| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.36496350364963503, | |
| "eval_steps": 9, | |
| "global_step": 75, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.004866180048661801, | |
| "grad_norm": 1.2298153638839722, | |
| "learning_rate": 1e-05, | |
| "loss": 1.6578, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.004866180048661801, | |
| "eval_loss": 1.3629190921783447, | |
| "eval_runtime": 19.7574, | |
| "eval_samples_per_second": 8.756, | |
| "eval_steps_per_second": 1.114, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.009732360097323601, | |
| "grad_norm": 0.9850460886955261, | |
| "learning_rate": 2e-05, | |
| "loss": 1.32, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.014598540145985401, | |
| "grad_norm": 1.2559576034545898, | |
| "learning_rate": 3e-05, | |
| "loss": 1.7337, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.019464720194647202, | |
| "grad_norm": 1.5054972171783447, | |
| "learning_rate": 4e-05, | |
| "loss": 1.5134, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.024330900243309004, | |
| "grad_norm": 1.3522080183029175, | |
| "learning_rate": 5e-05, | |
| "loss": 2.0051, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.029197080291970802, | |
| "grad_norm": 1.0276397466659546, | |
| "learning_rate": 6e-05, | |
| "loss": 1.1049, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0340632603406326, | |
| "grad_norm": 0.992424488067627, | |
| "learning_rate": 7e-05, | |
| "loss": 1.2848, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.038929440389294405, | |
| "grad_norm": 1.1700958013534546, | |
| "learning_rate": 8e-05, | |
| "loss": 1.2834, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.043795620437956206, | |
| "grad_norm": 1.5791586637496948, | |
| "learning_rate": 9e-05, | |
| "loss": 1.2998, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.043795620437956206, | |
| "eval_loss": 1.24652099609375, | |
| "eval_runtime": 19.0477, | |
| "eval_samples_per_second": 9.082, | |
| "eval_steps_per_second": 1.155, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.04866180048661801, | |
| "grad_norm": 1.362705111503601, | |
| "learning_rate": 0.0001, | |
| "loss": 1.4453, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0535279805352798, | |
| "grad_norm": 1.109766960144043, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 1.083, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.058394160583941604, | |
| "grad_norm": 1.4562040567398071, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 1.232, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.06326034063260341, | |
| "grad_norm": 1.8894864320755005, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 1.3801, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0681265206812652, | |
| "grad_norm": 1.7615281343460083, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 1.4841, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.072992700729927, | |
| "grad_norm": 2.0172765254974365, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 1.7566, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.07785888077858881, | |
| "grad_norm": 1.9831361770629883, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 1.3183, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0827250608272506, | |
| "grad_norm": 1.6174662113189697, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 1.1416, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.08759124087591241, | |
| "grad_norm": 1.2913496494293213, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 1.0978, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.08759124087591241, | |
| "eval_loss": 1.1273767948150635, | |
| "eval_runtime": 19.0434, | |
| "eval_samples_per_second": 9.085, | |
| "eval_steps_per_second": 1.155, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.09245742092457421, | |
| "grad_norm": 1.3372280597686768, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 1.3552, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.09732360097323602, | |
| "grad_norm": 1.3020259141921997, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 1.1153, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.10218978102189781, | |
| "grad_norm": 1.0988091230392456, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 1.1972, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.1070559610705596, | |
| "grad_norm": 1.1756395101547241, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 0.8024, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.11192214111922141, | |
| "grad_norm": 1.6778271198272705, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 1.1661, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.11678832116788321, | |
| "grad_norm": 1.229446530342102, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 1.1406, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.12165450121654502, | |
| "grad_norm": 1.4911742210388184, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 1.2144, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.12652068126520682, | |
| "grad_norm": 1.994720697402954, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 1.5322, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.13138686131386862, | |
| "grad_norm": 1.6444065570831299, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 1.2071, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.13138686131386862, | |
| "eval_loss": 1.0755081176757812, | |
| "eval_runtime": 19.0541, | |
| "eval_samples_per_second": 9.079, | |
| "eval_steps_per_second": 1.155, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.1362530413625304, | |
| "grad_norm": 1.2277100086212158, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 1.2957, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.1411192214111922, | |
| "grad_norm": 1.3212194442749023, | |
| "learning_rate": 8.940053768033609e-05, | |
| "loss": 1.1548, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.145985401459854, | |
| "grad_norm": 1.3625956773757935, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 1.3376, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.15085158150851583, | |
| "grad_norm": 1.485705852508545, | |
| "learning_rate": 8.715724127386972e-05, | |
| "loss": 1.1851, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.15571776155717762, | |
| "grad_norm": 1.6349189281463623, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 1.3334, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.16058394160583941, | |
| "grad_norm": 1.17620849609375, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 0.9048, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.1654501216545012, | |
| "grad_norm": 1.4480947256088257, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 1.2862, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.170316301703163, | |
| "grad_norm": 1.0143804550170898, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 0.8808, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.17518248175182483, | |
| "grad_norm": 1.331067681312561, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.7566, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.17518248175182483, | |
| "eval_loss": 1.0483735799789429, | |
| "eval_runtime": 19.1079, | |
| "eval_samples_per_second": 9.054, | |
| "eval_steps_per_second": 1.151, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.18004866180048662, | |
| "grad_norm": 1.112722396850586, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.8664, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.18491484184914841, | |
| "grad_norm": 1.5237404108047485, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 1.7592, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.1897810218978102, | |
| "grad_norm": 1.3811498880386353, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 1.114, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.19464720194647203, | |
| "grad_norm": 1.8014518022537231, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 1.0599, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.19951338199513383, | |
| "grad_norm": 1.097190022468567, | |
| "learning_rate": 7.347357813929454e-05, | |
| "loss": 1.014, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.20437956204379562, | |
| "grad_norm": 1.0327268838882446, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.9165, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.20924574209245742, | |
| "grad_norm": 1.215659260749817, | |
| "learning_rate": 7.033683215379002e-05, | |
| "loss": 0.9118, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.2141119221411192, | |
| "grad_norm": 1.2937884330749512, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 1.1294, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.21897810218978103, | |
| "grad_norm": 1.3080188035964966, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 1.472, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.21897810218978103, | |
| "eval_loss": 1.0341076850891113, | |
| "eval_runtime": 19.0928, | |
| "eval_samples_per_second": 9.061, | |
| "eval_steps_per_second": 1.152, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.22384428223844283, | |
| "grad_norm": 1.4121612310409546, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 1.3627, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.22871046228710462, | |
| "grad_norm": 1.0651960372924805, | |
| "learning_rate": 6.378186779084995e-05, | |
| "loss": 1.377, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.23357664233576642, | |
| "grad_norm": 1.1841092109680176, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 1.4281, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.2384428223844282, | |
| "grad_norm": 1.1239032745361328, | |
| "learning_rate": 6.0395584540887963e-05, | |
| "loss": 1.0758, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.24330900243309003, | |
| "grad_norm": 1.1765258312225342, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 1.1838, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.24817518248175183, | |
| "grad_norm": 1.2569478750228882, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 1.3188, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.25304136253041365, | |
| "grad_norm": 0.9744678735733032, | |
| "learning_rate": 5.522642316338268e-05, | |
| "loss": 1.0088, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.25790754257907544, | |
| "grad_norm": 1.0326542854309082, | |
| "learning_rate": 5.348782368720626e-05, | |
| "loss": 1.3964, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.26277372262773724, | |
| "grad_norm": 1.4304847717285156, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 1.2375, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.26277372262773724, | |
| "eval_loss": 1.0286833047866821, | |
| "eval_runtime": 19.0435, | |
| "eval_samples_per_second": 9.084, | |
| "eval_steps_per_second": 1.155, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.26763990267639903, | |
| "grad_norm": 1.2375237941741943, | |
| "learning_rate": 5e-05, | |
| "loss": 1.2123, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.2725060827250608, | |
| "grad_norm": 1.014127254486084, | |
| "learning_rate": 4.825502516487497e-05, | |
| "loss": 1.3642, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.2773722627737226, | |
| "grad_norm": 1.0220732688903809, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 0.8832, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.2822384428223844, | |
| "grad_norm": 1.0092318058013916, | |
| "learning_rate": 4.477357683661734e-05, | |
| "loss": 0.7852, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.2871046228710462, | |
| "grad_norm": 0.9593155980110168, | |
| "learning_rate": 4.3041344951996746e-05, | |
| "loss": 0.9424, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.291970802919708, | |
| "grad_norm": 1.1366311311721802, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 1.3309, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.29683698296836986, | |
| "grad_norm": 1.061318039894104, | |
| "learning_rate": 3.960441545911204e-05, | |
| "loss": 0.8699, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.30170316301703165, | |
| "grad_norm": 1.5996885299682617, | |
| "learning_rate": 3.790390522001662e-05, | |
| "loss": 1.4451, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.30656934306569344, | |
| "grad_norm": 1.3761976957321167, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 1.2438, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.30656934306569344, | |
| "eval_loss": 1.0191401243209839, | |
| "eval_runtime": 19.0485, | |
| "eval_samples_per_second": 9.082, | |
| "eval_steps_per_second": 1.155, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.31143552311435524, | |
| "grad_norm": 1.4792134761810303, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 1.4994, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.31630170316301703, | |
| "grad_norm": 1.02501380443573, | |
| "learning_rate": 3.289899283371657e-05, | |
| "loss": 0.8248, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.32116788321167883, | |
| "grad_norm": 1.9706014394760132, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 1.2776, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.3260340632603406, | |
| "grad_norm": 1.3999546766281128, | |
| "learning_rate": 2.9663167846209998e-05, | |
| "loss": 1.1599, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.3309002433090024, | |
| "grad_norm": 1.561223030090332, | |
| "learning_rate": 2.8081442660546125e-05, | |
| "loss": 0.8594, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.3357664233576642, | |
| "grad_norm": 1.1113779544830322, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 1.0403, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.340632603406326, | |
| "grad_norm": 1.1394065618515015, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 1.4539, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.34549878345498786, | |
| "grad_norm": 1.26051926612854, | |
| "learning_rate": 2.350403678833976e-05, | |
| "loss": 1.5036, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.35036496350364965, | |
| "grad_norm": 0.9440858960151672, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 1.1182, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.35036496350364965, | |
| "eval_loss": 1.0112653970718384, | |
| "eval_runtime": 19.0703, | |
| "eval_samples_per_second": 9.072, | |
| "eval_steps_per_second": 1.154, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.35523114355231145, | |
| "grad_norm": 1.110795021057129, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.8803, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.36009732360097324, | |
| "grad_norm": 1.3027766942977905, | |
| "learning_rate": 1.9216926233717085e-05, | |
| "loss": 1.238, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.36496350364963503, | |
| "grad_norm": 1.2464637756347656, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 0.8754, | |
| "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": 5.675437719237427e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |