|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 1.0, |
|
"eval_steps": 500, |
|
"global_step": 325, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.0, |
|
"grad_norm": 6.116353883934073, |
|
"learning_rate": 6.060606060606061e-06, |
|
"loss": 10.9077, |
|
"step": 1 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"grad_norm": 5.276408725156457, |
|
"learning_rate": 3.0303030303030306e-05, |
|
"loss": 10.8028, |
|
"step": 5 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"grad_norm": 7.553162631560424, |
|
"learning_rate": 6.060606060606061e-05, |
|
"loss": 10.1528, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.05, |
|
"grad_norm": 0.4291891781746429, |
|
"learning_rate": 9.090909090909092e-05, |
|
"loss": 8.605, |
|
"step": 15 |
|
}, |
|
{ |
|
"epoch": 0.06, |
|
"grad_norm": 0.10898194818740227, |
|
"learning_rate": 0.00012121212121212122, |
|
"loss": 8.4331, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.08, |
|
"grad_norm": 0.03189661005888827, |
|
"learning_rate": 0.00015151515151515152, |
|
"loss": 8.4675, |
|
"step": 25 |
|
}, |
|
{ |
|
"epoch": 0.09, |
|
"grad_norm": 0.02334602250201191, |
|
"learning_rate": 0.00018181818181818183, |
|
"loss": 8.472, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.11, |
|
"grad_norm": 0.01196690593776891, |
|
"learning_rate": 0.00019997685019798912, |
|
"loss": 8.4563, |
|
"step": 35 |
|
}, |
|
{ |
|
"epoch": 0.12, |
|
"grad_norm": 3.2907094840511184, |
|
"learning_rate": 0.0001997165380022878, |
|
"loss": 8.4385, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.14, |
|
"grad_norm": 0.0025910382161876, |
|
"learning_rate": 0.000199167731989929, |
|
"loss": 8.4094, |
|
"step": 45 |
|
}, |
|
{ |
|
"epoch": 0.15, |
|
"grad_norm": 0.0042103789632999345, |
|
"learning_rate": 0.0001983320199330545, |
|
"loss": 8.3828, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.17, |
|
"grad_norm": 0.002849586529470528, |
|
"learning_rate": 0.00019721181966290613, |
|
"loss": 8.3641, |
|
"step": 55 |
|
}, |
|
{ |
|
"epoch": 0.18, |
|
"grad_norm": 0.0011645283550178753, |
|
"learning_rate": 0.00019581037207470382, |
|
"loss": 8.3578, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 0.2, |
|
"grad_norm": 0.000614824705284003, |
|
"learning_rate": 0.00019413173175128473, |
|
"loss": 8.3426, |
|
"step": 65 |
|
}, |
|
{ |
|
"epoch": 0.22, |
|
"grad_norm": 0.0006912900930412507, |
|
"learning_rate": 0.00019218075523263104, |
|
"loss": 8.3484, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 0.23, |
|
"grad_norm": 0.005367571192250064, |
|
"learning_rate": 0.00018996308696522433, |
|
"loss": 8.4161, |
|
"step": 75 |
|
}, |
|
{ |
|
"epoch": 0.25, |
|
"grad_norm": 0.0606263367497006, |
|
"learning_rate": 0.00018748514297187648, |
|
"loss": 8.2675, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 0.26, |
|
"grad_norm": 0.01793815434279694, |
|
"learning_rate": 0.00018475409228928312, |
|
"loss": 8.2659, |
|
"step": 85 |
|
}, |
|
{ |
|
"epoch": 0.28, |
|
"grad_norm": 0.008717132473841635, |
|
"learning_rate": 0.00018177783622700327, |
|
"loss": 8.2502, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 0.29, |
|
"grad_norm": 0.0047497997520662555, |
|
"learning_rate": 0.00017856498550787144, |
|
"loss": 8.2204, |
|
"step": 95 |
|
}, |
|
{ |
|
"epoch": 0.31, |
|
"grad_norm": 0.0025777265518574194, |
|
"learning_rate": 0.00017512483535597867, |
|
"loss": 8.1985, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.32, |
|
"grad_norm": 0.0018423967258739478, |
|
"learning_rate": 0.00017146733860429612, |
|
"loss": 8.1782, |
|
"step": 105 |
|
}, |
|
{ |
|
"epoch": 0.34, |
|
"grad_norm": 0.0013575215364357142, |
|
"learning_rate": 0.0001676030768997445, |
|
"loss": 8.1844, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 0.35, |
|
"grad_norm": 0.0009592164974413616, |
|
"learning_rate": 0.00016354323008901776, |
|
"loss": 8.1625, |
|
"step": 115 |
|
}, |
|
{ |
|
"epoch": 0.37, |
|
"grad_norm": 0.0008181054725389062, |
|
"learning_rate": 0.00015929954387373103, |
|
"loss": 8.1688, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 0.38, |
|
"grad_norm": 0.0009746727182352604, |
|
"learning_rate": 0.00015488429582847192, |
|
"loss": 8.194, |
|
"step": 125 |
|
}, |
|
{ |
|
"epoch": 0.4, |
|
"grad_norm": 0.0015308543324832951, |
|
"learning_rate": 0.00015031025988006936, |
|
"loss": 8.1547, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 0.42, |
|
"grad_norm": 0.0017756159985463486, |
|
"learning_rate": 0.00014559066935084588, |
|
"loss": 8.1391, |
|
"step": 135 |
|
}, |
|
{ |
|
"epoch": 0.43, |
|
"grad_norm": 0.0015224483018365321, |
|
"learning_rate": 0.00014073917867277557, |
|
"loss": 8.1406, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 0.45, |
|
"grad_norm": 0.0015693594443860022, |
|
"learning_rate": 0.0001357698238833126, |
|
"loss": 8.1266, |
|
"step": 145 |
|
}, |
|
{ |
|
"epoch": 0.46, |
|
"grad_norm": 0.0010677737954625072, |
|
"learning_rate": 0.000130696982017182, |
|
"loss": 8.125, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 0.48, |
|
"grad_norm": 0.0007802127959643028, |
|
"learning_rate": 0.0001255353295116187, |
|
"loss": 8.1391, |
|
"step": 155 |
|
}, |
|
{ |
|
"epoch": 0.49, |
|
"grad_norm": 0.0006989059044981654, |
|
"learning_rate": 0.00012029979974539234, |
|
"loss": 8.1328, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 0.51, |
|
"grad_norm": 0.0009119182874070918, |
|
"learning_rate": 0.00011500553983446527, |
|
"loss": 8.1359, |
|
"step": 165 |
|
}, |
|
{ |
|
"epoch": 0.52, |
|
"grad_norm": 0.0005539230642084518, |
|
"learning_rate": 0.00010966786680927874, |
|
"loss": 8.1313, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 0.54, |
|
"grad_norm": 0.0005005591061251601, |
|
"learning_rate": 0.00010430222330045304, |
|
"loss": 8.1422, |
|
"step": 175 |
|
}, |
|
{ |
|
"epoch": 0.55, |
|
"grad_norm": 0.0005132965742146893, |
|
"learning_rate": 9.892413286110886e-05, |
|
"loss": 8.1822, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 0.57, |
|
"grad_norm": 0.0014694386938598014, |
|
"learning_rate": 9.354915505506839e-05, |
|
"loss": 8.1359, |
|
"step": 185 |
|
}, |
|
{ |
|
"epoch": 0.58, |
|
"grad_norm": 0.0024110842942386573, |
|
"learning_rate": 8.81928404408726e-05, |
|
"loss": 8.1313, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 0.6, |
|
"grad_norm": 0.0023927225333756773, |
|
"learning_rate": 8.287068558185225e-05, |
|
"loss": 8.1328, |
|
"step": 195 |
|
}, |
|
{ |
|
"epoch": 0.62, |
|
"grad_norm": 0.032000474157054606, |
|
"learning_rate": 7.759808821241406e-05, |
|
"loss": 8.1577, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.63, |
|
"grad_norm": 0.014912360829388252, |
|
"learning_rate": 7.239030269025311e-05, |
|
"loss": 8.1126, |
|
"step": 205 |
|
}, |
|
{ |
|
"epoch": 0.65, |
|
"grad_norm": 0.006980691127090787, |
|
"learning_rate": 6.726239586337408e-05, |
|
"loss": 8.111, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 0.66, |
|
"grad_norm": 0.0027905033268489166, |
|
"learning_rate": 6.22292034796035e-05, |
|
"loss": 8.1094, |
|
"step": 215 |
|
}, |
|
{ |
|
"epoch": 0.68, |
|
"grad_norm": 0.0017283607289640386, |
|
"learning_rate": 5.730528726470792e-05, |
|
"loss": 8.1047, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 0.69, |
|
"grad_norm": 0.002264196853860243, |
|
"learning_rate": 5.2504892793295e-05, |
|
"loss": 8.1047, |
|
"step": 225 |
|
}, |
|
{ |
|
"epoch": 0.71, |
|
"grad_norm": 0.0012025000857218745, |
|
"learning_rate": 4.7841908274384616e-05, |
|
"loss": 8.1156, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 0.72, |
|
"grad_norm": 0.0008764729176935806, |
|
"learning_rate": 4.332982437088825e-05, |
|
"loss": 8.1047, |
|
"step": 235 |
|
}, |
|
{ |
|
"epoch": 0.74, |
|
"grad_norm": 0.0011503436093413008, |
|
"learning_rate": 3.898169516924398e-05, |
|
"loss": 8.1109, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 0.75, |
|
"grad_norm": 0.0007409622754668202, |
|
"learning_rate": 3.4810100412128747e-05, |
|
"loss": 8.0875, |
|
"step": 245 |
|
}, |
|
{ |
|
"epoch": 0.77, |
|
"grad_norm": 0.0006627805954961152, |
|
"learning_rate": 3.0827109103512643e-05, |
|
"loss": 8.1016, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 0.78, |
|
"grad_norm": 0.0008662071988656678, |
|
"learning_rate": 2.7044244591351232e-05, |
|
"loss": 8.1047, |
|
"step": 255 |
|
}, |
|
{ |
|
"epoch": 0.8, |
|
"grad_norm": 0.0005382928432685694, |
|
"learning_rate": 2.3472451228937253e-05, |
|
"loss": 8.1, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 0.82, |
|
"grad_norm": 0.0005771465089242338, |
|
"learning_rate": 2.0122062711363532e-05, |
|
"loss": 8.0938, |
|
"step": 265 |
|
}, |
|
{ |
|
"epoch": 0.83, |
|
"grad_norm": 0.0007199434560267525, |
|
"learning_rate": 1.7002772178705716e-05, |
|
"loss": 8.1109, |
|
"step": 270 |
|
}, |
|
{ |
|
"epoch": 0.85, |
|
"grad_norm": 0.000623592568507816, |
|
"learning_rate": 1.4123604172419713e-05, |
|
"loss": 8.0984, |
|
"step": 275 |
|
}, |
|
{ |
|
"epoch": 0.86, |
|
"grad_norm": 0.0007635159116452594, |
|
"learning_rate": 1.149288852608743e-05, |
|
"loss": 8.0984, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 0.88, |
|
"grad_norm": 0.0005875388330864554, |
|
"learning_rate": 9.118236266049707e-06, |
|
"loss": 8.1047, |
|
"step": 285 |
|
}, |
|
{ |
|
"epoch": 0.89, |
|
"grad_norm": 0.00048181738292063823, |
|
"learning_rate": 7.0065175916482095e-06, |
|
"loss": 8.1016, |
|
"step": 290 |
|
}, |
|
{ |
|
"epoch": 0.91, |
|
"grad_norm": 0.0006907931371687696, |
|
"learning_rate": 5.163841998782837e-06, |
|
"loss": 8.1016, |
|
"step": 295 |
|
}, |
|
{ |
|
"epoch": 0.92, |
|
"grad_norm": 0.0005165153083703728, |
|
"learning_rate": 3.595540604290437e-06, |
|
"loss": 8.0984, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.94, |
|
"grad_norm": 0.0004680021239548418, |
|
"learning_rate": 2.30615072228183e-06, |
|
"loss": 8.0969, |
|
"step": 305 |
|
}, |
|
{ |
|
"epoch": 0.95, |
|
"grad_norm": 0.00047907728646944946, |
|
"learning_rate": 1.2994027370611173e-06, |
|
"loss": 8.0969, |
|
"step": 310 |
|
}, |
|
{ |
|
"epoch": 0.97, |
|
"grad_norm": 0.0004836942034123741, |
|
"learning_rate": 5.782093106048159e-07, |
|
"loss": 8.1016, |
|
"step": 315 |
|
}, |
|
{ |
|
"epoch": 0.98, |
|
"grad_norm": 0.0005764125024224334, |
|
"learning_rate": 1.446569558255395e-07, |
|
"loss": 8.1032, |
|
"step": 320 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"grad_norm": 0.0006775781098562137, |
|
"learning_rate": 0.0, |
|
"loss": 8.0938, |
|
"step": 325 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_loss": NaN, |
|
"eval_runtime": 747.3678, |
|
"eval_samples_per_second": 1.547, |
|
"eval_steps_per_second": 0.387, |
|
"step": 325 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"step": 325, |
|
"total_flos": 5242797101678592.0, |
|
"train_loss": 8.265161878145658, |
|
"train_runtime": 5949.9241, |
|
"train_samples_per_second": 1.747, |
|
"train_steps_per_second": 0.055 |
|
} |
|
], |
|
"logging_steps": 5, |
|
"max_steps": 325, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 1, |
|
"save_steps": 40, |
|
"total_flos": 5242797101678592.0, |
|
"train_batch_size": 4, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|