3035 lines
81 KiB
JSON
3035 lines
81 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.033516931636631775,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.692034435272216,
|
|
"epoch": 0.00011172310545543924,
|
|
"grad_norm": 16.25,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.6039,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 4250.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.692024230957031,
|
|
"epoch": 0.00022344621091087847,
|
|
"grad_norm": 17.125,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.5084,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 8228.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.691641712188721,
|
|
"epoch": 0.0003351693163663177,
|
|
"grad_norm": 13.125,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.2399,
|
|
"mean_token_accuracy": 0.01585084570106119,
|
|
"num_tokens": 12209.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.6864333152771,
|
|
"epoch": 0.00044689242182175694,
|
|
"grad_norm": 8.875,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 9.9231,
|
|
"mean_token_accuracy": 0.04045895729213953,
|
|
"num_tokens": 16225.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.658738136291504,
|
|
"epoch": 0.0005586155272771962,
|
|
"grad_norm": 5.71875,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 9.6193,
|
|
"mean_token_accuracy": 0.05384383201599121,
|
|
"num_tokens": 20212.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.617362594604492,
|
|
"epoch": 0.0006703386327326354,
|
|
"grad_norm": 4.46875,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.3987,
|
|
"mean_token_accuracy": 0.06703231446444988,
|
|
"num_tokens": 24445.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.580356121063232,
|
|
"epoch": 0.0007820617381880747,
|
|
"grad_norm": 4.09375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.2682,
|
|
"mean_token_accuracy": 0.06737326122820378,
|
|
"num_tokens": 28365.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.578466510772705,
|
|
"epoch": 0.0008937848436435139,
|
|
"grad_norm": 3.9375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.3685,
|
|
"mean_token_accuracy": 0.05515027567744255,
|
|
"num_tokens": 33055.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.575786018371582,
|
|
"epoch": 0.0010055079490989532,
|
|
"grad_norm": 3.5625,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.2537,
|
|
"mean_token_accuracy": 0.0627420760691166,
|
|
"num_tokens": 37599.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.472075748443604,
|
|
"epoch": 0.0011172310545543925,
|
|
"grad_norm": 3.5,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.1143,
|
|
"mean_token_accuracy": 0.07111909165978432,
|
|
"num_tokens": 41934.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.315015411376953,
|
|
"epoch": 0.0012289541600098315,
|
|
"grad_norm": 3.21875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.0802,
|
|
"mean_token_accuracy": 0.06700236462056637,
|
|
"num_tokens": 46178.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.27945785522461,
|
|
"epoch": 0.0013406772654652708,
|
|
"grad_norm": 3.546875,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 8.9726,
|
|
"mean_token_accuracy": 0.0730497345328331,
|
|
"num_tokens": 50513.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.237419605255127,
|
|
"epoch": 0.0014524003709207101,
|
|
"grad_norm": 2.859375,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 8.9616,
|
|
"mean_token_accuracy": 0.07224201261997223,
|
|
"num_tokens": 54924.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.185122776031495,
|
|
"epoch": 0.0015641234763761494,
|
|
"grad_norm": 3.125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 8.8671,
|
|
"mean_token_accuracy": 0.06654836386442184,
|
|
"num_tokens": 59083.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.172991943359374,
|
|
"epoch": 0.0016758465818315885,
|
|
"grad_norm": 3.9375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 8.7827,
|
|
"mean_token_accuracy": 0.06427431814372539,
|
|
"num_tokens": 63324.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.14193344116211,
|
|
"epoch": 0.0017875696872870278,
|
|
"grad_norm": 2.640625,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 8.8298,
|
|
"mean_token_accuracy": 0.07168133780360222,
|
|
"num_tokens": 67738.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.173482418060303,
|
|
"epoch": 0.001899292792742467,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 8.7488,
|
|
"mean_token_accuracy": 0.0716201275587082,
|
|
"num_tokens": 72305.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.12889585494995,
|
|
"epoch": 0.0020110158981979064,
|
|
"grad_norm": 3.046875,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 8.5249,
|
|
"mean_token_accuracy": 0.0720902569591999,
|
|
"num_tokens": 76579.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.03916826248169,
|
|
"epoch": 0.0021227390036533456,
|
|
"grad_norm": 2.203125,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.5006,
|
|
"mean_token_accuracy": 0.07380848824977874,
|
|
"num_tokens": 80812.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 9.937397384643555,
|
|
"epoch": 0.002234462109108785,
|
|
"grad_norm": 1.953125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.4135,
|
|
"mean_token_accuracy": 0.07211422435939312,
|
|
"num_tokens": 85090.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 9.917206573486329,
|
|
"epoch": 0.0023461852145642242,
|
|
"grad_norm": 2.53125,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.344,
|
|
"mean_token_accuracy": 0.07151557318866253,
|
|
"num_tokens": 89578.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 9.800644588470458,
|
|
"epoch": 0.002457908320019663,
|
|
"grad_norm": 1.9140625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.2336,
|
|
"mean_token_accuracy": 0.07592308931052685,
|
|
"num_tokens": 94117.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 9.670217800140382,
|
|
"epoch": 0.0025696314254751024,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.0337,
|
|
"mean_token_accuracy": 0.08074008114635944,
|
|
"num_tokens": 98082.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.533781147003173,
|
|
"epoch": 0.0026813545309305417,
|
|
"grad_norm": 1.875,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.0226,
|
|
"mean_token_accuracy": 0.07543974481523037,
|
|
"num_tokens": 102327.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.384562969207764,
|
|
"epoch": 0.002793077636385981,
|
|
"grad_norm": 1.9296875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 7.8842,
|
|
"mean_token_accuracy": 0.07998752668499946,
|
|
"num_tokens": 106567.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.304523754119874,
|
|
"epoch": 0.0029048007418414202,
|
|
"grad_norm": 1.9140625,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 7.7414,
|
|
"mean_token_accuracy": 0.07796499505639076,
|
|
"num_tokens": 110654.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.14549970626831,
|
|
"epoch": 0.0030165238472968595,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.7396,
|
|
"mean_token_accuracy": 0.0759395856410265,
|
|
"num_tokens": 114679.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 8.918598842620849,
|
|
"epoch": 0.003128246952752299,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.7208,
|
|
"mean_token_accuracy": 0.0734277281910181,
|
|
"num_tokens": 118817.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 8.880987071990967,
|
|
"epoch": 0.003239970058207738,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.63,
|
|
"mean_token_accuracy": 0.08075562492012978,
|
|
"num_tokens": 123188.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.665369319915772,
|
|
"epoch": 0.003351693163663177,
|
|
"grad_norm": 1.96875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.571,
|
|
"mean_token_accuracy": 0.0803100574761629,
|
|
"num_tokens": 127767.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.561956405639648,
|
|
"epoch": 0.0034634162691186163,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.3994,
|
|
"mean_token_accuracy": 0.0842840738594532,
|
|
"num_tokens": 131837.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.48619909286499,
|
|
"epoch": 0.0035751393745740555,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.4235,
|
|
"mean_token_accuracy": 0.08158716633915901,
|
|
"num_tokens": 136247.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.290351867675781,
|
|
"epoch": 0.003686862480029495,
|
|
"grad_norm": 2.203125,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.264,
|
|
"mean_token_accuracy": 0.0918136142194271,
|
|
"num_tokens": 140170.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.1635573387146,
|
|
"epoch": 0.003798585585484934,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.3583,
|
|
"mean_token_accuracy": 0.09022970125079155,
|
|
"num_tokens": 144139.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.112592220306396,
|
|
"epoch": 0.003910308690940373,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.3255,
|
|
"mean_token_accuracy": 0.08751063793897629,
|
|
"num_tokens": 148792.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.065679502487182,
|
|
"epoch": 0.004022031796395813,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.2901,
|
|
"mean_token_accuracy": 0.08218263126909733,
|
|
"num_tokens": 152844.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.083935976028442,
|
|
"epoch": 0.0041337549018512516,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.3158,
|
|
"mean_token_accuracy": 0.08304233588278294,
|
|
"num_tokens": 157366.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.946479368209839,
|
|
"epoch": 0.004245478007306691,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.1574,
|
|
"mean_token_accuracy": 0.093728469312191,
|
|
"num_tokens": 161348.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.886520957946777,
|
|
"epoch": 0.00435720111276213,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.1876,
|
|
"mean_token_accuracy": 0.10094638615846634,
|
|
"num_tokens": 165647.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.8108118057250975,
|
|
"epoch": 0.00446892421821757,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.1137,
|
|
"mean_token_accuracy": 0.09259389042854309,
|
|
"num_tokens": 169521.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.7959239959716795,
|
|
"epoch": 0.004580647323673009,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.1787,
|
|
"mean_token_accuracy": 0.09544989094138145,
|
|
"num_tokens": 173466.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.768160486221314,
|
|
"epoch": 0.0046923704291284484,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.1178,
|
|
"mean_token_accuracy": 0.08695701211690902,
|
|
"num_tokens": 177663.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.66025218963623,
|
|
"epoch": 0.004804093534583887,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.1533,
|
|
"mean_token_accuracy": 0.08611816093325615,
|
|
"num_tokens": 181778.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.759128427505493,
|
|
"epoch": 0.004915816640039326,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.1497,
|
|
"mean_token_accuracy": 0.09380927979946137,
|
|
"num_tokens": 185525.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.712249279022217,
|
|
"epoch": 0.005027539745494766,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.0063,
|
|
"mean_token_accuracy": 0.10432608500123024,
|
|
"num_tokens": 189418.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.597976541519165,
|
|
"epoch": 0.005139262850950205,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.0884,
|
|
"mean_token_accuracy": 0.10287595391273499,
|
|
"num_tokens": 193494.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.700113201141358,
|
|
"epoch": 0.0052509859564056445,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.1323,
|
|
"mean_token_accuracy": 0.09622926861047745,
|
|
"num_tokens": 198018.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.695586109161377,
|
|
"epoch": 0.005362709061861083,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.0591,
|
|
"mean_token_accuracy": 0.09503638371825218,
|
|
"num_tokens": 202296.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.5984455108642575,
|
|
"epoch": 0.005474432167316523,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.0644,
|
|
"mean_token_accuracy": 0.09454682543873787,
|
|
"num_tokens": 206694.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.6198304176330565,
|
|
"epoch": 0.005586155272771962,
|
|
"grad_norm": 2.03125,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.0863,
|
|
"mean_token_accuracy": 0.10327245965600014,
|
|
"num_tokens": 210810.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.671643877029419,
|
|
"epoch": 0.005697878378227402,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.0545,
|
|
"mean_token_accuracy": 0.09972267299890518,
|
|
"num_tokens": 215044.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.580405235290527,
|
|
"epoch": 0.0058096014836828405,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.1836,
|
|
"mean_token_accuracy": 0.09796730950474739,
|
|
"num_tokens": 219235.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.618637609481811,
|
|
"epoch": 0.005921324589138279,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.000132,
|
|
"loss": 6.9848,
|
|
"mean_token_accuracy": 0.10349723398685455,
|
|
"num_tokens": 223639.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.4419373035430905,
|
|
"epoch": 0.006033047694593719,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.0238,
|
|
"mean_token_accuracy": 0.10185197666287422,
|
|
"num_tokens": 227873.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.483202838897705,
|
|
"epoch": 0.006144770800049158,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 6.9487,
|
|
"mean_token_accuracy": 0.11078577488660812,
|
|
"num_tokens": 232147.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.483991575241089,
|
|
"epoch": 0.006256493905504598,
|
|
"grad_norm": 1.8203125,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.0367,
|
|
"mean_token_accuracy": 0.1014625109732151,
|
|
"num_tokens": 236456.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.457619524002075,
|
|
"epoch": 0.0063682170109600365,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.0498,
|
|
"mean_token_accuracy": 0.10427944287657738,
|
|
"num_tokens": 241039.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.737891387939453,
|
|
"epoch": 0.006479940116415476,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.097,
|
|
"mean_token_accuracy": 0.09794913604855537,
|
|
"num_tokens": 245132.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.422194004058838,
|
|
"epoch": 0.006591663221870915,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.0166,
|
|
"mean_token_accuracy": 0.1079220362007618,
|
|
"num_tokens": 249152.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.482765865325928,
|
|
"epoch": 0.006703386327326354,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.9395,
|
|
"mean_token_accuracy": 0.10692858770489692,
|
|
"num_tokens": 253439.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.519821882247925,
|
|
"epoch": 0.006815109432781794,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.9877,
|
|
"mean_token_accuracy": 0.10241441354155541,
|
|
"num_tokens": 258066.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.4823202133178714,
|
|
"epoch": 0.0069268325382372325,
|
|
"grad_norm": 1.890625,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.9284,
|
|
"mean_token_accuracy": 0.1084162712097168,
|
|
"num_tokens": 261954.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.320609712600708,
|
|
"epoch": 0.007038555643692672,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.9852,
|
|
"mean_token_accuracy": 0.10961857140064239,
|
|
"num_tokens": 266650.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.539988470077515,
|
|
"epoch": 0.007150278749148111,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.0467,
|
|
"mean_token_accuracy": 0.10235239341855049,
|
|
"num_tokens": 271069.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.4312175750732425,
|
|
"epoch": 0.007262001854603551,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.9195,
|
|
"mean_token_accuracy": 0.10588012635707855,
|
|
"num_tokens": 275084.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.405316781997681,
|
|
"epoch": 0.00737372496005899,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.9708,
|
|
"mean_token_accuracy": 0.11100029796361924,
|
|
"num_tokens": 279721.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.45182785987854,
|
|
"epoch": 0.007485448065514429,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.9745,
|
|
"mean_token_accuracy": 0.11023727878928184,
|
|
"num_tokens": 284317.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.376049518585205,
|
|
"epoch": 0.007597171170969868,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.9627,
|
|
"mean_token_accuracy": 0.11432994231581688,
|
|
"num_tokens": 288870.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.399447011947632,
|
|
"epoch": 0.007708894276425307,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.8883,
|
|
"mean_token_accuracy": 0.11889545023441314,
|
|
"num_tokens": 292996.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.417701625823975,
|
|
"epoch": 0.007820617381880746,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.8748,
|
|
"mean_token_accuracy": 0.11174583658576012,
|
|
"num_tokens": 297238.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.387735509872437,
|
|
"epoch": 0.007932340487336187,
|
|
"grad_norm": 1.890625,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.8957,
|
|
"mean_token_accuracy": 0.1190306767821312,
|
|
"num_tokens": 301453.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.298361539840698,
|
|
"epoch": 0.008044063592791625,
|
|
"grad_norm": 1.8984375,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.8362,
|
|
"mean_token_accuracy": 0.11696951016783715,
|
|
"num_tokens": 305949.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.339731693267822,
|
|
"epoch": 0.008155786698247064,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.8502,
|
|
"mean_token_accuracy": 0.12272856459021568,
|
|
"num_tokens": 310097.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.378586864471435,
|
|
"epoch": 0.008267509803702503,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.8746,
|
|
"mean_token_accuracy": 0.1073436789214611,
|
|
"num_tokens": 314567.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.258294773101807,
|
|
"epoch": 0.008379232909157944,
|
|
"grad_norm": 1.828125,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.9539,
|
|
"mean_token_accuracy": 0.1148001030087471,
|
|
"num_tokens": 319166.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.30250129699707,
|
|
"epoch": 0.008490956014613383,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.8336,
|
|
"mean_token_accuracy": 0.11580941900610923,
|
|
"num_tokens": 323682.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.259005832672119,
|
|
"epoch": 0.008602679120068821,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.7239,
|
|
"mean_token_accuracy": 0.12464385628700256,
|
|
"num_tokens": 327994.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.200961637496948,
|
|
"epoch": 0.00871440222552426,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.8242,
|
|
"mean_token_accuracy": 0.11740145906805992,
|
|
"num_tokens": 332026.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.481966304779053,
|
|
"epoch": 0.0088261253309797,
|
|
"grad_norm": 3.078125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.8603,
|
|
"mean_token_accuracy": 0.12176617532968521,
|
|
"num_tokens": 336552.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.178363084793091,
|
|
"epoch": 0.00893784843643514,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.8005,
|
|
"mean_token_accuracy": 0.11857323870062828,
|
|
"num_tokens": 340643.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.28194899559021,
|
|
"epoch": 0.009049571541890579,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.7174,
|
|
"mean_token_accuracy": 0.12609147280454636,
|
|
"num_tokens": 344886.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.182888126373291,
|
|
"epoch": 0.009161294647346017,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.7831,
|
|
"mean_token_accuracy": 0.12051703482866287,
|
|
"num_tokens": 349115.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.436312055587768,
|
|
"epoch": 0.009273017752801456,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.7448,
|
|
"mean_token_accuracy": 0.12795047089457512,
|
|
"num_tokens": 353368.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.033882093429566,
|
|
"epoch": 0.009384740858256897,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.7951,
|
|
"mean_token_accuracy": 0.11617444008588791,
|
|
"num_tokens": 357382.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.290721368789673,
|
|
"epoch": 0.009496463963712336,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.7152,
|
|
"mean_token_accuracy": 0.12455029636621476,
|
|
"num_tokens": 361542.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.23753170967102,
|
|
"epoch": 0.009608187069167775,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.7497,
|
|
"mean_token_accuracy": 0.12337451875209808,
|
|
"num_tokens": 366006.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.083537578582764,
|
|
"epoch": 0.009719910174623213,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.6902,
|
|
"mean_token_accuracy": 0.1326639696955681,
|
|
"num_tokens": 370021.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.248496341705322,
|
|
"epoch": 0.009831633280078652,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.805,
|
|
"mean_token_accuracy": 0.12277387753129006,
|
|
"num_tokens": 374434.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.251986265182495,
|
|
"epoch": 0.009943356385534093,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.8176,
|
|
"mean_token_accuracy": 0.11638971790671349,
|
|
"num_tokens": 378634.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.187275457382202,
|
|
"epoch": 0.010055079490989532,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.6952,
|
|
"mean_token_accuracy": 0.12703127264976502,
|
|
"num_tokens": 382904.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.244022417068481,
|
|
"epoch": 0.01016680259644497,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.6803,
|
|
"mean_token_accuracy": 0.12515532448887826,
|
|
"num_tokens": 386970.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.082424831390381,
|
|
"epoch": 0.01027852570190041,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.6572,
|
|
"mean_token_accuracy": 0.13228979632258414,
|
|
"num_tokens": 391043.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.1885302543640135,
|
|
"epoch": 0.01039024880735585,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.6428,
|
|
"mean_token_accuracy": 0.13703735694289207,
|
|
"num_tokens": 395413.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.166730117797852,
|
|
"epoch": 0.010501971912811289,
|
|
"grad_norm": 1.9296875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.7767,
|
|
"mean_token_accuracy": 0.12424237206578255,
|
|
"num_tokens": 399821.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.0755476474761965,
|
|
"epoch": 0.010613695018266728,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.5815,
|
|
"mean_token_accuracy": 0.13624327406287193,
|
|
"num_tokens": 404043.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.202919864654541,
|
|
"epoch": 0.010725418123722167,
|
|
"grad_norm": 1.9921875,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.7277,
|
|
"mean_token_accuracy": 0.1173846572637558,
|
|
"num_tokens": 408339.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.004957008361816,
|
|
"epoch": 0.010837141229177606,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.6233,
|
|
"mean_token_accuracy": 0.1330234356224537,
|
|
"num_tokens": 412384.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.170296764373779,
|
|
"epoch": 0.010948864334633046,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.7419,
|
|
"mean_token_accuracy": 0.122147136926651,
|
|
"num_tokens": 416891.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.086709833145141,
|
|
"epoch": 0.011060587440088485,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.6048,
|
|
"mean_token_accuracy": 0.13360075876116753,
|
|
"num_tokens": 420980.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 6.981275844573974,
|
|
"epoch": 0.011172310545543924,
|
|
"grad_norm": 1.921875,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.6585,
|
|
"mean_token_accuracy": 0.12793864756822587,
|
|
"num_tokens": 425454.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.063224363327026,
|
|
"epoch": 0.011284033650999363,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.6792,
|
|
"mean_token_accuracy": 0.12707910239696502,
|
|
"num_tokens": 430087.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.057527685165406,
|
|
"epoch": 0.011395756756454803,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.6015,
|
|
"mean_token_accuracy": 0.13783879056572915,
|
|
"num_tokens": 434062.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 6.918734359741211,
|
|
"epoch": 0.011507479861910242,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.5727,
|
|
"mean_token_accuracy": 0.1411380872130394,
|
|
"num_tokens": 437882.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.062899255752564,
|
|
"epoch": 0.011619202967365681,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.6023,
|
|
"mean_token_accuracy": 0.13055641129612922,
|
|
"num_tokens": 442248.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.028418397903442,
|
|
"epoch": 0.01173092607282112,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.5713,
|
|
"mean_token_accuracy": 0.13718463256955146,
|
|
"num_tokens": 446492.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.917926502227783,
|
|
"epoch": 0.011842649178276559,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.6019,
|
|
"mean_token_accuracy": 0.13765231221914292,
|
|
"num_tokens": 450359.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.029662656784057,
|
|
"epoch": 0.011954372283732,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.6264,
|
|
"mean_token_accuracy": 0.12923550009727477,
|
|
"num_tokens": 454508.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.076862907409668,
|
|
"epoch": 0.012066095389187438,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.6009,
|
|
"mean_token_accuracy": 0.13477242439985276,
|
|
"num_tokens": 459028.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.926972484588623,
|
|
"epoch": 0.012177818494642877,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.5556,
|
|
"mean_token_accuracy": 0.1286642447113991,
|
|
"num_tokens": 462911.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.978501319885254,
|
|
"epoch": 0.012289541600098316,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.5891,
|
|
"mean_token_accuracy": 0.13105733469128608,
|
|
"num_tokens": 467129.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.02958312034607,
|
|
"epoch": 0.012401264705553756,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.4952,
|
|
"mean_token_accuracy": 0.13785838335752487,
|
|
"num_tokens": 471324.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.861265659332275,
|
|
"epoch": 0.012512987811009195,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.6661,
|
|
"mean_token_accuracy": 0.1318783350288868,
|
|
"num_tokens": 475761.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.988262891769409,
|
|
"epoch": 0.012624710916464634,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.4608,
|
|
"mean_token_accuracy": 0.1332152046263218,
|
|
"num_tokens": 479532.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.955751943588257,
|
|
"epoch": 0.012736434021920073,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.6147,
|
|
"mean_token_accuracy": 0.1329347386956215,
|
|
"num_tokens": 483844.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.067316389083862,
|
|
"epoch": 0.012848157127375512,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.6526,
|
|
"mean_token_accuracy": 0.12774468660354615,
|
|
"num_tokens": 488402.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.788709592819214,
|
|
"epoch": 0.012959880232830952,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.3961,
|
|
"mean_token_accuracy": 0.13899129256606102,
|
|
"num_tokens": 492473.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.816238927841186,
|
|
"epoch": 0.013071603338286391,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.5152,
|
|
"mean_token_accuracy": 0.1357610009610653,
|
|
"num_tokens": 496830.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.027258968353271,
|
|
"epoch": 0.01318332644374183,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.5867,
|
|
"mean_token_accuracy": 0.13579039499163628,
|
|
"num_tokens": 500996.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.797857189178467,
|
|
"epoch": 0.013295049549197269,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.5446,
|
|
"mean_token_accuracy": 0.13526690602302552,
|
|
"num_tokens": 505174.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.864916706085205,
|
|
"epoch": 0.013406772654652708,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.548,
|
|
"mean_token_accuracy": 0.1401192255318165,
|
|
"num_tokens": 509425.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.837484741210938,
|
|
"epoch": 0.013518495760108148,
|
|
"grad_norm": 1.859375,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.4098,
|
|
"mean_token_accuracy": 0.14489975348114967,
|
|
"num_tokens": 513271.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.8677136421203615,
|
|
"epoch": 0.013630218865563587,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.6053,
|
|
"mean_token_accuracy": 0.14355736672878266,
|
|
"num_tokens": 517879.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.855578374862671,
|
|
"epoch": 0.013741941971019026,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.5708,
|
|
"mean_token_accuracy": 0.13215027675032615,
|
|
"num_tokens": 522321.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.7266088962554935,
|
|
"epoch": 0.013853665076474465,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.3619,
|
|
"mean_token_accuracy": 0.1518048860132694,
|
|
"num_tokens": 525884.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.896519756317138,
|
|
"epoch": 0.013965388181929906,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.4982,
|
|
"mean_token_accuracy": 0.1376729391515255,
|
|
"num_tokens": 530373.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.760099697113037,
|
|
"epoch": 0.014077111287385344,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.3254,
|
|
"mean_token_accuracy": 0.1519668973982334,
|
|
"num_tokens": 534571.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.845727109909058,
|
|
"epoch": 0.014188834392840783,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.496,
|
|
"mean_token_accuracy": 0.1373538002371788,
|
|
"num_tokens": 538938.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.782427406311035,
|
|
"epoch": 0.014300557498296222,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.4534,
|
|
"mean_token_accuracy": 0.15009932219982147,
|
|
"num_tokens": 542977.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.756459665298462,
|
|
"epoch": 0.014412280603751661,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.4253,
|
|
"mean_token_accuracy": 0.14535169675946236,
|
|
"num_tokens": 547005.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.681589221954345,
|
|
"epoch": 0.014524003709207102,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.468,
|
|
"mean_token_accuracy": 0.1390581026673317,
|
|
"num_tokens": 551361.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.782884311676026,
|
|
"epoch": 0.01463572681466254,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.4134,
|
|
"mean_token_accuracy": 0.13999306932091712,
|
|
"num_tokens": 555574.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.632973098754883,
|
|
"epoch": 0.01474744992011798,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.462,
|
|
"mean_token_accuracy": 0.14115875586867332,
|
|
"num_tokens": 560368.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.901856851577759,
|
|
"epoch": 0.014859173025573418,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.4939,
|
|
"mean_token_accuracy": 0.13747724890708923,
|
|
"num_tokens": 564884.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.668564367294311,
|
|
"epoch": 0.014970896131028859,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.4878,
|
|
"mean_token_accuracy": 0.1415390208363533,
|
|
"num_tokens": 569494.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.764341497421265,
|
|
"epoch": 0.015082619236484298,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.506,
|
|
"mean_token_accuracy": 0.1322600431740284,
|
|
"num_tokens": 574161.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.753030443191529,
|
|
"epoch": 0.015194342341939737,
|
|
"grad_norm": 1.828125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.4248,
|
|
"mean_token_accuracy": 0.14356208443641663,
|
|
"num_tokens": 578510.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.674463367462158,
|
|
"epoch": 0.015306065447395175,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.4388,
|
|
"mean_token_accuracy": 0.1411791481077671,
|
|
"num_tokens": 582695.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.776868438720703,
|
|
"epoch": 0.015417788552850614,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.4011,
|
|
"mean_token_accuracy": 0.1373072750866413,
|
|
"num_tokens": 586622.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.6929234027862545,
|
|
"epoch": 0.015529511658306055,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.4018,
|
|
"mean_token_accuracy": 0.14830828532576562,
|
|
"num_tokens": 591704.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.481786489486694,
|
|
"epoch": 0.015641234763761492,
|
|
"grad_norm": 1.859375,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.2342,
|
|
"mean_token_accuracy": 0.15492519065737725,
|
|
"num_tokens": 596029.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.6203367710113525,
|
|
"epoch": 0.015752957869216933,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.391,
|
|
"mean_token_accuracy": 0.14638458639383317,
|
|
"num_tokens": 600115.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.719356822967529,
|
|
"epoch": 0.015864680974672373,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.3892,
|
|
"mean_token_accuracy": 0.14423182979226112,
|
|
"num_tokens": 604488.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.73731255531311,
|
|
"epoch": 0.01597640408012781,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.4097,
|
|
"mean_token_accuracy": 0.1440168045461178,
|
|
"num_tokens": 608776.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.5224494457244875,
|
|
"epoch": 0.01608812718558325,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.4047,
|
|
"mean_token_accuracy": 0.14601838812232018,
|
|
"num_tokens": 612771.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.620055294036865,
|
|
"epoch": 0.01619985029103869,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.3194,
|
|
"mean_token_accuracy": 0.15284316167235373,
|
|
"num_tokens": 616947.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.584221696853637,
|
|
"epoch": 0.01631157339649413,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.4028,
|
|
"mean_token_accuracy": 0.14883917048573495,
|
|
"num_tokens": 621064.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.636436128616333,
|
|
"epoch": 0.01642329650194957,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.3876,
|
|
"mean_token_accuracy": 0.1397322230041027,
|
|
"num_tokens": 625349.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.571576929092407,
|
|
"epoch": 0.016535019607405006,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.3184,
|
|
"mean_token_accuracy": 0.15138321444392205,
|
|
"num_tokens": 629754.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.620907211303711,
|
|
"epoch": 0.016646742712860447,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.3518,
|
|
"mean_token_accuracy": 0.14085613414645196,
|
|
"num_tokens": 634111.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.450802040100098,
|
|
"epoch": 0.016758465818315887,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.2157,
|
|
"mean_token_accuracy": 0.16091271564364434,
|
|
"num_tokens": 638527.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.442024230957031,
|
|
"epoch": 0.016870188923771325,
|
|
"grad_norm": 1.796875,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.2322,
|
|
"mean_token_accuracy": 0.1605646252632141,
|
|
"num_tokens": 642530.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.5630576610565186,
|
|
"epoch": 0.016981912029226765,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.3145,
|
|
"mean_token_accuracy": 0.15974024683237076,
|
|
"num_tokens": 646707.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.436832427978516,
|
|
"epoch": 0.017093635134682202,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.3549,
|
|
"mean_token_accuracy": 0.15013475716114044,
|
|
"num_tokens": 651326.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.54889612197876,
|
|
"epoch": 0.017205358240137643,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.4225,
|
|
"mean_token_accuracy": 0.1450472079217434,
|
|
"num_tokens": 655651.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.598716402053833,
|
|
"epoch": 0.017317081345593083,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.3493,
|
|
"mean_token_accuracy": 0.15063260793685912,
|
|
"num_tokens": 659868.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.613773393630981,
|
|
"epoch": 0.01742880445104852,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.415,
|
|
"mean_token_accuracy": 0.14450336694717408,
|
|
"num_tokens": 664404.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.383858156204224,
|
|
"epoch": 0.01754052755650396,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.2007,
|
|
"mean_token_accuracy": 0.15557888001203538,
|
|
"num_tokens": 668543.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.472045183181763,
|
|
"epoch": 0.0176522506619594,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.203,
|
|
"mean_token_accuracy": 0.1600804843008518,
|
|
"num_tokens": 672637.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.512582492828369,
|
|
"epoch": 0.01776397376741484,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.2935,
|
|
"mean_token_accuracy": 0.15304349958896638,
|
|
"num_tokens": 676929.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.386533451080322,
|
|
"epoch": 0.01787569687287028,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.3513,
|
|
"mean_token_accuracy": 0.15382541790604592,
|
|
"num_tokens": 681287.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.473147869110107,
|
|
"epoch": 0.017987419978325717,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.2439,
|
|
"mean_token_accuracy": 0.15017452612519264,
|
|
"num_tokens": 685543.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.567328834533692,
|
|
"epoch": 0.018099143083781157,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.2352,
|
|
"mean_token_accuracy": 0.14613626077771186,
|
|
"num_tokens": 689479.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.370502090454101,
|
|
"epoch": 0.018210866189236594,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.1598,
|
|
"mean_token_accuracy": 0.16322519332170488,
|
|
"num_tokens": 693564.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.502647733688354,
|
|
"epoch": 0.018322589294692035,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.2557,
|
|
"mean_token_accuracy": 0.1501608557999134,
|
|
"num_tokens": 697719.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.316255664825439,
|
|
"epoch": 0.018434312400147475,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.4171,
|
|
"mean_token_accuracy": 0.15079098790884018,
|
|
"num_tokens": 701937.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.541279220581055,
|
|
"epoch": 0.018546035505602913,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.3393,
|
|
"mean_token_accuracy": 0.14786507934331894,
|
|
"num_tokens": 706457.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.479259777069092,
|
|
"epoch": 0.018657758611058353,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.3349,
|
|
"mean_token_accuracy": 0.15288059860467912,
|
|
"num_tokens": 710538.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.2955930709838865,
|
|
"epoch": 0.018769481716513794,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.222,
|
|
"mean_token_accuracy": 0.15921739637851715,
|
|
"num_tokens": 714663.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.474664878845215,
|
|
"epoch": 0.01888120482196923,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.2263,
|
|
"mean_token_accuracy": 0.15058805868029596,
|
|
"num_tokens": 718828.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.231873464584351,
|
|
"epoch": 0.01899292792742467,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.2178,
|
|
"mean_token_accuracy": 0.15931910127401352,
|
|
"num_tokens": 723155.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.52202787399292,
|
|
"epoch": 0.01910465103288011,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.2688,
|
|
"mean_token_accuracy": 0.157795250415802,
|
|
"num_tokens": 727605.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.3318055152893065,
|
|
"epoch": 0.01921637413833555,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.2337,
|
|
"mean_token_accuracy": 0.15254703313112258,
|
|
"num_tokens": 731510.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.418648195266724,
|
|
"epoch": 0.01932809724379099,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.467,
|
|
"mean_token_accuracy": 0.13948606923222542,
|
|
"num_tokens": 736048.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.461650276184082,
|
|
"epoch": 0.019439820349246427,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.2286,
|
|
"mean_token_accuracy": 0.1571047306060791,
|
|
"num_tokens": 740714.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.340702772140503,
|
|
"epoch": 0.019551543454701868,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.3055,
|
|
"mean_token_accuracy": 0.1546173483133316,
|
|
"num_tokens": 745167.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.3855945587158205,
|
|
"epoch": 0.019663266560157305,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.1172,
|
|
"mean_token_accuracy": 0.16043180972337723,
|
|
"num_tokens": 749401.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.357611846923828,
|
|
"epoch": 0.019774989665612745,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.1698,
|
|
"mean_token_accuracy": 0.16577810943126678,
|
|
"num_tokens": 753554.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.337348079681396,
|
|
"epoch": 0.019886712771068186,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.2451,
|
|
"mean_token_accuracy": 0.15055422931909562,
|
|
"num_tokens": 757744.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.3222455978393555,
|
|
"epoch": 0.019998435876523623,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.1133,
|
|
"mean_token_accuracy": 0.15861648172140122,
|
|
"num_tokens": 761858.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.288092851638794,
|
|
"epoch": 0.020110158981979064,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.1429,
|
|
"mean_token_accuracy": 0.16834202855825425,
|
|
"num_tokens": 765852.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.2346821308135985,
|
|
"epoch": 0.0202218820874345,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.1649,
|
|
"mean_token_accuracy": 0.1537023276090622,
|
|
"num_tokens": 770274.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.315171623229981,
|
|
"epoch": 0.02033360519288994,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.2175,
|
|
"mean_token_accuracy": 0.1551866799592972,
|
|
"num_tokens": 774775.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.4123328685760494,
|
|
"epoch": 0.020445328298345382,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.2295,
|
|
"mean_token_accuracy": 0.16605380773544312,
|
|
"num_tokens": 779374.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.239690780639648,
|
|
"epoch": 0.02055705140380082,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.2391,
|
|
"mean_token_accuracy": 0.1497175522148609,
|
|
"num_tokens": 783537.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.284845304489136,
|
|
"epoch": 0.02066877450925626,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.3045,
|
|
"mean_token_accuracy": 0.15461198166012763,
|
|
"num_tokens": 787562.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.4740015983581545,
|
|
"epoch": 0.0207804976147117,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.3361,
|
|
"mean_token_accuracy": 0.14666738733649254,
|
|
"num_tokens": 792303.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.263271856307983,
|
|
"epoch": 0.020892220720167137,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.1427,
|
|
"mean_token_accuracy": 0.16184145957231522,
|
|
"num_tokens": 796402.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.380016136169433,
|
|
"epoch": 0.021003943825622578,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.2087,
|
|
"mean_token_accuracy": 0.1643604129552841,
|
|
"num_tokens": 800004.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.157491159439087,
|
|
"epoch": 0.021115666931078015,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.0888,
|
|
"mean_token_accuracy": 0.1660729944705963,
|
|
"num_tokens": 803782.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.282210779190064,
|
|
"epoch": 0.021227390036533456,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.1556,
|
|
"mean_token_accuracy": 0.16121889799833297,
|
|
"num_tokens": 808013.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.159784889221191,
|
|
"epoch": 0.021339113141988896,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.1523,
|
|
"mean_token_accuracy": 0.1605899378657341,
|
|
"num_tokens": 812130.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.311656665802002,
|
|
"epoch": 0.021450836247444333,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.1163,
|
|
"mean_token_accuracy": 0.16060882434248924,
|
|
"num_tokens": 816246.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.1442811489105225,
|
|
"epoch": 0.021562559352899774,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.1295,
|
|
"mean_token_accuracy": 0.15766306221485138,
|
|
"num_tokens": 820628.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.282287883758545,
|
|
"epoch": 0.02167428245835521,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.1044,
|
|
"mean_token_accuracy": 0.16587121784687042,
|
|
"num_tokens": 824864.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.298398399353028,
|
|
"epoch": 0.02178600556381065,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.1733,
|
|
"mean_token_accuracy": 0.1572745218873024,
|
|
"num_tokens": 828887.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.119819927215576,
|
|
"epoch": 0.021897728669266092,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.0463,
|
|
"mean_token_accuracy": 0.16708520650863648,
|
|
"num_tokens": 832784.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.1874528408050535,
|
|
"epoch": 0.02200945177472153,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.0893,
|
|
"mean_token_accuracy": 0.15289360731840135,
|
|
"num_tokens": 837408.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.194441604614258,
|
|
"epoch": 0.02212117488017697,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.1339,
|
|
"mean_token_accuracy": 0.16136935949325562,
|
|
"num_tokens": 841536.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.170191717147827,
|
|
"epoch": 0.022232897985632407,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.0497,
|
|
"mean_token_accuracy": 0.16993979513645172,
|
|
"num_tokens": 846044.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.106483221054077,
|
|
"epoch": 0.022344621091087848,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.0682,
|
|
"mean_token_accuracy": 0.16326213181018828,
|
|
"num_tokens": 850254.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.112219524383545,
|
|
"epoch": 0.022456344196543288,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.000499998026082006,
|
|
"loss": 6.0539,
|
|
"mean_token_accuracy": 0.16882037818431855,
|
|
"num_tokens": 854548.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.08699860572815,
|
|
"epoch": 0.022568067301998725,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004999900070995136,
|
|
"loss": 6.2141,
|
|
"mean_token_accuracy": 0.1636434242129326,
|
|
"num_tokens": 858837.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.176929616928101,
|
|
"epoch": 0.022679790407454166,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0004999758199023239,
|
|
"loss": 6.0198,
|
|
"mean_token_accuracy": 0.15852618217468262,
|
|
"num_tokens": 862772.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.200733661651611,
|
|
"epoch": 0.022791513512909607,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004999554648793858,
|
|
"loss": 6.098,
|
|
"mean_token_accuracy": 0.1593088060617447,
|
|
"num_tokens": 866909.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.177529811859131,
|
|
"epoch": 0.022903236618365044,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 6.2321,
|
|
"mean_token_accuracy": 0.1547963500022888,
|
|
"num_tokens": 871127.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.2023234367370605,
|
|
"epoch": 0.023014959723820484,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.0004998962537575161,
|
|
"loss": 6.1609,
|
|
"mean_token_accuracy": 0.16332077234983444,
|
|
"num_tokens": 875292.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.16438570022583,
|
|
"epoch": 0.02312668282927592,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0004998573992818874,
|
|
"loss": 6.0549,
|
|
"mean_token_accuracy": 0.16016169339418412,
|
|
"num_tokens": 879286.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.2433970928192135,
|
|
"epoch": 0.023238405934731362,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004998123802270715,
|
|
"loss": 6.0876,
|
|
"mean_token_accuracy": 0.17015287280082703,
|
|
"num_tokens": 883490.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.14397964477539,
|
|
"epoch": 0.023350129040186803,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004997611978272886,
|
|
"loss": 6.0609,
|
|
"mean_token_accuracy": 0.16376899629831315,
|
|
"num_tokens": 887786.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.1847413063049315,
|
|
"epoch": 0.02346185214564224,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004997038534857298,
|
|
"loss": 6.0817,
|
|
"mean_token_accuracy": 0.17141184508800505,
|
|
"num_tokens": 892035.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.1673887252807615,
|
|
"epoch": 0.02357357525109768,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0004996403487745194,
|
|
"loss": 6.0719,
|
|
"mean_token_accuracy": 0.16404020488262178,
|
|
"num_tokens": 896099.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.215155744552613,
|
|
"epoch": 0.023685298356553117,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.000499570685434671,
|
|
"loss": 5.9941,
|
|
"mean_token_accuracy": 0.17437216117978097,
|
|
"num_tokens": 900448.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.082410001754761,
|
|
"epoch": 0.023797021462008558,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0004994948653760405,
|
|
"loss": 6.0422,
|
|
"mean_token_accuracy": 0.16581026017665862,
|
|
"num_tokens": 904538.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.211330890655518,
|
|
"epoch": 0.023908744567464,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.0004994128906772729,
|
|
"loss": 6.2105,
|
|
"mean_token_accuracy": 0.15310067683458328,
|
|
"num_tokens": 908680.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.132528066635132,
|
|
"epoch": 0.024020467672919436,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.000499324763585746,
|
|
"loss": 6.1633,
|
|
"mean_token_accuracy": 0.154509137570858,
|
|
"num_tokens": 912868.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.17259407043457,
|
|
"epoch": 0.024132190778374876,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.0004992304865175085,
|
|
"loss": 6.192,
|
|
"mean_token_accuracy": 0.16236126869916917,
|
|
"num_tokens": 917312.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.335460948944092,
|
|
"epoch": 0.024243913883830313,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 6.1838,
|
|
"mean_token_accuracy": 0.16260358244180678,
|
|
"num_tokens": 921977.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.128043842315674,
|
|
"epoch": 0.024355636989285754,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004990234929580494,
|
|
"loss": 6.1085,
|
|
"mean_token_accuracy": 0.16749032437801362,
|
|
"num_tokens": 925964.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.223902463912964,
|
|
"epoch": 0.024467360094741195,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0004989107821416609,
|
|
"loss": 5.9766,
|
|
"mean_token_accuracy": 0.1657112054526806,
|
|
"num_tokens": 930077.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.065625762939453,
|
|
"epoch": 0.02457908320019663,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004987919326980723,
|
|
"loss": 6.0066,
|
|
"mean_token_accuracy": 0.1741485133767128,
|
|
"num_tokens": 934547.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.170744371414185,
|
|
"epoch": 0.024690806305652072,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0004986669478856011,
|
|
"loss": 6.0122,
|
|
"mean_token_accuracy": 0.17556725889444352,
|
|
"num_tokens": 938583.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.233295917510986,
|
|
"epoch": 0.024802529411107513,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004985358311307688,
|
|
"loss": 6.0892,
|
|
"mean_token_accuracy": 0.16876054555177689,
|
|
"num_tokens": 943007.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.1148707389831545,
|
|
"epoch": 0.02491425251656295,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004983985860282081,
|
|
"loss": 6.089,
|
|
"mean_token_accuracy": 0.16639824211597443,
|
|
"num_tokens": 947281.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.048860692977906,
|
|
"epoch": 0.02502597562201839,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004982552163405623,
|
|
"loss": 5.952,
|
|
"mean_token_accuracy": 0.17678665965795518,
|
|
"num_tokens": 951443.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.026111602783203,
|
|
"epoch": 0.025137698727473828,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0004981057259983839,
|
|
"loss": 6.0601,
|
|
"mean_token_accuracy": 0.16274188607931137,
|
|
"num_tokens": 955131.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.128307962417603,
|
|
"epoch": 0.02524942183292927,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004979501191000262,
|
|
"loss": 6.0103,
|
|
"mean_token_accuracy": 0.16576859056949617,
|
|
"num_tokens": 959408.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 5.998388624191284,
|
|
"epoch": 0.02536114493838471,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004977883999115311,
|
|
"loss": 6.0554,
|
|
"mean_token_accuracy": 0.16911191642284393,
|
|
"num_tokens": 963657.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.107620096206665,
|
|
"epoch": 0.025472868043840146,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004976205728665113,
|
|
"loss": 6.0781,
|
|
"mean_token_accuracy": 0.1575411856174469,
|
|
"num_tokens": 967973.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.082073783874511,
|
|
"epoch": 0.025584591149295587,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 5.8803,
|
|
"mean_token_accuracy": 0.18416397273540497,
|
|
"num_tokens": 971957.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 5.985405111312867,
|
|
"epoch": 0.025696314254751024,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004972666137784759,
|
|
"loss": 5.8823,
|
|
"mean_token_accuracy": 0.1817352905869484,
|
|
"num_tokens": 975679.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.04285044670105,
|
|
"epoch": 0.025808037360206464,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0004970804914394271,
|
|
"loss": 6.1732,
|
|
"mean_token_accuracy": 0.15990505665540694,
|
|
"num_tokens": 979806.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.167063045501709,
|
|
"epoch": 0.025919760465661905,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004968882806515225,
|
|
"loss": 6.0201,
|
|
"mean_token_accuracy": 0.1767844155430794,
|
|
"num_tokens": 984221.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.029861211776733,
|
|
"epoch": 0.026031483571117342,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0004966899866843177,
|
|
"loss": 5.9773,
|
|
"mean_token_accuracy": 0.16563293039798738,
|
|
"num_tokens": 988323.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.0827898502349855,
|
|
"epoch": 0.026143206676572783,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000496485614974142,
|
|
"loss": 6.038,
|
|
"mean_token_accuracy": 0.1660424754023552,
|
|
"num_tokens": 992643.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 5.991740322113037,
|
|
"epoch": 0.02625492978202822,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0004962751711239492,
|
|
"loss": 5.8579,
|
|
"mean_token_accuracy": 0.1826507866382599,
|
|
"num_tokens": 996994.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 5.962495994567871,
|
|
"epoch": 0.02636665288748366,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0004960586609031636,
|
|
"loss": 6.0109,
|
|
"mean_token_accuracy": 0.17780142575502395,
|
|
"num_tokens": 1001345.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.076908683776855,
|
|
"epoch": 0.0264783759929391,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004958360902475224,
|
|
"loss": 5.9154,
|
|
"mean_token_accuracy": 0.17819885313510894,
|
|
"num_tokens": 1005839.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.0804304599761965,
|
|
"epoch": 0.026590099098394538,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004956074652589125,
|
|
"loss": 6.0655,
|
|
"mean_token_accuracy": 0.1642119474709034,
|
|
"num_tokens": 1010698.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.093788766860962,
|
|
"epoch": 0.02670182220384998,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004953727922052035,
|
|
"loss": 6.0287,
|
|
"mean_token_accuracy": 0.1619931563735008,
|
|
"num_tokens": 1014982.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 5.86436014175415,
|
|
"epoch": 0.026813545309305416,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.0004951320775200756,
|
|
"loss": 5.9534,
|
|
"mean_token_accuracy": 0.18322545439004898,
|
|
"num_tokens": 1019235.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.108639049530029,
|
|
"epoch": 0.026925268414760856,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 6.0397,
|
|
"mean_token_accuracy": 0.1752349078655243,
|
|
"num_tokens": 1023855.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.110106515884399,
|
|
"epoch": 0.027036991520216297,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004946325498182755,
|
|
"loss": 6.0001,
|
|
"mean_token_accuracy": 0.17471726685762407,
|
|
"num_tokens": 1028464.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 5.944786262512207,
|
|
"epoch": 0.027148714625671734,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004943737504964076,
|
|
"loss": 5.9103,
|
|
"mean_token_accuracy": 0.18010717928409575,
|
|
"num_tokens": 1033110.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 5.954358720779419,
|
|
"epoch": 0.027260437731127175,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000494108936932354,
|
|
"loss": 6.0102,
|
|
"mean_token_accuracy": 0.16354090273380278,
|
|
"num_tokens": 1037470.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.164341068267822,
|
|
"epoch": 0.027372160836582615,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0004938381163861124,
|
|
"loss": 6.0126,
|
|
"mean_token_accuracy": 0.16803643256425857,
|
|
"num_tokens": 1041662.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.096904563903808,
|
|
"epoch": 0.027483883942038052,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004935612962823645,
|
|
"loss": 5.92,
|
|
"mean_token_accuracy": 0.17529453635215758,
|
|
"num_tokens": 1046041.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 5.942513656616211,
|
|
"epoch": 0.027595607047493493,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004932784842102739,
|
|
"loss": 5.9454,
|
|
"mean_token_accuracy": 0.1719459719955921,
|
|
"num_tokens": 1049893.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.186945247650146,
|
|
"epoch": 0.02770733015294893,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0004929896879232758,
|
|
"loss": 6.1042,
|
|
"mean_token_accuracy": 0.15955148935317992,
|
|
"num_tokens": 1054048.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.0783411979675295,
|
|
"epoch": 0.02781905325840437,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004926949153388668,
|
|
"loss": 5.9941,
|
|
"mean_token_accuracy": 0.16463877633213997,
|
|
"num_tokens": 1058176.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.107759857177735,
|
|
"epoch": 0.02793077636385981,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004923941745383859,
|
|
"loss": 5.9543,
|
|
"mean_token_accuracy": 0.16195504367351532,
|
|
"num_tokens": 1062031.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 5.933470392227173,
|
|
"epoch": 0.02804249946931525,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000492087473766794,
|
|
"loss": 5.9244,
|
|
"mean_token_accuracy": 0.17964119464159012,
|
|
"num_tokens": 1066705.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.079675674438477,
|
|
"epoch": 0.02815422257477069,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000491774821432448,
|
|
"loss": 5.9448,
|
|
"mean_token_accuracy": 0.18084608763456345,
|
|
"num_tokens": 1070960.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.124145603179931,
|
|
"epoch": 0.028265945680226126,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 6.0945,
|
|
"mean_token_accuracy": 0.15580475330352783,
|
|
"num_tokens": 1075576.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.012061548233032,
|
|
"epoch": 0.028377668785681567,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004911316965245098,
|
|
"loss": 5.8603,
|
|
"mean_token_accuracy": 0.16867389604449273,
|
|
"num_tokens": 1079538.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 5.922763729095459,
|
|
"epoch": 0.028489391891137007,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.000490801241582512,
|
|
"loss": 5.9411,
|
|
"mean_token_accuracy": 0.1767105147242546,
|
|
"num_tokens": 1083296.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 5.973857307434082,
|
|
"epoch": 0.028601114996592444,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000490464870340465,
|
|
"loss": 5.9737,
|
|
"mean_token_accuracy": 0.17723702639341354,
|
|
"num_tokens": 1087688.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.082131719589233,
|
|
"epoch": 0.028712838102047885,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004901225920201563,
|
|
"loss": 5.9207,
|
|
"mean_token_accuracy": 0.17827319502830505,
|
|
"num_tokens": 1091988.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.049498653411865,
|
|
"epoch": 0.028824561207503322,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000489774416005319,
|
|
"loss": 5.9449,
|
|
"mean_token_accuracy": 0.17476076930761336,
|
|
"num_tokens": 1095949.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 5.945538663864136,
|
|
"epoch": 0.028936284312958763,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004894203518413742,
|
|
"loss": 5.8761,
|
|
"mean_token_accuracy": 0.17729702293872834,
|
|
"num_tokens": 1099795.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 5.840633201599121,
|
|
"epoch": 0.029048007418414203,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0004890604092351701,
|
|
"loss": 5.7914,
|
|
"mean_token_accuracy": 0.18237984031438828,
|
|
"num_tokens": 1103894.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.022551822662353,
|
|
"epoch": 0.02915973052386964,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.000488694598054715,
|
|
"loss": 6.0166,
|
|
"mean_token_accuracy": 0.1801459826529026,
|
|
"num_tokens": 1108382.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 5.830238723754883,
|
|
"epoch": 0.02927145362932508,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004883229283289071,
|
|
"loss": 5.772,
|
|
"mean_token_accuracy": 0.1923062965273857,
|
|
"num_tokens": 1112841.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.063643503189087,
|
|
"epoch": 0.02938317673478052,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00048794541024725993,
|
|
"loss": 6.0292,
|
|
"mean_token_accuracy": 0.16797914803028108,
|
|
"num_tokens": 1117319.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.104041051864624,
|
|
"epoch": 0.02949489984023596,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0004875620541596221,
|
|
"loss": 5.976,
|
|
"mean_token_accuracy": 0.1713699594140053,
|
|
"num_tokens": 1121371.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 5.860101127624512,
|
|
"epoch": 0.0296066229456914,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 5.8614,
|
|
"mean_token_accuracy": 0.17473118454217912,
|
|
"num_tokens": 1125496.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.027236604690552,
|
|
"epoch": 0.029718346051146836,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004867778701657417,
|
|
"loss": 5.9316,
|
|
"mean_token_accuracy": 0.17779530882835387,
|
|
"num_tokens": 1129929.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 5.962581586837769,
|
|
"epoch": 0.029830069156602277,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00048637706375829955,
|
|
"loss": 5.8692,
|
|
"mean_token_accuracy": 0.18104994744062425,
|
|
"num_tokens": 1133982.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 5.87730565071106,
|
|
"epoch": 0.029941792262057718,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.000485970462341878,
|
|
"loss": 5.8023,
|
|
"mean_token_accuracy": 0.17854024320840836,
|
|
"num_tokens": 1138270.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.029954719543457,
|
|
"epoch": 0.030053515367513155,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00048555807706366044,
|
|
"loss": 5.9336,
|
|
"mean_token_accuracy": 0.17757426649332048,
|
|
"num_tokens": 1142197.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 5.986188697814941,
|
|
"epoch": 0.030165238472968595,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00048513991922939756,
|
|
"loss": 5.9806,
|
|
"mean_token_accuracy": 0.17655822336673738,
|
|
"num_tokens": 1146439.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 5.9268670082092285,
|
|
"epoch": 0.030276961578424032,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00048471600030309744,
|
|
"loss": 5.9419,
|
|
"mean_token_accuracy": 0.1782355859875679,
|
|
"num_tokens": 1150809.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.041894149780274,
|
|
"epoch": 0.030388684683879473,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00048428633190671186,
|
|
"loss": 6.0754,
|
|
"mean_token_accuracy": 0.16220128163695335,
|
|
"num_tokens": 1155132.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.196034908294678,
|
|
"epoch": 0.030500407789334914,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004838509258198167,
|
|
"loss": 6.0108,
|
|
"mean_token_accuracy": 0.16138193309307097,
|
|
"num_tokens": 1159694.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 5.919046926498413,
|
|
"epoch": 0.03061213089479035,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00048340979397929,
|
|
"loss": 5.9393,
|
|
"mean_token_accuracy": 0.16888010501861572,
|
|
"num_tokens": 1164070.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 5.922135591506958,
|
|
"epoch": 0.03072385400024579,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00048296294847898386,
|
|
"loss": 5.7818,
|
|
"mean_token_accuracy": 0.1902960404753685,
|
|
"num_tokens": 1168097.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 5.983519697189331,
|
|
"epoch": 0.03083557710570123,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004825104015693934,
|
|
"loss": 5.9272,
|
|
"mean_token_accuracy": 0.1677720718085766,
|
|
"num_tokens": 1172375.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 5.890913963317871,
|
|
"epoch": 0.03094730021115667,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 5.9675,
|
|
"mean_token_accuracy": 0.16840586364269255,
|
|
"num_tokens": 1176628.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 5.930252313613892,
|
|
"epoch": 0.03105902331661211,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00048158825330553505,
|
|
"loss": 5.9445,
|
|
"mean_token_accuracy": 0.16537177711725234,
|
|
"num_tokens": 1180500.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 5.924455261230468,
|
|
"epoch": 0.031170746422067547,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00048111867723242763,
|
|
"loss": 5.9551,
|
|
"mean_token_accuracy": 0.17216705679893493,
|
|
"num_tokens": 1184883.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.142921876907349,
|
|
"epoch": 0.031282469527522984,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004806434503116637,
|
|
"loss": 5.9368,
|
|
"mean_token_accuracy": 0.1701400026679039,
|
|
"num_tokens": 1188723.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.0046467781066895,
|
|
"epoch": 0.03139419263297843,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004801625855718296,
|
|
"loss": 6.0485,
|
|
"mean_token_accuracy": 0.1668444462120533,
|
|
"num_tokens": 1193318.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.041379928588867,
|
|
"epoch": 0.031505915738433865,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00047967609619607477,
|
|
"loss": 5.9308,
|
|
"mean_token_accuracy": 0.16749139726161957,
|
|
"num_tokens": 1197663.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 5.913714504241943,
|
|
"epoch": 0.0316176388438893,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004791839955217513,
|
|
"loss": 5.7884,
|
|
"mean_token_accuracy": 0.18735090345144273,
|
|
"num_tokens": 1201949.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.008267402648926,
|
|
"epoch": 0.031729361949344746,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00047868629704004786,
|
|
"loss": 6.0329,
|
|
"mean_token_accuracy": 0.16291153579950332,
|
|
"num_tokens": 1206360.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 5.969942045211792,
|
|
"epoch": 0.03184108505480018,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00047818301439561965,
|
|
"loss": 5.8897,
|
|
"mean_token_accuracy": 0.1797643095254898,
|
|
"num_tokens": 1210738.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 5.823257780075073,
|
|
"epoch": 0.03195280816025562,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00047767416138621454,
|
|
"loss": 5.8901,
|
|
"mean_token_accuracy": 0.17570878565311432,
|
|
"num_tokens": 1214722.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.028759336471557,
|
|
"epoch": 0.032064531265711065,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000477159751962295,
|
|
"loss": 5.9947,
|
|
"mean_token_accuracy": 0.17196338325738908,
|
|
"num_tokens": 1218811.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.077054071426391,
|
|
"epoch": 0.0321762543711665,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00047663980022665507,
|
|
"loss": 5.9492,
|
|
"mean_token_accuracy": 0.1680076539516449,
|
|
"num_tokens": 1223089.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 5.646877670288086,
|
|
"epoch": 0.03228797747662194,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 5.7538,
|
|
"mean_token_accuracy": 0.18203833252191542,
|
|
"num_tokens": 1226970.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 5.893757677078247,
|
|
"epoch": 0.03239970058207738,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004755833269907267,
|
|
"loss": 5.8528,
|
|
"mean_token_accuracy": 0.17612469345331191,
|
|
"num_tokens": 1230734.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 5.9495970726013185,
|
|
"epoch": 0.03251142368753282,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004750468344541857,
|
|
"loss": 5.8897,
|
|
"mean_token_accuracy": 0.16842119693756102,
|
|
"num_tokens": 1235363.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 5.881724309921265,
|
|
"epoch": 0.03262314679298826,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00047450485753262525,
|
|
"loss": 5.739,
|
|
"mean_token_accuracy": 0.17950696647167205,
|
|
"num_tokens": 1239826.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 5.911339473724365,
|
|
"epoch": 0.032734869898443694,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00047395741108461633,
|
|
"loss": 5.9883,
|
|
"mean_token_accuracy": 0.16461902111768723,
|
|
"num_tokens": 1244067.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 5.944296741485596,
|
|
"epoch": 0.03284659300389914,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00047340451011867985,
|
|
"loss": 5.7769,
|
|
"mean_token_accuracy": 0.18020680397748948,
|
|
"num_tokens": 1248307.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 5.810064172744751,
|
|
"epoch": 0.032958316109354575,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00047284616979287515,
|
|
"loss": 5.8516,
|
|
"mean_token_accuracy": 0.1767789214849472,
|
|
"num_tokens": 1252504.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 5.784211921691894,
|
|
"epoch": 0.03307003921481001,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00047228240541438433,
|
|
"loss": 5.8911,
|
|
"mean_token_accuracy": 0.17289571911096574,
|
|
"num_tokens": 1256823.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 5.894031572341919,
|
|
"epoch": 0.03318176232026546,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00047171323243909257,
|
|
"loss": 5.8256,
|
|
"mean_token_accuracy": 0.17667611539363862,
|
|
"num_tokens": 1261236.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 5.957602310180664,
|
|
"epoch": 0.033293485425720894,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00047113866647116457,
|
|
"loss": 5.836,
|
|
"mean_token_accuracy": 0.18172842562198638,
|
|
"num_tokens": 1265360.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 5.874336624145508,
|
|
"epoch": 0.03340520853117633,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004705587232626164,
|
|
"loss": 5.8134,
|
|
"mean_token_accuracy": 0.18341318368911744,
|
|
"num_tokens": 1269683.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 5.891045808792114,
|
|
"epoch": 0.033516931636631775,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00046997341871288424,
|
|
"loss": 5.8988,
|
|
"mean_token_accuracy": 0.1770934522151947,
|
|
"num_tokens": 1273843.0,
|
|
"step": 1500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 4000,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 1,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 1835102932992000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|