| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.273224043715847, | |
| "eval_steps": 9, | |
| "global_step": 75, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0036429872495446266, | |
| "grad_norm": 0.22306445240974426, | |
| "learning_rate": 1e-05, | |
| "loss": 2.9329, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0036429872495446266, | |
| "eval_loss": 1.4070076942443848, | |
| "eval_runtime": 36.0987, | |
| "eval_samples_per_second": 6.399, | |
| "eval_steps_per_second": 0.803, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.007285974499089253, | |
| "grad_norm": 0.3016580641269684, | |
| "learning_rate": 2e-05, | |
| "loss": 3.0567, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.01092896174863388, | |
| "grad_norm": 0.21999327838420868, | |
| "learning_rate": 3e-05, | |
| "loss": 2.6965, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.014571948998178506, | |
| "grad_norm": 0.24598363041877747, | |
| "learning_rate": 4e-05, | |
| "loss": 3.2788, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.018214936247723135, | |
| "grad_norm": 0.328949898481369, | |
| "learning_rate": 5e-05, | |
| "loss": 3.1415, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.02185792349726776, | |
| "grad_norm": 0.3219306468963623, | |
| "learning_rate": 6e-05, | |
| "loss": 3.211, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.025500910746812388, | |
| "grad_norm": 0.30114132165908813, | |
| "learning_rate": 7e-05, | |
| "loss": 2.6, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.029143897996357013, | |
| "grad_norm": 0.2939751446247101, | |
| "learning_rate": 8e-05, | |
| "loss": 2.7389, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.03278688524590164, | |
| "grad_norm": 0.31080543994903564, | |
| "learning_rate": 9e-05, | |
| "loss": 2.8584, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.03278688524590164, | |
| "eval_loss": 1.3745203018188477, | |
| "eval_runtime": 36.5458, | |
| "eval_samples_per_second": 6.321, | |
| "eval_steps_per_second": 0.794, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.03642987249544627, | |
| "grad_norm": 0.3594372868537903, | |
| "learning_rate": 0.0001, | |
| "loss": 2.5839, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.04007285974499089, | |
| "grad_norm": 0.3138103485107422, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 2.5284, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.04371584699453552, | |
| "grad_norm": 0.40280646085739136, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 3.1069, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.04735883424408015, | |
| "grad_norm": 0.4165895879268646, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 2.2823, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.051001821493624776, | |
| "grad_norm": 0.46014565229415894, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 2.5827, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0546448087431694, | |
| "grad_norm": 0.416968435049057, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 2.9045, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.058287795992714025, | |
| "grad_norm": 0.44254550337791443, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 2.578, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.061930783242258654, | |
| "grad_norm": 0.6331019997596741, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 2.513, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.06557377049180328, | |
| "grad_norm": 0.3989606201648712, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 2.4668, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.06557377049180328, | |
| "eval_loss": 1.2810860872268677, | |
| "eval_runtime": 36.4865, | |
| "eval_samples_per_second": 6.331, | |
| "eval_steps_per_second": 0.795, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0692167577413479, | |
| "grad_norm": 0.4963116943836212, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 2.5629, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.07285974499089254, | |
| "grad_norm": 0.39597874879837036, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 2.8487, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.07650273224043716, | |
| "grad_norm": 0.49734658002853394, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 2.7251, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.08014571948998178, | |
| "grad_norm": 0.5086115002632141, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 2.6009, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.08378870673952642, | |
| "grad_norm": 0.5011266469955444, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 2.6599, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.08743169398907104, | |
| "grad_norm": 0.44377079606056213, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 2.3917, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.09107468123861566, | |
| "grad_norm": 0.4632920026779175, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 2.5564, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0947176684881603, | |
| "grad_norm": 0.5033605098724365, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 2.5636, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.09836065573770492, | |
| "grad_norm": 0.3843136429786682, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 2.7431, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.09836065573770492, | |
| "eval_loss": 1.2466179132461548, | |
| "eval_runtime": 36.1779, | |
| "eval_samples_per_second": 6.385, | |
| "eval_steps_per_second": 0.802, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.10200364298724955, | |
| "grad_norm": 0.4826091229915619, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 2.3946, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.10564663023679417, | |
| "grad_norm": 0.48955684900283813, | |
| "learning_rate": 8.940053768033609e-05, | |
| "loss": 2.5544, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.1092896174863388, | |
| "grad_norm": 0.5087964534759521, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 2.7766, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.11293260473588343, | |
| "grad_norm": 0.463750422000885, | |
| "learning_rate": 8.715724127386972e-05, | |
| "loss": 2.5906, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.11657559198542805, | |
| "grad_norm": 0.3833313286304474, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 2.4696, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.12021857923497267, | |
| "grad_norm": 0.4325249493122101, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 2.4887, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.12386156648451731, | |
| "grad_norm": 1.51544988155365, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 2.427, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.12750455373406194, | |
| "grad_norm": 0.47455018758773804, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 2.3536, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.13114754098360656, | |
| "grad_norm": 0.45129895210266113, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 2.3617, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.13114754098360656, | |
| "eval_loss": 1.231567144393921, | |
| "eval_runtime": 36.1382, | |
| "eval_samples_per_second": 6.392, | |
| "eval_steps_per_second": 0.802, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.13479052823315119, | |
| "grad_norm": 0.47395437955856323, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 2.424, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.1384335154826958, | |
| "grad_norm": 0.4069020748138428, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 2.3032, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.14207650273224043, | |
| "grad_norm": 0.47907042503356934, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 2.6569, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.14571948998178508, | |
| "grad_norm": 0.5535211563110352, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 2.442, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.1493624772313297, | |
| "grad_norm": 0.4240165650844574, | |
| "learning_rate": 7.347357813929454e-05, | |
| "loss": 2.474, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.15300546448087432, | |
| "grad_norm": 0.48536279797554016, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 2.7711, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.15664845173041894, | |
| "grad_norm": 0.44794994592666626, | |
| "learning_rate": 7.033683215379002e-05, | |
| "loss": 2.4307, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.16029143897996356, | |
| "grad_norm": 0.38578876852989197, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 2.5097, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.16393442622950818, | |
| "grad_norm": 0.4519205093383789, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 2.4849, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.16393442622950818, | |
| "eval_loss": 1.223783016204834, | |
| "eval_runtime": 36.2336, | |
| "eval_samples_per_second": 6.375, | |
| "eval_steps_per_second": 0.8, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.16757741347905283, | |
| "grad_norm": 0.5709835290908813, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 2.6203, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.17122040072859745, | |
| "grad_norm": 0.5105118155479431, | |
| "learning_rate": 6.378186779084995e-05, | |
| "loss": 2.4076, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.17486338797814208, | |
| "grad_norm": 0.4190482199192047, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 2.6536, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.1785063752276867, | |
| "grad_norm": 0.5616876482963562, | |
| "learning_rate": 6.0395584540887963e-05, | |
| "loss": 2.171, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.18214936247723132, | |
| "grad_norm": 0.47931408882141113, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 2.3936, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.18579234972677597, | |
| "grad_norm": 0.4086602032184601, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 2.0796, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.1894353369763206, | |
| "grad_norm": 0.4565342962741852, | |
| "learning_rate": 5.522642316338268e-05, | |
| "loss": 2.4687, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.1930783242258652, | |
| "grad_norm": 0.44736742973327637, | |
| "learning_rate": 5.348782368720626e-05, | |
| "loss": 2.6026, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.19672131147540983, | |
| "grad_norm": 0.49989748001098633, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 2.8232, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.19672131147540983, | |
| "eval_loss": 1.2164099216461182, | |
| "eval_runtime": 36.1482, | |
| "eval_samples_per_second": 6.39, | |
| "eval_steps_per_second": 0.802, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.20036429872495445, | |
| "grad_norm": 0.5327097773551941, | |
| "learning_rate": 5e-05, | |
| "loss": 2.1804, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.2040072859744991, | |
| "grad_norm": 0.46552613377571106, | |
| "learning_rate": 4.825502516487497e-05, | |
| "loss": 2.9111, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.20765027322404372, | |
| "grad_norm": 0.47907155752182007, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 2.5331, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.21129326047358835, | |
| "grad_norm": 0.5192755460739136, | |
| "learning_rate": 4.477357683661734e-05, | |
| "loss": 2.452, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.21493624772313297, | |
| "grad_norm": 0.5046710968017578, | |
| "learning_rate": 4.3041344951996746e-05, | |
| "loss": 2.2555, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.2185792349726776, | |
| "grad_norm": 0.4994800388813019, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 2.4397, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.2222222222222222, | |
| "grad_norm": 0.5281521081924438, | |
| "learning_rate": 3.960441545911204e-05, | |
| "loss": 2.9933, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.22586520947176686, | |
| "grad_norm": 0.45675769448280334, | |
| "learning_rate": 3.790390522001662e-05, | |
| "loss": 2.4611, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.22950819672131148, | |
| "grad_norm": 0.4723397493362427, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 2.2367, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.22950819672131148, | |
| "eval_loss": 1.2124544382095337, | |
| "eval_runtime": 36.1593, | |
| "eval_samples_per_second": 6.388, | |
| "eval_steps_per_second": 0.802, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.2331511839708561, | |
| "grad_norm": 0.4308946132659912, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 2.298, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.23679417122040072, | |
| "grad_norm": 0.5040618181228638, | |
| "learning_rate": 3.289899283371657e-05, | |
| "loss": 2.4357, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.24043715846994534, | |
| "grad_norm": 0.46097809076309204, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 2.6279, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.24408014571949, | |
| "grad_norm": 0.4577733278274536, | |
| "learning_rate": 2.9663167846209998e-05, | |
| "loss": 2.6152, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.24772313296903462, | |
| "grad_norm": 0.46525999903678894, | |
| "learning_rate": 2.8081442660546125e-05, | |
| "loss": 2.3304, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.25136612021857924, | |
| "grad_norm": 0.49677371978759766, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 2.5507, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.2550091074681239, | |
| "grad_norm": 0.5101350545883179, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 2.5659, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.2586520947176685, | |
| "grad_norm": 0.4523501992225647, | |
| "learning_rate": 2.350403678833976e-05, | |
| "loss": 2.2424, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.26229508196721313, | |
| "grad_norm": 0.5566951036453247, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 2.3174, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.26229508196721313, | |
| "eval_loss": 1.209464430809021, | |
| "eval_runtime": 36.1745, | |
| "eval_samples_per_second": 6.386, | |
| "eval_steps_per_second": 0.802, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.2659380692167577, | |
| "grad_norm": 0.4237559139728546, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 2.4986, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.26958105646630237, | |
| "grad_norm": 0.5177941918373108, | |
| "learning_rate": 1.9216926233717085e-05, | |
| "loss": 2.2564, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.273224043715847, | |
| "grad_norm": 0.503273606300354, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 2.3239, | |
| "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": 9.5217449435136e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |