| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 2.293205499649048, | |
| "learning_rate": 1.3333333333333334e-06, | |
| "loss": 1.4831, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 1.8984652757644653, | |
| "learning_rate": 2.666666666666667e-06, | |
| "loss": 1.099, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 2.401423454284668, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.316, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 2.0406532287597656, | |
| "learning_rate": 5.333333333333334e-06, | |
| "loss": 1.2658, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 2.89574933052063, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 1.1211, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.924607515335083, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.9488, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 2.5476768016815186, | |
| "learning_rate": 9.333333333333334e-06, | |
| "loss": 0.6864, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 2.355384349822998, | |
| "learning_rate": 9.99864620589731e-06, | |
| "loss": 0.3153, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.7077730894088745, | |
| "learning_rate": 9.987820251299121e-06, | |
| "loss": 0.2615, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.7769581079483032, | |
| "learning_rate": 9.966191788709716e-06, | |
| "loss": 0.1628, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 0.9826279878616333, | |
| "learning_rate": 9.933807660562898e-06, | |
| "loss": 0.1485, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.7460083961486816, | |
| "learning_rate": 9.890738003669029e-06, | |
| "loss": 0.2204, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.6706728935241699, | |
| "learning_rate": 9.83707609731432e-06, | |
| "loss": 0.1615, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.6467829346656799, | |
| "learning_rate": 9.77293816123866e-06, | |
| "loss": 0.1887, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.5178117752075195, | |
| "learning_rate": 9.698463103929542e-06, | |
| "loss": 0.1629, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.34681811928749084, | |
| "learning_rate": 9.613812221777212e-06, | |
| "loss": 0.1072, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 0.11321849375963211, | |
| "learning_rate": 9.519168849742603e-06, | |
| "loss": 0.1404, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 1.2257877588272095, | |
| "learning_rate": 9.414737964294636e-06, | |
| "loss": 0.1061, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 0.6019449234008789, | |
| "learning_rate": 9.30074573947683e-06, | |
| "loss": 0.1364, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.4469185173511505, | |
| "learning_rate": 9.177439057064684e-06, | |
| "loss": 0.1396, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 0.5408800840377808, | |
| "learning_rate": 9.045084971874738e-06, | |
| "loss": 0.1369, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.5450402498245239, | |
| "learning_rate": 8.903970133383297e-06, | |
| "loss": 0.1007, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 0.40399786829948425, | |
| "learning_rate": 8.754400164907496e-06, | |
| "loss": 0.0916, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.6208029389381409, | |
| "learning_rate": 8.596699001693257e-06, | |
| "loss": 0.0827, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.8427360653877258, | |
| "learning_rate": 8.43120818934367e-06, | |
| "loss": 0.0777, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 0.4341322183609009, | |
| "learning_rate": 8.258286144107277e-06, | |
| "loss": 0.0688, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 0.08183720707893372, | |
| "learning_rate": 8.078307376628292e-06, | |
| "loss": 0.0835, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 1.1468851566314697, | |
| "learning_rate": 7.891661680839932e-06, | |
| "loss": 0.0996, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 0.7030799388885498, | |
| "learning_rate": 7.698753289757565e-06, | |
| "loss": 0.1009, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 0.3389611840248108, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.0629, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "grad_norm": 0.27755409479141235, | |
| "learning_rate": 7.295832266935059e-06, | |
| "loss": 0.0931, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 0.31552398204803467, | |
| "learning_rate": 7.08669227240909e-06, | |
| "loss": 0.0795, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "grad_norm": 0.362184077501297, | |
| "learning_rate": 6.873032967079562e-06, | |
| "loss": 0.0542, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.3599999999999999, | |
| "grad_norm": 0.7962085604667664, | |
| "learning_rate": 6.655317089424791e-06, | |
| "loss": 0.0408, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 0.6351883411407471, | |
| "learning_rate": 6.434016163555452e-06, | |
| "loss": 0.0547, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 0.0651586651802063, | |
| "learning_rate": 6.209609477998339e-06, | |
| "loss": 0.0859, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "grad_norm": 1.0576623678207397, | |
| "learning_rate": 5.982583047664151e-06, | |
| "loss": 0.0766, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 0.042333897203207016, | |
| "learning_rate": 5.753428561247416e-06, | |
| "loss": 0.0711, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "grad_norm": 0.024764494970440865, | |
| "learning_rate": 5.522642316338268e-06, | |
| "loss": 0.0397, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.4914109706878662, | |
| "learning_rate": 5.290724144552379e-06, | |
| "loss": 0.0457, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.6400000000000001, | |
| "grad_norm": 0.31925445795059204, | |
| "learning_rate": 5.0581763290069865e-06, | |
| "loss": 0.0611, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.6800000000000002, | |
| "grad_norm": 0.8055039048194885, | |
| "learning_rate": 4.825502516487497e-06, | |
| "loss": 0.0922, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "grad_norm": 0.5628638863563538, | |
| "learning_rate": 4.59320662666071e-06, | |
| "loss": 0.0406, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 0.20605020225048065, | |
| "learning_rate": 4.361791760697027e-06, | |
| "loss": 0.0487, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 0.09808018803596497, | |
| "learning_rate": 4.131759111665349e-06, | |
| "loss": 0.066, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.8399999999999999, | |
| "grad_norm": 0.03515284135937691, | |
| "learning_rate": 3.903606879060483e-06, | |
| "loss": 0.0375, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 0.28039515018463135, | |
| "learning_rate": 3.6778291898139907e-06, | |
| "loss": 0.0745, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.22163686156272888, | |
| "learning_rate": 3.4549150281252635e-06, | |
| "loss": 0.064, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "grad_norm": 0.6988674402236938, | |
| "learning_rate": 3.2353471764306567e-06, | |
| "loss": 0.0828, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.21347041428089142, | |
| "learning_rate": 3.019601169804216e-06, | |
| "loss": 0.07, | |
| "step": 500 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 750, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 10, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.162145881587712e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |