| "eval_steps_per_second": 22.7, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0011409013120365088, | |
| "grad_norm": 2.2813360691070557, | |
| "learning_rate": 2e-05, | |
| "loss": 4.3676, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0017113519680547634, | |
| "grad_norm": 2.062041759490967, | |
| "learning_rate": 3e-05, | |
| "loss": 4.2251, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0022818026240730175, | |
| "grad_norm": 2.039959192276001, | |
| "learning_rate": 4e-05, | |
| "loss": 4.2274, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.002852253280091272, | |
| "grad_norm": 2.0442728996276855, | |
| "learning_rate": 5e-05, | |
| "loss": 4.0562, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0034227039361095267, | |
| "grad_norm": 2.0887296199798584, | |
| "learning_rate": 6e-05, | |
| "loss": 3.9563, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.003993154592127781, | |
| "grad_norm": 2.1998727321624756, | |
| "learning_rate": 7e-05, | |
| "loss": 4.432, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.004563605248146035, | |
| "grad_norm": 2.0811333656311035, | |
| "learning_rate": 8e-05, | |
| "loss": 4.1355, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.00513405590416429, | |
| "grad_norm": 1.9647859334945679, | |
| "learning_rate": 9e-05, | |
| "loss": 4.2561, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.00513405590416429, | |
| "eval_loss": 3.972166061401367, | |
| "eval_runtime": 8.067, | |
| "eval_samples_per_second": 731.87, | |
| "eval_steps_per_second": 22.933, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.005704506560182544, | |
| "grad_norm": 1.784217357635498, | |
| "learning_rate": 0.0001, | |
| "loss": 3.9204, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.006274957216200799, | |
| "grad_norm": 1.632825255393982, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 3.7814, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0068454078722190535, | |
| "grad_norm": 1.467962622642517, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 3.6729, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.007415858528237307, | |
| "grad_norm": 1.360505223274231, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 3.8295, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.007986309184255563, | |
| "grad_norm": 1.0902290344238281, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 3.622, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.008556759840273816, | |
| "grad_norm": 1.2049509286880493, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 3.3118, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.00912721049629207, | |
| "grad_norm": 1.5156885385513306, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 3.3338, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.009697661152310326, | |
| "grad_norm": 1.9360144138336182, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 3.5442, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.01026811180832858, | |
| "grad_norm": 1.753395438194275, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 3.3793, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.01026811180832858, | |
| "eval_loss": 3.318361282348633, | |
| "eval_runtime": 8.0828, | |
| "eval_samples_per_second": 730.436, | |
| "eval_steps_per_second": 22.888, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.010838562464346835, | |
| "grad_norm": 1.6109830141067505, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 3.2638, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.011409013120365089, | |
| "grad_norm": 1.2425326108932495, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 3.1719, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.011979463776383342, | |
| "grad_norm": 1.1413416862487793, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 3.2818, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.012549914432401598, | |
| "grad_norm": 0.9109447598457336, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 3.3284, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.013120365088419851, | |
| "grad_norm": 0.802269697189331, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 3.0046, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.013690815744438107, | |
| "grad_norm": 0.8611403703689575, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 3.0243, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.01426126640045636, | |
| "grad_norm": 0.8646649718284607, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 3.0422, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.014831717056474614, | |
| "grad_norm": 0.9183709621429443, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 2.9668, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.01540216771249287, | |
| "grad_norm": 0.8766528367996216, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 3.0137, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.01540216771249287, | |
| "eval_loss": 2.9818027019500732, | |
| "eval_runtime": 8.1382, | |
| "eval_samples_per_second": 725.47, | |
| "eval_steps_per_second": 22.732, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.015972618368511125, | |
| "grad_norm": 0.8579592108726501, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 2.9181, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.01654306902452938, | |
| "grad_norm": 0.8335272073745728, | |
| "learning_rate": 8.940053768033609e-05, | |
| "loss": 2.8452, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.017113519680547633, | |
| "grad_norm": 0.7539997696876526, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 2.9104, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.017683970336565887, | |
| "grad_norm": 0.7295988202095032, | |
| "learning_rate": 8.715724127386972e-05, | |
| "loss": 2.738, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.01825442099258414, | |
| "grad_norm": 0.745151162147522, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 2.845, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.018824871648602397, | |
| "grad_norm": 0.7539723515510559, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 2.9025, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.01939532230462065, | |
| "grad_norm": 0.7784116864204407, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 2.8647, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.019965772960638905, | |
| "grad_norm": 0.7879242897033691, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 2.8638, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.02053622361665716, | |
| "grad_norm": 0.8056729435920715, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 2.7922, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.02053622361665716, | |
| "eval_loss": 2.719639778137207, | |
| "eval_runtime": 8.0538, | |
| "eval_samples_per_second": 733.071, | |
| "eval_steps_per_second": 22.971, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.021106674272675412, | |
| "grad_norm": 0.7197538018226624, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 2.7332, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.02167712492869367, | |
| "grad_norm": 0.6454269289970398, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 2.6288, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.022247575584711923, | |
| "grad_norm": 0.9242546558380127, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 2.8319, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.022818026240730177, | |
| "grad_norm": 0.7071914076805115, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 2.7255, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.02338847689674843, | |
| "grad_norm": 0.6768962144851685, | |
| "learning_rate": 7.347357813929454e-05, | |
| "loss": 2.6548, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.023958927552766685, | |
| "grad_norm": 0.7336259484291077, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 2.5842, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.02452937820878494, | |
| "grad_norm": 0.6378612518310547, | |
| "learning_rate": 7.033683215379002e-05, | |
| "loss": 2.6391, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.025099828864803195, | |
| "grad_norm": 0.6894816756248474, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 2.5276, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.02567027952082145, | |
| "grad_norm": 0.6339684724807739, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 2.5678, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.02567027952082145, | |
| "eval_loss": 2.5278868675231934, | |
| "eval_runtime": 8.0833, | |
| "eval_samples_per_second": 730.395, | |
| "eval_steps_per_second": 22.887, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.026240730176839703, | |
| "grad_norm": 0.6821937561035156, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 2.6051, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.026811180832857957, | |
| "grad_norm": 0.625619113445282, | |
| "learning_rate": 6.378186779084995e-05, | |
| "loss": 2.512, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.027381631488876214, | |
| "grad_norm": 0.6462172865867615, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 2.4584, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.027952082144894468, | |
| "grad_norm": 0.6954731941223145, | |
| "learning_rate": 6.0395584540887963e-05, | |
| "loss": 2.4418, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.02852253280091272, | |
| "grad_norm": 0.6652643084526062, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 2.3786, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.029092983456930975, | |
| "grad_norm": 0.6005697250366211, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 2.4816, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.02966343411294923, | |
| "grad_norm": 0.7842492461204529, | |
| "learning_rate": 5.522642316338268e-05, | |
| "loss": 2.3731, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.030233884768967486, | |
| "grad_norm": 0.6030087471008301, | |
| "learning_rate": 5.348782368720626e-05, | |
| "loss": 2.3131, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.03080433542498574, | |
| "grad_norm": 0.6377494931221008, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 2.3366, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.03080433542498574, | |
| "eval_loss": 2.388643503189087, | |
| "eval_runtime": 8.0803, | |
| "eval_samples_per_second": 730.666, | |
| "eval_steps_per_second": 22.895, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.03137478608100399, | |
| "grad_norm": 0.7002130746841431, | |
| "learning_rate": 5e-05, | |
| "loss": 2.2085, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.03194523673702225, | |
| "grad_norm": 0.7296257615089417, | |
| "learning_rate": 4.825502516487497e-05, | |
| "loss": 2.3402, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.032515687393040504, | |
| "grad_norm": 0.6167798638343811, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 2.315, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.03308613804905876, | |
| "grad_norm": 0.6555004119873047, | |
| "learning_rate": 4.477357683661734e-05, | |
| "loss": 2.4566, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.03365658870507701, | |
| "grad_norm": 0.6642969846725464, | |
| "learning_rate": 4.3041344951996746e-05, | |
| "loss": 2.3569, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.034227039361095266, | |
| "grad_norm": 0.736000657081604, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 2.2542, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.03479749001711352, | |
| "grad_norm": 0.6817083358764648, | |
| "learning_rate": 3.960441545911204e-05, | |
| "loss": 2.4191, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.03536794067313177, | |
| "grad_norm": 0.6803595423698425, | |
| "learning_rate": 3.790390522001662e-05, | |
| "loss": 2.3597, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.03593839132915003, | |
| "grad_norm": 0.724940299987793, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 2.2966, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.03593839132915003, | |
| "eval_loss": 2.291182518005371, | |
| "eval_runtime": 8.0432, | |
| "eval_samples_per_second": 734.038, | |
| "eval_steps_per_second": 23.001, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.03650884198516828, | |
| "grad_norm": 0.6571291089057922, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 2.1965, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.037079292641186534, | |
| "grad_norm": 0.6792440414428711, | |
| "learning_rate": 3.289899283371657e-05, | |
| "loss": 2.2352, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.037649743297204795, | |
| "grad_norm": 0.6271291971206665, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 2.262, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.03822019395322305, | |
| "grad_norm": 0.6394158005714417, | |
| "learning_rate": 2.9663167846209998e-05, | |
| "loss": 2.3448, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.0387906446092413, | |
| "grad_norm": 0.6799787878990173, | |
| "learning_rate": 2.8081442660546125e-05, | |
| "loss": 2.1793, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.039361095265259556, | |
| "grad_norm": 0.6452582478523254, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 2.2576, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.03993154592127781, | |
| "grad_norm": 0.6168031692504883, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 2.1982, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.040501996577296064, | |
| "grad_norm": 0.6837731003761292, | |
| "learning_rate": 2.350403678833976e-05, | |
| "loss": 2.1536, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.04107244723331432, | |
| "grad_norm": 0.7503871917724609, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 2.2486, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.04107244723331432, | |
| "eval_loss": 2.22768497467041, | |
| "eval_runtime": 8.0532, | |
| "eval_samples_per_second": 733.125, | |
| "eval_steps_per_second": 22.972, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.04164289788933257, | |
| "grad_norm": 0.6214445233345032, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 2.3014, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.042213348545350825, | |
| "grad_norm": 0.5839588046073914, | |
| "learning_rate": 1.9216926233717085e-05, | |
| "loss": 2.1236, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.04278379920136908, | |
| "grad_norm": 0.6537484526634216, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 2.0327, | |
| "step": 75 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4172069461819392.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |