| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0002243326104838, | |
| "eval_steps": 300, | |
| "global_step": 3344, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00029911014731174755, | |
| "eval_loss": 2.900763511657715, | |
| "eval_runtime": 20.57, | |
| "eval_samples_per_second": 21.39, | |
| "eval_steps_per_second": 21.39, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.007477753682793689, | |
| "grad_norm": 9.5, | |
| "learning_rate": 5e-06, | |
| "loss": 2.3534, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.014955507365587378, | |
| "grad_norm": 9.0, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.022433261048381066, | |
| "grad_norm": 8.1875, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 1.5749, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.029911014731174756, | |
| "grad_norm": 7.0625, | |
| "learning_rate": 2e-05, | |
| "loss": 1.3312, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.037388768413968446, | |
| "grad_norm": 6.84375, | |
| "learning_rate": 1.9999824904190002e-05, | |
| "loss": 1.2434, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.04486652209676213, | |
| "grad_norm": 6.375, | |
| "learning_rate": 1.999929962289171e-05, | |
| "loss": 1.1479, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.05234427577955582, | |
| "grad_norm": 6.625, | |
| "learning_rate": 1.9998424174500043e-05, | |
| "loss": 1.103, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.05982202946234951, | |
| "grad_norm": 6.96875, | |
| "learning_rate": 1.9997198589672462e-05, | |
| "loss": 1.0688, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.0672997831451432, | |
| "grad_norm": 6.34375, | |
| "learning_rate": 1.9995622911327924e-05, | |
| "loss": 1.0343, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.07477753682793689, | |
| "grad_norm": 6.0, | |
| "learning_rate": 1.9993697194645362e-05, | |
| "loss": 0.9735, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.08225529051073058, | |
| "grad_norm": 6.40625, | |
| "learning_rate": 1.9991421507061763e-05, | |
| "loss": 0.975, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.08973304419352426, | |
| "grad_norm": 6.03125, | |
| "learning_rate": 1.99887959282698e-05, | |
| "loss": 0.9508, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.08973304419352426, | |
| "eval_loss": 1.0376813411712646, | |
| "eval_runtime": 19.5994, | |
| "eval_samples_per_second": 22.45, | |
| "eval_steps_per_second": 22.45, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.09721079787631795, | |
| "grad_norm": 5.3125, | |
| "learning_rate": 1.9985820550215044e-05, | |
| "loss": 0.9618, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.10468855155911164, | |
| "grad_norm": 5.3125, | |
| "learning_rate": 1.998249547709273e-05, | |
| "loss": 0.9224, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.11216630524190534, | |
| "grad_norm": 5.3125, | |
| "learning_rate": 1.9978820825344147e-05, | |
| "loss": 0.8982, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.11964405892469902, | |
| "grad_norm": 5.125, | |
| "learning_rate": 1.9974796723652508e-05, | |
| "loss": 0.9116, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.1271218126074927, | |
| "grad_norm": 5.5625, | |
| "learning_rate": 1.9970423312938488e-05, | |
| "loss": 0.8981, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.1345995662902864, | |
| "grad_norm": 5.21875, | |
| "learning_rate": 1.996570074635527e-05, | |
| "loss": 0.8831, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.1420773199730801, | |
| "grad_norm": 5.125, | |
| "learning_rate": 1.9960629189283165e-05, | |
| "loss": 0.8393, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.14955507365587378, | |
| "grad_norm": 5.21875, | |
| "learning_rate": 1.9955208819323864e-05, | |
| "loss": 0.851, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.15703282733866747, | |
| "grad_norm": 4.9375, | |
| "learning_rate": 1.9949439826294178e-05, | |
| "loss": 0.8164, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.16451058102146116, | |
| "grad_norm": 5.28125, | |
| "learning_rate": 1.9943322412219398e-05, | |
| "loss": 0.8388, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.17198833470425484, | |
| "grad_norm": 4.90625, | |
| "learning_rate": 1.9936856791326255e-05, | |
| "loss": 0.8275, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.17946608838704853, | |
| "grad_norm": 5.125, | |
| "learning_rate": 1.9930043190035364e-05, | |
| "loss": 0.8071, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.17946608838704853, | |
| "eval_loss": 0.913748562335968, | |
| "eval_runtime": 20.7611, | |
| "eval_samples_per_second": 21.193, | |
| "eval_steps_per_second": 21.193, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.18694384206984221, | |
| "grad_norm": 4.625, | |
| "learning_rate": 1.9922881846953333e-05, | |
| "loss": 0.8231, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.1944215957526359, | |
| "grad_norm": 4.96875, | |
| "learning_rate": 1.99153730128644e-05, | |
| "loss": 0.8186, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.2018993494354296, | |
| "grad_norm": 4.90625, | |
| "learning_rate": 1.9907516950721638e-05, | |
| "loss": 0.7854, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 0.20937710311822327, | |
| "grad_norm": 4.65625, | |
| "learning_rate": 1.9899313935637764e-05, | |
| "loss": 0.7851, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.216854856801017, | |
| "grad_norm": 4.4375, | |
| "learning_rate": 1.9890764254875488e-05, | |
| "loss": 0.7724, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 0.22433261048381067, | |
| "grad_norm": 4.5, | |
| "learning_rate": 1.9881868207837466e-05, | |
| "loss": 0.7909, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.23181036416660436, | |
| "grad_norm": 4.5625, | |
| "learning_rate": 1.987262610605581e-05, | |
| "loss": 0.7608, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 0.23928811784939805, | |
| "grad_norm": 4.78125, | |
| "learning_rate": 1.9863038273181187e-05, | |
| "loss": 0.765, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.24676587153219173, | |
| "grad_norm": 4.6875, | |
| "learning_rate": 1.985310504497146e-05, | |
| "loss": 0.7695, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 0.2542436252149854, | |
| "grad_norm": 4.3125, | |
| "learning_rate": 1.9842826769279965e-05, | |
| "loss": 0.7542, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.26172137889777913, | |
| "grad_norm": 4.53125, | |
| "learning_rate": 1.9832203806043296e-05, | |
| "loss": 0.7329, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 0.2691991325805728, | |
| "grad_norm": 4.4375, | |
| "learning_rate": 1.9821236527268727e-05, | |
| "loss": 0.7359, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.2691991325805728, | |
| "eval_loss": 0.8687529563903809, | |
| "eval_runtime": 19.5776, | |
| "eval_samples_per_second": 22.475, | |
| "eval_steps_per_second": 22.475, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.2766768862633665, | |
| "grad_norm": 4.375, | |
| "learning_rate": 1.980992531702117e-05, | |
| "loss": 0.731, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 0.2841546399461602, | |
| "grad_norm": 4.34375, | |
| "learning_rate": 1.979827057140973e-05, | |
| "loss": 0.7352, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.2916323936289539, | |
| "grad_norm": 4.875, | |
| "learning_rate": 1.978627269857383e-05, | |
| "loss": 0.72, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 0.29911014731174757, | |
| "grad_norm": 4.375, | |
| "learning_rate": 1.9773932118668924e-05, | |
| "loss": 0.7262, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.30658790099454125, | |
| "grad_norm": 4.40625, | |
| "learning_rate": 1.9761249263851777e-05, | |
| "loss": 0.7234, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 0.31406565467733494, | |
| "grad_norm": 4.40625, | |
| "learning_rate": 1.9748224578265338e-05, | |
| "loss": 0.7164, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.3215434083601286, | |
| "grad_norm": 5.1875, | |
| "learning_rate": 1.973485851802318e-05, | |
| "loss": 0.7072, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 0.3290211620429223, | |
| "grad_norm": 4.9375, | |
| "learning_rate": 1.9721151551193534e-05, | |
| "loss": 0.7091, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.336498915725716, | |
| "grad_norm": 4.25, | |
| "learning_rate": 1.970710415778289e-05, | |
| "loss": 0.715, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 0.3439766694085097, | |
| "grad_norm": 4.59375, | |
| "learning_rate": 1.9692716829719197e-05, | |
| "loss": 0.691, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.35145442309130337, | |
| "grad_norm": 4.5625, | |
| "learning_rate": 1.967799007083462e-05, | |
| "loss": 0.7097, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 0.35893217677409706, | |
| "grad_norm": 3.96875, | |
| "learning_rate": 1.9662924396847923e-05, | |
| "loss": 0.7124, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.35893217677409706, | |
| "eval_loss": 0.8409842848777771, | |
| "eval_runtime": 20.5718, | |
| "eval_samples_per_second": 21.389, | |
| "eval_steps_per_second": 21.389, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.36640993045689074, | |
| "grad_norm": 4.25, | |
| "learning_rate": 1.9647520335346377e-05, | |
| "loss": 0.6843, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 0.37388768413968443, | |
| "grad_norm": 4.03125, | |
| "learning_rate": 1.963177842576731e-05, | |
| "loss": 0.6928, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.3813654378224781, | |
| "grad_norm": 4.15625, | |
| "learning_rate": 1.961569921937921e-05, | |
| "loss": 0.6937, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 0.3888431915052718, | |
| "grad_norm": 4.0, | |
| "learning_rate": 1.9599283279262393e-05, | |
| "loss": 0.6796, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.3963209451880655, | |
| "grad_norm": 4.21875, | |
| "learning_rate": 1.9582531180289342e-05, | |
| "loss": 0.6784, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 0.4037986988708592, | |
| "grad_norm": 4.34375, | |
| "learning_rate": 1.956544350910452e-05, | |
| "loss": 0.6916, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.41127645255365286, | |
| "grad_norm": 4.25, | |
| "learning_rate": 1.954802086410385e-05, | |
| "loss": 0.6677, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 0.41875420623644655, | |
| "grad_norm": 4.5, | |
| "learning_rate": 1.9530263855413763e-05, | |
| "loss": 0.6672, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.42623195991924023, | |
| "grad_norm": 3.96875, | |
| "learning_rate": 1.951217310486982e-05, | |
| "loss": 0.6689, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 0.433709713602034, | |
| "grad_norm": 4.15625, | |
| "learning_rate": 1.9493749245994946e-05, | |
| "loss": 0.6642, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.44118746728482766, | |
| "grad_norm": 3.953125, | |
| "learning_rate": 1.947499292397724e-05, | |
| "loss": 0.652, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 0.44866522096762135, | |
| "grad_norm": 3.84375, | |
| "learning_rate": 1.945590479564738e-05, | |
| "loss": 0.648, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.44866522096762135, | |
| "eval_loss": 0.8143633008003235, | |
| "eval_runtime": 20.3858, | |
| "eval_samples_per_second": 21.584, | |
| "eval_steps_per_second": 21.584, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.45614297465041503, | |
| "grad_norm": 4.125, | |
| "learning_rate": 1.9436485529455628e-05, | |
| "loss": 0.6777, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 0.4636207283332087, | |
| "grad_norm": 4.375, | |
| "learning_rate": 1.941673580544841e-05, | |
| "loss": 0.6505, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 0.4710984820160024, | |
| "grad_norm": 4.09375, | |
| "learning_rate": 1.9396656315244507e-05, | |
| "loss": 0.6435, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 0.4785762356987961, | |
| "grad_norm": 3.734375, | |
| "learning_rate": 1.9376247762010844e-05, | |
| "loss": 0.6271, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.4860539893815898, | |
| "grad_norm": 3.953125, | |
| "learning_rate": 1.9355510860437852e-05, | |
| "loss": 0.6428, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 0.49353174306438347, | |
| "grad_norm": 3.984375, | |
| "learning_rate": 1.9334446336714446e-05, | |
| "loss": 0.6559, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 0.5010094967471771, | |
| "grad_norm": 3.96875, | |
| "learning_rate": 1.9313054928502596e-05, | |
| "loss": 0.6709, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 0.5084872504299708, | |
| "grad_norm": 4.21875, | |
| "learning_rate": 1.929133738491149e-05, | |
| "loss": 0.6513, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.5159650041127646, | |
| "grad_norm": 4.21875, | |
| "learning_rate": 1.9269294466471306e-05, | |
| "loss": 0.6478, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 0.5234427577955583, | |
| "grad_norm": 4.21875, | |
| "learning_rate": 1.9246926945106574e-05, | |
| "loss": 0.6424, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.530920511478352, | |
| "grad_norm": 4.53125, | |
| "learning_rate": 1.9224235604109153e-05, | |
| "loss": 0.6362, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 0.5383982651611456, | |
| "grad_norm": 3.9375, | |
| "learning_rate": 1.9201221238110783e-05, | |
| "loss": 0.6278, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.5383982651611456, | |
| "eval_loss": 0.785617470741272, | |
| "eval_runtime": 20.5638, | |
| "eval_samples_per_second": 21.397, | |
| "eval_steps_per_second": 21.397, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.5458760188439393, | |
| "grad_norm": 3.78125, | |
| "learning_rate": 1.917788465305528e-05, | |
| "loss": 0.6042, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 0.553353772526733, | |
| "grad_norm": 4.0, | |
| "learning_rate": 1.9154226666170296e-05, | |
| "loss": 0.6188, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 0.5608315262095267, | |
| "grad_norm": 5.84375, | |
| "learning_rate": 1.9130248105938705e-05, | |
| "loss": 0.6206, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 0.5683092798923204, | |
| "grad_norm": 3.8125, | |
| "learning_rate": 1.9105949812069592e-05, | |
| "loss": 0.644, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.5757870335751141, | |
| "grad_norm": 3.890625, | |
| "learning_rate": 1.9081332635468844e-05, | |
| "loss": 0.6196, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 0.5832647872579078, | |
| "grad_norm": 4.59375, | |
| "learning_rate": 1.9056397438209366e-05, | |
| "loss": 0.6196, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 0.5907425409407014, | |
| "grad_norm": 3.859375, | |
| "learning_rate": 1.9031145093500855e-05, | |
| "loss": 0.6264, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 0.5982202946234951, | |
| "grad_norm": 4.09375, | |
| "learning_rate": 1.9005576485659274e-05, | |
| "loss": 0.5989, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.6056980483062888, | |
| "grad_norm": 3.796875, | |
| "learning_rate": 1.897969251007584e-05, | |
| "loss": 0.6121, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 0.6131758019890825, | |
| "grad_norm": 4.1875, | |
| "learning_rate": 1.8953494073185684e-05, | |
| "loss": 0.6285, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 0.6206535556718762, | |
| "grad_norm": 4.28125, | |
| "learning_rate": 1.8926982092436117e-05, | |
| "loss": 0.6158, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 0.6281313093546699, | |
| "grad_norm": 3.375, | |
| "learning_rate": 1.890015749625448e-05, | |
| "loss": 0.6174, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.6281313093546699, | |
| "eval_loss": 0.7781485915184021, | |
| "eval_runtime": 21.1506, | |
| "eval_samples_per_second": 20.803, | |
| "eval_steps_per_second": 20.803, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.6356090630374636, | |
| "grad_norm": 3.921875, | |
| "learning_rate": 1.8873021224015662e-05, | |
| "loss": 0.5945, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 0.6430868167202572, | |
| "grad_norm": 3.921875, | |
| "learning_rate": 1.884557422600917e-05, | |
| "loss": 0.6124, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 0.6505645704030509, | |
| "grad_norm": 3.71875, | |
| "learning_rate": 1.8817817463405872e-05, | |
| "loss": 0.6007, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 0.6580423240858446, | |
| "grad_norm": 4.3125, | |
| "learning_rate": 1.878975190822434e-05, | |
| "loss": 0.6045, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.6655200777686383, | |
| "grad_norm": 3.84375, | |
| "learning_rate": 1.8761378543296795e-05, | |
| "loss": 0.5837, | |
| "step": 2225 | |
| }, | |
| { | |
| "epoch": 0.672997831451432, | |
| "grad_norm": 3.515625, | |
| "learning_rate": 1.8732698362234696e-05, | |
| "loss": 0.5839, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 0.6804755851342257, | |
| "grad_norm": 3.875, | |
| "learning_rate": 1.8703712369393953e-05, | |
| "loss": 0.5932, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 0.6879533388170194, | |
| "grad_norm": 4.0625, | |
| "learning_rate": 1.867442157983975e-05, | |
| "loss": 0.5795, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.695431092499813, | |
| "grad_norm": 4.125, | |
| "learning_rate": 1.8644827019310984e-05, | |
| "loss": 0.5887, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 0.7029088461826067, | |
| "grad_norm": 4.21875, | |
| "learning_rate": 1.861492972418437e-05, | |
| "loss": 0.6041, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 0.7103865998654004, | |
| "grad_norm": 3.53125, | |
| "learning_rate": 1.8584730741438128e-05, | |
| "loss": 0.5676, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 0.7178643535481941, | |
| "grad_norm": 4.3125, | |
| "learning_rate": 1.855423112861532e-05, | |
| "loss": 0.5752, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.7178643535481941, | |
| "eval_loss": 0.7577213644981384, | |
| "eval_runtime": 19.9112, | |
| "eval_samples_per_second": 22.098, | |
| "eval_steps_per_second": 22.098, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.7253421072309878, | |
| "grad_norm": 3.6875, | |
| "learning_rate": 1.8523431953786838e-05, | |
| "loss": 0.5731, | |
| "step": 2425 | |
| }, | |
| { | |
| "epoch": 0.7328198609137815, | |
| "grad_norm": 3.890625, | |
| "learning_rate": 1.8492334295513968e-05, | |
| "loss": 0.5611, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 0.7402976145965752, | |
| "grad_norm": 3.859375, | |
| "learning_rate": 1.846093924281065e-05, | |
| "loss": 0.5787, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 0.7477753682793689, | |
| "grad_norm": 3.984375, | |
| "learning_rate": 1.8429247895105314e-05, | |
| "loss": 0.5611, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.7552531219621625, | |
| "grad_norm": 4.15625, | |
| "learning_rate": 1.8397261362202402e-05, | |
| "loss": 0.5805, | |
| "step": 2525 | |
| }, | |
| { | |
| "epoch": 0.7627308756449562, | |
| "grad_norm": 4.03125, | |
| "learning_rate": 1.836498076424349e-05, | |
| "loss": 0.5648, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 0.7702086293277499, | |
| "grad_norm": 3.78125, | |
| "learning_rate": 1.833240723166807e-05, | |
| "loss": 0.5809, | |
| "step": 2575 | |
| }, | |
| { | |
| "epoch": 0.7776863830105436, | |
| "grad_norm": 3.484375, | |
| "learning_rate": 1.8299541905173955e-05, | |
| "loss": 0.5789, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.7851641366933373, | |
| "grad_norm": 3.75, | |
| "learning_rate": 1.8266385935677338e-05, | |
| "loss": 0.5672, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 0.792641890376131, | |
| "grad_norm": 3.796875, | |
| "learning_rate": 1.8232940484272482e-05, | |
| "loss": 0.5657, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 0.8001196440589247, | |
| "grad_norm": 3.875, | |
| "learning_rate": 1.819920672219108e-05, | |
| "loss": 0.5514, | |
| "step": 2675 | |
| }, | |
| { | |
| "epoch": 0.8075973977417183, | |
| "grad_norm": 4.09375, | |
| "learning_rate": 1.8165185830761193e-05, | |
| "loss": 0.5625, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.8075973977417183, | |
| "eval_loss": 0.736329972743988, | |
| "eval_runtime": 19.4873, | |
| "eval_samples_per_second": 22.579, | |
| "eval_steps_per_second": 22.579, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.815075151424512, | |
| "grad_norm": 3.703125, | |
| "learning_rate": 1.8130879001365944e-05, | |
| "loss": 0.5391, | |
| "step": 2725 | |
| }, | |
| { | |
| "epoch": 0.8225529051073057, | |
| "grad_norm": 3.71875, | |
| "learning_rate": 1.8096287435401744e-05, | |
| "loss": 0.5438, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 0.8300306587900994, | |
| "grad_norm": 3.859375, | |
| "learning_rate": 1.8061412344236245e-05, | |
| "loss": 0.5504, | |
| "step": 2775 | |
| }, | |
| { | |
| "epoch": 0.8375084124728931, | |
| "grad_norm": 3.828125, | |
| "learning_rate": 1.8026254949165915e-05, | |
| "loss": 0.5569, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.8449861661556868, | |
| "grad_norm": 3.46875, | |
| "learning_rate": 1.7990816481373267e-05, | |
| "loss": 0.5397, | |
| "step": 2825 | |
| }, | |
| { | |
| "epoch": 0.8524639198384805, | |
| "grad_norm": 3.609375, | |
| "learning_rate": 1.795509818188375e-05, | |
| "loss": 0.5519, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 0.8599416735212743, | |
| "grad_norm": 3.609375, | |
| "learning_rate": 1.791910130152227e-05, | |
| "loss": 0.539, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 0.867419427204068, | |
| "grad_norm": 4.125, | |
| "learning_rate": 1.788282710086942e-05, | |
| "loss": 0.5517, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.8748971808868616, | |
| "grad_norm": 3.421875, | |
| "learning_rate": 1.784627685021731e-05, | |
| "loss": 0.5283, | |
| "step": 2925 | |
| }, | |
| { | |
| "epoch": 0.8823749345696553, | |
| "grad_norm": 3.6875, | |
| "learning_rate": 1.7809451829525083e-05, | |
| "loss": 0.5439, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 0.889852688252449, | |
| "grad_norm": 3.75, | |
| "learning_rate": 1.777235332837411e-05, | |
| "loss": 0.5311, | |
| "step": 2975 | |
| }, | |
| { | |
| "epoch": 0.8973304419352427, | |
| "grad_norm": 3.984375, | |
| "learning_rate": 1.773498264592281e-05, | |
| "loss": 0.5298, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.8973304419352427, | |
| "eval_loss": 0.7269648313522339, | |
| "eval_runtime": 20.5732, | |
| "eval_samples_per_second": 21.387, | |
| "eval_steps_per_second": 21.387, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.9048081956180364, | |
| "grad_norm": 4.125, | |
| "learning_rate": 1.7697341090861163e-05, | |
| "loss": 0.5296, | |
| "step": 3025 | |
| }, | |
| { | |
| "epoch": 0.9122859493008301, | |
| "grad_norm": 3.375, | |
| "learning_rate": 1.7659429981364887e-05, | |
| "loss": 0.5318, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 0.9197637029836238, | |
| "grad_norm": 4.125, | |
| "learning_rate": 1.7621250645049267e-05, | |
| "loss": 0.5406, | |
| "step": 3075 | |
| }, | |
| { | |
| "epoch": 0.9272414566664174, | |
| "grad_norm": 3.6875, | |
| "learning_rate": 1.7582804418922666e-05, | |
| "loss": 0.5264, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 0.9347192103492111, | |
| "grad_norm": 4.25, | |
| "learning_rate": 1.7544092649339704e-05, | |
| "loss": 0.5119, | |
| "step": 3125 | |
| }, | |
| { | |
| "epoch": 0.9421969640320048, | |
| "grad_norm": 3.546875, | |
| "learning_rate": 1.7505116691954117e-05, | |
| "loss": 0.5283, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 0.9496747177147985, | |
| "grad_norm": 3.8125, | |
| "learning_rate": 1.746587791167126e-05, | |
| "loss": 0.5297, | |
| "step": 3175 | |
| }, | |
| { | |
| "epoch": 0.9571524713975922, | |
| "grad_norm": 3.609375, | |
| "learning_rate": 1.7426377682600345e-05, | |
| "loss": 0.52, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.9646302250803859, | |
| "grad_norm": 3.828125, | |
| "learning_rate": 1.738661738800629e-05, | |
| "loss": 0.515, | |
| "step": 3225 | |
| }, | |
| { | |
| "epoch": 0.9721079787631796, | |
| "grad_norm": 3.671875, | |
| "learning_rate": 1.7346598420261294e-05, | |
| "loss": 0.5171, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 0.9795857324459732, | |
| "grad_norm": 3.65625, | |
| "learning_rate": 1.730632218079607e-05, | |
| "loss": 0.5142, | |
| "step": 3275 | |
| }, | |
| { | |
| "epoch": 0.9870634861287669, | |
| "grad_norm": 3.375, | |
| "learning_rate": 1.7265790080050772e-05, | |
| "loss": 0.5355, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 0.9870634861287669, | |
| "eval_loss": 0.706967294216156, | |
| "eval_runtime": 20.875, | |
| "eval_samples_per_second": 21.078, | |
| "eval_steps_per_second": 21.078, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 0.9945412398115606, | |
| "grad_norm": 3.5625, | |
| "learning_rate": 1.7225003537425603e-05, | |
| "loss": 0.5135, | |
| "step": 3325 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 13372, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 1672, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.2890860801163264e+18, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |