| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.02258403732012167, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00022584037320121672, | |
| "eval_loss": 10.381133079528809, | |
| "eval_runtime": 113.7963, | |
| "eval_samples_per_second": 131.068, | |
| "eval_steps_per_second": 65.538, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0006775211196036502, | |
| "grad_norm": 0.031333375722169876, | |
| "learning_rate": 3e-05, | |
| "loss": 10.3808, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0013550422392073003, | |
| "grad_norm": 0.031410422176122665, | |
| "learning_rate": 6e-05, | |
| "loss": 10.3818, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0020325633588109504, | |
| "grad_norm": 0.028049860149621964, | |
| "learning_rate": 9e-05, | |
| "loss": 10.3804, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0020325633588109504, | |
| "eval_loss": 10.380571365356445, | |
| "eval_runtime": 114.7582, | |
| "eval_samples_per_second": 129.969, | |
| "eval_steps_per_second": 64.989, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0027100844784146007, | |
| "grad_norm": 0.030115395784378052, | |
| "learning_rate": 0.00012, | |
| "loss": 10.3802, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.003387605598018251, | |
| "grad_norm": 0.030468817800283432, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 10.3799, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.004065126717621901, | |
| "grad_norm": 0.03643874451518059, | |
| "learning_rate": 0.00018, | |
| "loss": 10.3789, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.004065126717621901, | |
| "eval_loss": 10.378609657287598, | |
| "eval_runtime": 114.6566, | |
| "eval_samples_per_second": 130.084, | |
| "eval_steps_per_second": 65.046, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.004742647837225551, | |
| "grad_norm": 0.03757527098059654, | |
| "learning_rate": 0.0001999229036240723, | |
| "loss": 10.3778, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.005420168956829201, | |
| "grad_norm": 0.05077025294303894, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 10.3773, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.006097690076432852, | |
| "grad_norm": 0.06361916661262512, | |
| "learning_rate": 0.00019624552364536473, | |
| "loss": 10.3755, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.006097690076432852, | |
| "eval_loss": 10.374011993408203, | |
| "eval_runtime": 113.7536, | |
| "eval_samples_per_second": 131.117, | |
| "eval_steps_per_second": 65.563, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.006775211196036502, | |
| "grad_norm": 0.06706535816192627, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 10.3735, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.007452732315640151, | |
| "grad_norm": 0.09487693011760712, | |
| "learning_rate": 0.00018724960070727972, | |
| "loss": 10.3707, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.008130253435243802, | |
| "grad_norm": 0.09645885974168777, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 10.3681, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.008130253435243802, | |
| "eval_loss": 10.36545181274414, | |
| "eval_runtime": 113.0342, | |
| "eval_samples_per_second": 131.951, | |
| "eval_steps_per_second": 65.98, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.008807774554847452, | |
| "grad_norm": 0.09872834384441376, | |
| "learning_rate": 0.00017343225094356855, | |
| "loss": 10.3647, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.009485295674451102, | |
| "grad_norm": 0.082485131919384, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 10.3616, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.010162816794054752, | |
| "grad_norm": 0.0801018700003624, | |
| "learning_rate": 0.00015555702330196023, | |
| "loss": 10.3585, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.010162816794054752, | |
| "eval_loss": 10.357089042663574, | |
| "eval_runtime": 113.8095, | |
| "eval_samples_per_second": 131.052, | |
| "eval_steps_per_second": 65.531, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.010840337913658403, | |
| "grad_norm": 0.06634080410003662, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 10.3567, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.011517859033262053, | |
| "grad_norm": 0.06341370195150375, | |
| "learning_rate": 0.0001346117057077493, | |
| "loss": 10.3552, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.012195380152865703, | |
| "grad_norm": 0.05498109757900238, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 10.3544, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.012195380152865703, | |
| "eval_loss": 10.353216171264648, | |
| "eval_runtime": 114.8408, | |
| "eval_samples_per_second": 129.875, | |
| "eval_steps_per_second": 64.942, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.012872901272469353, | |
| "grad_norm": 0.048390697687864304, | |
| "learning_rate": 0.00011175373974578378, | |
| "loss": 10.3538, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.013550422392073004, | |
| "grad_norm": 0.0455036386847496, | |
| "learning_rate": 0.0001, | |
| "loss": 10.3531, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.014227943511676652, | |
| "grad_norm": 0.041428498923778534, | |
| "learning_rate": 8.824626025421626e-05, | |
| "loss": 10.3526, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.014227943511676652, | |
| "eval_loss": 10.351635932922363, | |
| "eval_runtime": 114.7051, | |
| "eval_samples_per_second": 130.029, | |
| "eval_steps_per_second": 65.019, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.014905464631280303, | |
| "grad_norm": 0.040886059403419495, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 10.3521, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.015582985750883953, | |
| "grad_norm": 0.031167220324277878, | |
| "learning_rate": 6.538829429225069e-05, | |
| "loss": 10.3522, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.016260506870487603, | |
| "grad_norm": 0.030341865494847298, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 10.3524, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.016260506870487603, | |
| "eval_loss": 10.350973129272461, | |
| "eval_runtime": 114.5051, | |
| "eval_samples_per_second": 130.256, | |
| "eval_steps_per_second": 65.132, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.016938027990091253, | |
| "grad_norm": 0.02817876823246479, | |
| "learning_rate": 4.444297669803981e-05, | |
| "loss": 10.3521, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.017615549109694904, | |
| "grad_norm": 0.025949513539671898, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 10.351, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.018293070229298554, | |
| "grad_norm": 0.022331327199935913, | |
| "learning_rate": 2.6567749056431467e-05, | |
| "loss": 10.3507, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.018293070229298554, | |
| "eval_loss": 10.35069751739502, | |
| "eval_runtime": 113.8246, | |
| "eval_samples_per_second": 131.035, | |
| "eval_steps_per_second": 65.522, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.018970591348902204, | |
| "grad_norm": 0.02584969997406006, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 10.3516, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.019648112468505854, | |
| "grad_norm": 0.029781047254800797, | |
| "learning_rate": 1.2750399292720283e-05, | |
| "loss": 10.3506, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.020325633588109505, | |
| "grad_norm": 0.022765731438994408, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 10.3504, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.020325633588109505, | |
| "eval_loss": 10.350595474243164, | |
| "eval_runtime": 113.4837, | |
| "eval_samples_per_second": 131.429, | |
| "eval_steps_per_second": 65.719, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.021003154707713155, | |
| "grad_norm": 0.026100726798176765, | |
| "learning_rate": 3.7544763546352834e-06, | |
| "loss": 10.3511, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.021680675827316805, | |
| "grad_norm": 0.0227440744638443, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 10.3517, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.022358196946920456, | |
| "grad_norm": 0.02963266521692276, | |
| "learning_rate": 7.709637592770991e-08, | |
| "loss": 10.3508, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.022358196946920456, | |
| "eval_loss": 10.350578308105469, | |
| "eval_runtime": 114.4993, | |
| "eval_samples_per_second": 130.263, | |
| "eval_steps_per_second": 65.136, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 42768059596800.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |