21266 lines
581 KiB
JSON
21266 lines
581 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 8.786610878661088,
|
|
"eval_steps": 500,
|
|
"global_step": 10500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.742597675323486,
|
|
"epoch": 0.0041841004184100415,
|
|
"grad_norm": 5.46875,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.7608,
|
|
"mean_token_accuracy": 9.661835501901805e-05,
|
|
"num_tokens": 9848.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.74262809753418,
|
|
"epoch": 0.008368200836820083,
|
|
"grad_norm": 5.5,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.723,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 18196.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.742603778839111,
|
|
"epoch": 0.012552301255230125,
|
|
"grad_norm": 5.34375,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.703,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 26927.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.742668437957764,
|
|
"epoch": 0.016736401673640166,
|
|
"grad_norm": 5.65625,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.6357,
|
|
"mean_token_accuracy": 0.0005675206426531077,
|
|
"num_tokens": 34987.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.742566776275634,
|
|
"epoch": 0.02092050209205021,
|
|
"grad_norm": 5.03125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.5588,
|
|
"mean_token_accuracy": 0.010127833811566234,
|
|
"num_tokens": 44936.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.742234230041504,
|
|
"epoch": 0.02510460251046025,
|
|
"grad_norm": 4.3125,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.427,
|
|
"mean_token_accuracy": 0.0348518967628479,
|
|
"num_tokens": 54562.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.741086196899413,
|
|
"epoch": 0.029288702928870293,
|
|
"grad_norm": 3.359375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.2814,
|
|
"mean_token_accuracy": 0.03771395143121481,
|
|
"num_tokens": 63844.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.738179874420165,
|
|
"epoch": 0.03347280334728033,
|
|
"grad_norm": 2.734375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.1587,
|
|
"mean_token_accuracy": 0.04109239801764488,
|
|
"num_tokens": 74331.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.734289169311523,
|
|
"epoch": 0.03765690376569038,
|
|
"grad_norm": 2.484375,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.995,
|
|
"mean_token_accuracy": 0.04218972772359848,
|
|
"num_tokens": 83028.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.730850219726562,
|
|
"epoch": 0.04184100418410042,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.9353,
|
|
"mean_token_accuracy": 0.040206428803503516,
|
|
"num_tokens": 91395.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.729630088806152,
|
|
"epoch": 0.04602510460251046,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.8674,
|
|
"mean_token_accuracy": 0.048422401398420335,
|
|
"num_tokens": 101868.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.727583599090575,
|
|
"epoch": 0.0502092050209205,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.7933,
|
|
"mean_token_accuracy": 0.04613714665174484,
|
|
"num_tokens": 110403.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.723635578155518,
|
|
"epoch": 0.05439330543933055,
|
|
"grad_norm": 1.921875,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.8021,
|
|
"mean_token_accuracy": 0.041691455990076065,
|
|
"num_tokens": 119557.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.719999694824219,
|
|
"epoch": 0.058577405857740586,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.676,
|
|
"mean_token_accuracy": 0.04646212197840214,
|
|
"num_tokens": 128903.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.71647481918335,
|
|
"epoch": 0.06276150627615062,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.6628,
|
|
"mean_token_accuracy": 0.041718531772494315,
|
|
"num_tokens": 137630.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.713119316101075,
|
|
"epoch": 0.06694560669456066,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.5313,
|
|
"mean_token_accuracy": 0.0421627514064312,
|
|
"num_tokens": 146335.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.708419799804688,
|
|
"epoch": 0.07112970711297072,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.4686,
|
|
"mean_token_accuracy": 0.04617839008569717,
|
|
"num_tokens": 155766.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.702109909057617,
|
|
"epoch": 0.07531380753138076,
|
|
"grad_norm": 1.859375,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.412,
|
|
"mean_token_accuracy": 0.04476847723126411,
|
|
"num_tokens": 165844.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.693535137176514,
|
|
"epoch": 0.0794979079497908,
|
|
"grad_norm": 1.90625,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.3394,
|
|
"mean_token_accuracy": 0.05125464424490929,
|
|
"num_tokens": 174858.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.678814315795899,
|
|
"epoch": 0.08368200836820083,
|
|
"grad_norm": 1.9140625,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.2166,
|
|
"mean_token_accuracy": 0.06135832369327545,
|
|
"num_tokens": 184020.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.653692531585694,
|
|
"epoch": 0.08786610878661087,
|
|
"grad_norm": 1.8828125,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.115,
|
|
"mean_token_accuracy": 0.05314192697405815,
|
|
"num_tokens": 193714.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.62697982788086,
|
|
"epoch": 0.09205020920502092,
|
|
"grad_norm": 1.90625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.0408,
|
|
"mean_token_accuracy": 0.05419124960899353,
|
|
"num_tokens": 203139.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.581832027435302,
|
|
"epoch": 0.09623430962343096,
|
|
"grad_norm": 1.875,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.8865,
|
|
"mean_token_accuracy": 0.05798161998391151,
|
|
"num_tokens": 212446.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.510433387756347,
|
|
"epoch": 0.100418410041841,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.7392,
|
|
"mean_token_accuracy": 0.06025223545730114,
|
|
"num_tokens": 220910.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.451266765594482,
|
|
"epoch": 0.10460251046025104,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.6729,
|
|
"mean_token_accuracy": 0.055778010934591296,
|
|
"num_tokens": 230522.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.406130599975587,
|
|
"epoch": 0.1087866108786611,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.5065,
|
|
"mean_token_accuracy": 0.05544214770197868,
|
|
"num_tokens": 240092.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.331532001495361,
|
|
"epoch": 0.11297071129707113,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.4083,
|
|
"mean_token_accuracy": 0.06165078543126583,
|
|
"num_tokens": 249478.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.237152481079102,
|
|
"epoch": 0.11715481171548117,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.2,
|
|
"mean_token_accuracy": 0.06240398585796356,
|
|
"num_tokens": 257714.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.116984844207764,
|
|
"epoch": 0.12133891213389121,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.1928,
|
|
"mean_token_accuracy": 0.05669947192072868,
|
|
"num_tokens": 266569.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 10.02002773284912,
|
|
"epoch": 0.12552301255230125,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.0619,
|
|
"mean_token_accuracy": 0.06153124496340752,
|
|
"num_tokens": 276688.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.869840335845947,
|
|
"epoch": 0.1297071129707113,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.8676,
|
|
"mean_token_accuracy": 0.06409889869391919,
|
|
"num_tokens": 285430.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.67565851211548,
|
|
"epoch": 0.13389121338912133,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.789,
|
|
"mean_token_accuracy": 0.06454115808010101,
|
|
"num_tokens": 295727.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.464604949951172,
|
|
"epoch": 0.13807531380753138,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.6957,
|
|
"mean_token_accuracy": 0.06615287810564041,
|
|
"num_tokens": 304854.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.24670057296753,
|
|
"epoch": 0.14225941422594143,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.6093,
|
|
"mean_token_accuracy": 0.07150843031704426,
|
|
"num_tokens": 313579.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 9.048341655731202,
|
|
"epoch": 0.14644351464435146,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.6128,
|
|
"mean_token_accuracy": 0.07077031470835209,
|
|
"num_tokens": 323310.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.823628616333007,
|
|
"epoch": 0.1506276150627615,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.4519,
|
|
"mean_token_accuracy": 0.06803311333060265,
|
|
"num_tokens": 331720.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.658255577087402,
|
|
"epoch": 0.15481171548117154,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.3797,
|
|
"mean_token_accuracy": 0.07455825209617614,
|
|
"num_tokens": 340331.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.487533950805664,
|
|
"epoch": 0.1589958158995816,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.3713,
|
|
"mean_token_accuracy": 0.07890852093696595,
|
|
"num_tokens": 349677.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.328230381011963,
|
|
"epoch": 0.16317991631799164,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.3358,
|
|
"mean_token_accuracy": 0.06953225210309029,
|
|
"num_tokens": 358674.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.279506969451905,
|
|
"epoch": 0.16736401673640167,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.33,
|
|
"mean_token_accuracy": 0.07827548906207085,
|
|
"num_tokens": 368258.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.202318572998047,
|
|
"epoch": 0.17154811715481172,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.2291,
|
|
"mean_token_accuracy": 0.08247470743954181,
|
|
"num_tokens": 377083.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.081873416900635,
|
|
"epoch": 0.17573221757322174,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.2799,
|
|
"mean_token_accuracy": 0.07478422187268734,
|
|
"num_tokens": 385456.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.072468662261963,
|
|
"epoch": 0.1799163179916318,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.1697,
|
|
"mean_token_accuracy": 0.0813635230064392,
|
|
"num_tokens": 395295.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.06587061882019,
|
|
"epoch": 0.18410041841004185,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.2016,
|
|
"mean_token_accuracy": 0.07997569367289543,
|
|
"num_tokens": 405023.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.917844343185425,
|
|
"epoch": 0.18828451882845187,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.2075,
|
|
"mean_token_accuracy": 0.07806592285633088,
|
|
"num_tokens": 414224.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.906620788574219,
|
|
"epoch": 0.19246861924686193,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.0979,
|
|
"mean_token_accuracy": 0.07944381982088089,
|
|
"num_tokens": 423951.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.995381021499634,
|
|
"epoch": 0.19665271966527198,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.1259,
|
|
"mean_token_accuracy": 0.08254904597997666,
|
|
"num_tokens": 432427.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.85433292388916,
|
|
"epoch": 0.200836820083682,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.0351,
|
|
"mean_token_accuracy": 0.08393027260899544,
|
|
"num_tokens": 441560.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.8878974437713625,
|
|
"epoch": 0.20502092050209206,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.1383,
|
|
"mean_token_accuracy": 0.0793025080114603,
|
|
"num_tokens": 450364.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.774558973312378,
|
|
"epoch": 0.20920502092050208,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.0707,
|
|
"mean_token_accuracy": 0.08014305979013443,
|
|
"num_tokens": 460076.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.792851877212525,
|
|
"epoch": 0.21338912133891214,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.023,
|
|
"mean_token_accuracy": 0.08774993643164634,
|
|
"num_tokens": 468718.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.835711526870727,
|
|
"epoch": 0.2175732217573222,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.0841,
|
|
"mean_token_accuracy": 0.08281839713454246,
|
|
"num_tokens": 478436.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.807744693756104,
|
|
"epoch": 0.2217573221757322,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.0932,
|
|
"mean_token_accuracy": 0.08015878945589065,
|
|
"num_tokens": 486833.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.68460259437561,
|
|
"epoch": 0.22594142259414227,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 6.9589,
|
|
"mean_token_accuracy": 0.09470265135169029,
|
|
"num_tokens": 495119.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.6987611770629885,
|
|
"epoch": 0.2301255230125523,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.1042,
|
|
"mean_token_accuracy": 0.09113300144672394,
|
|
"num_tokens": 505396.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.621922779083252,
|
|
"epoch": 0.23430962343096234,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 6.9717,
|
|
"mean_token_accuracy": 0.08859988525509835,
|
|
"num_tokens": 515259.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.760473251342773,
|
|
"epoch": 0.2384937238493724,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 6.9769,
|
|
"mean_token_accuracy": 0.09006874561309815,
|
|
"num_tokens": 523608.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.605395030975342,
|
|
"epoch": 0.24267782426778242,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.9736,
|
|
"mean_token_accuracy": 0.08525220081210136,
|
|
"num_tokens": 534613.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.54643816947937,
|
|
"epoch": 0.24686192468619247,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000147,
|
|
"loss": 6.9884,
|
|
"mean_token_accuracy": 0.09405484721064568,
|
|
"num_tokens": 543347.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.628460121154785,
|
|
"epoch": 0.2510460251046025,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.956,
|
|
"mean_token_accuracy": 0.08765949010848999,
|
|
"num_tokens": 552371.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.5912059307098385,
|
|
"epoch": 0.25523012552301255,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.8822,
|
|
"mean_token_accuracy": 0.08999307677149773,
|
|
"num_tokens": 560742.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.553208541870117,
|
|
"epoch": 0.2594142259414226,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.9822,
|
|
"mean_token_accuracy": 0.09437551945447922,
|
|
"num_tokens": 569960.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.580251216888428,
|
|
"epoch": 0.26359832635983266,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.9347,
|
|
"mean_token_accuracy": 0.08774123415350914,
|
|
"num_tokens": 578859.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.558048486709595,
|
|
"epoch": 0.26778242677824265,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.9406,
|
|
"mean_token_accuracy": 0.09432831630110741,
|
|
"num_tokens": 588146.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.445785140991211,
|
|
"epoch": 0.2719665271966527,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.8822,
|
|
"mean_token_accuracy": 0.0896220214664936,
|
|
"num_tokens": 597268.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.3907698631286625,
|
|
"epoch": 0.27615062761506276,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.8154,
|
|
"mean_token_accuracy": 0.0945595309138298,
|
|
"num_tokens": 606199.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.577031898498535,
|
|
"epoch": 0.2803347280334728,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.9307,
|
|
"mean_token_accuracy": 0.08274017758667469,
|
|
"num_tokens": 615966.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.524800825119018,
|
|
"epoch": 0.28451882845188287,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.9039,
|
|
"mean_token_accuracy": 0.08914825171232224,
|
|
"num_tokens": 625689.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.424820232391357,
|
|
"epoch": 0.28870292887029286,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.9367,
|
|
"mean_token_accuracy": 0.08863812088966369,
|
|
"num_tokens": 634877.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.479735708236694,
|
|
"epoch": 0.2928870292887029,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.8576,
|
|
"mean_token_accuracy": 0.09228913635015487,
|
|
"num_tokens": 644042.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.3995614528656,
|
|
"epoch": 0.29707112970711297,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.8397,
|
|
"mean_token_accuracy": 0.09121191054582596,
|
|
"num_tokens": 653072.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.390188837051392,
|
|
"epoch": 0.301255230125523,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.8429,
|
|
"mean_token_accuracy": 0.0936873622238636,
|
|
"num_tokens": 663686.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.433347082138061,
|
|
"epoch": 0.3054393305439331,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.7824,
|
|
"mean_token_accuracy": 0.09806219711899758,
|
|
"num_tokens": 672293.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.224409723281861,
|
|
"epoch": 0.30962343096234307,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.7495,
|
|
"mean_token_accuracy": 0.09524724259972572,
|
|
"num_tokens": 681758.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.373961687088013,
|
|
"epoch": 0.3138075313807531,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.7705,
|
|
"mean_token_accuracy": 0.0979244202375412,
|
|
"num_tokens": 690727.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.338280773162841,
|
|
"epoch": 0.3179916317991632,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.7246,
|
|
"mean_token_accuracy": 0.09411645382642746,
|
|
"num_tokens": 701210.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.166968154907226,
|
|
"epoch": 0.32217573221757323,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.7198,
|
|
"mean_token_accuracy": 0.0960633248090744,
|
|
"num_tokens": 710869.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.326888370513916,
|
|
"epoch": 0.3263598326359833,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.8285,
|
|
"mean_token_accuracy": 0.09366477653384209,
|
|
"num_tokens": 721455.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.2980750560760494,
|
|
"epoch": 0.3305439330543933,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.7142,
|
|
"mean_token_accuracy": 0.10093756765127182,
|
|
"num_tokens": 730917.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.274618291854859,
|
|
"epoch": 0.33472803347280333,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.7402,
|
|
"mean_token_accuracy": 0.09866241216659546,
|
|
"num_tokens": 739302.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.29223780632019,
|
|
"epoch": 0.3389121338912134,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.7097,
|
|
"mean_token_accuracy": 0.09633656591176987,
|
|
"num_tokens": 748785.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.214037179946899,
|
|
"epoch": 0.34309623430962344,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.6545,
|
|
"mean_token_accuracy": 0.10300204828381539,
|
|
"num_tokens": 757215.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.234053421020508,
|
|
"epoch": 0.3472803347280335,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.5793,
|
|
"mean_token_accuracy": 0.10455078259110451,
|
|
"num_tokens": 765682.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.289216184616089,
|
|
"epoch": 0.3514644351464435,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.7796,
|
|
"mean_token_accuracy": 0.0947409562766552,
|
|
"num_tokens": 774814.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.198604440689087,
|
|
"epoch": 0.35564853556485354,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.6991,
|
|
"mean_token_accuracy": 0.1013828955590725,
|
|
"num_tokens": 784019.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.1404036521911625,
|
|
"epoch": 0.3598326359832636,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.5758,
|
|
"mean_token_accuracy": 0.10198531225323677,
|
|
"num_tokens": 793237.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.1974271774292,
|
|
"epoch": 0.36401673640167365,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.6162,
|
|
"mean_token_accuracy": 0.10484573319554329,
|
|
"num_tokens": 801548.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.14229302406311,
|
|
"epoch": 0.3682008368200837,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.6647,
|
|
"mean_token_accuracy": 0.10372263044118882,
|
|
"num_tokens": 810974.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.127859115600586,
|
|
"epoch": 0.3723849372384937,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.5601,
|
|
"mean_token_accuracy": 0.1046792708337307,
|
|
"num_tokens": 820023.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.167815637588501,
|
|
"epoch": 0.37656903765690375,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.644,
|
|
"mean_token_accuracy": 0.10097603350877762,
|
|
"num_tokens": 828297.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.183742904663086,
|
|
"epoch": 0.3807531380753138,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.693,
|
|
"mean_token_accuracy": 0.0922306977212429,
|
|
"num_tokens": 838268.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.183096885681152,
|
|
"epoch": 0.38493723849372385,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.6117,
|
|
"mean_token_accuracy": 0.10094281509518624,
|
|
"num_tokens": 846491.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.094890165328979,
|
|
"epoch": 0.3891213389121339,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.5546,
|
|
"mean_token_accuracy": 0.10523919835686683,
|
|
"num_tokens": 856501.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.092893457412719,
|
|
"epoch": 0.39330543933054396,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.5958,
|
|
"mean_token_accuracy": 0.10569338127970695,
|
|
"num_tokens": 867054.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.0160187721252445,
|
|
"epoch": 0.39748953974895396,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.6239,
|
|
"mean_token_accuracy": 0.10482471361756325,
|
|
"num_tokens": 875956.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.013614892959595,
|
|
"epoch": 0.401673640167364,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.5334,
|
|
"mean_token_accuracy": 0.1062661923468113,
|
|
"num_tokens": 884427.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.170659828186035,
|
|
"epoch": 0.40585774058577406,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.6249,
|
|
"mean_token_accuracy": 0.10283279418945312,
|
|
"num_tokens": 894181.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 6.948815584182739,
|
|
"epoch": 0.4100418410041841,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.4707,
|
|
"mean_token_accuracy": 0.10783380344510078,
|
|
"num_tokens": 904076.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.047258281707764,
|
|
"epoch": 0.41422594142259417,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.6131,
|
|
"mean_token_accuracy": 0.10789759010076523,
|
|
"num_tokens": 913915.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.123381805419922,
|
|
"epoch": 0.41841004184100417,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.6688,
|
|
"mean_token_accuracy": 0.1050674706697464,
|
|
"num_tokens": 922876.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.41841004184100417,
|
|
"eval_entropy": 7.012504460621465,
|
|
"eval_loss": 6.6101908683776855,
|
|
"eval_mean_token_accuracy": 0.10924081287226795,
|
|
"eval_num_tokens": 922876.0,
|
|
"eval_runtime": 2.0948,
|
|
"eval_samples_per_second": 1865.089,
|
|
"eval_steps_per_second": 233.434,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.033219575881958,
|
|
"epoch": 0.4225941422594142,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.4668,
|
|
"mean_token_accuracy": 0.11332256868481635,
|
|
"num_tokens": 931998.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.079867506027222,
|
|
"epoch": 0.42677824267782427,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.5197,
|
|
"mean_token_accuracy": 0.11432325989007949,
|
|
"num_tokens": 941387.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 6.93487229347229,
|
|
"epoch": 0.4309623430962343,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.5834,
|
|
"mean_token_accuracy": 0.10249554514884948,
|
|
"num_tokens": 951866.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 6.953582286834717,
|
|
"epoch": 0.4351464435146444,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.5018,
|
|
"mean_token_accuracy": 0.10587486550211907,
|
|
"num_tokens": 961252.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.056652307510376,
|
|
"epoch": 0.4393305439330544,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.5132,
|
|
"mean_token_accuracy": 0.1105636864900589,
|
|
"num_tokens": 969857.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.941860485076904,
|
|
"epoch": 0.4435146443514644,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.5023,
|
|
"mean_token_accuracy": 0.10602044016122818,
|
|
"num_tokens": 978754.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.038395500183105,
|
|
"epoch": 0.4476987447698745,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.4919,
|
|
"mean_token_accuracy": 0.110679891705513,
|
|
"num_tokens": 987935.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.966084718704224,
|
|
"epoch": 0.45188284518828453,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.4664,
|
|
"mean_token_accuracy": 0.10865552499890327,
|
|
"num_tokens": 996873.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.99370813369751,
|
|
"epoch": 0.4560669456066946,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.5515,
|
|
"mean_token_accuracy": 0.1058430440723896,
|
|
"num_tokens": 1006487.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.852114534378051,
|
|
"epoch": 0.4602510460251046,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.3853,
|
|
"mean_token_accuracy": 0.11713317707180977,
|
|
"num_tokens": 1015715.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.969930839538574,
|
|
"epoch": 0.46443514644351463,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.5013,
|
|
"mean_token_accuracy": 0.11022178530693054,
|
|
"num_tokens": 1025091.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.90451455116272,
|
|
"epoch": 0.4686192468619247,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.4309,
|
|
"mean_token_accuracy": 0.10915803164243698,
|
|
"num_tokens": 1034174.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.026072406768799,
|
|
"epoch": 0.47280334728033474,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.6341,
|
|
"mean_token_accuracy": 0.1005947694182396,
|
|
"num_tokens": 1043777.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.970922946929932,
|
|
"epoch": 0.4769874476987448,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.472,
|
|
"mean_token_accuracy": 0.11195325776934624,
|
|
"num_tokens": 1052749.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.93495135307312,
|
|
"epoch": 0.4811715481171548,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.4883,
|
|
"mean_token_accuracy": 0.1101924866437912,
|
|
"num_tokens": 1061564.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.911145734786987,
|
|
"epoch": 0.48535564853556484,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.4042,
|
|
"mean_token_accuracy": 0.1145945630967617,
|
|
"num_tokens": 1070006.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.821494293212891,
|
|
"epoch": 0.4895397489539749,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.4755,
|
|
"mean_token_accuracy": 0.10867053493857384,
|
|
"num_tokens": 1079799.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.9094232559204105,
|
|
"epoch": 0.49372384937238495,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.3056,
|
|
"mean_token_accuracy": 0.12089367508888245,
|
|
"num_tokens": 1088842.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.798662376403809,
|
|
"epoch": 0.497907949790795,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.3798,
|
|
"mean_token_accuracy": 0.12022144198417664,
|
|
"num_tokens": 1097960.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.814728689193726,
|
|
"epoch": 0.502092050209205,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.3724,
|
|
"mean_token_accuracy": 0.11288540661334992,
|
|
"num_tokens": 1106797.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.8617775440216064,
|
|
"epoch": 0.5062761506276151,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.5201,
|
|
"mean_token_accuracy": 0.1136062651872635,
|
|
"num_tokens": 1116384.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.71177659034729,
|
|
"epoch": 0.5104602510460251,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.28,
|
|
"mean_token_accuracy": 0.11779255121946335,
|
|
"num_tokens": 1125206.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.948096656799317,
|
|
"epoch": 0.5146443514644351,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.5012,
|
|
"mean_token_accuracy": 0.11721058636903763,
|
|
"num_tokens": 1134110.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.975963926315307,
|
|
"epoch": 0.5188284518828452,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.498,
|
|
"mean_token_accuracy": 0.11025303602218628,
|
|
"num_tokens": 1142833.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.840864133834839,
|
|
"epoch": 0.5230125523012552,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.411,
|
|
"mean_token_accuracy": 0.11678725853562355,
|
|
"num_tokens": 1152006.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.82530255317688,
|
|
"epoch": 0.5271966527196653,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.3572,
|
|
"mean_token_accuracy": 0.11277682185173035,
|
|
"num_tokens": 1161061.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.856273174285889,
|
|
"epoch": 0.5313807531380753,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.3817,
|
|
"mean_token_accuracy": 0.11423582211136818,
|
|
"num_tokens": 1170754.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.753575038909912,
|
|
"epoch": 0.5355648535564853,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.3207,
|
|
"mean_token_accuracy": 0.11631909161806106,
|
|
"num_tokens": 1180854.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.778096914291382,
|
|
"epoch": 0.5397489539748954,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.399,
|
|
"mean_token_accuracy": 0.11601272895932198,
|
|
"num_tokens": 1190567.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.753851079940796,
|
|
"epoch": 0.5439330543933054,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.2537,
|
|
"mean_token_accuracy": 0.12245947867631912,
|
|
"num_tokens": 1200247.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.746957540512085,
|
|
"epoch": 0.5481171548117155,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.3527,
|
|
"mean_token_accuracy": 0.1171742670238018,
|
|
"num_tokens": 1209358.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.7102185726165775,
|
|
"epoch": 0.5523012552301255,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.3628,
|
|
"mean_token_accuracy": 0.11740919798612595,
|
|
"num_tokens": 1217746.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.741930294036865,
|
|
"epoch": 0.5564853556485355,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.3404,
|
|
"mean_token_accuracy": 0.12082380875945091,
|
|
"num_tokens": 1226829.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.716346597671508,
|
|
"epoch": 0.5606694560669456,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.2708,
|
|
"mean_token_accuracy": 0.1217438243329525,
|
|
"num_tokens": 1236642.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.851208734512329,
|
|
"epoch": 0.5648535564853556,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.3657,
|
|
"mean_token_accuracy": 0.12304765060544014,
|
|
"num_tokens": 1244934.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.569645929336548,
|
|
"epoch": 0.5690376569037657,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.2686,
|
|
"mean_token_accuracy": 0.11759066805243493,
|
|
"num_tokens": 1254786.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.745736312866211,
|
|
"epoch": 0.5732217573221757,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.2219,
|
|
"mean_token_accuracy": 0.12392263934016227,
|
|
"num_tokens": 1263745.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.629066610336304,
|
|
"epoch": 0.5774058577405857,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.3768,
|
|
"mean_token_accuracy": 0.12062641084194184,
|
|
"num_tokens": 1273101.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.80110445022583,
|
|
"epoch": 0.5815899581589958,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.3366,
|
|
"mean_token_accuracy": 0.1182647056877613,
|
|
"num_tokens": 1282547.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.721027660369873,
|
|
"epoch": 0.5857740585774058,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.361,
|
|
"mean_token_accuracy": 0.11650118827819825,
|
|
"num_tokens": 1291594.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.686480140686035,
|
|
"epoch": 0.5899581589958159,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.2755,
|
|
"mean_token_accuracy": 0.12708859369158745,
|
|
"num_tokens": 1300282.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.703722047805786,
|
|
"epoch": 0.5941422594142259,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.3668,
|
|
"mean_token_accuracy": 0.12298200204968453,
|
|
"num_tokens": 1309304.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.666914319992065,
|
|
"epoch": 0.5983263598326359,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.2053,
|
|
"mean_token_accuracy": 0.12288029938936233,
|
|
"num_tokens": 1317638.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.816176033020019,
|
|
"epoch": 0.602510460251046,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.4085,
|
|
"mean_token_accuracy": 0.1202589988708496,
|
|
"num_tokens": 1327242.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.689037799835205,
|
|
"epoch": 0.606694560669456,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.2151,
|
|
"mean_token_accuracy": 0.12098749801516533,
|
|
"num_tokens": 1335880.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.657412004470825,
|
|
"epoch": 0.6108786610878661,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.298,
|
|
"mean_token_accuracy": 0.12196487411856652,
|
|
"num_tokens": 1344909.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.623217487335205,
|
|
"epoch": 0.6150627615062761,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.1665,
|
|
"mean_token_accuracy": 0.12766727730631827,
|
|
"num_tokens": 1353997.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.612822961807251,
|
|
"epoch": 0.6192468619246861,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.3168,
|
|
"mean_token_accuracy": 0.12191536054015159,
|
|
"num_tokens": 1363293.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.85352053642273,
|
|
"epoch": 0.6234309623430963,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.335,
|
|
"mean_token_accuracy": 0.11951717808842659,
|
|
"num_tokens": 1372174.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.609316825866699,
|
|
"epoch": 0.6276150627615062,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.2327,
|
|
"mean_token_accuracy": 0.12466636523604394,
|
|
"num_tokens": 1381067.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.585425615310669,
|
|
"epoch": 0.6317991631799164,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.2179,
|
|
"mean_token_accuracy": 0.12963995933532715,
|
|
"num_tokens": 1390072.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.66300778388977,
|
|
"epoch": 0.6359832635983264,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.2147,
|
|
"mean_token_accuracy": 0.13023070693016053,
|
|
"num_tokens": 1399391.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.558363771438598,
|
|
"epoch": 0.6401673640167364,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.2236,
|
|
"mean_token_accuracy": 0.12857722416520118,
|
|
"num_tokens": 1408054.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.632496404647827,
|
|
"epoch": 0.6443514644351465,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.2303,
|
|
"mean_token_accuracy": 0.11898199543356895,
|
|
"num_tokens": 1417406.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.595701122283936,
|
|
"epoch": 0.6485355648535565,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.2822,
|
|
"mean_token_accuracy": 0.11752806529402733,
|
|
"num_tokens": 1426907.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.572923707962036,
|
|
"epoch": 0.6527196652719666,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.3368,
|
|
"mean_token_accuracy": 0.12002972811460495,
|
|
"num_tokens": 1436085.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.790693521499634,
|
|
"epoch": 0.6569037656903766,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.233,
|
|
"mean_token_accuracy": 0.1243402674794197,
|
|
"num_tokens": 1444823.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.5481942653656,
|
|
"epoch": 0.6610878661087866,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.2919,
|
|
"mean_token_accuracy": 0.12223824709653855,
|
|
"num_tokens": 1454621.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.545780181884766,
|
|
"epoch": 0.6652719665271967,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.1532,
|
|
"mean_token_accuracy": 0.1284099817276001,
|
|
"num_tokens": 1463784.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.611534309387207,
|
|
"epoch": 0.6694560669456067,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.2818,
|
|
"mean_token_accuracy": 0.12604539170861245,
|
|
"num_tokens": 1473257.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.62378249168396,
|
|
"epoch": 0.6736401673640168,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.23,
|
|
"mean_token_accuracy": 0.12763578295707703,
|
|
"num_tokens": 1482195.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.484483909606934,
|
|
"epoch": 0.6778242677824268,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.2622,
|
|
"mean_token_accuracy": 0.12974331825971602,
|
|
"num_tokens": 1491590.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.641168451309204,
|
|
"epoch": 0.6820083682008368,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.1307,
|
|
"mean_token_accuracy": 0.13457018807530402,
|
|
"num_tokens": 1500208.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.524597692489624,
|
|
"epoch": 0.6861924686192469,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.2619,
|
|
"mean_token_accuracy": 0.12744015082716942,
|
|
"num_tokens": 1509412.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.619950962066651,
|
|
"epoch": 0.6903765690376569,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.2651,
|
|
"mean_token_accuracy": 0.12576703503727912,
|
|
"num_tokens": 1518426.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.689215564727784,
|
|
"epoch": 0.694560669456067,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.2173,
|
|
"mean_token_accuracy": 0.12354342117905617,
|
|
"num_tokens": 1528605.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.413666105270385,
|
|
"epoch": 0.698744769874477,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.1746,
|
|
"mean_token_accuracy": 0.13070485070347787,
|
|
"num_tokens": 1537449.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.594534254074096,
|
|
"epoch": 0.702928870292887,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.1963,
|
|
"mean_token_accuracy": 0.13175581246614457,
|
|
"num_tokens": 1546743.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.484565734863281,
|
|
"epoch": 0.7071129707112971,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.1817,
|
|
"mean_token_accuracy": 0.12574852034449577,
|
|
"num_tokens": 1555691.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.489085245132446,
|
|
"epoch": 0.7112970711297071,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.1334,
|
|
"mean_token_accuracy": 0.1334412842988968,
|
|
"num_tokens": 1564957.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.588404178619385,
|
|
"epoch": 0.7154811715481172,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.1974,
|
|
"mean_token_accuracy": 0.13589535504579545,
|
|
"num_tokens": 1573901.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.4834239959716795,
|
|
"epoch": 0.7196652719665272,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.1916,
|
|
"mean_token_accuracy": 0.131468053907156,
|
|
"num_tokens": 1582644.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.61418342590332,
|
|
"epoch": 0.7238493723849372,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.3198,
|
|
"mean_token_accuracy": 0.12452186271548271,
|
|
"num_tokens": 1592363.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.546977949142456,
|
|
"epoch": 0.7280334728033473,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.2498,
|
|
"mean_token_accuracy": 0.12673747837543486,
|
|
"num_tokens": 1602012.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.463743782043457,
|
|
"epoch": 0.7322175732217573,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.141,
|
|
"mean_token_accuracy": 0.13073898032307624,
|
|
"num_tokens": 1611625.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.395287895202637,
|
|
"epoch": 0.7364016736401674,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.1141,
|
|
"mean_token_accuracy": 0.12709902375936508,
|
|
"num_tokens": 1620941.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.560024118423462,
|
|
"epoch": 0.7405857740585774,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.2078,
|
|
"mean_token_accuracy": 0.1278709888458252,
|
|
"num_tokens": 1629225.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.4868265151977536,
|
|
"epoch": 0.7447698744769874,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.1722,
|
|
"mean_token_accuracy": 0.12686478942632676,
|
|
"num_tokens": 1639551.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.418560361862182,
|
|
"epoch": 0.7489539748953975,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.1591,
|
|
"mean_token_accuracy": 0.11997205838561058,
|
|
"num_tokens": 1648963.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.467530012130737,
|
|
"epoch": 0.7531380753138075,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.0847,
|
|
"mean_token_accuracy": 0.1315011478960514,
|
|
"num_tokens": 1657735.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.564916038513184,
|
|
"epoch": 0.7573221757322176,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.1211,
|
|
"mean_token_accuracy": 0.13231319710612297,
|
|
"num_tokens": 1666349.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.3584685802459715,
|
|
"epoch": 0.7615062761506276,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.0491,
|
|
"mean_token_accuracy": 0.1330125942826271,
|
|
"num_tokens": 1675920.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.345976114273071,
|
|
"epoch": 0.7656903765690377,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.0834,
|
|
"mean_token_accuracy": 0.13028152361512185,
|
|
"num_tokens": 1685626.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.5751933574676515,
|
|
"epoch": 0.7698744769874477,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.2104,
|
|
"mean_token_accuracy": 0.13214068561792375,
|
|
"num_tokens": 1694972.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.388721895217896,
|
|
"epoch": 0.7740585774058577,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.1102,
|
|
"mean_token_accuracy": 0.14101462960243225,
|
|
"num_tokens": 1703806.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.467752361297608,
|
|
"epoch": 0.7782426778242678,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.1868,
|
|
"mean_token_accuracy": 0.13027679920196533,
|
|
"num_tokens": 1712949.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.603727626800537,
|
|
"epoch": 0.7824267782426778,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.3193,
|
|
"mean_token_accuracy": 0.12327147945761681,
|
|
"num_tokens": 1722328.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.54327802658081,
|
|
"epoch": 0.7866108786610879,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.2759,
|
|
"mean_token_accuracy": 0.1266297660768032,
|
|
"num_tokens": 1731880.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.298931503295899,
|
|
"epoch": 0.7907949790794979,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.0979,
|
|
"mean_token_accuracy": 0.1360059678554535,
|
|
"num_tokens": 1740719.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.476912546157837,
|
|
"epoch": 0.7949790794979079,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.1403,
|
|
"mean_token_accuracy": 0.13098946809768677,
|
|
"num_tokens": 1749753.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.520664882659912,
|
|
"epoch": 0.799163179916318,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.1454,
|
|
"mean_token_accuracy": 0.12673419713974,
|
|
"num_tokens": 1758913.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.467524242401123,
|
|
"epoch": 0.803347280334728,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.1628,
|
|
"mean_token_accuracy": 0.12942118048667908,
|
|
"num_tokens": 1768536.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.351734304428101,
|
|
"epoch": 0.8075313807531381,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.1028,
|
|
"mean_token_accuracy": 0.129199555516243,
|
|
"num_tokens": 1777291.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.46312575340271,
|
|
"epoch": 0.8117154811715481,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.2004,
|
|
"mean_token_accuracy": 0.13277156054973602,
|
|
"num_tokens": 1787186.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.409244060516357,
|
|
"epoch": 0.8158995815899581,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000487,
|
|
"loss": 5.9866,
|
|
"mean_token_accuracy": 0.13967311680316924,
|
|
"num_tokens": 1795581.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.252563381195069,
|
|
"epoch": 0.8200836820083682,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 5.9885,
|
|
"mean_token_accuracy": 0.141060072183609,
|
|
"num_tokens": 1804627.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.4267253398895265,
|
|
"epoch": 0.8242677824267782,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.1131,
|
|
"mean_token_accuracy": 0.13135791718959808,
|
|
"num_tokens": 1813808.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.388511610031128,
|
|
"epoch": 0.8284518828451883,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.1245,
|
|
"mean_token_accuracy": 0.1352509319782257,
|
|
"num_tokens": 1823422.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.442827367782593,
|
|
"epoch": 0.8326359832635983,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.1643,
|
|
"mean_token_accuracy": 0.13328976929187775,
|
|
"num_tokens": 1832437.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.325851535797119,
|
|
"epoch": 0.8368200836820083,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.0927,
|
|
"mean_token_accuracy": 0.13116609752178193,
|
|
"num_tokens": 1842063.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.8368200836820083,
|
|
"eval_entropy": 6.231119002788832,
|
|
"eval_loss": 6.111203670501709,
|
|
"eval_mean_token_accuracy": 0.1427788064493236,
|
|
"eval_num_tokens": 1842063.0,
|
|
"eval_runtime": 2.2784,
|
|
"eval_samples_per_second": 1714.779,
|
|
"eval_steps_per_second": 214.622,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.437012720108032,
|
|
"epoch": 0.8410041841004184,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004999998518355666,
|
|
"loss": 6.2293,
|
|
"mean_token_accuracy": 0.13318051919341087,
|
|
"num_tokens": 1851043.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.348101234436035,
|
|
"epoch": 0.8451882845188284,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999992499178905,
|
|
"loss": 6.0751,
|
|
"mean_token_accuracy": 0.13775536939501762,
|
|
"num_tokens": 1860268.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.429720544815064,
|
|
"epoch": 0.8493723849372385,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999981849879322,
|
|
"loss": 6.1844,
|
|
"mean_token_accuracy": 0.13298494592309,
|
|
"num_tokens": 1870096.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.338539743423462,
|
|
"epoch": 0.8535564853556485,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999966570478833,
|
|
"loss": 6.0722,
|
|
"mean_token_accuracy": 0.13555127531290054,
|
|
"num_tokens": 1879164.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.346892070770264,
|
|
"epoch": 0.8577405857740585,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004999946661008878,
|
|
"loss": 6.0087,
|
|
"mean_token_accuracy": 0.1377266153693199,
|
|
"num_tokens": 1888135.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.362658977508545,
|
|
"epoch": 0.8619246861924686,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000499992212151043,
|
|
"loss": 6.0797,
|
|
"mean_token_accuracy": 0.1345998913049698,
|
|
"num_tokens": 1898302.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.250446605682373,
|
|
"epoch": 0.8661087866108786,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004999892952033987,
|
|
"loss": 6.0507,
|
|
"mean_token_accuracy": 0.1350796677172184,
|
|
"num_tokens": 1907753.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.385110569000244,
|
|
"epoch": 0.8702928870292888,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004999859152639573,
|
|
"loss": 6.1031,
|
|
"mean_token_accuracy": 0.13561567589640616,
|
|
"num_tokens": 1917543.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.414516115188599,
|
|
"epoch": 0.8744769874476988,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999820723396745,
|
|
"loss": 6.1416,
|
|
"mean_token_accuracy": 0.13240931779146195,
|
|
"num_tokens": 1927335.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.312007570266724,
|
|
"epoch": 0.8786610878661087,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999777664384581,
|
|
"loss": 6.0641,
|
|
"mean_token_accuracy": 0.13389642611145974,
|
|
"num_tokens": 1936381.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.345766687393189,
|
|
"epoch": 0.8828451882845189,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004999729975691692,
|
|
"loss": 6.0975,
|
|
"mean_token_accuracy": 0.13339410722255707,
|
|
"num_tokens": 1945559.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.25928635597229,
|
|
"epoch": 0.8870292887029289,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999677657416212,
|
|
"loss": 5.9715,
|
|
"mean_token_accuracy": 0.13333128094673158,
|
|
"num_tokens": 1954709.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.3643087387084964,
|
|
"epoch": 0.891213389121339,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004999620709665804,
|
|
"loss": 6.0994,
|
|
"mean_token_accuracy": 0.13211975023150443,
|
|
"num_tokens": 1963664.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.317386150360107,
|
|
"epoch": 0.895397489539749,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999559132557659,
|
|
"loss": 6.1559,
|
|
"mean_token_accuracy": 0.1293097920715809,
|
|
"num_tokens": 1974103.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.498347997665405,
|
|
"epoch": 0.899581589958159,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004999492926218489,
|
|
"loss": 6.025,
|
|
"mean_token_accuracy": 0.13773725554347038,
|
|
"num_tokens": 1982924.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.243421602249145,
|
|
"epoch": 0.9037656903765691,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000499942209078454,
|
|
"loss": 5.9663,
|
|
"mean_token_accuracy": 0.1428417220711708,
|
|
"num_tokens": 1992033.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.1911012649536135,
|
|
"epoch": 0.9079497907949791,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999346626401578,
|
|
"loss": 5.9586,
|
|
"mean_token_accuracy": 0.14146680235862732,
|
|
"num_tokens": 2001218.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.288367033004761,
|
|
"epoch": 0.9121338912133892,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999266533224897,
|
|
"loss": 6.111,
|
|
"mean_token_accuracy": 0.1368482619524002,
|
|
"num_tokens": 2010463.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.326322317123413,
|
|
"epoch": 0.9163179916317992,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004999181811419314,
|
|
"loss": 5.9808,
|
|
"mean_token_accuracy": 0.13955054804682732,
|
|
"num_tokens": 2019246.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.388986015319825,
|
|
"epoch": 0.9205020920502092,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999092461159177,
|
|
"loss": 6.0683,
|
|
"mean_token_accuracy": 0.14024771228432656,
|
|
"num_tokens": 2028083.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.223995351791382,
|
|
"epoch": 0.9246861924686193,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004998998482628351,
|
|
"loss": 5.8752,
|
|
"mean_token_accuracy": 0.1448050245642662,
|
|
"num_tokens": 2036631.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.252813816070557,
|
|
"epoch": 0.9288702928870293,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004998899876020229,
|
|
"loss": 6.0088,
|
|
"mean_token_accuracy": 0.13815638795495033,
|
|
"num_tokens": 2045910.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.219067764282227,
|
|
"epoch": 0.9330543933054394,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000499879664153773,
|
|
"loss": 6.0618,
|
|
"mean_token_accuracy": 0.14408401846885682,
|
|
"num_tokens": 2055959.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.39411563873291,
|
|
"epoch": 0.9372384937238494,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004998688779393291,
|
|
"loss": 6.1148,
|
|
"mean_token_accuracy": 0.1365758113563061,
|
|
"num_tokens": 2065475.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.1112316131591795,
|
|
"epoch": 0.9414225941422594,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004998576289808877,
|
|
"loss": 5.885,
|
|
"mean_token_accuracy": 0.14670267850160598,
|
|
"num_tokens": 2074469.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.235973262786866,
|
|
"epoch": 0.9456066945606695,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998459173015972,
|
|
"loss": 6.0622,
|
|
"mean_token_accuracy": 0.14163898378610612,
|
|
"num_tokens": 2084604.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.387408065795898,
|
|
"epoch": 0.9497907949790795,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004998337429255584,
|
|
"loss": 6.0332,
|
|
"mean_token_accuracy": 0.141568361222744,
|
|
"num_tokens": 2093377.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.254647159576416,
|
|
"epoch": 0.9539748953974896,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998211058778243,
|
|
"loss": 6.0613,
|
|
"mean_token_accuracy": 0.1345399834215641,
|
|
"num_tokens": 2102798.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.338140487670898,
|
|
"epoch": 0.9581589958158996,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004998080061843999,
|
|
"loss": 6.034,
|
|
"mean_token_accuracy": 0.1373809665441513,
|
|
"num_tokens": 2112516.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.290405893325806,
|
|
"epoch": 0.9623430962343096,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000499794443872242,
|
|
"loss": 6.1161,
|
|
"mean_token_accuracy": 0.13683309629559517,
|
|
"num_tokens": 2121677.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.194741535186767,
|
|
"epoch": 0.9665271966527197,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004997804189692601,
|
|
"loss": 5.9541,
|
|
"mean_token_accuracy": 0.14311629980802537,
|
|
"num_tokens": 2131220.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.297001314163208,
|
|
"epoch": 0.9707112970711297,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004997659315043149,
|
|
"loss": 6.1041,
|
|
"mean_token_accuracy": 0.134016689658165,
|
|
"num_tokens": 2142104.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.216556453704834,
|
|
"epoch": 0.9748953974895398,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004997509815072191,
|
|
"loss": 6.0076,
|
|
"mean_token_accuracy": 0.1369858868420124,
|
|
"num_tokens": 2151970.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.243878126144409,
|
|
"epoch": 0.9790794979079498,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004997355690087379,
|
|
"loss": 5.9732,
|
|
"mean_token_accuracy": 0.13861552774906158,
|
|
"num_tokens": 2161371.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.119789552688599,
|
|
"epoch": 0.9832635983263598,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004997196940405874,
|
|
"loss": 5.9401,
|
|
"mean_token_accuracy": 0.13875976651906968,
|
|
"num_tokens": 2170650.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.23285436630249,
|
|
"epoch": 0.9874476987447699,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004997033566354358,
|
|
"loss": 5.9983,
|
|
"mean_token_accuracy": 0.14316384941339494,
|
|
"num_tokens": 2180934.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.242339372634888,
|
|
"epoch": 0.9916317991631799,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004996865568269028,
|
|
"loss": 5.9674,
|
|
"mean_token_accuracy": 0.1459835335612297,
|
|
"num_tokens": 2189894.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.221574783325195,
|
|
"epoch": 0.99581589958159,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004996692946495597,
|
|
"loss": 5.9109,
|
|
"mean_token_accuracy": 0.15525714308023453,
|
|
"num_tokens": 2200208.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.222945690155029,
|
|
"epoch": 1.0,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004996515701389295,
|
|
"loss": 5.9637,
|
|
"mean_token_accuracy": 0.1387474164366722,
|
|
"num_tokens": 2209322.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.12794132232666,
|
|
"epoch": 1.00418410041841,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004996333833314862,
|
|
"loss": 5.7841,
|
|
"mean_token_accuracy": 0.1540491595864296,
|
|
"num_tokens": 2217663.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.149629306793213,
|
|
"epoch": 1.00836820083682,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004996147342646554,
|
|
"loss": 5.8162,
|
|
"mean_token_accuracy": 0.14273207560181617,
|
|
"num_tokens": 2226559.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.246478652954101,
|
|
"epoch": 1.0125523012552302,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004995956229768137,
|
|
"loss": 5.8536,
|
|
"mean_token_accuracy": 0.15705856084823608,
|
|
"num_tokens": 2235468.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.064305543899536,
|
|
"epoch": 1.0167364016736402,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004995760495072893,
|
|
"loss": 5.6822,
|
|
"mean_token_accuracy": 0.1609800159931183,
|
|
"num_tokens": 2245176.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.180075407028198,
|
|
"epoch": 1.0209205020920502,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004995560138963611,
|
|
"loss": 5.8354,
|
|
"mean_token_accuracy": 0.15163842141628264,
|
|
"num_tokens": 2255123.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.143266677856445,
|
|
"epoch": 1.0251046025104602,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004995355161852593,
|
|
"loss": 5.7615,
|
|
"mean_token_accuracy": 0.14811853021383287,
|
|
"num_tokens": 2263596.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.131743717193603,
|
|
"epoch": 1.0292887029288702,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004995145564161648,
|
|
"loss": 5.7905,
|
|
"mean_token_accuracy": 0.14918514639139174,
|
|
"num_tokens": 2272228.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.156940174102783,
|
|
"epoch": 1.0334728033472804,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004994931346322094,
|
|
"loss": 5.8357,
|
|
"mean_token_accuracy": 0.1530950203537941,
|
|
"num_tokens": 2281248.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.178971242904663,
|
|
"epoch": 1.0376569037656904,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004994712508774757,
|
|
"loss": 5.813,
|
|
"mean_token_accuracy": 0.14753878712654114,
|
|
"num_tokens": 2290958.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.062367248535156,
|
|
"epoch": 1.0418410041841004,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000499448905196997,
|
|
"loss": 5.7476,
|
|
"mean_token_accuracy": 0.1606424257159233,
|
|
"num_tokens": 2299763.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.122745561599731,
|
|
"epoch": 1.0460251046025104,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004994260976367573,
|
|
"loss": 5.7576,
|
|
"mean_token_accuracy": 0.1608661964535713,
|
|
"num_tokens": 2309847.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.17245512008667,
|
|
"epoch": 1.0502092050209204,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004994028282436905,
|
|
"loss": 5.8114,
|
|
"mean_token_accuracy": 0.14902583807706832,
|
|
"num_tokens": 2317556.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.0724709033966064,
|
|
"epoch": 1.0543933054393306,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004993790970656814,
|
|
"loss": 5.7246,
|
|
"mean_token_accuracy": 0.15202396363019943,
|
|
"num_tokens": 2327151.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.026160478591919,
|
|
"epoch": 1.0585774058577406,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004993549041515652,
|
|
"loss": 5.7251,
|
|
"mean_token_accuracy": 0.15616882890462874,
|
|
"num_tokens": 2336119.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.0959837436676025,
|
|
"epoch": 1.0627615062761506,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004993302495511268,
|
|
"loss": 5.7422,
|
|
"mean_token_accuracy": 0.15904839485883712,
|
|
"num_tokens": 2345262.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.198559761047363,
|
|
"epoch": 1.0669456066945606,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004993051333151014,
|
|
"loss": 5.8823,
|
|
"mean_token_accuracy": 0.14653945490717887,
|
|
"num_tokens": 2355210.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.012920284271241,
|
|
"epoch": 1.0711297071129706,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004992795554951742,
|
|
"loss": 5.6244,
|
|
"mean_token_accuracy": 0.15309498757123946,
|
|
"num_tokens": 2364599.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 5.942229127883911,
|
|
"epoch": 1.0753138075313808,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 0.0004992535161439803,
|
|
"loss": 5.8123,
|
|
"mean_token_accuracy": 0.14621696844697,
|
|
"num_tokens": 2374143.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.222422313690186,
|
|
"epoch": 1.0794979079497908,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004992270153151045,
|
|
"loss": 5.8022,
|
|
"mean_token_accuracy": 0.1560079947113991,
|
|
"num_tokens": 2383155.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.197408390045166,
|
|
"epoch": 1.0836820083682008,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004992000530630812,
|
|
"loss": 5.8354,
|
|
"mean_token_accuracy": 0.1492285281419754,
|
|
"num_tokens": 2392542.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.031621360778809,
|
|
"epoch": 1.0878661087866108,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004991726294433943,
|
|
"loss": 5.8021,
|
|
"mean_token_accuracy": 0.14778647273778917,
|
|
"num_tokens": 2401851.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.112837648391723,
|
|
"epoch": 1.0920502092050208,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004991447445124774,
|
|
"loss": 5.7584,
|
|
"mean_token_accuracy": 0.15441246777772905,
|
|
"num_tokens": 2410726.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.012867641448975,
|
|
"epoch": 1.096234309623431,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000499116398327713,
|
|
"loss": 5.8041,
|
|
"mean_token_accuracy": 0.14813676327466965,
|
|
"num_tokens": 2420702.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.022567844390869,
|
|
"epoch": 1.100418410041841,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004990875909474331,
|
|
"loss": 5.7489,
|
|
"mean_token_accuracy": 0.15360454171895982,
|
|
"num_tokens": 2431335.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.095766305923462,
|
|
"epoch": 1.104602510460251,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004990583224309187,
|
|
"loss": 5.7173,
|
|
"mean_token_accuracy": 0.15707379579544067,
|
|
"num_tokens": 2440722.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.091929769515991,
|
|
"epoch": 1.108786610878661,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004990285928383997,
|
|
"loss": 5.7968,
|
|
"mean_token_accuracy": 0.14727665334939957,
|
|
"num_tokens": 2451230.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.078146171569824,
|
|
"epoch": 1.112970711297071,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004989984022310545,
|
|
"loss": 5.8262,
|
|
"mean_token_accuracy": 0.14682078659534453,
|
|
"num_tokens": 2461442.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.035668563842774,
|
|
"epoch": 1.1171548117154813,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004989677506710109,
|
|
"loss": 5.6587,
|
|
"mean_token_accuracy": 0.15804804116487503,
|
|
"num_tokens": 2471381.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.023805093765259,
|
|
"epoch": 1.1213389121338913,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004989366382213449,
|
|
"loss": 5.7038,
|
|
"mean_token_accuracy": 0.1592436045408249,
|
|
"num_tokens": 2479937.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.033054256439209,
|
|
"epoch": 1.1255230125523012,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004989050649460806,
|
|
"loss": 5.6689,
|
|
"mean_token_accuracy": 0.16124916672706605,
|
|
"num_tokens": 2488974.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.041873598098755,
|
|
"epoch": 1.1297071129707112,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000498873030910191,
|
|
"loss": 5.8188,
|
|
"mean_token_accuracy": 0.1526327461004257,
|
|
"num_tokens": 2498800.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 5.99496603012085,
|
|
"epoch": 1.1338912133891212,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004988405361795969,
|
|
"loss": 5.7115,
|
|
"mean_token_accuracy": 0.15621206909418106,
|
|
"num_tokens": 2506580.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.047221612930298,
|
|
"epoch": 1.1380753138075315,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004988075808211672,
|
|
"loss": 5.756,
|
|
"mean_token_accuracy": 0.1572654902935028,
|
|
"num_tokens": 2516342.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 5.983310604095459,
|
|
"epoch": 1.1422594142259415,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004987741649027189,
|
|
"loss": 5.8001,
|
|
"mean_token_accuracy": 0.1560211196541786,
|
|
"num_tokens": 2526054.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.1061523914337155,
|
|
"epoch": 1.1464435146443515,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004987402884930163,
|
|
"loss": 5.7612,
|
|
"mean_token_accuracy": 0.1552029624581337,
|
|
"num_tokens": 2535556.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.062144947052002,
|
|
"epoch": 1.1506276150627615,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004987059516617717,
|
|
"loss": 5.7273,
|
|
"mean_token_accuracy": 0.15809612423181535,
|
|
"num_tokens": 2544930.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 5.931625747680664,
|
|
"epoch": 1.1548117154811715,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004986711544796449,
|
|
"loss": 5.6373,
|
|
"mean_token_accuracy": 0.1627542957663536,
|
|
"num_tokens": 2554362.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.023519992828369,
|
|
"epoch": 1.1589958158995817,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000498635897018243,
|
|
"loss": 5.7605,
|
|
"mean_token_accuracy": 0.1554619088768959,
|
|
"num_tokens": 2563499.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.004778623580933,
|
|
"epoch": 1.1631799163179917,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004986001793501198,
|
|
"loss": 5.7618,
|
|
"mean_token_accuracy": 0.15196048617362976,
|
|
"num_tokens": 2573346.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.028854703903198,
|
|
"epoch": 1.1673640167364017,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004985640015487768,
|
|
"loss": 5.6471,
|
|
"mean_token_accuracy": 0.16079026758670806,
|
|
"num_tokens": 2582541.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.001078462600708,
|
|
"epoch": 1.1715481171548117,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004985273636886621,
|
|
"loss": 5.7365,
|
|
"mean_token_accuracy": 0.15553900450468064,
|
|
"num_tokens": 2592015.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.000535202026367,
|
|
"epoch": 1.1757322175732217,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004984902658451704,
|
|
"loss": 5.7402,
|
|
"mean_token_accuracy": 0.1547416031360626,
|
|
"num_tokens": 2601998.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.00449857711792,
|
|
"epoch": 1.1799163179916319,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004984527080946432,
|
|
"loss": 5.6785,
|
|
"mean_token_accuracy": 0.16148286759853364,
|
|
"num_tokens": 2611435.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.018415307998657,
|
|
"epoch": 1.1841004184100419,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004984146905143683,
|
|
"loss": 5.7759,
|
|
"mean_token_accuracy": 0.15475187003612517,
|
|
"num_tokens": 2621639.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.0044273853302,
|
|
"epoch": 1.1882845188284519,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004983762131825797,
|
|
"loss": 5.7389,
|
|
"mean_token_accuracy": 0.1585055634379387,
|
|
"num_tokens": 2631390.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 5.9931258201599125,
|
|
"epoch": 1.1924686192468619,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004983372761784576,
|
|
"loss": 5.7802,
|
|
"mean_token_accuracy": 0.15404947847127914,
|
|
"num_tokens": 2640769.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.096167278289795,
|
|
"epoch": 1.196652719665272,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004982978795821281,
|
|
"loss": 5.779,
|
|
"mean_token_accuracy": 0.15546282827854158,
|
|
"num_tokens": 2651137.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 5.995174264907837,
|
|
"epoch": 1.200836820083682,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004982580234746632,
|
|
"loss": 5.7072,
|
|
"mean_token_accuracy": 0.15209258496761321,
|
|
"num_tokens": 2660530.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 5.99799337387085,
|
|
"epoch": 1.205020920502092,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00049821770793808,
|
|
"loss": 5.7481,
|
|
"mean_token_accuracy": 0.1549163356423378,
|
|
"num_tokens": 2669541.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.023744249343872,
|
|
"epoch": 1.209205020920502,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004981769330553418,
|
|
"loss": 5.7502,
|
|
"mean_token_accuracy": 0.15705211609601974,
|
|
"num_tokens": 2678542.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.029311323165894,
|
|
"epoch": 1.213389121338912,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004981356989103566,
|
|
"loss": 5.8299,
|
|
"mean_token_accuracy": 0.15689381957054138,
|
|
"num_tokens": 2687720.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.031181144714355,
|
|
"epoch": 1.217573221757322,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004980940055879774,
|
|
"loss": 5.6934,
|
|
"mean_token_accuracy": 0.15815790593624116,
|
|
"num_tokens": 2697641.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 5.9612993717193605,
|
|
"epoch": 1.2217573221757323,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004980518531740027,
|
|
"loss": 5.7072,
|
|
"mean_token_accuracy": 0.1583000048995018,
|
|
"num_tokens": 2706600.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.063562631607056,
|
|
"epoch": 1.2259414225941423,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004980092417551752,
|
|
"loss": 5.8514,
|
|
"mean_token_accuracy": 0.14898592680692674,
|
|
"num_tokens": 2716633.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.047478342056275,
|
|
"epoch": 1.2301255230125523,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004979661714191825,
|
|
"loss": 5.628,
|
|
"mean_token_accuracy": 0.16600804030895233,
|
|
"num_tokens": 2724829.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 5.925992202758789,
|
|
"epoch": 1.2343096234309623,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004979226422546563,
|
|
"loss": 5.7374,
|
|
"mean_token_accuracy": 0.1585901975631714,
|
|
"num_tokens": 2734276.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.035736894607544,
|
|
"epoch": 1.2384937238493725,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004978786543511726,
|
|
"loss": 5.8273,
|
|
"mean_token_accuracy": 0.157758666574955,
|
|
"num_tokens": 2744413.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.014146232604981,
|
|
"epoch": 1.2426778242677825,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004978342077992515,
|
|
"loss": 5.7065,
|
|
"mean_token_accuracy": 0.15562080591917038,
|
|
"num_tokens": 2753242.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 5.832571172714234,
|
|
"epoch": 1.2468619246861925,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004977893026903568,
|
|
"loss": 5.6248,
|
|
"mean_token_accuracy": 0.16905127316713334,
|
|
"num_tokens": 2761811.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.082716751098633,
|
|
"epoch": 1.2510460251046025,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004977439391168963,
|
|
"loss": 5.7302,
|
|
"mean_token_accuracy": 0.15641461163759232,
|
|
"num_tokens": 2770560.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.041548204421997,
|
|
"epoch": 1.2552301255230125,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004976981171722205,
|
|
"loss": 5.7379,
|
|
"mean_token_accuracy": 0.15544732064008712,
|
|
"num_tokens": 2780131.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.2552301255230125,
|
|
"eval_entropy": 5.6769954109972,
|
|
"eval_loss": 5.8483991622924805,
|
|
"eval_mean_token_accuracy": 0.15941667107290042,
|
|
"eval_num_tokens": 2780131.0,
|
|
"eval_runtime": 2.0884,
|
|
"eval_samples_per_second": 1870.827,
|
|
"eval_steps_per_second": 234.153,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.092925977706909,
|
|
"epoch": 1.2594142259414225,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004976518369506237,
|
|
"loss": 5.8435,
|
|
"mean_token_accuracy": 0.14601506665349007,
|
|
"num_tokens": 2789319.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 5.92581090927124,
|
|
"epoch": 1.2635983263598327,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004976050985473433,
|
|
"loss": 5.7037,
|
|
"mean_token_accuracy": 0.1522383913397789,
|
|
"num_tokens": 2799253.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.032972526550293,
|
|
"epoch": 1.2677824267782427,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004975579020585593,
|
|
"loss": 5.8124,
|
|
"mean_token_accuracy": 0.15160835534334183,
|
|
"num_tokens": 2808968.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.074542236328125,
|
|
"epoch": 1.2719665271966527,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004975102475813947,
|
|
"loss": 5.6999,
|
|
"mean_token_accuracy": 0.1575356364250183,
|
|
"num_tokens": 2817282.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.00976939201355,
|
|
"epoch": 1.2761506276150627,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004974621352139144,
|
|
"loss": 5.7538,
|
|
"mean_token_accuracy": 0.15369689762592315,
|
|
"num_tokens": 2826395.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 5.989823675155639,
|
|
"epoch": 1.280334728033473,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004974135650551262,
|
|
"loss": 5.8344,
|
|
"mean_token_accuracy": 0.15231696516275406,
|
|
"num_tokens": 2836195.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.010791635513305,
|
|
"epoch": 1.284518828451883,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004973645372049796,
|
|
"loss": 5.6702,
|
|
"mean_token_accuracy": 0.16355077773332596,
|
|
"num_tokens": 2846020.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 5.891250371932983,
|
|
"epoch": 1.288702928870293,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004973150517643659,
|
|
"loss": 5.6186,
|
|
"mean_token_accuracy": 0.16435685753822327,
|
|
"num_tokens": 2854895.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 5.977266883850097,
|
|
"epoch": 1.292887029288703,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004972651088351185,
|
|
"loss": 5.6632,
|
|
"mean_token_accuracy": 0.16027058362960817,
|
|
"num_tokens": 2864491.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 5.980721950531006,
|
|
"epoch": 1.297071129707113,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004972147085200116,
|
|
"loss": 5.6816,
|
|
"mean_token_accuracy": 0.15872606933116912,
|
|
"num_tokens": 2873962.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.017466497421265,
|
|
"epoch": 1.301255230125523,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004971638509227611,
|
|
"loss": 5.7151,
|
|
"mean_token_accuracy": 0.16112489104270936,
|
|
"num_tokens": 2883530.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 5.925843286514282,
|
|
"epoch": 1.3054393305439331,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004971125361480236,
|
|
"loss": 5.6906,
|
|
"mean_token_accuracy": 0.15688816756010054,
|
|
"num_tokens": 2892867.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 5.94919638633728,
|
|
"epoch": 1.3096234309623431,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004970607643013969,
|
|
"loss": 5.6645,
|
|
"mean_token_accuracy": 0.15854619294404984,
|
|
"num_tokens": 2902320.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 5.903445148468018,
|
|
"epoch": 1.3138075313807531,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000497008535489419,
|
|
"loss": 5.5946,
|
|
"mean_token_accuracy": 0.16695220172405242,
|
|
"num_tokens": 2911390.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 5.889070701599121,
|
|
"epoch": 1.3179916317991631,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004969558498195684,
|
|
"loss": 5.6532,
|
|
"mean_token_accuracy": 0.16906433552503586,
|
|
"num_tokens": 2921295.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 5.958902788162232,
|
|
"epoch": 1.3221757322175733,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004969027074002638,
|
|
"loss": 5.6603,
|
|
"mean_token_accuracy": 0.15776332914829255,
|
|
"num_tokens": 2931328.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 5.8630468368530275,
|
|
"epoch": 1.3263598326359833,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004968491083408636,
|
|
"loss": 5.6133,
|
|
"mean_token_accuracy": 0.1640845373272896,
|
|
"num_tokens": 2940193.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 5.861121034622192,
|
|
"epoch": 1.3305439330543933,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004967950527516663,
|
|
"loss": 5.6416,
|
|
"mean_token_accuracy": 0.16769741475582123,
|
|
"num_tokens": 2949759.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 5.887998342514038,
|
|
"epoch": 1.3347280334728033,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004967405407439092,
|
|
"loss": 5.577,
|
|
"mean_token_accuracy": 0.15625791251659393,
|
|
"num_tokens": 2959185.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 5.965415334701538,
|
|
"epoch": 1.3389121338912133,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004966855724297698,
|
|
"loss": 5.7328,
|
|
"mean_token_accuracy": 0.15423317551612853,
|
|
"num_tokens": 2968945.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 5.944570827484131,
|
|
"epoch": 1.3430962343096233,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004966301479223637,
|
|
"loss": 5.6716,
|
|
"mean_token_accuracy": 0.159438119828701,
|
|
"num_tokens": 2978395.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 5.858027172088623,
|
|
"epoch": 1.3472803347280335,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004965742673357456,
|
|
"loss": 5.5804,
|
|
"mean_token_accuracy": 0.1700463503599167,
|
|
"num_tokens": 2987030.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 5.9461335182189945,
|
|
"epoch": 1.3514644351464435,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004965179307849088,
|
|
"loss": 5.6511,
|
|
"mean_token_accuracy": 0.16362962871789932,
|
|
"num_tokens": 2995736.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 5.962363815307617,
|
|
"epoch": 1.3556485355648535,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000496461138385785,
|
|
"loss": 5.7402,
|
|
"mean_token_accuracy": 0.15967344343662263,
|
|
"num_tokens": 3005105.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 5.873168802261352,
|
|
"epoch": 1.3598326359832635,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004964038902552438,
|
|
"loss": 5.6571,
|
|
"mean_token_accuracy": 0.16216485053300858,
|
|
"num_tokens": 3013857.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 5.950975131988526,
|
|
"epoch": 1.3640167364016738,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004963461865110926,
|
|
"loss": 5.689,
|
|
"mean_token_accuracy": 0.1605420783162117,
|
|
"num_tokens": 3023972.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 5.836272764205932,
|
|
"epoch": 1.3682008368200838,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004962880272720764,
|
|
"loss": 5.6589,
|
|
"mean_token_accuracy": 0.16736868917942047,
|
|
"num_tokens": 3032705.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 5.930795669555664,
|
|
"epoch": 1.3723849372384938,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004962294126578775,
|
|
"loss": 5.5945,
|
|
"mean_token_accuracy": 0.16802308410406114,
|
|
"num_tokens": 3040900.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 5.917369699478149,
|
|
"epoch": 1.3765690376569037,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004961703427891154,
|
|
"loss": 5.6341,
|
|
"mean_token_accuracy": 0.16583087146282197,
|
|
"num_tokens": 3050477.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 5.9194018840789795,
|
|
"epoch": 1.3807531380753137,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004961108177873466,
|
|
"loss": 5.7088,
|
|
"mean_token_accuracy": 0.1571326047182083,
|
|
"num_tokens": 3059743.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 5.843382024765015,
|
|
"epoch": 1.3849372384937237,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004960508377750638,
|
|
"loss": 5.5053,
|
|
"mean_token_accuracy": 0.16926025450229645,
|
|
"num_tokens": 3067801.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 5.825117397308349,
|
|
"epoch": 1.389121338912134,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004959904028756961,
|
|
"loss": 5.6868,
|
|
"mean_token_accuracy": 0.156205852329731,
|
|
"num_tokens": 3077404.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 5.940052938461304,
|
|
"epoch": 1.393305439330544,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004959295132136091,
|
|
"loss": 5.7066,
|
|
"mean_token_accuracy": 0.15989601761102676,
|
|
"num_tokens": 3087130.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 5.978323650360108,
|
|
"epoch": 1.397489539748954,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004958681689141036,
|
|
"loss": 5.7219,
|
|
"mean_token_accuracy": 0.1560539923608303,
|
|
"num_tokens": 3096038.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 5.858187627792359,
|
|
"epoch": 1.401673640167364,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004958063701034167,
|
|
"loss": 5.5735,
|
|
"mean_token_accuracy": 0.16297442018985747,
|
|
"num_tokens": 3105668.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 5.857545661926269,
|
|
"epoch": 1.4058577405857742,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00049574411690872,
|
|
"loss": 5.5643,
|
|
"mean_token_accuracy": 0.15730429887771608,
|
|
"num_tokens": 3114020.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 5.732494449615478,
|
|
"epoch": 1.4100418410041842,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004956814094581208,
|
|
"loss": 5.602,
|
|
"mean_token_accuracy": 0.16669758707284926,
|
|
"num_tokens": 3124096.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 5.935622072219848,
|
|
"epoch": 1.4142259414225942,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000495618247880661,
|
|
"loss": 5.6677,
|
|
"mean_token_accuracy": 0.16170076429843902,
|
|
"num_tokens": 3132806.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 5.8426909923553465,
|
|
"epoch": 1.4184100418410042,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004955546323063168,
|
|
"loss": 5.5876,
|
|
"mean_token_accuracy": 0.17422180920839309,
|
|
"num_tokens": 3142001.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 5.8333580017089846,
|
|
"epoch": 1.4225941422594142,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004954905628659989,
|
|
"loss": 5.5666,
|
|
"mean_token_accuracy": 0.16144170612096786,
|
|
"num_tokens": 3151658.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 5.846378564834595,
|
|
"epoch": 1.4267782426778242,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004954260396915521,
|
|
"loss": 5.591,
|
|
"mean_token_accuracy": 0.1657014384865761,
|
|
"num_tokens": 3159573.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 5.669739723205566,
|
|
"epoch": 1.4309623430962344,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004953610629157544,
|
|
"loss": 5.4647,
|
|
"mean_token_accuracy": 0.17399210333824158,
|
|
"num_tokens": 3168312.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 5.927958869934082,
|
|
"epoch": 1.4351464435146444,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004952956326723177,
|
|
"loss": 5.5834,
|
|
"mean_token_accuracy": 0.17099785804748535,
|
|
"num_tokens": 3176811.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 5.8302654266357425,
|
|
"epoch": 1.4393305439330544,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000495229749095887,
|
|
"loss": 5.6826,
|
|
"mean_token_accuracy": 0.15766623616218567,
|
|
"num_tokens": 3186599.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 5.99775767326355,
|
|
"epoch": 1.4435146443514644,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00049516341232204,
|
|
"loss": 5.7751,
|
|
"mean_token_accuracy": 0.15921125411987305,
|
|
"num_tokens": 3195905.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 5.9371254444122314,
|
|
"epoch": 1.4476987447698746,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000495096622487287,
|
|
"loss": 5.7281,
|
|
"mean_token_accuracy": 0.15743090063333512,
|
|
"num_tokens": 3205132.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 5.935694646835327,
|
|
"epoch": 1.4518828451882846,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000495029379729071,
|
|
"loss": 5.7187,
|
|
"mean_token_accuracy": 0.15283734053373338,
|
|
"num_tokens": 3214362.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 5.887416410446167,
|
|
"epoch": 1.4560669456066946,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004949616841857666,
|
|
"loss": 5.7136,
|
|
"mean_token_accuracy": 0.15853825807571412,
|
|
"num_tokens": 3224248.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 5.913235187530518,
|
|
"epoch": 1.4602510460251046,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004948935359966806,
|
|
"loss": 5.5927,
|
|
"mean_token_accuracy": 0.16305283904075624,
|
|
"num_tokens": 3233416.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 5.725965213775635,
|
|
"epoch": 1.4644351464435146,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004948249353020506,
|
|
"loss": 5.5373,
|
|
"mean_token_accuracy": 0.1675902485847473,
|
|
"num_tokens": 3242354.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.873963499069214,
|
|
"epoch": 1.4686192468619246,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004947558822430461,
|
|
"loss": 5.6023,
|
|
"mean_token_accuracy": 0.1656225636601448,
|
|
"num_tokens": 3251026.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 5.83093957901001,
|
|
"epoch": 1.4728033472803348,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004946863769617672,
|
|
"loss": 5.5668,
|
|
"mean_token_accuracy": 0.1644663542509079,
|
|
"num_tokens": 3259429.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.83068356513977,
|
|
"epoch": 1.4769874476987448,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004946164196012445,
|
|
"loss": 5.6217,
|
|
"mean_token_accuracy": 0.16164354085922242,
|
|
"num_tokens": 3269305.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.877197027206421,
|
|
"epoch": 1.4811715481171548,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004945460103054389,
|
|
"loss": 5.6843,
|
|
"mean_token_accuracy": 0.15862189531326293,
|
|
"num_tokens": 3278846.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 5.899416828155518,
|
|
"epoch": 1.4853556485355648,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004944751492192416,
|
|
"loss": 5.611,
|
|
"mean_token_accuracy": 0.16517737805843352,
|
|
"num_tokens": 3287200.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.9020170211792,
|
|
"epoch": 1.489539748953975,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004944038364884734,
|
|
"loss": 5.6415,
|
|
"mean_token_accuracy": 0.15735085159540177,
|
|
"num_tokens": 3296651.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.842372989654541,
|
|
"epoch": 1.493723849372385,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004943320722598842,
|
|
"loss": 5.524,
|
|
"mean_token_accuracy": 0.17010669261217118,
|
|
"num_tokens": 3304887.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.83018159866333,
|
|
"epoch": 1.497907949790795,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004942598566811532,
|
|
"loss": 5.543,
|
|
"mean_token_accuracy": 0.1716176301240921,
|
|
"num_tokens": 3313925.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 5.847691488265991,
|
|
"epoch": 1.502092050209205,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004941871899008887,
|
|
"loss": 5.5298,
|
|
"mean_token_accuracy": 0.16718751192092896,
|
|
"num_tokens": 3322943.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 5.730998134613037,
|
|
"epoch": 1.506276150627615,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000494114072068627,
|
|
"loss": 5.5467,
|
|
"mean_token_accuracy": 0.16466595083475113,
|
|
"num_tokens": 3332742.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.813080930709839,
|
|
"epoch": 1.510460251046025,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004940405033348332,
|
|
"loss": 5.6352,
|
|
"mean_token_accuracy": 0.16255978345870972,
|
|
"num_tokens": 3342228.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.814670515060425,
|
|
"epoch": 1.514644351464435,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004939664838508996,
|
|
"loss": 5.6105,
|
|
"mean_token_accuracy": 0.1650616854429245,
|
|
"num_tokens": 3351854.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.786225271224976,
|
|
"epoch": 1.5188284518828452,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004938920137691465,
|
|
"loss": 5.586,
|
|
"mean_token_accuracy": 0.16562919616699218,
|
|
"num_tokens": 3360809.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.833091688156128,
|
|
"epoch": 1.5230125523012552,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004938170932428214,
|
|
"loss": 5.6233,
|
|
"mean_token_accuracy": 0.1586744472384453,
|
|
"num_tokens": 3371033.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.90294919013977,
|
|
"epoch": 1.5271966527196654,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004937417224260984,
|
|
"loss": 5.4947,
|
|
"mean_token_accuracy": 0.17257912904024125,
|
|
"num_tokens": 3379943.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.752410268783569,
|
|
"epoch": 1.5313807531380754,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004936659014740791,
|
|
"loss": 5.6229,
|
|
"mean_token_accuracy": 0.16704918593168258,
|
|
"num_tokens": 3389316.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.85987696647644,
|
|
"epoch": 1.5355648535564854,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004935896305427904,
|
|
"loss": 5.6551,
|
|
"mean_token_accuracy": 0.16151361167430878,
|
|
"num_tokens": 3398914.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.975553369522094,
|
|
"epoch": 1.5397489539748954,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004935129097891857,
|
|
"loss": 5.6227,
|
|
"mean_token_accuracy": 0.16726224571466447,
|
|
"num_tokens": 3407967.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.9012768268585205,
|
|
"epoch": 1.5439330543933054,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000493435739371144,
|
|
"loss": 5.6502,
|
|
"mean_token_accuracy": 0.1688822716474533,
|
|
"num_tokens": 3416530.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 5.824185705184936,
|
|
"epoch": 1.5481171548117154,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004933581194474695,
|
|
"loss": 5.5298,
|
|
"mean_token_accuracy": 0.1656462609767914,
|
|
"num_tokens": 3425762.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.774158477783203,
|
|
"epoch": 1.5523012552301254,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004932800501778918,
|
|
"loss": 5.6079,
|
|
"mean_token_accuracy": 0.1647947371006012,
|
|
"num_tokens": 3434799.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.902548933029175,
|
|
"epoch": 1.5564853556485354,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004932015317230646,
|
|
"loss": 5.6221,
|
|
"mean_token_accuracy": 0.1627484917640686,
|
|
"num_tokens": 3442986.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.7775719165802,
|
|
"epoch": 1.5606694560669456,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004931225642445665,
|
|
"loss": 5.6517,
|
|
"mean_token_accuracy": 0.16315302401781082,
|
|
"num_tokens": 3452914.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.883452749252319,
|
|
"epoch": 1.5648535564853556,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004930431479048998,
|
|
"loss": 5.5036,
|
|
"mean_token_accuracy": 0.16831693202257156,
|
|
"num_tokens": 3462093.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.7762549877166744,
|
|
"epoch": 1.5690376569037658,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004929632828674904,
|
|
"loss": 5.6812,
|
|
"mean_token_accuracy": 0.15995465219020844,
|
|
"num_tokens": 3471259.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.880064201354981,
|
|
"epoch": 1.5732217573221758,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004928829692966879,
|
|
"loss": 5.6262,
|
|
"mean_token_accuracy": 0.17060019671916962,
|
|
"num_tokens": 3480071.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.924511051177978,
|
|
"epoch": 1.5774058577405858,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004928022073577646,
|
|
"loss": 5.6393,
|
|
"mean_token_accuracy": 0.17406027764081955,
|
|
"num_tokens": 3489075.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.866768455505371,
|
|
"epoch": 1.5815899581589958,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004927209972169157,
|
|
"loss": 5.475,
|
|
"mean_token_accuracy": 0.17666401118040084,
|
|
"num_tokens": 3498495.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.721832513809204,
|
|
"epoch": 1.5857740585774058,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004926393390412586,
|
|
"loss": 5.6119,
|
|
"mean_token_accuracy": 0.16383255571126937,
|
|
"num_tokens": 3508590.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.934581184387207,
|
|
"epoch": 1.5899581589958158,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004925572329988325,
|
|
"loss": 5.6307,
|
|
"mean_token_accuracy": 0.16441386789083481,
|
|
"num_tokens": 3518397.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.789706468582153,
|
|
"epoch": 1.5941422594142258,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004924746792585987,
|
|
"loss": 5.608,
|
|
"mean_token_accuracy": 0.1638333410024643,
|
|
"num_tokens": 3526620.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.928402853012085,
|
|
"epoch": 1.5983263598326358,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004923916779904394,
|
|
"loss": 5.6147,
|
|
"mean_token_accuracy": 0.16687250435352324,
|
|
"num_tokens": 3535534.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.735380840301514,
|
|
"epoch": 1.602510460251046,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004923082293651578,
|
|
"loss": 5.5745,
|
|
"mean_token_accuracy": 0.16308026015758514,
|
|
"num_tokens": 3544771.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.762562894821167,
|
|
"epoch": 1.606694560669456,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000492224333554478,
|
|
"loss": 5.5449,
|
|
"mean_token_accuracy": 0.1586107388138771,
|
|
"num_tokens": 3553970.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.760586595535278,
|
|
"epoch": 1.6108786610878663,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004921399907310437,
|
|
"loss": 5.4547,
|
|
"mean_token_accuracy": 0.16664935201406478,
|
|
"num_tokens": 3563116.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.777907133102417,
|
|
"epoch": 1.6150627615062763,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004920552010684191,
|
|
"loss": 5.6021,
|
|
"mean_token_accuracy": 0.16663048416376114,
|
|
"num_tokens": 3571605.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.784053897857666,
|
|
"epoch": 1.6192468619246863,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004919699647410878,
|
|
"loss": 5.6637,
|
|
"mean_token_accuracy": 0.15977739542722702,
|
|
"num_tokens": 3581837.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.804090404510498,
|
|
"epoch": 1.6234309623430963,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004918842819244523,
|
|
"loss": 5.6153,
|
|
"mean_token_accuracy": 0.1600283220410347,
|
|
"num_tokens": 3590853.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.812429189682007,
|
|
"epoch": 1.6276150627615062,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004917981527948342,
|
|
"loss": 5.5107,
|
|
"mean_token_accuracy": 0.16755159795284272,
|
|
"num_tokens": 3599708.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.736073541641235,
|
|
"epoch": 1.6317991631799162,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004917115775294731,
|
|
"loss": 5.538,
|
|
"mean_token_accuracy": 0.16955381333827974,
|
|
"num_tokens": 3608634.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.845906925201416,
|
|
"epoch": 1.6359832635983262,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004916245563065273,
|
|
"loss": 5.6043,
|
|
"mean_token_accuracy": 0.16488172560930253,
|
|
"num_tokens": 3618776.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.776925754547119,
|
|
"epoch": 1.6401673640167362,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004915370893050722,
|
|
"loss": 5.6009,
|
|
"mean_token_accuracy": 0.1616799533367157,
|
|
"num_tokens": 3628595.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.918238115310669,
|
|
"epoch": 1.6443514644351465,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004914491767051012,
|
|
"loss": 5.7267,
|
|
"mean_token_accuracy": 0.15665314570069314,
|
|
"num_tokens": 3638133.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.785878467559814,
|
|
"epoch": 1.6485355648535565,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004913608186875241,
|
|
"loss": 5.5189,
|
|
"mean_token_accuracy": 0.17225270271301268,
|
|
"num_tokens": 3646679.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.835364151000976,
|
|
"epoch": 1.6527196652719667,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004912720154341674,
|
|
"loss": 5.6365,
|
|
"mean_token_accuracy": 0.16081463247537614,
|
|
"num_tokens": 3656131.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.834744548797607,
|
|
"epoch": 1.6569037656903767,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004911827671277739,
|
|
"loss": 5.5349,
|
|
"mean_token_accuracy": 0.16812530755996705,
|
|
"num_tokens": 3665187.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.789411354064941,
|
|
"epoch": 1.6610878661087867,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004910930739520025,
|
|
"loss": 5.6057,
|
|
"mean_token_accuracy": 0.16710528880357742,
|
|
"num_tokens": 3674649.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.87918701171875,
|
|
"epoch": 1.6652719665271967,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004910029360914271,
|
|
"loss": 5.6179,
|
|
"mean_token_accuracy": 0.16003223210573198,
|
|
"num_tokens": 3684680.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.727990341186524,
|
|
"epoch": 1.6694560669456067,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000490912353731537,
|
|
"loss": 5.562,
|
|
"mean_token_accuracy": 0.16933501809835433,
|
|
"num_tokens": 3693560.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.739227724075318,
|
|
"epoch": 1.6736401673640167,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004908213270587363,
|
|
"loss": 5.554,
|
|
"mean_token_accuracy": 0.16775781214237212,
|
|
"num_tokens": 3702632.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.6736401673640167,
|
|
"eval_entropy": 5.588579570832672,
|
|
"eval_loss": 5.659614562988281,
|
|
"eval_mean_token_accuracy": 0.16905502562081642,
|
|
"eval_num_tokens": 3702632.0,
|
|
"eval_runtime": 2.2974,
|
|
"eval_samples_per_second": 1700.65,
|
|
"eval_steps_per_second": 212.853,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.733205652236938,
|
|
"epoch": 1.6778242677824267,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004907298562603431,
|
|
"loss": 5.4419,
|
|
"mean_token_accuracy": 0.179792420566082,
|
|
"num_tokens": 3711022.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.765319538116455,
|
|
"epoch": 1.6820083682008367,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004906379415245896,
|
|
"loss": 5.5635,
|
|
"mean_token_accuracy": 0.17425089180469513,
|
|
"num_tokens": 3719778.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.72466812133789,
|
|
"epoch": 1.6861924686192469,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004905455830406215,
|
|
"loss": 5.541,
|
|
"mean_token_accuracy": 0.16978248208761215,
|
|
"num_tokens": 3728773.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.749242210388184,
|
|
"epoch": 1.6903765690376569,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004904527809984978,
|
|
"loss": 5.5758,
|
|
"mean_token_accuracy": 0.17408233433961867,
|
|
"num_tokens": 3737781.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.907482528686524,
|
|
"epoch": 1.694560669456067,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004903595355891902,
|
|
"loss": 5.6325,
|
|
"mean_token_accuracy": 0.15966468602418898,
|
|
"num_tokens": 3748010.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.757623910903931,
|
|
"epoch": 1.698744769874477,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004902658470045826,
|
|
"loss": 5.6104,
|
|
"mean_token_accuracy": 0.16668835431337356,
|
|
"num_tokens": 3757894.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.913280773162842,
|
|
"epoch": 1.702928870292887,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004901717154374713,
|
|
"loss": 5.7001,
|
|
"mean_token_accuracy": 0.1627259835600853,
|
|
"num_tokens": 3768760.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.737261056900024,
|
|
"epoch": 1.707112970711297,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004900771410815638,
|
|
"loss": 5.5313,
|
|
"mean_token_accuracy": 0.17250364124774933,
|
|
"num_tokens": 3778273.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.72805814743042,
|
|
"epoch": 1.711297071129707,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004899821241314788,
|
|
"loss": 5.5411,
|
|
"mean_token_accuracy": 0.1667292982339859,
|
|
"num_tokens": 3787012.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.802943515777588,
|
|
"epoch": 1.715481171548117,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000489886664782746,
|
|
"loss": 5.495,
|
|
"mean_token_accuracy": 0.17157293409109114,
|
|
"num_tokens": 3795144.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.84842586517334,
|
|
"epoch": 1.719665271966527,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004897907632318055,
|
|
"loss": 5.7066,
|
|
"mean_token_accuracy": 0.16221378445625306,
|
|
"num_tokens": 3804534.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.764593982696534,
|
|
"epoch": 1.723849372384937,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004896944196760071,
|
|
"loss": 5.4946,
|
|
"mean_token_accuracy": 0.17056726664304733,
|
|
"num_tokens": 3813757.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.6439008712768555,
|
|
"epoch": 1.7280334728033473,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004895976343136106,
|
|
"loss": 5.5233,
|
|
"mean_token_accuracy": 0.1781034454703331,
|
|
"num_tokens": 3823157.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.68931622505188,
|
|
"epoch": 1.7322175732217573,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004895004073437844,
|
|
"loss": 5.4059,
|
|
"mean_token_accuracy": 0.1745637461543083,
|
|
"num_tokens": 3831661.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.6525579452514645,
|
|
"epoch": 1.7364016736401675,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004894027389666061,
|
|
"loss": 5.4101,
|
|
"mean_token_accuracy": 0.17606336176395415,
|
|
"num_tokens": 3840591.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.792623472213745,
|
|
"epoch": 1.7405857740585775,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004893046293830616,
|
|
"loss": 5.5278,
|
|
"mean_token_accuracy": 0.16255860030651093,
|
|
"num_tokens": 3850094.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.742485284805298,
|
|
"epoch": 1.7447698744769875,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004892060787950447,
|
|
"loss": 5.6075,
|
|
"mean_token_accuracy": 0.16815222799777985,
|
|
"num_tokens": 3858973.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.63767294883728,
|
|
"epoch": 1.7489539748953975,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004891070874053564,
|
|
"loss": 5.3706,
|
|
"mean_token_accuracy": 0.1753972977399826,
|
|
"num_tokens": 3868315.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.736858320236206,
|
|
"epoch": 1.7531380753138075,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004890076554177051,
|
|
"loss": 5.5362,
|
|
"mean_token_accuracy": 0.16723127663135529,
|
|
"num_tokens": 3877605.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.74579815864563,
|
|
"epoch": 1.7573221757322175,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004889077830367061,
|
|
"loss": 5.5509,
|
|
"mean_token_accuracy": 0.171517913043499,
|
|
"num_tokens": 3886998.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.767471218109131,
|
|
"epoch": 1.7615062761506275,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004888074704678806,
|
|
"loss": 5.5082,
|
|
"mean_token_accuracy": 0.17135567963123322,
|
|
"num_tokens": 3896018.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.735943555831909,
|
|
"epoch": 1.7656903765690377,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004887067179176558,
|
|
"loss": 5.4913,
|
|
"mean_token_accuracy": 0.17377638518810273,
|
|
"num_tokens": 3904957.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.741572475433349,
|
|
"epoch": 1.7698744769874477,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004886055255933639,
|
|
"loss": 5.6059,
|
|
"mean_token_accuracy": 0.16303271129727365,
|
|
"num_tokens": 3913427.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.74202823638916,
|
|
"epoch": 1.7740585774058577,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004885038937032429,
|
|
"loss": 5.5313,
|
|
"mean_token_accuracy": 0.16900549381971358,
|
|
"num_tokens": 3923134.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.753329849243164,
|
|
"epoch": 1.778242677824268,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004884018224564345,
|
|
"loss": 5.4325,
|
|
"mean_token_accuracy": 0.17378120124340057,
|
|
"num_tokens": 3931867.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.744208335876465,
|
|
"epoch": 1.782426778242678,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004882993120629852,
|
|
"loss": 5.5386,
|
|
"mean_token_accuracy": 0.1757911577820778,
|
|
"num_tokens": 3941057.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.839227104187012,
|
|
"epoch": 1.786610878661088,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004881963627338449,
|
|
"loss": 5.6526,
|
|
"mean_token_accuracy": 0.1612655907869339,
|
|
"num_tokens": 3950511.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.7714259147644045,
|
|
"epoch": 1.790794979079498,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048809297468086654,
|
|
"loss": 5.5708,
|
|
"mean_token_accuracy": 0.17301054149866105,
|
|
"num_tokens": 3959769.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.754560470581055,
|
|
"epoch": 1.794979079497908,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004879891481168063,
|
|
"loss": 5.4885,
|
|
"mean_token_accuracy": 0.17386019825935364,
|
|
"num_tokens": 3968462.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.694684267044067,
|
|
"epoch": 1.799163179916318,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00048788488325532235,
|
|
"loss": 5.525,
|
|
"mean_token_accuracy": 0.17304963022470474,
|
|
"num_tokens": 3977464.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.735943078994751,
|
|
"epoch": 1.803347280334728,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004877801803109753,
|
|
"loss": 5.449,
|
|
"mean_token_accuracy": 0.176297365128994,
|
|
"num_tokens": 3986303.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.791642427444458,
|
|
"epoch": 1.8075313807531381,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00048767503949922683,
|
|
"loss": 5.4987,
|
|
"mean_token_accuracy": 0.173785038292408,
|
|
"num_tokens": 3994970.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.815527820587159,
|
|
"epoch": 1.8117154811715481,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004875694610364398,
|
|
"loss": 5.5822,
|
|
"mean_token_accuracy": 0.1655818536877632,
|
|
"num_tokens": 4004679.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.808756685256958,
|
|
"epoch": 1.8158995815899581,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048746344513987775,
|
|
"loss": 5.5967,
|
|
"mean_token_accuracy": 0.1684105336666107,
|
|
"num_tokens": 4013516.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.781185150146484,
|
|
"epoch": 1.8200836820083683,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004873569920277043,
|
|
"loss": 5.466,
|
|
"mean_token_accuracy": 0.17886558920145035,
|
|
"num_tokens": 4022335.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.631121635437012,
|
|
"epoch": 1.8242677824267783,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048725010191898296,
|
|
"loss": 5.4482,
|
|
"mean_token_accuracy": 0.17270184606313704,
|
|
"num_tokens": 4031295.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.809648323059082,
|
|
"epoch": 1.8284518828451883,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00048714277503367627,
|
|
"loss": 5.6615,
|
|
"mean_token_accuracy": 0.1627314418554306,
|
|
"num_tokens": 4040378.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.838449716567993,
|
|
"epoch": 1.8326359832635983,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048703501159264576,
|
|
"loss": 5.5254,
|
|
"mean_token_accuracy": 0.16325923651456833,
|
|
"num_tokens": 4050217.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.614929103851319,
|
|
"epoch": 1.8368200836820083,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004869268118176512,
|
|
"loss": 5.3981,
|
|
"mean_token_accuracy": 0.16747482419013976,
|
|
"num_tokens": 4059088.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.792963457107544,
|
|
"epoch": 1.8410041841004183,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048681817593135046,
|
|
"loss": 5.5137,
|
|
"mean_token_accuracy": 0.1806218609213829,
|
|
"num_tokens": 4067609.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.725982236862182,
|
|
"epoch": 1.8451882845188283,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048670910415729855,
|
|
"loss": 5.4925,
|
|
"mean_token_accuracy": 0.17104198634624482,
|
|
"num_tokens": 4076763.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.796593093872071,
|
|
"epoch": 1.8493723849372385,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048659959671994786,
|
|
"loss": 5.6414,
|
|
"mean_token_accuracy": 0.16381513625383376,
|
|
"num_tokens": 4086552.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.615056848526001,
|
|
"epoch": 1.8535564853556485,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048648965384464693,
|
|
"loss": 5.4132,
|
|
"mean_token_accuracy": 0.17252946645021439,
|
|
"num_tokens": 4096200.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.649881219863891,
|
|
"epoch": 1.8577405857740585,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048637927575764057,
|
|
"loss": 5.4644,
|
|
"mean_token_accuracy": 0.17159941494464875,
|
|
"num_tokens": 4105080.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.693881797790527,
|
|
"epoch": 1.8619246861924688,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048626846268606913,
|
|
"loss": 5.3967,
|
|
"mean_token_accuracy": 0.1740979105234146,
|
|
"num_tokens": 4113972.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.787312078475952,
|
|
"epoch": 1.8661087866108788,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048615721485796803,
|
|
"loss": 5.6322,
|
|
"mean_token_accuracy": 0.16151763051748275,
|
|
"num_tokens": 4123615.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.778605556488037,
|
|
"epoch": 1.8702928870292888,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004860455325022676,
|
|
"loss": 5.4188,
|
|
"mean_token_accuracy": 0.17042351365089417,
|
|
"num_tokens": 4132128.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.806691884994507,
|
|
"epoch": 1.8744769874476988,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000485933415848792,
|
|
"loss": 5.6195,
|
|
"mean_token_accuracy": 0.16596710234880446,
|
|
"num_tokens": 4141575.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.74137282371521,
|
|
"epoch": 1.8786610878661087,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004858208651282593,
|
|
"loss": 5.4838,
|
|
"mean_token_accuracy": 0.16864698082208635,
|
|
"num_tokens": 4150372.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.717060470581055,
|
|
"epoch": 1.8828451882845187,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000485707880572281,
|
|
"loss": 5.5708,
|
|
"mean_token_accuracy": 0.17850400805473327,
|
|
"num_tokens": 4159007.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.713363838195801,
|
|
"epoch": 1.8870292887029287,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004855944624133608,
|
|
"loss": 5.606,
|
|
"mean_token_accuracy": 0.16776737868785857,
|
|
"num_tokens": 4168646.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.706373119354248,
|
|
"epoch": 1.891213389121339,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048548061088489523,
|
|
"loss": 5.5191,
|
|
"mean_token_accuracy": 0.16962461769580842,
|
|
"num_tokens": 4177979.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.747662687301636,
|
|
"epoch": 1.895397489539749,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048536632622117246,
|
|
"loss": 5.4532,
|
|
"mean_token_accuracy": 0.1754816398024559,
|
|
"num_tokens": 4186781.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.674567317962646,
|
|
"epoch": 1.899581589958159,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048525160865737186,
|
|
"loss": 5.5135,
|
|
"mean_token_accuracy": 0.17038754671812056,
|
|
"num_tokens": 4195661.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.711926651000977,
|
|
"epoch": 1.9037656903765692,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048513645842956384,
|
|
"loss": 5.4171,
|
|
"mean_token_accuracy": 0.17328426092863083,
|
|
"num_tokens": 4203902.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.65511646270752,
|
|
"epoch": 1.9079497907949792,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048502087577470895,
|
|
"loss": 5.4801,
|
|
"mean_token_accuracy": 0.1693297103047371,
|
|
"num_tokens": 4213663.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.7536944389343265,
|
|
"epoch": 1.9121338912133892,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048490486093065774,
|
|
"loss": 5.5155,
|
|
"mean_token_accuracy": 0.17362722307443618,
|
|
"num_tokens": 4222152.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.7442272186279295,
|
|
"epoch": 1.9163179916317992,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00048478841413615017,
|
|
"loss": 5.5351,
|
|
"mean_token_accuracy": 0.17005095928907393,
|
|
"num_tokens": 4232347.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.7341259002685545,
|
|
"epoch": 1.9205020920502092,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048467153563081487,
|
|
"loss": 5.5971,
|
|
"mean_token_accuracy": 0.17492114007472992,
|
|
"num_tokens": 4240607.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.781266546249389,
|
|
"epoch": 1.9246861924686192,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004845542256551692,
|
|
"loss": 5.59,
|
|
"mean_token_accuracy": 0.16862932443618775,
|
|
"num_tokens": 4251131.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.73166675567627,
|
|
"epoch": 1.9288702928870292,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004844364844506181,
|
|
"loss": 5.5218,
|
|
"mean_token_accuracy": 0.17353187799453734,
|
|
"num_tokens": 4259955.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.801835727691651,
|
|
"epoch": 1.9330543933054394,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004843183122594541,
|
|
"loss": 5.5077,
|
|
"mean_token_accuracy": 0.1740906909108162,
|
|
"num_tokens": 4268686.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.691273832321167,
|
|
"epoch": 1.9372384937238494,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004841997093248566,
|
|
"loss": 5.4908,
|
|
"mean_token_accuracy": 0.17442866414785385,
|
|
"num_tokens": 4277377.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.642513227462769,
|
|
"epoch": 1.9414225941422594,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004840806758908913,
|
|
"loss": 5.4783,
|
|
"mean_token_accuracy": 0.1721930593252182,
|
|
"num_tokens": 4287037.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.82609486579895,
|
|
"epoch": 1.9456066945606696,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00048396121220250983,
|
|
"loss": 5.6235,
|
|
"mean_token_accuracy": 0.15890175402164458,
|
|
"num_tokens": 4296923.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.645568037033081,
|
|
"epoch": 1.9497907949790796,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004838413185055494,
|
|
"loss": 5.3844,
|
|
"mean_token_accuracy": 0.1826627552509308,
|
|
"num_tokens": 4305873.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.69848518371582,
|
|
"epoch": 1.9539748953974896,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004837209950467318,
|
|
"loss": 5.5444,
|
|
"mean_token_accuracy": 0.16894280910491943,
|
|
"num_tokens": 4314912.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.754768705368042,
|
|
"epoch": 1.9581589958158996,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004836002420736635,
|
|
"loss": 5.5253,
|
|
"mean_token_accuracy": 0.16813971996307372,
|
|
"num_tokens": 4324606.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.638680171966553,
|
|
"epoch": 1.9623430962343096,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004834790598348347,
|
|
"loss": 5.403,
|
|
"mean_token_accuracy": 0.18076600432395934,
|
|
"num_tokens": 4333076.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.783349657058716,
|
|
"epoch": 1.9665271966527196,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004833574485796188,
|
|
"loss": 5.5358,
|
|
"mean_token_accuracy": 0.17332175523042678,
|
|
"num_tokens": 4341893.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.636692905426026,
|
|
"epoch": 1.9707112970711296,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048323540855827244,
|
|
"loss": 5.5143,
|
|
"mean_token_accuracy": 0.1677964821457863,
|
|
"num_tokens": 4352303.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.820708417892456,
|
|
"epoch": 1.9748953974895398,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004831129400219341,
|
|
"loss": 5.4715,
|
|
"mean_token_accuracy": 0.16791264712810516,
|
|
"num_tokens": 4361166.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.729510307312012,
|
|
"epoch": 1.9790794979079498,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048299004322262457,
|
|
"loss": 5.5136,
|
|
"mean_token_accuracy": 0.17077897489070892,
|
|
"num_tokens": 4369999.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.690209627151489,
|
|
"epoch": 1.9832635983263598,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048286671841324565,
|
|
"loss": 5.4774,
|
|
"mean_token_accuracy": 0.17686031758785248,
|
|
"num_tokens": 4379373.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.721845436096191,
|
|
"epoch": 1.98744769874477,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004827429658475799,
|
|
"loss": 5.4771,
|
|
"mean_token_accuracy": 0.1777857258915901,
|
|
"num_tokens": 4389035.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.727389669418335,
|
|
"epoch": 1.99163179916318,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048261878578029026,
|
|
"loss": 5.6408,
|
|
"mean_token_accuracy": 0.1699129104614258,
|
|
"num_tokens": 4399366.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.701954936981201,
|
|
"epoch": 1.99581589958159,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004824941784669193,
|
|
"loss": 5.4662,
|
|
"mean_token_accuracy": 0.17321588844060898,
|
|
"num_tokens": 4408183.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.705373239517212,
|
|
"epoch": 2.0,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004823691441638889,
|
|
"loss": 5.5908,
|
|
"mean_token_accuracy": 0.15743454173207283,
|
|
"num_tokens": 4418644.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.676972484588623,
|
|
"epoch": 2.00418410041841,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004822436831284995,
|
|
"loss": 5.2176,
|
|
"mean_token_accuracy": 0.18418388664722443,
|
|
"num_tokens": 4427828.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.63955602645874,
|
|
"epoch": 2.00836820083682,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004821177956189297,
|
|
"loss": 5.1468,
|
|
"mean_token_accuracy": 0.18794921636581421,
|
|
"num_tokens": 4436194.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.572412300109863,
|
|
"epoch": 2.01255230125523,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004819914818942359,
|
|
"loss": 5.2065,
|
|
"mean_token_accuracy": 0.19006218761205673,
|
|
"num_tokens": 4444468.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.723620510101318,
|
|
"epoch": 2.01673640167364,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048186474221435127,
|
|
"loss": 5.1262,
|
|
"mean_token_accuracy": 0.1960455819964409,
|
|
"num_tokens": 4452760.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.634727859497071,
|
|
"epoch": 2.02092050209205,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048173757684008583,
|
|
"loss": 5.1789,
|
|
"mean_token_accuracy": 0.19306624829769134,
|
|
"num_tokens": 4462154.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.666573143005371,
|
|
"epoch": 2.0251046025104604,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004816099860331255,
|
|
"loss": 5.2761,
|
|
"mean_token_accuracy": 0.18292328268289565,
|
|
"num_tokens": 4471987.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.602439880371094,
|
|
"epoch": 2.0292887029288704,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048148197005603145,
|
|
"loss": 5.1894,
|
|
"mean_token_accuracy": 0.18424307256937028,
|
|
"num_tokens": 4481645.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.563078451156616,
|
|
"epoch": 2.0334728033472804,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048135352917224005,
|
|
"loss": 5.1851,
|
|
"mean_token_accuracy": 0.19173673838376998,
|
|
"num_tokens": 4491166.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.69947361946106,
|
|
"epoch": 2.0376569037656904,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00048122466364606214,
|
|
"loss": 5.1809,
|
|
"mean_token_accuracy": 0.18710510134696962,
|
|
"num_tokens": 4500159.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.537801504135132,
|
|
"epoch": 2.0418410041841004,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048109537374268216,
|
|
"loss": 5.1395,
|
|
"mean_token_accuracy": 0.1943747103214264,
|
|
"num_tokens": 4509737.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.414461278915406,
|
|
"epoch": 2.0460251046025104,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00048096565972815793,
|
|
"loss": 5.068,
|
|
"mean_token_accuracy": 0.19497113823890685,
|
|
"num_tokens": 4518379.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.585501289367675,
|
|
"epoch": 2.0502092050209204,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004808355218694199,
|
|
"loss": 5.2011,
|
|
"mean_token_accuracy": 0.19233749955892562,
|
|
"num_tokens": 4527885.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.642935276031494,
|
|
"epoch": 2.0543933054393304,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048070496043427116,
|
|
"loss": 5.2326,
|
|
"mean_token_accuracy": 0.18954924792051314,
|
|
"num_tokens": 4537894.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.6098230361938475,
|
|
"epoch": 2.0585774058577404,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00048057397569138596,
|
|
"loss": 5.1323,
|
|
"mean_token_accuracy": 0.19356006979942322,
|
|
"num_tokens": 4546462.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.581704616546631,
|
|
"epoch": 2.062761506276151,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004804425679103099,
|
|
"loss": 5.1662,
|
|
"mean_token_accuracy": 0.18637972325086594,
|
|
"num_tokens": 4555970.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.614779233932495,
|
|
"epoch": 2.066945606694561,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004803107373614589,
|
|
"loss": 5.2263,
|
|
"mean_token_accuracy": 0.18251074105501175,
|
|
"num_tokens": 4565799.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.591456508636474,
|
|
"epoch": 2.071129707112971,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004801784843161194,
|
|
"loss": 5.2588,
|
|
"mean_token_accuracy": 0.18001774698495865,
|
|
"num_tokens": 4575571.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.648578453063965,
|
|
"epoch": 2.075313807531381,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004800458090464465,
|
|
"loss": 5.1558,
|
|
"mean_token_accuracy": 0.19157422780990602,
|
|
"num_tokens": 4584263.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.549553346633911,
|
|
"epoch": 2.079497907949791,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047991271182546476,
|
|
"loss": 5.1255,
|
|
"mean_token_accuracy": 0.19961231499910354,
|
|
"num_tokens": 4592979.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.545702266693115,
|
|
"epoch": 2.083682008368201,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047977919292706696,
|
|
"loss": 5.1558,
|
|
"mean_token_accuracy": 0.1967400923371315,
|
|
"num_tokens": 4602042.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.575488901138305,
|
|
"epoch": 2.087866108786611,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004796452526260134,
|
|
"loss": 5.2003,
|
|
"mean_token_accuracy": 0.19144488722085953,
|
|
"num_tokens": 4611197.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.467491197586059,
|
|
"epoch": 2.092050209205021,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047951089119793164,
|
|
"loss": 5.1512,
|
|
"mean_token_accuracy": 0.18873886466026307,
|
|
"num_tokens": 4621612.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.092050209205021,
|
|
"eval_entropy": 5.31002169103954,
|
|
"eval_loss": 5.553900241851807,
|
|
"eval_mean_token_accuracy": 0.1790137020813907,
|
|
"eval_num_tokens": 4621612.0,
|
|
"eval_runtime": 2.0824,
|
|
"eval_samples_per_second": 1876.19,
|
|
"eval_steps_per_second": 234.824,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.466035270690918,
|
|
"epoch": 2.096234309623431,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000479376108919316,
|
|
"loss": 5.0416,
|
|
"mean_token_accuracy": 0.2022264838218689,
|
|
"num_tokens": 4630198.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.5752606868743895,
|
|
"epoch": 2.100418410041841,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004792409060675267,
|
|
"loss": 5.1309,
|
|
"mean_token_accuracy": 0.18947514593601228,
|
|
"num_tokens": 4638729.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.543802404403687,
|
|
"epoch": 2.104602510460251,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004791052829207895,
|
|
"loss": 5.281,
|
|
"mean_token_accuracy": 0.18752007484436034,
|
|
"num_tokens": 4648759.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.539657258987427,
|
|
"epoch": 2.1087866108786613,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000478969239758195,
|
|
"loss": 5.2249,
|
|
"mean_token_accuracy": 0.18912642896175386,
|
|
"num_tokens": 4657744.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.583879613876343,
|
|
"epoch": 2.1129707112970713,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047883277685969815,
|
|
"loss": 5.1756,
|
|
"mean_token_accuracy": 0.19520145207643508,
|
|
"num_tokens": 4666154.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.605193853378296,
|
|
"epoch": 2.1171548117154813,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047869589450611773,
|
|
"loss": 5.1974,
|
|
"mean_token_accuracy": 0.18547933399677277,
|
|
"num_tokens": 4675677.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.5078535079956055,
|
|
"epoch": 2.1213389121338913,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004785585929791355,
|
|
"loss": 5.2027,
|
|
"mean_token_accuracy": 0.18887757360935212,
|
|
"num_tokens": 4684733.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.568530702590943,
|
|
"epoch": 2.1255230125523012,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000478420872561296,
|
|
"loss": 5.1059,
|
|
"mean_token_accuracy": 0.19232923835515975,
|
|
"num_tokens": 4694155.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.501318836212159,
|
|
"epoch": 2.1297071129707112,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047828273353600577,
|
|
"loss": 5.1311,
|
|
"mean_token_accuracy": 0.1967197060585022,
|
|
"num_tokens": 4703286.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.541385889053345,
|
|
"epoch": 2.1338912133891212,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047814417618753256,
|
|
"loss": 5.1594,
|
|
"mean_token_accuracy": 0.18664800822734834,
|
|
"num_tokens": 4711666.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.5342179298400875,
|
|
"epoch": 2.1380753138075312,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00047800520080100526,
|
|
"loss": 5.1789,
|
|
"mean_token_accuracy": 0.18600201457738877,
|
|
"num_tokens": 4721644.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.53387713432312,
|
|
"epoch": 2.1422594142259412,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047786580766241265,
|
|
"loss": 5.1499,
|
|
"mean_token_accuracy": 0.19156892448663712,
|
|
"num_tokens": 4730885.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.480189800262451,
|
|
"epoch": 2.1464435146443517,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004777259970586036,
|
|
"loss": 5.1329,
|
|
"mean_token_accuracy": 0.1930587887763977,
|
|
"num_tokens": 4740871.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.532391023635864,
|
|
"epoch": 2.1506276150627617,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004775857692772859,
|
|
"loss": 5.117,
|
|
"mean_token_accuracy": 0.1901528760790825,
|
|
"num_tokens": 4750603.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.529564762115479,
|
|
"epoch": 2.1548117154811717,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004774451246070257,
|
|
"loss": 5.2211,
|
|
"mean_token_accuracy": 0.1945394530892372,
|
|
"num_tokens": 4760681.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.55622820854187,
|
|
"epoch": 2.1589958158995817,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004773040633372471,
|
|
"loss": 5.2048,
|
|
"mean_token_accuracy": 0.19357844293117524,
|
|
"num_tokens": 4770237.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.567328310012817,
|
|
"epoch": 2.1631799163179917,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00047716258575823147,
|
|
"loss": 5.2308,
|
|
"mean_token_accuracy": 0.19111842960119246,
|
|
"num_tokens": 4779022.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.59231653213501,
|
|
"epoch": 2.1673640167364017,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000477020692161117,
|
|
"loss": 5.2432,
|
|
"mean_token_accuracy": 0.1882817640900612,
|
|
"num_tokens": 4788644.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.468843793869018,
|
|
"epoch": 2.1715481171548117,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004768783828378978,
|
|
"loss": 5.248,
|
|
"mean_token_accuracy": 0.18198808878660203,
|
|
"num_tokens": 4798628.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.709801912307739,
|
|
"epoch": 2.1757322175732217,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004767356580814237,
|
|
"loss": 5.3477,
|
|
"mean_token_accuracy": 0.18395230323076248,
|
|
"num_tokens": 4807727.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.627778053283691,
|
|
"epoch": 2.1799163179916317,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004765925181853993,
|
|
"loss": 5.2205,
|
|
"mean_token_accuracy": 0.18238564878702163,
|
|
"num_tokens": 4817896.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.565152502059936,
|
|
"epoch": 2.1841004184100417,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004764489634443834,
|
|
"loss": 5.1815,
|
|
"mean_token_accuracy": 0.18810514807701112,
|
|
"num_tokens": 4827135.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.571887063980102,
|
|
"epoch": 2.1882845188284517,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047630499415378863,
|
|
"loss": 5.2606,
|
|
"mean_token_accuracy": 0.185622002184391,
|
|
"num_tokens": 4837554.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.454201078414917,
|
|
"epoch": 2.192468619246862,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004761606106098806,
|
|
"loss": 5.1374,
|
|
"mean_token_accuracy": 0.19262371063232422,
|
|
"num_tokens": 4846739.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.549982786178589,
|
|
"epoch": 2.196652719665272,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047601581310977747,
|
|
"loss": 5.2102,
|
|
"mean_token_accuracy": 0.1892186775803566,
|
|
"num_tokens": 4855707.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.534137153625489,
|
|
"epoch": 2.200836820083682,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004758706019514493,
|
|
"loss": 5.1952,
|
|
"mean_token_accuracy": 0.19004165083169938,
|
|
"num_tokens": 4865098.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.605195140838623,
|
|
"epoch": 2.205020920502092,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004757249774337172,
|
|
"loss": 5.2084,
|
|
"mean_token_accuracy": 0.19359517991542816,
|
|
"num_tokens": 4874426.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.523797559738159,
|
|
"epoch": 2.209205020920502,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047557893985625293,
|
|
"loss": 5.1694,
|
|
"mean_token_accuracy": 0.1928726240992546,
|
|
"num_tokens": 4883240.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.478337240219116,
|
|
"epoch": 2.213389121338912,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004754324895195786,
|
|
"loss": 5.1591,
|
|
"mean_token_accuracy": 0.1926814079284668,
|
|
"num_tokens": 4891989.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.5565934658050535,
|
|
"epoch": 2.217573221757322,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004752856267250652,
|
|
"loss": 5.2583,
|
|
"mean_token_accuracy": 0.18897372037172316,
|
|
"num_tokens": 4901546.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.444426965713501,
|
|
"epoch": 2.221757322175732,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00047513835177493286,
|
|
"loss": 5.0804,
|
|
"mean_token_accuracy": 0.19561816602945328,
|
|
"num_tokens": 4910222.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.5684233665466305,
|
|
"epoch": 2.225941422594142,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004749906649722498,
|
|
"loss": 5.3037,
|
|
"mean_token_accuracy": 0.1823274552822113,
|
|
"num_tokens": 4920184.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.530097723007202,
|
|
"epoch": 2.2301255230125525,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004748425666209316,
|
|
"loss": 5.1993,
|
|
"mean_token_accuracy": 0.1915793240070343,
|
|
"num_tokens": 4930014.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.52888674736023,
|
|
"epoch": 2.2343096234309625,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004746940570257409,
|
|
"loss": 5.1327,
|
|
"mean_token_accuracy": 0.19633713066577912,
|
|
"num_tokens": 4939556.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.456666135787964,
|
|
"epoch": 2.2384937238493725,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004745451364922867,
|
|
"loss": 5.1854,
|
|
"mean_token_accuracy": 0.18619713932275772,
|
|
"num_tokens": 4949066.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.60462965965271,
|
|
"epoch": 2.2426778242677825,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004743958053270233,
|
|
"loss": 5.1857,
|
|
"mean_token_accuracy": 0.19684374779462815,
|
|
"num_tokens": 4957881.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.516653776168823,
|
|
"epoch": 2.2468619246861925,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047424606383725037,
|
|
"loss": 5.2274,
|
|
"mean_token_accuracy": 0.1889663055539131,
|
|
"num_tokens": 4967105.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.500734186172485,
|
|
"epoch": 2.2510460251046025,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004740959123311118,
|
|
"loss": 5.1201,
|
|
"mean_token_accuracy": 0.19694772958755494,
|
|
"num_tokens": 4977676.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.448371934890747,
|
|
"epoch": 2.2552301255230125,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047394535111759526,
|
|
"loss": 5.1571,
|
|
"mean_token_accuracy": 0.19497282058000565,
|
|
"num_tokens": 4986163.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.516038846969605,
|
|
"epoch": 2.2594142259414225,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004737943805065314,
|
|
"loss": 5.1569,
|
|
"mean_token_accuracy": 0.18557292073965073,
|
|
"num_tokens": 4995179.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.57070837020874,
|
|
"epoch": 2.2635983263598325,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047364300080859363,
|
|
"loss": 5.2456,
|
|
"mean_token_accuracy": 0.18458279073238373,
|
|
"num_tokens": 5003856.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.54024863243103,
|
|
"epoch": 2.2677824267782425,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047349121233529695,
|
|
"loss": 5.2841,
|
|
"mean_token_accuracy": 0.18307160437107087,
|
|
"num_tokens": 5013506.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.517574882507324,
|
|
"epoch": 2.2719665271966525,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004733390153989976,
|
|
"loss": 5.2075,
|
|
"mean_token_accuracy": 0.19041822850704193,
|
|
"num_tokens": 5022159.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.544815492630005,
|
|
"epoch": 2.276150627615063,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047318641031289245,
|
|
"loss": 5.1628,
|
|
"mean_token_accuracy": 0.1895567938685417,
|
|
"num_tokens": 5031869.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.4250811576843265,
|
|
"epoch": 2.280334728033473,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004730333973910183,
|
|
"loss": 5.031,
|
|
"mean_token_accuracy": 0.20200449526309966,
|
|
"num_tokens": 5039904.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.529689121246338,
|
|
"epoch": 2.284518828451883,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004728799769482511,
|
|
"loss": 5.3029,
|
|
"mean_token_accuracy": 0.18466683924198152,
|
|
"num_tokens": 5049144.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.519205617904663,
|
|
"epoch": 2.288702928870293,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047272614930030533,
|
|
"loss": 5.1013,
|
|
"mean_token_accuracy": 0.19491563737392426,
|
|
"num_tokens": 5058302.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.497249746322632,
|
|
"epoch": 2.292887029288703,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047257191476373375,
|
|
"loss": 5.1137,
|
|
"mean_token_accuracy": 0.19153977185487747,
|
|
"num_tokens": 5067443.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.462966060638427,
|
|
"epoch": 2.297071129707113,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047241727365592603,
|
|
"loss": 5.2445,
|
|
"mean_token_accuracy": 0.1829087108373642,
|
|
"num_tokens": 5076415.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.544723129272461,
|
|
"epoch": 2.301255230125523,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004722622262951089,
|
|
"loss": 5.2121,
|
|
"mean_token_accuracy": 0.18837485909461976,
|
|
"num_tokens": 5085882.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.561155891418457,
|
|
"epoch": 2.305439330543933,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004721067730003448,
|
|
"loss": 5.1795,
|
|
"mean_token_accuracy": 0.1986195296049118,
|
|
"num_tokens": 5094685.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.522579193115234,
|
|
"epoch": 2.309623430962343,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004719509140915317,
|
|
"loss": 5.1946,
|
|
"mean_token_accuracy": 0.1885368198156357,
|
|
"num_tokens": 5103885.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.523617172241211,
|
|
"epoch": 2.3138075313807533,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047179464988940205,
|
|
"loss": 5.1232,
|
|
"mean_token_accuracy": 0.20171012580394745,
|
|
"num_tokens": 5112164.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.397410917282104,
|
|
"epoch": 2.3179916317991633,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047163798071552253,
|
|
"loss": 5.0884,
|
|
"mean_token_accuracy": 0.1930816426873207,
|
|
"num_tokens": 5121913.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.499872779846191,
|
|
"epoch": 2.3221757322175733,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004714809068922931,
|
|
"loss": 5.0993,
|
|
"mean_token_accuracy": 0.20085703879594802,
|
|
"num_tokens": 5130447.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.530040597915649,
|
|
"epoch": 2.3263598326359833,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004713234287429465,
|
|
"loss": 5.2262,
|
|
"mean_token_accuracy": 0.18991127461194993,
|
|
"num_tokens": 5139863.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.390472364425659,
|
|
"epoch": 2.3305439330543933,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004711655465915473,
|
|
"loss": 5.1751,
|
|
"mean_token_accuracy": 0.1947562500834465,
|
|
"num_tokens": 5149945.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.560927772521973,
|
|
"epoch": 2.3347280334728033,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004710072607629917,
|
|
"loss": 5.1167,
|
|
"mean_token_accuracy": 0.18760036081075668,
|
|
"num_tokens": 5160293.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.460022878646851,
|
|
"epoch": 2.3389121338912133,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047084857158300646,
|
|
"loss": 5.1393,
|
|
"mean_token_accuracy": 0.1829672560095787,
|
|
"num_tokens": 5169896.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.503097057342529,
|
|
"epoch": 2.3430962343096233,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00047068947937814844,
|
|
"loss": 5.1958,
|
|
"mean_token_accuracy": 0.18684695065021514,
|
|
"num_tokens": 5178898.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.561424970626831,
|
|
"epoch": 2.3472803347280333,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047052998447580374,
|
|
"loss": 5.2588,
|
|
"mean_token_accuracy": 0.18708261400461196,
|
|
"num_tokens": 5188733.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.4390233039855955,
|
|
"epoch": 2.3514644351464433,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047037008720418724,
|
|
"loss": 5.1555,
|
|
"mean_token_accuracy": 0.19035712778568267,
|
|
"num_tokens": 5198117.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.538933753967285,
|
|
"epoch": 2.3556485355648533,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004702097878923419,
|
|
"loss": 5.2019,
|
|
"mean_token_accuracy": 0.1884738877415657,
|
|
"num_tokens": 5206791.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.4940228939056395,
|
|
"epoch": 2.3598326359832638,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004700490868701377,
|
|
"loss": 5.1058,
|
|
"mean_token_accuracy": 0.19394215941429138,
|
|
"num_tokens": 5216427.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.3847336769104,
|
|
"epoch": 2.3640167364016738,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004698879844682718,
|
|
"loss": 5.0899,
|
|
"mean_token_accuracy": 0.1956726461648941,
|
|
"num_tokens": 5225983.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.439257955551147,
|
|
"epoch": 2.3682008368200838,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046972648101826695,
|
|
"loss": 5.1712,
|
|
"mean_token_accuracy": 0.19315690398216248,
|
|
"num_tokens": 5234547.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.596295785903931,
|
|
"epoch": 2.3723849372384938,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004695645768524712,
|
|
"loss": 5.2238,
|
|
"mean_token_accuracy": 0.19835204333066941,
|
|
"num_tokens": 5243281.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.490299701690674,
|
|
"epoch": 2.3765690376569037,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004694022723040573,
|
|
"loss": 5.1261,
|
|
"mean_token_accuracy": 0.18668307811021806,
|
|
"num_tokens": 5252318.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.515432500839234,
|
|
"epoch": 2.3807531380753137,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046923956770702206,
|
|
"loss": 5.204,
|
|
"mean_token_accuracy": 0.1940845176577568,
|
|
"num_tokens": 5261781.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.573498725891113,
|
|
"epoch": 2.3849372384937237,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004690764633961852,
|
|
"loss": 5.11,
|
|
"mean_token_accuracy": 0.19316617548465728,
|
|
"num_tokens": 5270409.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.364716100692749,
|
|
"epoch": 2.3891213389121337,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004689129597071893,
|
|
"loss": 5.1221,
|
|
"mean_token_accuracy": 0.20516035258769988,
|
|
"num_tokens": 5278766.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.515507650375366,
|
|
"epoch": 2.393305439330544,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046874905697649857,
|
|
"loss": 5.2754,
|
|
"mean_token_accuracy": 0.1816343769431114,
|
|
"num_tokens": 5287572.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.501306343078613,
|
|
"epoch": 2.397489539748954,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046858475554139857,
|
|
"loss": 5.1833,
|
|
"mean_token_accuracy": 0.19312700480222703,
|
|
"num_tokens": 5297348.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.549457693099976,
|
|
"epoch": 2.401673640167364,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004684200557399952,
|
|
"loss": 5.1952,
|
|
"mean_token_accuracy": 0.1903080239892006,
|
|
"num_tokens": 5305748.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.492236948013305,
|
|
"epoch": 2.405857740585774,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046825495791121414,
|
|
"loss": 5.1385,
|
|
"mean_token_accuracy": 0.19962307959794998,
|
|
"num_tokens": 5314482.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.50344614982605,
|
|
"epoch": 2.410041841004184,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00046808946239480017,
|
|
"loss": 5.2372,
|
|
"mean_token_accuracy": 0.1866622343659401,
|
|
"num_tokens": 5325161.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.4480592727661135,
|
|
"epoch": 2.414225941422594,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004679235695313165,
|
|
"loss": 5.2173,
|
|
"mean_token_accuracy": 0.18587664663791656,
|
|
"num_tokens": 5335262.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.471191024780273,
|
|
"epoch": 2.418410041841004,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00046775727966214393,
|
|
"loss": 5.1161,
|
|
"mean_token_accuracy": 0.18551510572433472,
|
|
"num_tokens": 5345145.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.438928413391113,
|
|
"epoch": 2.422594142259414,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004675905931294803,
|
|
"loss": 5.1008,
|
|
"mean_token_accuracy": 0.19111968725919723,
|
|
"num_tokens": 5354715.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.437091255187989,
|
|
"epoch": 2.426778242677824,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004674235102763398,
|
|
"loss": 5.1683,
|
|
"mean_token_accuracy": 0.1983390673995018,
|
|
"num_tokens": 5364180.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.454319906234741,
|
|
"epoch": 2.430962343096234,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00046725603144655195,
|
|
"loss": 5.1333,
|
|
"mean_token_accuracy": 0.19619197398424149,
|
|
"num_tokens": 5372919.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.462763833999634,
|
|
"epoch": 2.435146443514644,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004670881569847612,
|
|
"loss": 5.104,
|
|
"mean_token_accuracy": 0.20203736573457717,
|
|
"num_tokens": 5383002.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.55969090461731,
|
|
"epoch": 2.439330543933054,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046691988723642636,
|
|
"loss": 5.3128,
|
|
"mean_token_accuracy": 0.18525985181331633,
|
|
"num_tokens": 5393622.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.5230185985565186,
|
|
"epoch": 2.4435146443514646,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004667512225478195,
|
|
"loss": 5.1987,
|
|
"mean_token_accuracy": 0.18882321268320085,
|
|
"num_tokens": 5403106.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.513156461715698,
|
|
"epoch": 2.4476987447698746,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046658216326602526,
|
|
"loss": 5.1784,
|
|
"mean_token_accuracy": 0.1932978704571724,
|
|
"num_tokens": 5412596.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.55132269859314,
|
|
"epoch": 2.4518828451882846,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004664127097389406,
|
|
"loss": 5.2235,
|
|
"mean_token_accuracy": 0.19170084595680237,
|
|
"num_tokens": 5422301.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.533843612670898,
|
|
"epoch": 2.4560669456066946,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046624286231527367,
|
|
"loss": 5.1438,
|
|
"mean_token_accuracy": 0.1933741256594658,
|
|
"num_tokens": 5432178.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.372134637832642,
|
|
"epoch": 2.4602510460251046,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004660726213445431,
|
|
"loss": 5.1823,
|
|
"mean_token_accuracy": 0.19677756428718568,
|
|
"num_tokens": 5441162.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.468031597137451,
|
|
"epoch": 2.4644351464435146,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004659019871770774,
|
|
"loss": 5.1041,
|
|
"mean_token_accuracy": 0.20391910523176193,
|
|
"num_tokens": 5450566.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.5154331684112545,
|
|
"epoch": 2.4686192468619246,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046573096016401435,
|
|
"loss": 5.144,
|
|
"mean_token_accuracy": 0.20277026444673538,
|
|
"num_tokens": 5460289.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.427037620544434,
|
|
"epoch": 2.4728033472803346,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004655595406572999,
|
|
"loss": 5.1242,
|
|
"mean_token_accuracy": 0.19660421907901765,
|
|
"num_tokens": 5469581.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.518569612503052,
|
|
"epoch": 2.476987447698745,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000465387729009688,
|
|
"loss": 5.2243,
|
|
"mean_token_accuracy": 0.19369943886995317,
|
|
"num_tokens": 5478005.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.485342979431152,
|
|
"epoch": 2.481171548117155,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046521552557473927,
|
|
"loss": 5.2092,
|
|
"mean_token_accuracy": 0.19067464172840118,
|
|
"num_tokens": 5486963.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.396170234680175,
|
|
"epoch": 2.485355648535565,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046504293070682073,
|
|
"loss": 5.0971,
|
|
"mean_token_accuracy": 0.19292196929454802,
|
|
"num_tokens": 5496368.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.4054327487945555,
|
|
"epoch": 2.489539748953975,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046486994476110493,
|
|
"loss": 5.0894,
|
|
"mean_token_accuracy": 0.19642373919487,
|
|
"num_tokens": 5505706.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.394309616088867,
|
|
"epoch": 2.493723849372385,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046469656809356916,
|
|
"loss": 5.0808,
|
|
"mean_token_accuracy": 0.20244767069816588,
|
|
"num_tokens": 5513945.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.3953783988952635,
|
|
"epoch": 2.497907949790795,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046452280106099454,
|
|
"loss": 5.1325,
|
|
"mean_token_accuracy": 0.188616843521595,
|
|
"num_tokens": 5523831.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.527229118347168,
|
|
"epoch": 2.502092050209205,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004643486440209659,
|
|
"loss": 5.2699,
|
|
"mean_token_accuracy": 0.18961481899023055,
|
|
"num_tokens": 5532705.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.525477695465088,
|
|
"epoch": 2.506276150627615,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004641740973318702,
|
|
"loss": 5.1864,
|
|
"mean_token_accuracy": 0.19227942377328872,
|
|
"num_tokens": 5542793.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.5265655517578125,
|
|
"epoch": 2.510460251046025,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004639991613528967,
|
|
"loss": 5.186,
|
|
"mean_token_accuracy": 0.19159698039293288,
|
|
"num_tokens": 5551938.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.510460251046025,
|
|
"eval_entropy": 5.213770697697052,
|
|
"eval_loss": 5.475037574768066,
|
|
"eval_mean_token_accuracy": 0.1852224213823463,
|
|
"eval_num_tokens": 5551938.0,
|
|
"eval_runtime": 2.2909,
|
|
"eval_samples_per_second": 1705.412,
|
|
"eval_steps_per_second": 213.449,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.462968111038208,
|
|
"epoch": 2.514644351464435,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046382383644403543,
|
|
"loss": 5.2129,
|
|
"mean_token_accuracy": 0.18845061659812928,
|
|
"num_tokens": 5561374.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.3695238590240475,
|
|
"epoch": 2.518828451882845,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046364812296607676,
|
|
"loss": 5.0817,
|
|
"mean_token_accuracy": 0.19389346539974212,
|
|
"num_tokens": 5570459.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.408000421524048,
|
|
"epoch": 2.523012552301255,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046347202128061094,
|
|
"loss": 5.1683,
|
|
"mean_token_accuracy": 0.1907093733549118,
|
|
"num_tokens": 5579692.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.527827787399292,
|
|
"epoch": 2.5271966527196654,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004632955317500268,
|
|
"loss": 5.1324,
|
|
"mean_token_accuracy": 0.19062095284461975,
|
|
"num_tokens": 5588338.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.4718592166900635,
|
|
"epoch": 2.5313807531380754,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046311865473751154,
|
|
"loss": 5.22,
|
|
"mean_token_accuracy": 0.19249685406684874,
|
|
"num_tokens": 5598589.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.462061405181885,
|
|
"epoch": 2.5355648535564854,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004629413906070495,
|
|
"loss": 5.1907,
|
|
"mean_token_accuracy": 0.19549299776554108,
|
|
"num_tokens": 5608127.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.548971509933471,
|
|
"epoch": 2.5397489539748954,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046276373972342177,
|
|
"loss": 5.235,
|
|
"mean_token_accuracy": 0.19585680067539216,
|
|
"num_tokens": 5617728.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.430024433135986,
|
|
"epoch": 2.5439330543933054,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004625857024522054,
|
|
"loss": 5.1747,
|
|
"mean_token_accuracy": 0.1904665008187294,
|
|
"num_tokens": 5626823.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.46235408782959,
|
|
"epoch": 2.5481171548117154,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046240727915977236,
|
|
"loss": 5.158,
|
|
"mean_token_accuracy": 0.19430482536554336,
|
|
"num_tokens": 5636284.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.555997657775879,
|
|
"epoch": 2.5523012552301254,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004622284702132893,
|
|
"loss": 5.2855,
|
|
"mean_token_accuracy": 0.1841801658272743,
|
|
"num_tokens": 5645721.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.413286399841309,
|
|
"epoch": 2.5564853556485354,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00046204927598071595,
|
|
"loss": 5.2032,
|
|
"mean_token_accuracy": 0.19346165359020234,
|
|
"num_tokens": 5654734.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.527040958404541,
|
|
"epoch": 2.560669456066946,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046186969683080547,
|
|
"loss": 5.1896,
|
|
"mean_token_accuracy": 0.20113472640514374,
|
|
"num_tokens": 5664010.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.474124813079834,
|
|
"epoch": 2.564853556485356,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046168973313310294,
|
|
"loss": 5.17,
|
|
"mean_token_accuracy": 0.18840867280960083,
|
|
"num_tokens": 5673028.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.43609094619751,
|
|
"epoch": 2.569037656903766,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004615093852579445,
|
|
"loss": 5.1336,
|
|
"mean_token_accuracy": 0.19433332085609437,
|
|
"num_tokens": 5682517.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.3480634689331055,
|
|
"epoch": 2.573221757322176,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046132865357645717,
|
|
"loss": 5.0674,
|
|
"mean_token_accuracy": 0.20391639918088914,
|
|
"num_tokens": 5690920.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.414696264266968,
|
|
"epoch": 2.577405857740586,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046114753846055774,
|
|
"loss": 5.143,
|
|
"mean_token_accuracy": 0.19163515716791152,
|
|
"num_tokens": 5700336.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.515250396728516,
|
|
"epoch": 2.581589958158996,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000460966040282952,
|
|
"loss": 5.2814,
|
|
"mean_token_accuracy": 0.18952076584100724,
|
|
"num_tokens": 5711085.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.500215530395508,
|
|
"epoch": 2.585774058577406,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00046078415941713384,
|
|
"loss": 5.0908,
|
|
"mean_token_accuracy": 0.20170718282461167,
|
|
"num_tokens": 5719560.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.378318500518799,
|
|
"epoch": 2.589958158995816,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004606018962373852,
|
|
"loss": 5.0494,
|
|
"mean_token_accuracy": 0.20389935672283171,
|
|
"num_tokens": 5727865.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.4493427753448485,
|
|
"epoch": 2.594142259414226,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004604192511187741,
|
|
"loss": 5.1844,
|
|
"mean_token_accuracy": 0.1872186169028282,
|
|
"num_tokens": 5737705.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.503284740447998,
|
|
"epoch": 2.598326359832636,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00046023622443715483,
|
|
"loss": 5.2045,
|
|
"mean_token_accuracy": 0.1855546861886978,
|
|
"num_tokens": 5747057.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.434207344055176,
|
|
"epoch": 2.602510460251046,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046005281656916707,
|
|
"loss": 5.156,
|
|
"mean_token_accuracy": 0.19481234550476073,
|
|
"num_tokens": 5755793.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.47066297531128,
|
|
"epoch": 2.606694560669456,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004598690278922346,
|
|
"loss": 5.1581,
|
|
"mean_token_accuracy": 0.1878283441066742,
|
|
"num_tokens": 5765385.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.33690767288208,
|
|
"epoch": 2.6108786610878663,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045968485878456497,
|
|
"loss": 5.0776,
|
|
"mean_token_accuracy": 0.19226381778717042,
|
|
"num_tokens": 5773947.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.443906211853028,
|
|
"epoch": 2.6150627615062763,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004595003096251486,
|
|
"loss": 5.0726,
|
|
"mean_token_accuracy": 0.19814347773790358,
|
|
"num_tokens": 5782700.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.493331956863403,
|
|
"epoch": 2.6192468619246863,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045931538079375803,
|
|
"loss": 5.2399,
|
|
"mean_token_accuracy": 0.18730711191892624,
|
|
"num_tokens": 5792452.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.486668348312378,
|
|
"epoch": 2.6234309623430963,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00045913007267094704,
|
|
"loss": 5.1706,
|
|
"mean_token_accuracy": 0.19296336323022842,
|
|
"num_tokens": 5801270.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.532931137084961,
|
|
"epoch": 2.6276150627615062,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004589443856380499,
|
|
"loss": 5.211,
|
|
"mean_token_accuracy": 0.19481545835733413,
|
|
"num_tokens": 5810704.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.481737565994263,
|
|
"epoch": 2.6317991631799162,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004587583200771807,
|
|
"loss": 5.1578,
|
|
"mean_token_accuracy": 0.1936139225959778,
|
|
"num_tokens": 5819690.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.37274956703186,
|
|
"epoch": 2.6359832635983262,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004585718763712325,
|
|
"loss": 5.0867,
|
|
"mean_token_accuracy": 0.19192241877317429,
|
|
"num_tokens": 5828912.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.31561918258667,
|
|
"epoch": 2.6401673640167362,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004583850549038764,
|
|
"loss": 5.0132,
|
|
"mean_token_accuracy": 0.19629174768924712,
|
|
"num_tokens": 5836914.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.444818162918091,
|
|
"epoch": 2.6443514644351467,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045819785605956105,
|
|
"loss": 5.1609,
|
|
"mean_token_accuracy": 0.19689084142446517,
|
|
"num_tokens": 5846268.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.400204181671143,
|
|
"epoch": 2.6485355648535567,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004580102802235115,
|
|
"loss": 5.1714,
|
|
"mean_token_accuracy": 0.1971372663974762,
|
|
"num_tokens": 5855029.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.49607892036438,
|
|
"epoch": 2.6527196652719667,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004578223277817286,
|
|
"loss": 5.1717,
|
|
"mean_token_accuracy": 0.18653464615345,
|
|
"num_tokens": 5865049.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.5427673816680905,
|
|
"epoch": 2.6569037656903767,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045763399912098835,
|
|
"loss": 5.1744,
|
|
"mean_token_accuracy": 0.18617971241474152,
|
|
"num_tokens": 5873593.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.361571311950684,
|
|
"epoch": 2.6610878661087867,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045744529462884086,
|
|
"loss": 5.0781,
|
|
"mean_token_accuracy": 0.19762487709522247,
|
|
"num_tokens": 5882562.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.429044437408447,
|
|
"epoch": 2.6652719665271967,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045725621469360964,
|
|
"loss": 5.2668,
|
|
"mean_token_accuracy": 0.18453310579061508,
|
|
"num_tokens": 5892516.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.4789777278900145,
|
|
"epoch": 2.6694560669456067,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004570667597043907,
|
|
"loss": 5.1428,
|
|
"mean_token_accuracy": 0.19438394159078598,
|
|
"num_tokens": 5901494.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.456045532226563,
|
|
"epoch": 2.6736401673640167,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000456876930051052,
|
|
"loss": 5.1421,
|
|
"mean_token_accuracy": 0.19510222375392913,
|
|
"num_tokens": 5910750.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.541888809204101,
|
|
"epoch": 2.6778242677824267,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004566867261242325,
|
|
"loss": 5.183,
|
|
"mean_token_accuracy": 0.19630327373743056,
|
|
"num_tokens": 5920277.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.410405969619751,
|
|
"epoch": 2.6820083682008367,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045649614831534127,
|
|
"loss": 5.1542,
|
|
"mean_token_accuracy": 0.1965371772646904,
|
|
"num_tokens": 5929880.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.527627420425415,
|
|
"epoch": 2.6861924686192467,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045630519701655677,
|
|
"loss": 5.321,
|
|
"mean_token_accuracy": 0.18892762362957,
|
|
"num_tokens": 5939538.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.448320198059082,
|
|
"epoch": 2.6903765690376567,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045611387262082604,
|
|
"loss": 5.1572,
|
|
"mean_token_accuracy": 0.19353785663843154,
|
|
"num_tokens": 5948747.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.4134666442871096,
|
|
"epoch": 2.694560669456067,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045592217552186407,
|
|
"loss": 5.1274,
|
|
"mean_token_accuracy": 0.19890922158956528,
|
|
"num_tokens": 5958209.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.474023628234863,
|
|
"epoch": 2.698744769874477,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004557301061141525,
|
|
"loss": 5.1838,
|
|
"mean_token_accuracy": 0.19570582509040832,
|
|
"num_tokens": 5967184.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.3882392883300785,
|
|
"epoch": 2.702928870292887,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00045553766479293945,
|
|
"loss": 5.1882,
|
|
"mean_token_accuracy": 0.19017804712057113,
|
|
"num_tokens": 5976006.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.338523197174072,
|
|
"epoch": 2.707112970711297,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045534485195423806,
|
|
"loss": 5.032,
|
|
"mean_token_accuracy": 0.19907400161027908,
|
|
"num_tokens": 5984671.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.536633491516113,
|
|
"epoch": 2.711297071129707,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045515166799482633,
|
|
"loss": 5.1701,
|
|
"mean_token_accuracy": 0.1948863998055458,
|
|
"num_tokens": 5993190.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.443432140350342,
|
|
"epoch": 2.715481171548117,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00045495811331224564,
|
|
"loss": 5.1621,
|
|
"mean_token_accuracy": 0.19129905104637146,
|
|
"num_tokens": 6004080.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.404340171813965,
|
|
"epoch": 2.719665271966527,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004547641883048005,
|
|
"loss": 5.1551,
|
|
"mean_token_accuracy": 0.19158968031406404,
|
|
"num_tokens": 6013212.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.437668323516846,
|
|
"epoch": 2.723849372384937,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004545698933715573,
|
|
"loss": 5.0994,
|
|
"mean_token_accuracy": 0.19823037832975388,
|
|
"num_tokens": 6023444.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.4280712604522705,
|
|
"epoch": 2.7280334728033475,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045437522891234386,
|
|
"loss": 5.1034,
|
|
"mean_token_accuracy": 0.19958617091178893,
|
|
"num_tokens": 6032217.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.404166889190674,
|
|
"epoch": 2.7322175732217575,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045418019532774827,
|
|
"loss": 5.0588,
|
|
"mean_token_accuracy": 0.19934537559747695,
|
|
"num_tokens": 6040246.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.41451325416565,
|
|
"epoch": 2.7364016736401675,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045398479301911836,
|
|
"loss": 5.173,
|
|
"mean_token_accuracy": 0.1900328889489174,
|
|
"num_tokens": 6049447.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.332231903076172,
|
|
"epoch": 2.7405857740585775,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045378902238856064,
|
|
"loss": 5.0451,
|
|
"mean_token_accuracy": 0.19907548427581787,
|
|
"num_tokens": 6058294.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.449809646606445,
|
|
"epoch": 2.7447698744769875,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004535928838389396,
|
|
"loss": 5.1878,
|
|
"mean_token_accuracy": 0.19494476914405823,
|
|
"num_tokens": 6067237.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.51179518699646,
|
|
"epoch": 2.7489539748953975,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004533963777738768,
|
|
"loss": 5.1898,
|
|
"mean_token_accuracy": 0.19057921171188355,
|
|
"num_tokens": 6075599.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.460715055465698,
|
|
"epoch": 2.7531380753138075,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045319950459775024,
|
|
"loss": 5.1629,
|
|
"mean_token_accuracy": 0.18881246596574783,
|
|
"num_tokens": 6085131.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.487667608261108,
|
|
"epoch": 2.7573221757322175,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004530022647156931,
|
|
"loss": 5.1314,
|
|
"mean_token_accuracy": 0.19969360232353212,
|
|
"num_tokens": 6094745.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.3998754024505615,
|
|
"epoch": 2.7615062761506275,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004528046585335935,
|
|
"loss": 5.1953,
|
|
"mean_token_accuracy": 0.18483933806419373,
|
|
"num_tokens": 6104182.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.410027885437012,
|
|
"epoch": 2.7656903765690375,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045260668645809304,
|
|
"loss": 5.0517,
|
|
"mean_token_accuracy": 0.2025168314576149,
|
|
"num_tokens": 6113370.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.356280279159546,
|
|
"epoch": 2.7698744769874475,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004524083488965866,
|
|
"loss": 5.1293,
|
|
"mean_token_accuracy": 0.19631304293870927,
|
|
"num_tokens": 6122202.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.374682331085205,
|
|
"epoch": 2.7740585774058575,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004522096462572208,
|
|
"loss": 5.1901,
|
|
"mean_token_accuracy": 0.1911054402589798,
|
|
"num_tokens": 6131081.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.586391258239746,
|
|
"epoch": 2.778242677824268,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004520105789488939,
|
|
"loss": 5.2668,
|
|
"mean_token_accuracy": 0.18820440918207168,
|
|
"num_tokens": 6141707.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.4600382328033445,
|
|
"epoch": 2.782426778242678,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045181114738125427,
|
|
"loss": 5.1162,
|
|
"mean_token_accuracy": 0.19284327626228331,
|
|
"num_tokens": 6151651.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.379871606826782,
|
|
"epoch": 2.786610878661088,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045161135196470015,
|
|
"loss": 5.0777,
|
|
"mean_token_accuracy": 0.19958383738994598,
|
|
"num_tokens": 6159928.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.37453989982605,
|
|
"epoch": 2.790794979079498,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004514111931103783,
|
|
"loss": 5.1734,
|
|
"mean_token_accuracy": 0.19033128768205643,
|
|
"num_tokens": 6169115.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.4737201690673825,
|
|
"epoch": 2.794979079497908,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045121067123018353,
|
|
"loss": 5.1854,
|
|
"mean_token_accuracy": 0.1947602093219757,
|
|
"num_tokens": 6177995.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.435278463363647,
|
|
"epoch": 2.799163179916318,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045100978673675755,
|
|
"loss": 5.1761,
|
|
"mean_token_accuracy": 0.19039879590272904,
|
|
"num_tokens": 6188081.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.50736927986145,
|
|
"epoch": 2.803347280334728,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045080854004348836,
|
|
"loss": 5.1291,
|
|
"mean_token_accuracy": 0.1885787844657898,
|
|
"num_tokens": 6198010.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.363803195953369,
|
|
"epoch": 2.8075313807531384,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00045060693156450933,
|
|
"loss": 5.1165,
|
|
"mean_token_accuracy": 0.19714236706495286,
|
|
"num_tokens": 6207164.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.4973828315734865,
|
|
"epoch": 2.8117154811715483,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004504049617146983,
|
|
"loss": 5.2401,
|
|
"mean_token_accuracy": 0.18754298090934754,
|
|
"num_tokens": 6216512.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.427261829376221,
|
|
"epoch": 2.8158995815899583,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045020263090967674,
|
|
"loss": 5.1889,
|
|
"mean_token_accuracy": 0.1960294485092163,
|
|
"num_tokens": 6226196.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.4154823303222654,
|
|
"epoch": 2.8200836820083683,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004499999395658088,
|
|
"loss": 5.0659,
|
|
"mean_token_accuracy": 0.19427476078271866,
|
|
"num_tokens": 6234820.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.381675148010254,
|
|
"epoch": 2.8242677824267783,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000449796888100201,
|
|
"loss": 5.1385,
|
|
"mean_token_accuracy": 0.19204012900590897,
|
|
"num_tokens": 6243813.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.518025684356689,
|
|
"epoch": 2.8284518828451883,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00044959347693070026,
|
|
"loss": 5.2249,
|
|
"mean_token_accuracy": 0.1914212316274643,
|
|
"num_tokens": 6253376.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.555794095993042,
|
|
"epoch": 2.8326359832635983,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004493897064758943,
|
|
"loss": 5.2586,
|
|
"mean_token_accuracy": 0.1916716992855072,
|
|
"num_tokens": 6262640.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.519548797607422,
|
|
"epoch": 2.8368200836820083,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004491855771551098,
|
|
"loss": 5.2022,
|
|
"mean_token_accuracy": 0.1959479421377182,
|
|
"num_tokens": 6270908.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.38501124382019,
|
|
"epoch": 2.8410041841004183,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004489810893884121,
|
|
"loss": 5.1067,
|
|
"mean_token_accuracy": 0.19345175325870514,
|
|
"num_tokens": 6280361.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.398903322219849,
|
|
"epoch": 2.8451882845188283,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004487762435966042,
|
|
"loss": 5.1248,
|
|
"mean_token_accuracy": 0.19799116104841233,
|
|
"num_tokens": 6288832.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.410068607330322,
|
|
"epoch": 2.8493723849372383,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00044857104020122585,
|
|
"loss": 5.1068,
|
|
"mean_token_accuracy": 0.19581186324357985,
|
|
"num_tokens": 6298553.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.341890954971314,
|
|
"epoch": 2.8535564853556483,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004483654796245526,
|
|
"loss": 5.0606,
|
|
"mean_token_accuracy": 0.19626759737730026,
|
|
"num_tokens": 6307692.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.482541465759278,
|
|
"epoch": 2.8577405857740583,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000448159562289595,
|
|
"loss": 5.1455,
|
|
"mean_token_accuracy": 0.19332865476608277,
|
|
"num_tokens": 6316729.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.4931943893432615,
|
|
"epoch": 2.8619246861924688,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00044795328862009794,
|
|
"loss": 5.2146,
|
|
"mean_token_accuracy": 0.18952576369047164,
|
|
"num_tokens": 6326210.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.321086454391479,
|
|
"epoch": 2.8661087866108788,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00044774665904053943,
|
|
"loss": 5.1329,
|
|
"mean_token_accuracy": 0.1981959745287895,
|
|
"num_tokens": 6335156.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.361148929595947,
|
|
"epoch": 2.8702928870292888,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004475396739761299,
|
|
"loss": 5.0203,
|
|
"mean_token_accuracy": 0.20167842358350754,
|
|
"num_tokens": 6344264.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.422211742401123,
|
|
"epoch": 2.8744769874476988,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00044733233385281117,
|
|
"loss": 5.093,
|
|
"mean_token_accuracy": 0.200849287211895,
|
|
"num_tokens": 6352997.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.363988924026489,
|
|
"epoch": 2.8786610878661087,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00044712463909725626,
|
|
"loss": 5.0922,
|
|
"mean_token_accuracy": 0.1967838779091835,
|
|
"num_tokens": 6362312.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.528323650360107,
|
|
"epoch": 2.8828451882845187,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00044691659013686727,
|
|
"loss": 5.2201,
|
|
"mean_token_accuracy": 0.19050872921943665,
|
|
"num_tokens": 6372107.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.387990474700928,
|
|
"epoch": 2.8870292887029287,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004467081873997756,
|
|
"loss": 5.0526,
|
|
"mean_token_accuracy": 0.20210649967193603,
|
|
"num_tokens": 6381306.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.337385892868042,
|
|
"epoch": 2.891213389121339,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044649943131484056,
|
|
"loss": 5.0115,
|
|
"mean_token_accuracy": 0.19747427701950074,
|
|
"num_tokens": 6390404.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.489330101013183,
|
|
"epoch": 2.895397489539749,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044629032231164874,
|
|
"loss": 5.2212,
|
|
"mean_token_accuracy": 0.1843263640999794,
|
|
"num_tokens": 6400366.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.408727359771729,
|
|
"epoch": 2.899581589958159,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004460808608205127,
|
|
"loss": 5.0302,
|
|
"mean_token_accuracy": 0.1996001347899437,
|
|
"num_tokens": 6408955.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.262295055389404,
|
|
"epoch": 2.903765690376569,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004458710472724706,
|
|
"loss": 5.0918,
|
|
"mean_token_accuracy": 0.19133829474449157,
|
|
"num_tokens": 6417172.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.39056544303894,
|
|
"epoch": 2.907949790794979,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00044566088209928485,
|
|
"loss": 5.1034,
|
|
"mean_token_accuracy": 0.19498197138309478,
|
|
"num_tokens": 6425898.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.371723222732544,
|
|
"epoch": 2.912133891213389,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00044545036573344175,
|
|
"loss": 5.1406,
|
|
"mean_token_accuracy": 0.19467767626047133,
|
|
"num_tokens": 6435940.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.376276540756225,
|
|
"epoch": 2.916317991631799,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00044523949860815015,
|
|
"loss": 5.0623,
|
|
"mean_token_accuracy": 0.20263846069574357,
|
|
"num_tokens": 6445119.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.384667587280274,
|
|
"epoch": 2.920502092050209,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004450282811573405,
|
|
"loss": 5.1117,
|
|
"mean_token_accuracy": 0.19618954360485077,
|
|
"num_tokens": 6453986.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.376170206069946,
|
|
"epoch": 2.924686192468619,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044481671381566446,
|
|
"loss": 5.0843,
|
|
"mean_token_accuracy": 0.19363461285829545,
|
|
"num_tokens": 6463658.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.379592990875244,
|
|
"epoch": 2.928870292887029,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004446047970184937,
|
|
"loss": 5.1255,
|
|
"mean_token_accuracy": 0.19629420936107636,
|
|
"num_tokens": 6472370.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 2.928870292887029,
|
|
"eval_entropy": 5.151964174945418,
|
|
"eval_loss": 5.394122123718262,
|
|
"eval_mean_token_accuracy": 0.1907018796066565,
|
|
"eval_num_tokens": 6472370.0,
|
|
"eval_runtime": 2.0851,
|
|
"eval_samples_per_second": 1873.805,
|
|
"eval_steps_per_second": 234.525,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.438113689422607,
|
|
"epoch": 2.933054393305439,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044439253120191877,
|
|
"loss": 5.2539,
|
|
"mean_token_accuracy": 0.1858174741268158,
|
|
"num_tokens": 6482642.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.365430307388306,
|
|
"epoch": 2.937238493723849,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00044417991680274874,
|
|
"loss": 5.1493,
|
|
"mean_token_accuracy": 0.1935872480273247,
|
|
"num_tokens": 6491818.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.419385147094727,
|
|
"epoch": 2.941422594142259,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004439669542585097,
|
|
"loss": 5.1454,
|
|
"mean_token_accuracy": 0.18960228860378264,
|
|
"num_tokens": 6500908.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.352400112152099,
|
|
"epoch": 2.9456066945606696,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004437536440074444,
|
|
"loss": 5.1151,
|
|
"mean_token_accuracy": 0.19481378197669982,
|
|
"num_tokens": 6510182.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.35415678024292,
|
|
"epoch": 2.9497907949790796,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004435399864885111,
|
|
"loss": 5.0908,
|
|
"mean_token_accuracy": 0.1997893199324608,
|
|
"num_tokens": 6519051.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.330616283416748,
|
|
"epoch": 2.9539748953974896,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004433259821413825,
|
|
"loss": 5.0269,
|
|
"mean_token_accuracy": 0.20871360450983048,
|
|
"num_tokens": 6527949.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.506001710891724,
|
|
"epoch": 2.9581589958158996,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044311163140644524,
|
|
"loss": 5.3052,
|
|
"mean_token_accuracy": 0.1850181758403778,
|
|
"num_tokens": 6537227.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.38893723487854,
|
|
"epoch": 2.9623430962343096,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044289693472479867,
|
|
"loss": 5.0979,
|
|
"mean_token_accuracy": 0.1902851089835167,
|
|
"num_tokens": 6546938.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.363958740234375,
|
|
"epoch": 2.9665271966527196,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044268189253825397,
|
|
"loss": 5.1071,
|
|
"mean_token_accuracy": 0.19246285557746887,
|
|
"num_tokens": 6555750.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.310831022262573,
|
|
"epoch": 2.9707112970711296,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004424665052893334,
|
|
"loss": 5.0443,
|
|
"mean_token_accuracy": 0.19983767420053483,
|
|
"num_tokens": 6565680.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.266908311843872,
|
|
"epoch": 2.97489539748954,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00044225077342126935,
|
|
"loss": 5.014,
|
|
"mean_token_accuracy": 0.20443485230207442,
|
|
"num_tokens": 6574595.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.436822366714478,
|
|
"epoch": 2.97907949790795,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004420346973780033,
|
|
"loss": 5.1484,
|
|
"mean_token_accuracy": 0.19291843771934508,
|
|
"num_tokens": 6583333.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.461638736724853,
|
|
"epoch": 2.98326359832636,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000441818277604185,
|
|
"loss": 5.1758,
|
|
"mean_token_accuracy": 0.2010513201355934,
|
|
"num_tokens": 6592252.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.345468807220459,
|
|
"epoch": 2.98744769874477,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004416015145451715,
|
|
"loss": 4.9912,
|
|
"mean_token_accuracy": 0.20407403260469437,
|
|
"num_tokens": 6600675.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.34960904121399,
|
|
"epoch": 2.99163179916318,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004413844086470264,
|
|
"loss": 5.0959,
|
|
"mean_token_accuracy": 0.19900625348091125,
|
|
"num_tokens": 6609722.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.422952175140381,
|
|
"epoch": 2.99581589958159,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004411669603565187,
|
|
"loss": 5.0922,
|
|
"mean_token_accuracy": 0.20171553194522857,
|
|
"num_tokens": 6618415.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.397150373458862,
|
|
"epoch": 3.0,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000440949170121122,
|
|
"loss": 5.1008,
|
|
"mean_token_accuracy": 0.19668337553739548,
|
|
"num_tokens": 6627966.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.417689371109009,
|
|
"epoch": 3.00418410041841,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00044073103838901373,
|
|
"loss": 4.7726,
|
|
"mean_token_accuracy": 0.22511388063430787,
|
|
"num_tokens": 6636820.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.302271938323974,
|
|
"epoch": 3.00836820083682,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044051256560907374,
|
|
"loss": 4.8061,
|
|
"mean_token_accuracy": 0.21354635208845138,
|
|
"num_tokens": 6645792.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.42719087600708,
|
|
"epoch": 3.01255230125523,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00044029375223088394,
|
|
"loss": 4.9038,
|
|
"mean_token_accuracy": 0.21108814477920532,
|
|
"num_tokens": 6654850.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.293935489654541,
|
|
"epoch": 3.01673640167364,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00044007459870472706,
|
|
"loss": 4.8389,
|
|
"mean_token_accuracy": 0.21405893117189406,
|
|
"num_tokens": 6664524.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.289966678619384,
|
|
"epoch": 3.02092050209205,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004398551054815859,
|
|
"loss": 4.7767,
|
|
"mean_token_accuracy": 0.21690000742673873,
|
|
"num_tokens": 6674330.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.350748205184937,
|
|
"epoch": 3.0251046025104604,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004396352730131422,
|
|
"loss": 4.8596,
|
|
"mean_token_accuracy": 0.21818140000104905,
|
|
"num_tokens": 6683121.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.360624027252197,
|
|
"epoch": 3.0292887029288704,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004394151017517758,
|
|
"loss": 4.816,
|
|
"mean_token_accuracy": 0.2204051747918129,
|
|
"num_tokens": 6692566.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.379842805862427,
|
|
"epoch": 3.0334728033472804,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004391945921505638,
|
|
"loss": 4.8994,
|
|
"mean_token_accuracy": 0.20829516500234604,
|
|
"num_tokens": 6701577.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.25770583152771,
|
|
"epoch": 3.0376569037656904,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00043897374466327944,
|
|
"loss": 4.7579,
|
|
"mean_token_accuracy": 0.22355487197637558,
|
|
"num_tokens": 6710672.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.287049102783203,
|
|
"epoch": 3.0418410041841004,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00043875255974439145,
|
|
"loss": 4.7783,
|
|
"mean_token_accuracy": 0.21751174479722976,
|
|
"num_tokens": 6719412.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.277928113937378,
|
|
"epoch": 3.0460251046025104,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004385310378490627,
|
|
"loss": 4.7796,
|
|
"mean_token_accuracy": 0.21473814100027083,
|
|
"num_tokens": 6729169.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.305811214447021,
|
|
"epoch": 3.0502092050209204,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00043830917943314977,
|
|
"loss": 4.7221,
|
|
"mean_token_accuracy": 0.2207148939371109,
|
|
"num_tokens": 6738549.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.262117481231689,
|
|
"epoch": 3.0543933054393304,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00043808698495320154,
|
|
"loss": 4.7783,
|
|
"mean_token_accuracy": 0.22020159810781478,
|
|
"num_tokens": 6747909.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.2283553123474125,
|
|
"epoch": 3.0585774058577404,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00043786445486645846,
|
|
"loss": 4.835,
|
|
"mean_token_accuracy": 0.21614689081907273,
|
|
"num_tokens": 6757125.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.379630041122437,
|
|
"epoch": 3.062761506276151,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004376415896308518,
|
|
"loss": 4.8186,
|
|
"mean_token_accuracy": 0.22030971944332123,
|
|
"num_tokens": 6765774.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.341221857070923,
|
|
"epoch": 3.066945606694561,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00043741838970500234,
|
|
"loss": 4.874,
|
|
"mean_token_accuracy": 0.21571923047304153,
|
|
"num_tokens": 6774980.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.3157493591308596,
|
|
"epoch": 3.071129707112971,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004371948555482196,
|
|
"loss": 4.8663,
|
|
"mean_token_accuracy": 0.21201944798231126,
|
|
"num_tokens": 6784187.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.28399658203125,
|
|
"epoch": 3.075313807531381,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000436970987620501,
|
|
"loss": 4.8791,
|
|
"mean_token_accuracy": 0.21151311695575714,
|
|
"num_tokens": 6792953.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.325482034683228,
|
|
"epoch": 3.079497907949791,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00043674678638253066,
|
|
"loss": 4.7583,
|
|
"mean_token_accuracy": 0.22190648466348648,
|
|
"num_tokens": 6803004.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.262072944641114,
|
|
"epoch": 3.083682008368201,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004365222522956788,
|
|
"loss": 4.8085,
|
|
"mean_token_accuracy": 0.215910604596138,
|
|
"num_tokens": 6811883.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.257794713973999,
|
|
"epoch": 3.087866108786611,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004362973858220004,
|
|
"loss": 4.8195,
|
|
"mean_token_accuracy": 0.2243594378232956,
|
|
"num_tokens": 6820327.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.31516466140747,
|
|
"epoch": 3.092050209205021,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00043607218742423447,
|
|
"loss": 4.8363,
|
|
"mean_token_accuracy": 0.21631665378808976,
|
|
"num_tokens": 6829412.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.25584397315979,
|
|
"epoch": 3.096234309623431,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00043584665756580323,
|
|
"loss": 4.8056,
|
|
"mean_token_accuracy": 0.2158111184835434,
|
|
"num_tokens": 6838997.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.361407375335693,
|
|
"epoch": 3.100418410041841,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004356207967108106,
|
|
"loss": 4.8709,
|
|
"mean_token_accuracy": 0.20751748085021973,
|
|
"num_tokens": 6848957.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.308306169509888,
|
|
"epoch": 3.104602510460251,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004353946053240422,
|
|
"loss": 4.9086,
|
|
"mean_token_accuracy": 0.21284400522708893,
|
|
"num_tokens": 6859688.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.336264944076538,
|
|
"epoch": 3.1087866108786613,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004351680838709635,
|
|
"loss": 4.9385,
|
|
"mean_token_accuracy": 0.2013743668794632,
|
|
"num_tokens": 6869149.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.385293483734131,
|
|
"epoch": 3.1129707112970713,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00043494123281771915,
|
|
"loss": 4.909,
|
|
"mean_token_accuracy": 0.20783170610666274,
|
|
"num_tokens": 6879432.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.26263394355774,
|
|
"epoch": 3.1171548117154813,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00043471405263113214,
|
|
"loss": 4.8632,
|
|
"mean_token_accuracy": 0.21832842081785203,
|
|
"num_tokens": 6888398.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.298207092285156,
|
|
"epoch": 3.1213389121338913,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004344865437787027,
|
|
"loss": 4.7927,
|
|
"mean_token_accuracy": 0.21798287332057953,
|
|
"num_tokens": 6898082.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.222813844680786,
|
|
"epoch": 3.1255230125523012,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00043425870672860774,
|
|
"loss": 4.7684,
|
|
"mean_token_accuracy": 0.21932897120714187,
|
|
"num_tokens": 6907693.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.389808130264282,
|
|
"epoch": 3.1297071129707112,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004340305419496989,
|
|
"loss": 4.9751,
|
|
"mean_token_accuracy": 0.20066292434930802,
|
|
"num_tokens": 6917095.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.317314767837525,
|
|
"epoch": 3.1338912133891212,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004338020499115029,
|
|
"loss": 4.9058,
|
|
"mean_token_accuracy": 0.20590936839580537,
|
|
"num_tokens": 6926952.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.308876848220825,
|
|
"epoch": 3.1380753138075312,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004335732310842196,
|
|
"loss": 4.8358,
|
|
"mean_token_accuracy": 0.21241551339626313,
|
|
"num_tokens": 6936704.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.27081069946289,
|
|
"epoch": 3.1422594142259412,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004333440859387212,
|
|
"loss": 4.8492,
|
|
"mean_token_accuracy": 0.21257602721452712,
|
|
"num_tokens": 6946614.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.350625896453858,
|
|
"epoch": 3.1464435146443517,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00043311461494655165,
|
|
"loss": 4.8762,
|
|
"mean_token_accuracy": 0.2187810242176056,
|
|
"num_tokens": 6955980.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.359692811965942,
|
|
"epoch": 3.1506276150627617,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004328848185799254,
|
|
"loss": 4.8457,
|
|
"mean_token_accuracy": 0.21638285219669343,
|
|
"num_tokens": 6964108.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.319613885879517,
|
|
"epoch": 3.1548117154811717,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004326546973117264,
|
|
"loss": 4.8449,
|
|
"mean_token_accuracy": 0.22057590186595916,
|
|
"num_tokens": 6972895.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.270555591583252,
|
|
"epoch": 3.1589958158995817,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00043242425161550723,
|
|
"loss": 4.858,
|
|
"mean_token_accuracy": 0.2126601293683052,
|
|
"num_tokens": 6982222.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.32768440246582,
|
|
"epoch": 3.1631799163179917,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00043219348196548807,
|
|
"loss": 4.8495,
|
|
"mean_token_accuracy": 0.21487165689468385,
|
|
"num_tokens": 6990820.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.277839851379395,
|
|
"epoch": 3.1673640167364017,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004319623888365557,
|
|
"loss": 4.8912,
|
|
"mean_token_accuracy": 0.21117942333221434,
|
|
"num_tokens": 7000846.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.257175159454346,
|
|
"epoch": 3.1715481171548117,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00043173097270426275,
|
|
"loss": 4.8065,
|
|
"mean_token_accuracy": 0.21191611289978027,
|
|
"num_tokens": 7009086.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.408314180374146,
|
|
"epoch": 3.1757322175732217,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00043149923404482634,
|
|
"loss": 4.8774,
|
|
"mean_token_accuracy": 0.21050115823745727,
|
|
"num_tokens": 7017752.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.266625022888183,
|
|
"epoch": 3.1799163179916317,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00043126717333512733,
|
|
"loss": 4.8745,
|
|
"mean_token_accuracy": 0.22228188961744308,
|
|
"num_tokens": 7026535.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.29256157875061,
|
|
"epoch": 3.1841004184100417,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004310347910527094,
|
|
"loss": 4.9079,
|
|
"mean_token_accuracy": 0.21733937710523604,
|
|
"num_tokens": 7035462.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.292649555206299,
|
|
"epoch": 3.1882845188284517,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004308020876757778,
|
|
"loss": 4.8296,
|
|
"mean_token_accuracy": 0.21759999841451644,
|
|
"num_tokens": 7043978.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.2791407108306885,
|
|
"epoch": 3.192468619246862,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000430569063683199,
|
|
"loss": 4.8306,
|
|
"mean_token_accuracy": 0.22045563608407975,
|
|
"num_tokens": 7053757.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.215965270996094,
|
|
"epoch": 3.196652719665272,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004303357195544986,
|
|
"loss": 4.7617,
|
|
"mean_token_accuracy": 0.2122887060046196,
|
|
"num_tokens": 7062491.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.24834885597229,
|
|
"epoch": 3.200836820083682,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004301020557698615,
|
|
"loss": 4.8122,
|
|
"mean_token_accuracy": 0.21571405231952667,
|
|
"num_tokens": 7072157.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.243205547332764,
|
|
"epoch": 3.205020920502092,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00042986807281013013,
|
|
"loss": 4.771,
|
|
"mean_token_accuracy": 0.21973754316568375,
|
|
"num_tokens": 7080631.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.338223552703857,
|
|
"epoch": 3.209205020920502,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004296337711568037,
|
|
"loss": 4.8465,
|
|
"mean_token_accuracy": 0.2220018669962883,
|
|
"num_tokens": 7089944.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.306629037857055,
|
|
"epoch": 3.213389121338912,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00042939915129203745,
|
|
"loss": 4.8861,
|
|
"mean_token_accuracy": 0.21429578810930253,
|
|
"num_tokens": 7099190.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.15996675491333,
|
|
"epoch": 3.217573221757322,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004291642136986413,
|
|
"loss": 4.8527,
|
|
"mean_token_accuracy": 0.20626747012138366,
|
|
"num_tokens": 7107944.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.237857961654663,
|
|
"epoch": 3.221757322175732,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000428928958860079,
|
|
"loss": 4.7044,
|
|
"mean_token_accuracy": 0.21445917338132858,
|
|
"num_tokens": 7116953.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.225046586990357,
|
|
"epoch": 3.225941422594142,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004286933872604673,
|
|
"loss": 4.7621,
|
|
"mean_token_accuracy": 0.22059133797883987,
|
|
"num_tokens": 7126236.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.29739933013916,
|
|
"epoch": 3.2301255230125525,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.0004284574993845746,
|
|
"loss": 4.8464,
|
|
"mean_token_accuracy": 0.2139418512582779,
|
|
"num_tokens": 7136286.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.284384250640869,
|
|
"epoch": 3.2343096234309625,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00042822129571782025,
|
|
"loss": 4.8919,
|
|
"mean_token_accuracy": 0.2139872580766678,
|
|
"num_tokens": 7145551.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.313933515548706,
|
|
"epoch": 3.2384937238493725,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00042798477674627343,
|
|
"loss": 4.8427,
|
|
"mean_token_accuracy": 0.2191818654537201,
|
|
"num_tokens": 7155486.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.2608051776885985,
|
|
"epoch": 3.2426778242677825,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00042774794295665213,
|
|
"loss": 4.8475,
|
|
"mean_token_accuracy": 0.21361929476261138,
|
|
"num_tokens": 7163984.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.26825442314148,
|
|
"epoch": 3.2468619246861925,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004275107948363223,
|
|
"loss": 4.7978,
|
|
"mean_token_accuracy": 0.21317970901727676,
|
|
"num_tokens": 7172854.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.197178030014038,
|
|
"epoch": 3.2510460251046025,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004272733328732965,
|
|
"loss": 4.8196,
|
|
"mean_token_accuracy": 0.20986296534538268,
|
|
"num_tokens": 7182202.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.286785459518432,
|
|
"epoch": 3.2552301255230125,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00042703555755623337,
|
|
"loss": 4.9265,
|
|
"mean_token_accuracy": 0.2052895739674568,
|
|
"num_tokens": 7192062.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.4119648933410645,
|
|
"epoch": 3.2594142259414225,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004267974693744362,
|
|
"loss": 4.9313,
|
|
"mean_token_accuracy": 0.20530452877283095,
|
|
"num_tokens": 7201766.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.283129405975342,
|
|
"epoch": 3.2635983263598325,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004265590688178523,
|
|
"loss": 4.8391,
|
|
"mean_token_accuracy": 0.20876989662647247,
|
|
"num_tokens": 7211361.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.275751447677612,
|
|
"epoch": 3.2677824267782425,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004263203563770716,
|
|
"loss": 4.8111,
|
|
"mean_token_accuracy": 0.21774289458990098,
|
|
"num_tokens": 7220221.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.2844874382019045,
|
|
"epoch": 3.2719665271966525,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000426081332543326,
|
|
"loss": 4.9261,
|
|
"mean_token_accuracy": 0.20752736777067185,
|
|
"num_tokens": 7230134.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.192396020889282,
|
|
"epoch": 3.276150627615063,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004258419978084881,
|
|
"loss": 4.7405,
|
|
"mean_token_accuracy": 0.22151417583227156,
|
|
"num_tokens": 7238713.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.34380407333374,
|
|
"epoch": 3.280334728033473,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004256023526650702,
|
|
"loss": 4.8633,
|
|
"mean_token_accuracy": 0.21645605862140654,
|
|
"num_tokens": 7247603.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.252472591400147,
|
|
"epoch": 3.284518828451883,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004253623976062236,
|
|
"loss": 4.8991,
|
|
"mean_token_accuracy": 0.21033538281917571,
|
|
"num_tokens": 7257452.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.377154922485351,
|
|
"epoch": 3.288702928870293,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004251221331257372,
|
|
"loss": 4.8837,
|
|
"mean_token_accuracy": 0.21271614581346512,
|
|
"num_tokens": 7266401.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.290256547927856,
|
|
"epoch": 3.292887029288703,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004248815597180366,
|
|
"loss": 4.841,
|
|
"mean_token_accuracy": 0.2185330480337143,
|
|
"num_tokens": 7275836.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.27263913154602,
|
|
"epoch": 3.297071129707113,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00042464067787818333,
|
|
"loss": 4.8939,
|
|
"mean_token_accuracy": 0.2124941200017929,
|
|
"num_tokens": 7284854.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.299465465545654,
|
|
"epoch": 3.301255230125523,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004243994881018735,
|
|
"loss": 4.8587,
|
|
"mean_token_accuracy": 0.21343783140182496,
|
|
"num_tokens": 7293421.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.272430753707885,
|
|
"epoch": 3.305439330543933,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004241579908854367,
|
|
"loss": 4.8021,
|
|
"mean_token_accuracy": 0.2127429038286209,
|
|
"num_tokens": 7302184.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.383147096633911,
|
|
"epoch": 3.309623430962343,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00042391618672583546,
|
|
"loss": 4.9469,
|
|
"mean_token_accuracy": 0.209491965174675,
|
|
"num_tokens": 7311596.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.273970890045166,
|
|
"epoch": 3.3138075313807533,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00042367407612066383,
|
|
"loss": 4.7863,
|
|
"mean_token_accuracy": 0.21876903474330903,
|
|
"num_tokens": 7319866.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.2668585777282715,
|
|
"epoch": 3.3179916317991633,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004234316595681468,
|
|
"loss": 4.8559,
|
|
"mean_token_accuracy": 0.21466988176107407,
|
|
"num_tokens": 7329327.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.3854817867279055,
|
|
"epoch": 3.3221757322175733,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004231889375671382,
|
|
"loss": 5.0159,
|
|
"mean_token_accuracy": 0.20021829903125762,
|
|
"num_tokens": 7339717.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.314059782028198,
|
|
"epoch": 3.3263598326359833,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00042294591061712124,
|
|
"loss": 4.8441,
|
|
"mean_token_accuracy": 0.21653733253479004,
|
|
"num_tokens": 7348850.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.2205463409423825,
|
|
"epoch": 3.3305439330543933,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00042270257921820623,
|
|
"loss": 4.789,
|
|
"mean_token_accuracy": 0.2239212676882744,
|
|
"num_tokens": 7357081.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.1775803565979,
|
|
"epoch": 3.3347280334728033,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00042245894387112987,
|
|
"loss": 4.8069,
|
|
"mean_token_accuracy": 0.2220182329416275,
|
|
"num_tokens": 7365762.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.281820964813233,
|
|
"epoch": 3.3389121338912133,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004222150050772548,
|
|
"loss": 4.8727,
|
|
"mean_token_accuracy": 0.2108008921146393,
|
|
"num_tokens": 7374870.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.243097877502441,
|
|
"epoch": 3.3430962343096233,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004219707633385677,
|
|
"loss": 4.7353,
|
|
"mean_token_accuracy": 0.2225353464484215,
|
|
"num_tokens": 7383836.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.2761904239654545,
|
|
"epoch": 3.3472803347280333,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00042172621915767886,
|
|
"loss": 4.8533,
|
|
"mean_token_accuracy": 0.21824376583099364,
|
|
"num_tokens": 7394059.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 3.3472803347280333,
|
|
"eval_entropy": 5.010007921659629,
|
|
"eval_loss": 5.385568141937256,
|
|
"eval_mean_token_accuracy": 0.19328919281623114,
|
|
"eval_num_tokens": 7394059.0,
|
|
"eval_runtime": 2.2664,
|
|
"eval_samples_per_second": 1723.889,
|
|
"eval_steps_per_second": 215.762,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 5.173360347747803,
|
|
"epoch": 3.3514644351464433,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00042148137303782077,
|
|
"loss": 4.7823,
|
|
"mean_token_accuracy": 0.22140112221240998,
|
|
"num_tokens": 7402947.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 5.283971452713013,
|
|
"epoch": 3.3556485355648533,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004212362254828475,
|
|
"loss": 4.8695,
|
|
"mean_token_accuracy": 0.21366961747407914,
|
|
"num_tokens": 7413430.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 5.321456432342529,
|
|
"epoch": 3.3598326359832638,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00042099077699723326,
|
|
"loss": 4.8712,
|
|
"mean_token_accuracy": 0.21481242924928665,
|
|
"num_tokens": 7423239.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 5.2574036598205565,
|
|
"epoch": 3.3640167364016738,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004207450280860716,
|
|
"loss": 4.8548,
|
|
"mean_token_accuracy": 0.22157425284385682,
|
|
"num_tokens": 7432519.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 5.210657358169556,
|
|
"epoch": 3.3682008368200838,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004204989792550742,
|
|
"loss": 4.8383,
|
|
"mean_token_accuracy": 0.21389587968587875,
|
|
"num_tokens": 7442089.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 5.305828952789307,
|
|
"epoch": 3.3723849372384938,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00042025263101057016,
|
|
"loss": 4.8607,
|
|
"mean_token_accuracy": 0.2121138483285904,
|
|
"num_tokens": 7451059.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 5.260219287872315,
|
|
"epoch": 3.3765690376569037,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00042000598385950427,
|
|
"loss": 4.7951,
|
|
"mean_token_accuracy": 0.21677857488393784,
|
|
"num_tokens": 7458960.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 5.286827754974365,
|
|
"epoch": 3.3807531380753137,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000419759038309437,
|
|
"loss": 4.9756,
|
|
"mean_token_accuracy": 0.20775616616010667,
|
|
"num_tokens": 7468863.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 5.286834716796875,
|
|
"epoch": 3.3849372384937237,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00041951179486854253,
|
|
"loss": 4.928,
|
|
"mean_token_accuracy": 0.2115194469690323,
|
|
"num_tokens": 7477204.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 5.228511047363281,
|
|
"epoch": 3.3891213389121337,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000419264254045608,
|
|
"loss": 4.7634,
|
|
"mean_token_accuracy": 0.2170839086174965,
|
|
"num_tokens": 7487367.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 5.131165218353272,
|
|
"epoch": 3.393305439330544,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004190164163500326,
|
|
"loss": 4.753,
|
|
"mean_token_accuracy": 0.22677248418331147,
|
|
"num_tokens": 7495808.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 5.239573621749878,
|
|
"epoch": 3.397489539748954,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004187682822918266,
|
|
"loss": 4.9052,
|
|
"mean_token_accuracy": 0.2009483754634857,
|
|
"num_tokens": 7506119.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 5.2634800434112545,
|
|
"epoch": 3.401673640167364,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00041851985238161,
|
|
"loss": 4.8667,
|
|
"mean_token_accuracy": 0.21869323700666427,
|
|
"num_tokens": 7514931.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 5.142191410064697,
|
|
"epoch": 3.405857740585774,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00041827112713061155,
|
|
"loss": 4.6771,
|
|
"mean_token_accuracy": 0.23150861412286758,
|
|
"num_tokens": 7523753.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 5.224830055236817,
|
|
"epoch": 3.410041841004184,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00041802210705066775,
|
|
"loss": 4.7633,
|
|
"mean_token_accuracy": 0.21790978014469148,
|
|
"num_tokens": 7532447.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 5.220453262329102,
|
|
"epoch": 3.414225941422594,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000417772792654222,
|
|
"loss": 4.9113,
|
|
"mean_token_accuracy": 0.20994119048118592,
|
|
"num_tokens": 7541388.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 5.2316076278686525,
|
|
"epoch": 3.418410041841004,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00041752318445432323,
|
|
"loss": 4.9148,
|
|
"mean_token_accuracy": 0.201690736413002,
|
|
"num_tokens": 7550422.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 5.240323448181153,
|
|
"epoch": 3.422594142259414,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000417273282964625,
|
|
"loss": 4.7695,
|
|
"mean_token_accuracy": 0.22500985711812974,
|
|
"num_tokens": 7559522.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 5.27940182685852,
|
|
"epoch": 3.426778242677824,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004170230886993844,
|
|
"loss": 4.8866,
|
|
"mean_token_accuracy": 0.21124469935894014,
|
|
"num_tokens": 7568785.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 5.230301189422607,
|
|
"epoch": 3.430962343096234,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00041677260217346093,
|
|
"loss": 4.8188,
|
|
"mean_token_accuracy": 0.2161060705780983,
|
|
"num_tokens": 7579013.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 5.231803274154663,
|
|
"epoch": 3.435146443514644,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00041652182390231556,
|
|
"loss": 4.8345,
|
|
"mean_token_accuracy": 0.20797687470912934,
|
|
"num_tokens": 7588577.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.21132001876831,
|
|
"epoch": 3.439330543933054,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00041627075440200977,
|
|
"loss": 4.7739,
|
|
"mean_token_accuracy": 0.22572113573551178,
|
|
"num_tokens": 7597702.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 5.197178030014038,
|
|
"epoch": 3.4435146443514646,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004160193941892043,
|
|
"loss": 4.8335,
|
|
"mean_token_accuracy": 0.21552833914756775,
|
|
"num_tokens": 7607156.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 5.253040599822998,
|
|
"epoch": 3.4476987447698746,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004157677437811578,
|
|
"loss": 4.8997,
|
|
"mean_token_accuracy": 0.22069366574287413,
|
|
"num_tokens": 7615936.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 5.311299133300781,
|
|
"epoch": 3.4518828451882846,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004155158036957265,
|
|
"loss": 4.888,
|
|
"mean_token_accuracy": 0.21268777698278427,
|
|
"num_tokens": 7626568.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 5.3051958084106445,
|
|
"epoch": 3.4560669456066946,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004152635744513626,
|
|
"loss": 4.825,
|
|
"mean_token_accuracy": 0.21435431092977525,
|
|
"num_tokens": 7635102.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 5.1717061519622805,
|
|
"epoch": 3.4602510460251046,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004150110565671133,
|
|
"loss": 4.7556,
|
|
"mean_token_accuracy": 0.23037197589874267,
|
|
"num_tokens": 7644204.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 5.292326354980469,
|
|
"epoch": 3.4644351464435146,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00041475825056261985,
|
|
"loss": 4.8838,
|
|
"mean_token_accuracy": 0.21161860823631287,
|
|
"num_tokens": 7653585.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 5.276823663711548,
|
|
"epoch": 3.4686192468619246,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004145051569581162,
|
|
"loss": 4.8671,
|
|
"mean_token_accuracy": 0.20865801572799683,
|
|
"num_tokens": 7662734.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 5.2846942901611325,
|
|
"epoch": 3.4728033472803346,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004142517762744286,
|
|
"loss": 4.8393,
|
|
"mean_token_accuracy": 0.2154109835624695,
|
|
"num_tokens": 7671682.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 5.320446014404297,
|
|
"epoch": 3.476987447698745,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004139981090329735,
|
|
"loss": 4.9383,
|
|
"mean_token_accuracy": 0.21157864332199097,
|
|
"num_tokens": 7681551.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 5.3092334270477295,
|
|
"epoch": 3.481171548117155,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004137441557557575,
|
|
"loss": 4.8733,
|
|
"mean_token_accuracy": 0.2136368378996849,
|
|
"num_tokens": 7690943.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 5.213263320922851,
|
|
"epoch": 3.485355648535565,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00041348991696537533,
|
|
"loss": 4.7962,
|
|
"mean_token_accuracy": 0.22002696692943574,
|
|
"num_tokens": 7700376.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 5.288240528106689,
|
|
"epoch": 3.489539748953975,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004132353931850098,
|
|
"loss": 4.8826,
|
|
"mean_token_accuracy": 0.20910301208496093,
|
|
"num_tokens": 7709845.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 5.323158168792725,
|
|
"epoch": 3.493723849372385,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004129805849384299,
|
|
"loss": 4.9708,
|
|
"mean_token_accuracy": 0.20627950876951218,
|
|
"num_tokens": 7719908.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 5.260655260086059,
|
|
"epoch": 3.497907949790795,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004127254927499898,
|
|
"loss": 4.8862,
|
|
"mean_token_accuracy": 0.20989871323108672,
|
|
"num_tokens": 7729543.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 5.342302370071411,
|
|
"epoch": 3.502092050209205,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004124701171446286,
|
|
"loss": 4.9041,
|
|
"mean_token_accuracy": 0.2150196611881256,
|
|
"num_tokens": 7739371.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 5.285289907455445,
|
|
"epoch": 3.506276150627615,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00041221445864786794,
|
|
"loss": 4.8285,
|
|
"mean_token_accuracy": 0.21524600982666015,
|
|
"num_tokens": 7748027.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 5.1448150157928465,
|
|
"epoch": 3.510460251046025,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000411958517785812,
|
|
"loss": 4.7594,
|
|
"mean_token_accuracy": 0.21886340975761415,
|
|
"num_tokens": 7756588.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 5.237047815322876,
|
|
"epoch": 3.514644351464435,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00041170229508514584,
|
|
"loss": 4.7513,
|
|
"mean_token_accuracy": 0.22075403332710267,
|
|
"num_tokens": 7765222.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 5.249066638946533,
|
|
"epoch": 3.518828451882845,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004114457910731347,
|
|
"loss": 4.8439,
|
|
"mean_token_accuracy": 0.21644431799650193,
|
|
"num_tokens": 7774139.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 5.162914896011353,
|
|
"epoch": 3.523012552301255,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004111890062776226,
|
|
"loss": 4.7371,
|
|
"mean_token_accuracy": 0.21626648902893067,
|
|
"num_tokens": 7783597.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 5.146347522735596,
|
|
"epoch": 3.5271966527196654,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004109319412270314,
|
|
"loss": 4.8385,
|
|
"mean_token_accuracy": 0.21763765513896943,
|
|
"num_tokens": 7792711.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 5.371251773834229,
|
|
"epoch": 3.5313807531380754,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004106745964503594,
|
|
"loss": 4.9184,
|
|
"mean_token_accuracy": 0.21169726252555848,
|
|
"num_tokens": 7801878.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 5.267062950134277,
|
|
"epoch": 3.5355648535564854,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000410416972477181,
|
|
"loss": 4.796,
|
|
"mean_token_accuracy": 0.22080903947353364,
|
|
"num_tokens": 7810484.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 5.271413135528564,
|
|
"epoch": 3.5397489539748954,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00041015906983764495,
|
|
"loss": 4.9093,
|
|
"mean_token_accuracy": 0.21164571344852448,
|
|
"num_tokens": 7820521.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 5.266349649429321,
|
|
"epoch": 3.5439330543933054,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00040990088906247326,
|
|
"loss": 4.8307,
|
|
"mean_token_accuracy": 0.2146122321486473,
|
|
"num_tokens": 7829218.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 5.227874708175659,
|
|
"epoch": 3.5481171548117154,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004096424306829605,
|
|
"loss": 4.8507,
|
|
"mean_token_accuracy": 0.21140016913414,
|
|
"num_tokens": 7837795.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 5.28312087059021,
|
|
"epoch": 3.5523012552301254,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004093836952309726,
|
|
"loss": 4.8533,
|
|
"mean_token_accuracy": 0.2144578516483307,
|
|
"num_tokens": 7846973.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 5.332216262817383,
|
|
"epoch": 3.5564853556485354,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004091246832389453,
|
|
"loss": 4.8992,
|
|
"mean_token_accuracy": 0.21368567645549774,
|
|
"num_tokens": 7856627.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 5.186514616012573,
|
|
"epoch": 3.560669456066946,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004088653952398837,
|
|
"loss": 4.8131,
|
|
"mean_token_accuracy": 0.21810080260038375,
|
|
"num_tokens": 7866308.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 5.212180280685425,
|
|
"epoch": 3.564853556485356,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00040860583176736097,
|
|
"loss": 4.7955,
|
|
"mean_token_accuracy": 0.22124039381742477,
|
|
"num_tokens": 7875350.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 5.221352291107178,
|
|
"epoch": 3.569037656903766,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004083459933555169,
|
|
"loss": 4.8011,
|
|
"mean_token_accuracy": 0.22317755967378616,
|
|
"num_tokens": 7884276.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 5.251061534881591,
|
|
"epoch": 3.573221757322176,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004080858805390571,
|
|
"loss": 4.8766,
|
|
"mean_token_accuracy": 0.20903294533491135,
|
|
"num_tokens": 7893836.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 5.325833940505982,
|
|
"epoch": 3.577405857740586,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000407825493853252,
|
|
"loss": 4.8563,
|
|
"mean_token_accuracy": 0.21939553916454316,
|
|
"num_tokens": 7903249.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 5.2541382789611815,
|
|
"epoch": 3.581589958158996,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00040756483383393556,
|
|
"loss": 4.905,
|
|
"mean_token_accuracy": 0.21057581305503845,
|
|
"num_tokens": 7912217.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 5.259947729110718,
|
|
"epoch": 3.585774058577406,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004073039010175041,
|
|
"loss": 4.9428,
|
|
"mean_token_accuracy": 0.20682976990938187,
|
|
"num_tokens": 7922241.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 5.285345888137817,
|
|
"epoch": 3.589958158995816,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00040704269594091556,
|
|
"loss": 4.8478,
|
|
"mean_token_accuracy": 0.21154145747423173,
|
|
"num_tokens": 7931866.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 5.320349454879761,
|
|
"epoch": 3.594142259414226,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004067812191416878,
|
|
"loss": 4.8791,
|
|
"mean_token_accuracy": 0.20850423723459244,
|
|
"num_tokens": 7941787.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 5.254129457473755,
|
|
"epoch": 3.598326359832636,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004065194711578982,
|
|
"loss": 4.9103,
|
|
"mean_token_accuracy": 0.20791636109352113,
|
|
"num_tokens": 7950832.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 5.244665288925171,
|
|
"epoch": 3.602510460251046,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004062574525281819,
|
|
"loss": 4.8914,
|
|
"mean_token_accuracy": 0.20982639789581298,
|
|
"num_tokens": 7960059.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 5.304762125015259,
|
|
"epoch": 3.606694560669456,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00040599516379173113,
|
|
"loss": 4.8673,
|
|
"mean_token_accuracy": 0.20721402168273925,
|
|
"num_tokens": 7969411.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 5.262882709503174,
|
|
"epoch": 3.6108786610878663,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00040573260548829417,
|
|
"loss": 4.8703,
|
|
"mean_token_accuracy": 0.21179659217596053,
|
|
"num_tokens": 7978446.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 5.139433288574219,
|
|
"epoch": 3.6150627615062763,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004054697781581734,
|
|
"loss": 4.7747,
|
|
"mean_token_accuracy": 0.2160693421959877,
|
|
"num_tokens": 7987117.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 5.246527242660522,
|
|
"epoch": 3.6192468619246863,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004052066823422255,
|
|
"loss": 4.8534,
|
|
"mean_token_accuracy": 0.21306308209896088,
|
|
"num_tokens": 7995989.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 5.235064268112183,
|
|
"epoch": 3.6234309623430963,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004049433185818593,
|
|
"loss": 4.8986,
|
|
"mean_token_accuracy": 0.2070925772190094,
|
|
"num_tokens": 8005486.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 5.238435983657837,
|
|
"epoch": 3.6276150627615062,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00040467968741903507,
|
|
"loss": 4.7908,
|
|
"mean_token_accuracy": 0.2217899352312088,
|
|
"num_tokens": 8014853.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 5.321291446685791,
|
|
"epoch": 3.6317991631799162,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004044157893962633,
|
|
"loss": 4.837,
|
|
"mean_token_accuracy": 0.21470011174678802,
|
|
"num_tokens": 8023931.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 5.2389562129974365,
|
|
"epoch": 3.6359832635983262,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004041516250566037,
|
|
"loss": 4.8549,
|
|
"mean_token_accuracy": 0.21009364277124404,
|
|
"num_tokens": 8032659.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 5.242457866668701,
|
|
"epoch": 3.6401673640167362,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00040388719494366404,
|
|
"loss": 4.9418,
|
|
"mean_token_accuracy": 0.21339166313409805,
|
|
"num_tokens": 8041857.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 5.347074031829834,
|
|
"epoch": 3.6443514644351467,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00040362249960159904,
|
|
"loss": 4.9426,
|
|
"mean_token_accuracy": 0.20893017202615738,
|
|
"num_tokens": 8051186.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 5.265613460540772,
|
|
"epoch": 3.6485355648535567,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00040335753957510906,
|
|
"loss": 4.8691,
|
|
"mean_token_accuracy": 0.20750440210103988,
|
|
"num_tokens": 8060396.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 5.205458116531372,
|
|
"epoch": 3.6527196652719667,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00040309231540943937,
|
|
"loss": 4.8106,
|
|
"mean_token_accuracy": 0.22267701774835585,
|
|
"num_tokens": 8068106.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 5.270960330963135,
|
|
"epoch": 3.6569037656903767,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004028268276503784,
|
|
"loss": 4.8273,
|
|
"mean_token_accuracy": 0.2134488195180893,
|
|
"num_tokens": 8077878.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 5.311983108520508,
|
|
"epoch": 3.6610878661087867,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004025610768442576,
|
|
"loss": 4.9537,
|
|
"mean_token_accuracy": 0.2077968090772629,
|
|
"num_tokens": 8086968.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 5.205004119873047,
|
|
"epoch": 3.6652719665271967,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004022950635379492,
|
|
"loss": 4.7618,
|
|
"mean_token_accuracy": 0.22324050068855286,
|
|
"num_tokens": 8095999.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 5.138569498062134,
|
|
"epoch": 3.6694560669456067,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00040202878827886587,
|
|
"loss": 4.7798,
|
|
"mean_token_accuracy": 0.22284745723009108,
|
|
"num_tokens": 8105376.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 5.272082996368408,
|
|
"epoch": 3.6736401673640167,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004017622516149592,
|
|
"loss": 4.873,
|
|
"mean_token_accuracy": 0.21667338907718658,
|
|
"num_tokens": 8115024.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 5.278810453414917,
|
|
"epoch": 3.6778242677824267,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004014954540947189,
|
|
"loss": 4.8776,
|
|
"mean_token_accuracy": 0.21214754730463029,
|
|
"num_tokens": 8124306.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 5.223426628112793,
|
|
"epoch": 3.6820083682008367,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00040122839626717137,
|
|
"loss": 4.8656,
|
|
"mean_token_accuracy": 0.2172100856900215,
|
|
"num_tokens": 8133151.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 5.270011329650879,
|
|
"epoch": 3.6861924686192467,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004009610786818786,
|
|
"loss": 4.8786,
|
|
"mean_token_accuracy": 0.21756852716207503,
|
|
"num_tokens": 8141745.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 5.170894622802734,
|
|
"epoch": 3.6903765690376567,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00040069350188893735,
|
|
"loss": 4.7282,
|
|
"mean_token_accuracy": 0.22400207668542862,
|
|
"num_tokens": 8150884.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 5.231955909729004,
|
|
"epoch": 3.694560669456067,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00040042566643897745,
|
|
"loss": 4.8321,
|
|
"mean_token_accuracy": 0.20814217925071715,
|
|
"num_tokens": 8160138.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 5.329418134689331,
|
|
"epoch": 3.698744769874477,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004001575728831613,
|
|
"loss": 4.9026,
|
|
"mean_token_accuracy": 0.20905722826719284,
|
|
"num_tokens": 8169655.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 5.3440381526947025,
|
|
"epoch": 3.702928870292887,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003998892217731824,
|
|
"loss": 4.9092,
|
|
"mean_token_accuracy": 0.21718588024377822,
|
|
"num_tokens": 8178846.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 5.219350814819336,
|
|
"epoch": 3.707112970711297,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000399620613661264,
|
|
"loss": 4.8141,
|
|
"mean_token_accuracy": 0.21372939944267272,
|
|
"num_tokens": 8187525.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 5.212192153930664,
|
|
"epoch": 3.711297071129707,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003993517491001586,
|
|
"loss": 4.9017,
|
|
"mean_token_accuracy": 0.21221342533826829,
|
|
"num_tokens": 8196855.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 5.1624078273773195,
|
|
"epoch": 3.715481171548117,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003990826286431462,
|
|
"loss": 4.7712,
|
|
"mean_token_accuracy": 0.21273223906755448,
|
|
"num_tokens": 8205530.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 5.283079385757446,
|
|
"epoch": 3.719665271966527,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003988132528440333,
|
|
"loss": 4.8773,
|
|
"mean_token_accuracy": 0.20793134868144988,
|
|
"num_tokens": 8215066.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 5.316819524765014,
|
|
"epoch": 3.723849372384937,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003985436222571521,
|
|
"loss": 4.8527,
|
|
"mean_token_accuracy": 0.21046682745218276,
|
|
"num_tokens": 8224585.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 5.270925807952881,
|
|
"epoch": 3.7280334728033475,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000398273737437359,
|
|
"loss": 4.9225,
|
|
"mean_token_accuracy": 0.20935059040784837,
|
|
"num_tokens": 8233334.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 5.277789449691772,
|
|
"epoch": 3.7322175732217575,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00039800359894003355,
|
|
"loss": 4.8395,
|
|
"mean_token_accuracy": 0.22024863660335542,
|
|
"num_tokens": 8241636.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 5.270069599151611,
|
|
"epoch": 3.7364016736401675,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003977332073210774,
|
|
"loss": 4.8932,
|
|
"mean_token_accuracy": 0.21395074129104613,
|
|
"num_tokens": 8250273.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 5.257729196548462,
|
|
"epoch": 3.7405857740585775,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003974625631369129,
|
|
"loss": 4.9216,
|
|
"mean_token_accuracy": 0.20676147043704987,
|
|
"num_tokens": 8260349.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 5.321069717407227,
|
|
"epoch": 3.7447698744769875,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003971916669444824,
|
|
"loss": 4.9249,
|
|
"mean_token_accuracy": 0.2073482245206833,
|
|
"num_tokens": 8270505.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 5.292701578140258,
|
|
"epoch": 3.7489539748953975,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00039692051930124663,
|
|
"loss": 4.8989,
|
|
"mean_token_accuracy": 0.21353759318590165,
|
|
"num_tokens": 8278651.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 5.229554319381714,
|
|
"epoch": 3.7531380753138075,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003966491207651838,
|
|
"loss": 4.8387,
|
|
"mean_token_accuracy": 0.21973665952682495,
|
|
"num_tokens": 8287173.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 5.304041337966919,
|
|
"epoch": 3.7573221757322175,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00039637747189478866,
|
|
"loss": 4.8485,
|
|
"mean_token_accuracy": 0.2121449813246727,
|
|
"num_tokens": 8295785.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 5.243969678878784,
|
|
"epoch": 3.7615062761506275,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003961055732490708,
|
|
"loss": 4.9602,
|
|
"mean_token_accuracy": 0.21216156631708144,
|
|
"num_tokens": 8305534.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 5.253064680099487,
|
|
"epoch": 3.7656903765690375,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00039583342538755374,
|
|
"loss": 4.8518,
|
|
"mean_token_accuracy": 0.21743525713682174,
|
|
"num_tokens": 8314232.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"epoch": 3.7656903765690375,
|
|
"eval_entropy": 5.036744974873549,
|
|
"eval_loss": 5.336486339569092,
|
|
"eval_mean_token_accuracy": 0.19650313689360102,
|
|
"eval_num_tokens": 8314232.0,
|
|
"eval_runtime": 2.0921,
|
|
"eval_samples_per_second": 1867.477,
|
|
"eval_steps_per_second": 233.733,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 5.305580282211304,
|
|
"epoch": 3.7698744769874475,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00039556102887027423,
|
|
"loss": 4.9058,
|
|
"mean_token_accuracy": 0.20134300887584686,
|
|
"num_tokens": 8323230.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 5.316751289367676,
|
|
"epoch": 3.7740585774058575,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00039528838425778053,
|
|
"loss": 4.9286,
|
|
"mean_token_accuracy": 0.21118088960647582,
|
|
"num_tokens": 8333072.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 5.2931084632873535,
|
|
"epoch": 3.778242677824268,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003950154921111313,
|
|
"loss": 4.8819,
|
|
"mean_token_accuracy": 0.21092555075883865,
|
|
"num_tokens": 8342953.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 5.25038743019104,
|
|
"epoch": 3.782426778242678,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00039474235299189484,
|
|
"loss": 4.8838,
|
|
"mean_token_accuracy": 0.21132055521011353,
|
|
"num_tokens": 8351729.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 5.3162829875946045,
|
|
"epoch": 3.786610878661088,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003944689674621474,
|
|
"loss": 4.9374,
|
|
"mean_token_accuracy": 0.21075452566146852,
|
|
"num_tokens": 8360929.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 5.375373458862304,
|
|
"epoch": 3.790794979079498,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00039419533608447244,
|
|
"loss": 4.9124,
|
|
"mean_token_accuracy": 0.2174785166978836,
|
|
"num_tokens": 8369506.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 5.268008804321289,
|
|
"epoch": 3.794979079497908,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003939214594219595,
|
|
"loss": 4.8442,
|
|
"mean_token_accuracy": 0.21211714446544647,
|
|
"num_tokens": 8378583.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 5.285557508468628,
|
|
"epoch": 3.799163179916318,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00039364733803820263,
|
|
"loss": 4.8253,
|
|
"mean_token_accuracy": 0.21860430687665938,
|
|
"num_tokens": 8388078.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 5.171735143661499,
|
|
"epoch": 3.803347280334728,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003933729724972996,
|
|
"loss": 4.8174,
|
|
"mean_token_accuracy": 0.21998861879110337,
|
|
"num_tokens": 8397086.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 5.2138529300689695,
|
|
"epoch": 3.8075313807531384,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00039309836336385044,
|
|
"loss": 4.8163,
|
|
"mean_token_accuracy": 0.22575073838233947,
|
|
"num_tokens": 8405621.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 5.333823347091675,
|
|
"epoch": 3.8117154811715483,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003928235112029568,
|
|
"loss": 4.8673,
|
|
"mean_token_accuracy": 0.21636037528514862,
|
|
"num_tokens": 8415220.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 5.234338045120239,
|
|
"epoch": 3.8158995815899583,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003925484165802201,
|
|
"loss": 4.9254,
|
|
"mean_token_accuracy": 0.21351963132619858,
|
|
"num_tokens": 8424719.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 5.216107702255249,
|
|
"epoch": 3.8200836820083683,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003922730800617408,
|
|
"loss": 4.8393,
|
|
"mean_token_accuracy": 0.21145475804805755,
|
|
"num_tokens": 8433844.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 5.138389635086059,
|
|
"epoch": 3.8242677824267783,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003919975022141174,
|
|
"loss": 4.7636,
|
|
"mean_token_accuracy": 0.22463047057390212,
|
|
"num_tokens": 8442700.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 5.320693874359131,
|
|
"epoch": 3.8284518828451883,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00039172168360444463,
|
|
"loss": 4.9992,
|
|
"mean_token_accuracy": 0.20473110377788545,
|
|
"num_tokens": 8452246.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 5.288448333740234,
|
|
"epoch": 3.8326359832635983,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00039144562480031297,
|
|
"loss": 4.8449,
|
|
"mean_token_accuracy": 0.22275521159172057,
|
|
"num_tokens": 8461977.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 5.179286336898803,
|
|
"epoch": 3.8368200836820083,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00039116932636980685,
|
|
"loss": 4.7544,
|
|
"mean_token_accuracy": 0.22053902596235275,
|
|
"num_tokens": 8470124.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 5.194449520111084,
|
|
"epoch": 3.8410041841004183,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00039089278888150417,
|
|
"loss": 4.8962,
|
|
"mean_token_accuracy": 0.20818911790847777,
|
|
"num_tokens": 8479408.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 5.280982160568238,
|
|
"epoch": 3.8451882845188283,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003906160129044746,
|
|
"loss": 4.8478,
|
|
"mean_token_accuracy": 0.21387334614992143,
|
|
"num_tokens": 8488239.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 5.3347920894622805,
|
|
"epoch": 3.8493723849372383,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003903389990082787,
|
|
"loss": 4.8756,
|
|
"mean_token_accuracy": 0.21144764721393586,
|
|
"num_tokens": 8497690.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 5.299941730499268,
|
|
"epoch": 3.8535564853556483,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003900617477629663,
|
|
"loss": 4.9588,
|
|
"mean_token_accuracy": 0.20949174016714095,
|
|
"num_tokens": 8506974.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 5.256332492828369,
|
|
"epoch": 3.8577405857740583,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003897842597390761,
|
|
"loss": 4.8931,
|
|
"mean_token_accuracy": 0.215071003139019,
|
|
"num_tokens": 8516066.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 5.260136651992798,
|
|
"epoch": 3.8619246861924688,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003895065355076338,
|
|
"loss": 4.8581,
|
|
"mean_token_accuracy": 0.21104921400547028,
|
|
"num_tokens": 8524747.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 5.198331308364868,
|
|
"epoch": 3.8661087866108788,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003892285756401511,
|
|
"loss": 4.843,
|
|
"mean_token_accuracy": 0.21788447052240373,
|
|
"num_tokens": 8534274.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 5.298676156997681,
|
|
"epoch": 3.8702928870292888,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003889503807086248,
|
|
"loss": 4.8548,
|
|
"mean_token_accuracy": 0.21071720123291016,
|
|
"num_tokens": 8543453.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 5.241233253479004,
|
|
"epoch": 3.8744769874476988,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00038867195128553536,
|
|
"loss": 4.9282,
|
|
"mean_token_accuracy": 0.2084151566028595,
|
|
"num_tokens": 8553190.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 5.316149711608887,
|
|
"epoch": 3.8786610878661087,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003883932879438457,
|
|
"loss": 4.8293,
|
|
"mean_token_accuracy": 0.20948052555322647,
|
|
"num_tokens": 8562814.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 5.200613737106323,
|
|
"epoch": 3.8828451882845187,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003881143912570002,
|
|
"loss": 4.827,
|
|
"mean_token_accuracy": 0.2099359154701233,
|
|
"num_tokens": 8572131.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 5.189424180984497,
|
|
"epoch": 3.8870292887029287,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003878352617989235,
|
|
"loss": 4.9036,
|
|
"mean_token_accuracy": 0.20632086098194122,
|
|
"num_tokens": 8582325.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 5.229073572158813,
|
|
"epoch": 3.891213389121339,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003875559001440191,
|
|
"loss": 4.8236,
|
|
"mean_token_accuracy": 0.21966834664344786,
|
|
"num_tokens": 8591704.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 5.260281085968018,
|
|
"epoch": 3.895397489539749,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003872763068671682,
|
|
"loss": 4.877,
|
|
"mean_token_accuracy": 0.20855671912431717,
|
|
"num_tokens": 8601136.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 5.212862825393676,
|
|
"epoch": 3.899581589958159,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003869964825437291,
|
|
"loss": 4.8427,
|
|
"mean_token_accuracy": 0.2192463532090187,
|
|
"num_tokens": 8610491.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 5.27723240852356,
|
|
"epoch": 3.903765690376569,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00038671642774953507,
|
|
"loss": 4.8603,
|
|
"mean_token_accuracy": 0.21588762551546098,
|
|
"num_tokens": 8619199.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 5.263813638687134,
|
|
"epoch": 3.907949790794979,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003864361430608939,
|
|
"loss": 4.9394,
|
|
"mean_token_accuracy": 0.21231396794319152,
|
|
"num_tokens": 8629291.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 5.25251350402832,
|
|
"epoch": 3.912133891213389,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003861556290545864,
|
|
"loss": 4.89,
|
|
"mean_token_accuracy": 0.2131063461303711,
|
|
"num_tokens": 8639212.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 5.2548792362213135,
|
|
"epoch": 3.916317991631799,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00038587488630786534,
|
|
"loss": 4.8377,
|
|
"mean_token_accuracy": 0.2166909843683243,
|
|
"num_tokens": 8648351.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 5.27034559249878,
|
|
"epoch": 3.920502092050209,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00038559391539845413,
|
|
"loss": 4.8863,
|
|
"mean_token_accuracy": 0.20865781903266906,
|
|
"num_tokens": 8657267.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 5.260030364990234,
|
|
"epoch": 3.924686192468619,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00038531271690454567,
|
|
"loss": 5.041,
|
|
"mean_token_accuracy": 0.20459137260913848,
|
|
"num_tokens": 8666982.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 5.285803270339966,
|
|
"epoch": 3.928870292887029,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003850312914048015,
|
|
"loss": 4.8309,
|
|
"mean_token_accuracy": 0.22487906366586685,
|
|
"num_tokens": 8676067.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 5.308206987380982,
|
|
"epoch": 3.933054393305439,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003847496394783496,
|
|
"loss": 4.8749,
|
|
"mean_token_accuracy": 0.21968012899160386,
|
|
"num_tokens": 8685106.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 5.252278232574463,
|
|
"epoch": 3.937238493723849,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003844677617047848,
|
|
"loss": 4.9366,
|
|
"mean_token_accuracy": 0.2031252786517143,
|
|
"num_tokens": 8695203.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 5.114925670623779,
|
|
"epoch": 3.941422594142259,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000384185658664166,
|
|
"loss": 4.7547,
|
|
"mean_token_accuracy": 0.21928240954875947,
|
|
"num_tokens": 8703700.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 5.2834752082824705,
|
|
"epoch": 3.9456066945606696,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00038390333093701596,
|
|
"loss": 4.9164,
|
|
"mean_token_accuracy": 0.21281677931547166,
|
|
"num_tokens": 8713393.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 5.27287335395813,
|
|
"epoch": 3.9497907949790796,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00038362077910431973,
|
|
"loss": 4.8579,
|
|
"mean_token_accuracy": 0.21459901779890062,
|
|
"num_tokens": 8722263.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 5.211973476409912,
|
|
"epoch": 3.9539748953974896,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003833380037475236,
|
|
"loss": 4.859,
|
|
"mean_token_accuracy": 0.2133357673883438,
|
|
"num_tokens": 8732190.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 5.292964076995849,
|
|
"epoch": 3.9581589958158996,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00038305500544853375,
|
|
"loss": 4.832,
|
|
"mean_token_accuracy": 0.22431707233190537,
|
|
"num_tokens": 8740995.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 5.32289228439331,
|
|
"epoch": 3.9623430962343096,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00038277178478971527,
|
|
"loss": 4.9437,
|
|
"mean_token_accuracy": 0.2054638534784317,
|
|
"num_tokens": 8750768.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 5.199044847488404,
|
|
"epoch": 3.9665271966527196,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003824883423538905,
|
|
"loss": 4.8694,
|
|
"mean_token_accuracy": 0.2152554824948311,
|
|
"num_tokens": 8760070.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 5.171033477783203,
|
|
"epoch": 3.9707112970711296,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00038220467872433875,
|
|
"loss": 4.8578,
|
|
"mean_token_accuracy": 0.2238023355603218,
|
|
"num_tokens": 8769313.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 5.246708250045776,
|
|
"epoch": 3.97489539748954,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003819207944847939,
|
|
"loss": 4.8515,
|
|
"mean_token_accuracy": 0.21775915771722792,
|
|
"num_tokens": 8778328.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 5.2649068355560305,
|
|
"epoch": 3.97907949790795,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00038163669021944413,
|
|
"loss": 4.8698,
|
|
"mean_token_accuracy": 0.22113397419452668,
|
|
"num_tokens": 8787408.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 5.239938497543335,
|
|
"epoch": 3.98326359832636,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003813523665129304,
|
|
"loss": 4.818,
|
|
"mean_token_accuracy": 0.2159178912639618,
|
|
"num_tokens": 8796781.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 5.249154043197632,
|
|
"epoch": 3.98744769874477,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003810678239503452,
|
|
"loss": 4.8556,
|
|
"mean_token_accuracy": 0.21590107232332229,
|
|
"num_tokens": 8806965.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 5.200403213500977,
|
|
"epoch": 3.99163179916318,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003807830631172312,
|
|
"loss": 4.8738,
|
|
"mean_token_accuracy": 0.21220654249191284,
|
|
"num_tokens": 8817348.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 5.233676528930664,
|
|
"epoch": 3.99581589958159,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00038049808459958065,
|
|
"loss": 4.9178,
|
|
"mean_token_accuracy": 0.21693020761013032,
|
|
"num_tokens": 8827304.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 5.237826299667359,
|
|
"epoch": 4.0,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00038021288898383343,
|
|
"loss": 4.8289,
|
|
"mean_token_accuracy": 0.21403277516365052,
|
|
"num_tokens": 8837288.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 5.250961542129517,
|
|
"epoch": 4.00418410041841,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003799274768568761,
|
|
"loss": 4.6435,
|
|
"mean_token_accuracy": 0.23990754932165145,
|
|
"num_tokens": 8846447.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 5.241725587844849,
|
|
"epoch": 4.00836820083682,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00037964184880604125,
|
|
"loss": 4.5734,
|
|
"mean_token_accuracy": 0.23939525187015534,
|
|
"num_tokens": 8855990.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 5.129968547821045,
|
|
"epoch": 4.01255230125523,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00037935600541910523,
|
|
"loss": 4.5744,
|
|
"mean_token_accuracy": 0.23216097801923752,
|
|
"num_tokens": 8864949.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 5.117290401458741,
|
|
"epoch": 4.01673640167364,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003790699472842879,
|
|
"loss": 4.5257,
|
|
"mean_token_accuracy": 0.24570415467023848,
|
|
"num_tokens": 8873501.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 5.185815095901489,
|
|
"epoch": 4.02092050209205,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00037878367499025086,
|
|
"loss": 4.6358,
|
|
"mean_token_accuracy": 0.23204002231359483,
|
|
"num_tokens": 8882920.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 5.20503830909729,
|
|
"epoch": 4.02510460251046,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003784971891260965,
|
|
"loss": 4.5809,
|
|
"mean_token_accuracy": 0.23567768335342407,
|
|
"num_tokens": 8893617.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 5.212201929092407,
|
|
"epoch": 4.02928870292887,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00037821049028136664,
|
|
"loss": 4.5648,
|
|
"mean_token_accuracy": 0.2350827783346176,
|
|
"num_tokens": 8902558.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 5.199494552612305,
|
|
"epoch": 4.03347280334728,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003779235790460414,
|
|
"loss": 4.5951,
|
|
"mean_token_accuracy": 0.2367618441581726,
|
|
"num_tokens": 8912477.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 5.162327146530151,
|
|
"epoch": 4.03765690376569,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003776364560105379,
|
|
"loss": 4.6175,
|
|
"mean_token_accuracy": 0.23187377750873567,
|
|
"num_tokens": 8922313.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 5.218354606628418,
|
|
"epoch": 4.0418410041841,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00037734912176570935,
|
|
"loss": 4.5534,
|
|
"mean_token_accuracy": 0.24160847067832947,
|
|
"num_tokens": 8932011.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 5.187836647033691,
|
|
"epoch": 4.046025104602511,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00037706157690284335,
|
|
"loss": 4.6151,
|
|
"mean_token_accuracy": 0.22896585166454314,
|
|
"num_tokens": 8941022.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 5.168837547302246,
|
|
"epoch": 4.050209205020921,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00037677382201366103,
|
|
"loss": 4.6195,
|
|
"mean_token_accuracy": 0.2407027393579483,
|
|
"num_tokens": 8950825.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 5.1660277366638185,
|
|
"epoch": 4.054393305439331,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00037648585769031555,
|
|
"loss": 4.6306,
|
|
"mean_token_accuracy": 0.22982614934444429,
|
|
"num_tokens": 8961035.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 5.234085464477539,
|
|
"epoch": 4.058577405857741,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003761976845253915,
|
|
"loss": 4.665,
|
|
"mean_token_accuracy": 0.2341006025671959,
|
|
"num_tokens": 8970535.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 5.151261472702027,
|
|
"epoch": 4.062761506276151,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003759093031119025,
|
|
"loss": 4.6092,
|
|
"mean_token_accuracy": 0.23320264965295792,
|
|
"num_tokens": 8979404.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 5.2555201053619385,
|
|
"epoch": 4.066945606694561,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00037562071404329147,
|
|
"loss": 4.6228,
|
|
"mean_token_accuracy": 0.2442083552479744,
|
|
"num_tokens": 8989164.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 5.240410566329956,
|
|
"epoch": 4.071129707112971,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00037533191791342817,
|
|
"loss": 4.6634,
|
|
"mean_token_accuracy": 0.23153546452522278,
|
|
"num_tokens": 8998630.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 5.181739377975464,
|
|
"epoch": 4.075313807531381,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00037504291531660866,
|
|
"loss": 4.5906,
|
|
"mean_token_accuracy": 0.24348855167627334,
|
|
"num_tokens": 9008235.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 5.2158783912658695,
|
|
"epoch": 4.079497907949791,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003747537068475538,
|
|
"loss": 4.6987,
|
|
"mean_token_accuracy": 0.23010110408067702,
|
|
"num_tokens": 9018175.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 5.2824383735656735,
|
|
"epoch": 4.083682008368201,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003744642931014082,
|
|
"loss": 4.719,
|
|
"mean_token_accuracy": 0.2294529691338539,
|
|
"num_tokens": 9028159.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 5.099898386001587,
|
|
"epoch": 4.087866108786611,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003741746746737386,
|
|
"loss": 4.6183,
|
|
"mean_token_accuracy": 0.23134950250387193,
|
|
"num_tokens": 9038263.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 5.19078426361084,
|
|
"epoch": 4.092050209205021,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003738848521605333,
|
|
"loss": 4.6085,
|
|
"mean_token_accuracy": 0.23534238040447236,
|
|
"num_tokens": 9048023.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 5.194351005554199,
|
|
"epoch": 4.096234309623431,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003735948261582005,
|
|
"loss": 4.6812,
|
|
"mean_token_accuracy": 0.22248969972133636,
|
|
"num_tokens": 9056988.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 5.195067548751831,
|
|
"epoch": 4.100418410041841,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000373304597263567,
|
|
"loss": 4.6346,
|
|
"mean_token_accuracy": 0.23709141165018083,
|
|
"num_tokens": 9066595.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 5.174516916275024,
|
|
"epoch": 4.104602510460251,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00037301416607387716,
|
|
"loss": 4.6741,
|
|
"mean_token_accuracy": 0.22700658291578293,
|
|
"num_tokens": 9075034.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 5.14497594833374,
|
|
"epoch": 4.108786610878661,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00037272353318679185,
|
|
"loss": 4.6003,
|
|
"mean_token_accuracy": 0.2381413236260414,
|
|
"num_tokens": 9084574.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 5.138711023330688,
|
|
"epoch": 4.112970711297071,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003724326992003867,
|
|
"loss": 4.5722,
|
|
"mean_token_accuracy": 0.2353218138217926,
|
|
"num_tokens": 9093591.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 5.203644609451294,
|
|
"epoch": 4.117154811715481,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003721416647131512,
|
|
"loss": 4.6202,
|
|
"mean_token_accuracy": 0.23942174315452575,
|
|
"num_tokens": 9102439.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 5.117591905593872,
|
|
"epoch": 4.121338912133891,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003718504303239876,
|
|
"loss": 4.491,
|
|
"mean_token_accuracy": 0.2479431688785553,
|
|
"num_tokens": 9111363.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 5.176326274871826,
|
|
"epoch": 4.125523012552302,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003715589966322097,
|
|
"loss": 4.6885,
|
|
"mean_token_accuracy": 0.23365084677934647,
|
|
"num_tokens": 9120417.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 5.164815092086792,
|
|
"epoch": 4.129707112970712,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003712673642375409,
|
|
"loss": 4.6464,
|
|
"mean_token_accuracy": 0.22893903255462647,
|
|
"num_tokens": 9130622.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 5.162829875946045,
|
|
"epoch": 4.133891213389122,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003709755337401138,
|
|
"loss": 4.6451,
|
|
"mean_token_accuracy": 0.22939460277557372,
|
|
"num_tokens": 9139992.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 5.153492116928101,
|
|
"epoch": 4.138075313807532,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00037068350574046867,
|
|
"loss": 4.6016,
|
|
"mean_token_accuracy": 0.23627205789089203,
|
|
"num_tokens": 9149883.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 5.144874334335327,
|
|
"epoch": 4.142259414225942,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00037039128083955227,
|
|
"loss": 4.5866,
|
|
"mean_token_accuracy": 0.23919778913259507,
|
|
"num_tokens": 9159633.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 5.123017263412476,
|
|
"epoch": 4.146443514644352,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003700988596387163,
|
|
"loss": 4.5729,
|
|
"mean_token_accuracy": 0.23572549521923064,
|
|
"num_tokens": 9167780.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 5.179871892929077,
|
|
"epoch": 4.150627615062762,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003698062427397168,
|
|
"loss": 4.5972,
|
|
"mean_token_accuracy": 0.23665703684091569,
|
|
"num_tokens": 9176808.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 5.219782876968384,
|
|
"epoch": 4.154811715481172,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00036951343074471217,
|
|
"loss": 4.6205,
|
|
"mean_token_accuracy": 0.23705107271671294,
|
|
"num_tokens": 9186097.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 5.152626609802246,
|
|
"epoch": 4.158995815899582,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003692204242562624,
|
|
"loss": 4.6637,
|
|
"mean_token_accuracy": 0.22735612839460373,
|
|
"num_tokens": 9194428.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 5.166129636764526,
|
|
"epoch": 4.163179916317992,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003689272238773278,
|
|
"loss": 4.6506,
|
|
"mean_token_accuracy": 0.2307108908891678,
|
|
"num_tokens": 9203156.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 5.1565343856811525,
|
|
"epoch": 4.167364016736402,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00036863383021126756,
|
|
"loss": 4.6101,
|
|
"mean_token_accuracy": 0.23440367877483367,
|
|
"num_tokens": 9212102.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 5.144541835784912,
|
|
"epoch": 4.171548117154812,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00036834024386183875,
|
|
"loss": 4.6445,
|
|
"mean_token_accuracy": 0.2309114307165146,
|
|
"num_tokens": 9221507.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 5.17828369140625,
|
|
"epoch": 4.175732217573222,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003680464654331947,
|
|
"loss": 4.5811,
|
|
"mean_token_accuracy": 0.23645592778921126,
|
|
"num_tokens": 9230543.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 5.2377725601196286,
|
|
"epoch": 4.179916317991632,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003677524955298844,
|
|
"loss": 4.698,
|
|
"mean_token_accuracy": 0.23034105747938155,
|
|
"num_tokens": 9240092.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 5.183539724349975,
|
|
"epoch": 4.184100418410042,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003674583347568505,
|
|
"loss": 4.612,
|
|
"mean_token_accuracy": 0.23191826790571213,
|
|
"num_tokens": 9250145.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 4.184100418410042,
|
|
"eval_entropy": 4.974178003143435,
|
|
"eval_loss": 5.335689067840576,
|
|
"eval_mean_token_accuracy": 0.1969404085282168,
|
|
"eval_num_tokens": 9250145.0,
|
|
"eval_runtime": 2.2754,
|
|
"eval_samples_per_second": 1717.076,
|
|
"eval_steps_per_second": 214.909,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"entropy": 5.197928428649902,
|
|
"epoch": 4.188284518828452,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00036716398371942866,
|
|
"loss": 4.6229,
|
|
"mean_token_accuracy": 0.2342490077018738,
|
|
"num_tokens": 9259643.0,
|
|
"step": 5005
|
|
},
|
|
{
|
|
"entropy": 5.216267347335815,
|
|
"epoch": 4.192468619246862,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003668694430233459,
|
|
"loss": 4.6369,
|
|
"mean_token_accuracy": 0.23395116478204728,
|
|
"num_tokens": 9268471.0,
|
|
"step": 5010
|
|
},
|
|
{
|
|
"entropy": 5.1739284038543705,
|
|
"epoch": 4.196652719665272,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003665747132747198,
|
|
"loss": 4.5785,
|
|
"mean_token_accuracy": 0.23431975096464158,
|
|
"num_tokens": 9276916.0,
|
|
"step": 5015
|
|
},
|
|
{
|
|
"entropy": 5.080595064163208,
|
|
"epoch": 4.200836820083682,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00036627979508005654,
|
|
"loss": 4.5988,
|
|
"mean_token_accuracy": 0.23754304349422456,
|
|
"num_tokens": 9285867.0,
|
|
"step": 5020
|
|
},
|
|
{
|
|
"entropy": 5.215056419372559,
|
|
"epoch": 4.205020920502092,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003659846890462506,
|
|
"loss": 4.7019,
|
|
"mean_token_accuracy": 0.23110421597957612,
|
|
"num_tokens": 9295044.0,
|
|
"step": 5025
|
|
},
|
|
{
|
|
"entropy": 5.257772397994995,
|
|
"epoch": 4.209205020920502,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003656893957805826,
|
|
"loss": 4.6994,
|
|
"mean_token_accuracy": 0.23348197788000108,
|
|
"num_tokens": 9303744.0,
|
|
"step": 5030
|
|
},
|
|
{
|
|
"entropy": 5.161002922058105,
|
|
"epoch": 4.2133891213389125,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003653939158907186,
|
|
"loss": 4.5982,
|
|
"mean_token_accuracy": 0.23504361212253572,
|
|
"num_tokens": 9312996.0,
|
|
"step": 5035
|
|
},
|
|
{
|
|
"entropy": 5.138589191436767,
|
|
"epoch": 4.2175732217573225,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003650982499847088,
|
|
"loss": 4.6047,
|
|
"mean_token_accuracy": 0.23797947019338608,
|
|
"num_tokens": 9321912.0,
|
|
"step": 5040
|
|
},
|
|
{
|
|
"entropy": 5.157578659057617,
|
|
"epoch": 4.2217573221757325,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000364802398670986,
|
|
"loss": 4.6006,
|
|
"mean_token_accuracy": 0.2358361303806305,
|
|
"num_tokens": 9331798.0,
|
|
"step": 5045
|
|
},
|
|
{
|
|
"entropy": 5.176553392410279,
|
|
"epoch": 4.2259414225941425,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00036450636255836474,
|
|
"loss": 4.7134,
|
|
"mean_token_accuracy": 0.22264675945043563,
|
|
"num_tokens": 9342064.0,
|
|
"step": 5050
|
|
},
|
|
{
|
|
"entropy": 5.22673659324646,
|
|
"epoch": 4.2301255230125525,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00036421014225603964,
|
|
"loss": 4.5896,
|
|
"mean_token_accuracy": 0.22956815659999846,
|
|
"num_tokens": 9351494.0,
|
|
"step": 5055
|
|
},
|
|
{
|
|
"entropy": 5.082833337783813,
|
|
"epoch": 4.2343096234309625,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00036391373837358455,
|
|
"loss": 4.5563,
|
|
"mean_token_accuracy": 0.23747244477272034,
|
|
"num_tokens": 9360683.0,
|
|
"step": 5060
|
|
},
|
|
{
|
|
"entropy": 5.121048831939698,
|
|
"epoch": 4.2384937238493725,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00036361715152095095,
|
|
"loss": 4.6413,
|
|
"mean_token_accuracy": 0.22565638720989228,
|
|
"num_tokens": 9370784.0,
|
|
"step": 5065
|
|
},
|
|
{
|
|
"entropy": 5.190201616287231,
|
|
"epoch": 4.2426778242677825,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003633203823084669,
|
|
"loss": 4.6805,
|
|
"mean_token_accuracy": 0.23453124314546586,
|
|
"num_tokens": 9380673.0,
|
|
"step": 5070
|
|
},
|
|
{
|
|
"entropy": 5.245079708099365,
|
|
"epoch": 4.2468619246861925,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00036302343134683566,
|
|
"loss": 4.6626,
|
|
"mean_token_accuracy": 0.23210002481937408,
|
|
"num_tokens": 9390260.0,
|
|
"step": 5075
|
|
},
|
|
{
|
|
"entropy": 5.212691068649292,
|
|
"epoch": 4.2510460251046025,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003627262992471348,
|
|
"loss": 4.7242,
|
|
"mean_token_accuracy": 0.2273639753460884,
|
|
"num_tokens": 9400643.0,
|
|
"step": 5080
|
|
},
|
|
{
|
|
"entropy": 5.132342338562012,
|
|
"epoch": 4.2552301255230125,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003624289866208142,
|
|
"loss": 4.5897,
|
|
"mean_token_accuracy": 0.23307592868804933,
|
|
"num_tokens": 9408898.0,
|
|
"step": 5085
|
|
},
|
|
{
|
|
"entropy": 5.188813066482544,
|
|
"epoch": 4.2594142259414225,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00036213149407969547,
|
|
"loss": 4.6407,
|
|
"mean_token_accuracy": 0.22960346490144729,
|
|
"num_tokens": 9418024.0,
|
|
"step": 5090
|
|
},
|
|
{
|
|
"entropy": 5.28659987449646,
|
|
"epoch": 4.2635983263598325,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003618338222359705,
|
|
"loss": 4.7354,
|
|
"mean_token_accuracy": 0.2183732345700264,
|
|
"num_tokens": 9427548.0,
|
|
"step": 5095
|
|
},
|
|
{
|
|
"entropy": 5.156073760986328,
|
|
"epoch": 4.2677824267782425,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00036153597170220005,
|
|
"loss": 4.562,
|
|
"mean_token_accuracy": 0.23628815710544587,
|
|
"num_tokens": 9436370.0,
|
|
"step": 5100
|
|
},
|
|
{
|
|
"entropy": 5.147730684280395,
|
|
"epoch": 4.2719665271966525,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003612379430913126,
|
|
"loss": 4.6308,
|
|
"mean_token_accuracy": 0.2279965102672577,
|
|
"num_tokens": 9445506.0,
|
|
"step": 5105
|
|
},
|
|
{
|
|
"entropy": 5.178540420532227,
|
|
"epoch": 4.2761506276150625,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003609397370166031,
|
|
"loss": 4.5567,
|
|
"mean_token_accuracy": 0.24433823823928832,
|
|
"num_tokens": 9455288.0,
|
|
"step": 5110
|
|
},
|
|
{
|
|
"entropy": 5.148305368423462,
|
|
"epoch": 4.2803347280334725,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003606413540917319,
|
|
"loss": 4.6642,
|
|
"mean_token_accuracy": 0.23243563771247863,
|
|
"num_tokens": 9464366.0,
|
|
"step": 5115
|
|
},
|
|
{
|
|
"entropy": 5.12925157546997,
|
|
"epoch": 4.2845188284518825,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003603427949307228,
|
|
"loss": 4.5995,
|
|
"mean_token_accuracy": 0.24886890053749083,
|
|
"num_tokens": 9474205.0,
|
|
"step": 5120
|
|
},
|
|
{
|
|
"entropy": 5.1201306819915775,
|
|
"epoch": 4.2887029288702925,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00036004406014796274,
|
|
"loss": 4.604,
|
|
"mean_token_accuracy": 0.23724390119314193,
|
|
"num_tokens": 9483181.0,
|
|
"step": 5125
|
|
},
|
|
{
|
|
"entropy": 5.158036947250366,
|
|
"epoch": 4.292887029288703,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003597451503581998,
|
|
"loss": 4.6003,
|
|
"mean_token_accuracy": 0.23712555766105653,
|
|
"num_tokens": 9492394.0,
|
|
"step": 5130
|
|
},
|
|
{
|
|
"entropy": 5.196285915374756,
|
|
"epoch": 4.297071129707113,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00035944606617654236,
|
|
"loss": 4.6982,
|
|
"mean_token_accuracy": 0.2232271298766136,
|
|
"num_tokens": 9502184.0,
|
|
"step": 5135
|
|
},
|
|
{
|
|
"entropy": 5.129484605789185,
|
|
"epoch": 4.301255230125523,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00035914680821845746,
|
|
"loss": 4.6531,
|
|
"mean_token_accuracy": 0.22931210249662398,
|
|
"num_tokens": 9511399.0,
|
|
"step": 5140
|
|
},
|
|
{
|
|
"entropy": 5.181810331344605,
|
|
"epoch": 4.305439330543933,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00035884737709976995,
|
|
"loss": 4.6872,
|
|
"mean_token_accuracy": 0.23609948307275772,
|
|
"num_tokens": 9520710.0,
|
|
"step": 5145
|
|
},
|
|
{
|
|
"entropy": 5.154796838760376,
|
|
"epoch": 4.309623430962343,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00035854777343666093,
|
|
"loss": 4.5495,
|
|
"mean_token_accuracy": 0.24540453851222993,
|
|
"num_tokens": 9529992.0,
|
|
"step": 5150
|
|
},
|
|
{
|
|
"entropy": 5.114209127426148,
|
|
"epoch": 4.313807531380753,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003582479978456665,
|
|
"loss": 4.5586,
|
|
"mean_token_accuracy": 0.23949185758829117,
|
|
"num_tokens": 9538279.0,
|
|
"step": 5155
|
|
},
|
|
{
|
|
"entropy": 5.121897029876709,
|
|
"epoch": 4.317991631799163,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003579480509436766,
|
|
"loss": 4.5522,
|
|
"mean_token_accuracy": 0.24451960772275924,
|
|
"num_tokens": 9547425.0,
|
|
"step": 5160
|
|
},
|
|
{
|
|
"entropy": 5.090685129165649,
|
|
"epoch": 4.322175732217573,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00035764793334793393,
|
|
"loss": 4.5967,
|
|
"mean_token_accuracy": 0.23420415073633194,
|
|
"num_tokens": 9556657.0,
|
|
"step": 5165
|
|
},
|
|
{
|
|
"entropy": 5.176266813278199,
|
|
"epoch": 4.326359832635983,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00035734764567603216,
|
|
"loss": 4.6882,
|
|
"mean_token_accuracy": 0.23627186417579651,
|
|
"num_tokens": 9565989.0,
|
|
"step": 5170
|
|
},
|
|
{
|
|
"entropy": 5.19171175956726,
|
|
"epoch": 4.330543933054393,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003570471885459149,
|
|
"loss": 4.6134,
|
|
"mean_token_accuracy": 0.2376025587320328,
|
|
"num_tokens": 9575246.0,
|
|
"step": 5175
|
|
},
|
|
{
|
|
"entropy": 5.1593687534332275,
|
|
"epoch": 4.334728033472803,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003567465625758748,
|
|
"loss": 4.5491,
|
|
"mean_token_accuracy": 0.24256271868944168,
|
|
"num_tokens": 9586185.0,
|
|
"step": 5180
|
|
},
|
|
{
|
|
"entropy": 5.087054014205933,
|
|
"epoch": 4.338912133891213,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003564457683845516,
|
|
"loss": 4.5665,
|
|
"mean_token_accuracy": 0.23564370572566987,
|
|
"num_tokens": 9595222.0,
|
|
"step": 5185
|
|
},
|
|
{
|
|
"entropy": 5.19888219833374,
|
|
"epoch": 4.343096234309623,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003561448065909315,
|
|
"loss": 4.7098,
|
|
"mean_token_accuracy": 0.2337145149707794,
|
|
"num_tokens": 9604866.0,
|
|
"step": 5190
|
|
},
|
|
{
|
|
"entropy": 5.117510604858398,
|
|
"epoch": 4.347280334728033,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003558436778143455,
|
|
"loss": 4.5609,
|
|
"mean_token_accuracy": 0.2411576509475708,
|
|
"num_tokens": 9614522.0,
|
|
"step": 5195
|
|
},
|
|
{
|
|
"entropy": 5.184785842895508,
|
|
"epoch": 4.351464435146443,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003555423826744682,
|
|
"loss": 4.7054,
|
|
"mean_token_accuracy": 0.2224012494087219,
|
|
"num_tokens": 9623308.0,
|
|
"step": 5200
|
|
},
|
|
{
|
|
"entropy": 5.211578416824341,
|
|
"epoch": 4.355648535564853,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00035524092179131645,
|
|
"loss": 4.6844,
|
|
"mean_token_accuracy": 0.2289336010813713,
|
|
"num_tokens": 9632887.0,
|
|
"step": 5205
|
|
},
|
|
{
|
|
"entropy": 5.153478527069092,
|
|
"epoch": 4.359832635983263,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003549392957852485,
|
|
"loss": 4.6728,
|
|
"mean_token_accuracy": 0.2340501591563225,
|
|
"num_tokens": 9643218.0,
|
|
"step": 5210
|
|
},
|
|
{
|
|
"entropy": 5.167119884490967,
|
|
"epoch": 4.364016736401673,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003546375052769619,
|
|
"loss": 4.6724,
|
|
"mean_token_accuracy": 0.22194544672966005,
|
|
"num_tokens": 9653175.0,
|
|
"step": 5215
|
|
},
|
|
{
|
|
"entropy": 5.04668664932251,
|
|
"epoch": 4.368200836820083,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003543355508874932,
|
|
"loss": 4.5746,
|
|
"mean_token_accuracy": 0.2391060397028923,
|
|
"num_tokens": 9662431.0,
|
|
"step": 5220
|
|
},
|
|
{
|
|
"entropy": 5.1702228546142575,
|
|
"epoch": 4.372384937238493,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003540334332382159,
|
|
"loss": 4.6154,
|
|
"mean_token_accuracy": 0.2378630071878433,
|
|
"num_tokens": 9671191.0,
|
|
"step": 5225
|
|
},
|
|
{
|
|
"entropy": 5.101017618179322,
|
|
"epoch": 4.376569037656903,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003537311529508396,
|
|
"loss": 4.5837,
|
|
"mean_token_accuracy": 0.23958070427179337,
|
|
"num_tokens": 9680637.0,
|
|
"step": 5230
|
|
},
|
|
{
|
|
"entropy": 5.119145250320434,
|
|
"epoch": 4.380753138075314,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00035342871064740846,
|
|
"loss": 4.5449,
|
|
"mean_token_accuracy": 0.2313419222831726,
|
|
"num_tokens": 9689325.0,
|
|
"step": 5235
|
|
},
|
|
{
|
|
"entropy": 5.1736297607421875,
|
|
"epoch": 4.384937238493724,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00035312610695030017,
|
|
"loss": 4.6511,
|
|
"mean_token_accuracy": 0.22896308451890945,
|
|
"num_tokens": 9698335.0,
|
|
"step": 5240
|
|
},
|
|
{
|
|
"entropy": 5.128261661529541,
|
|
"epoch": 4.389121338912134,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003528233424822245,
|
|
"loss": 4.6104,
|
|
"mean_token_accuracy": 0.23348375260829926,
|
|
"num_tokens": 9706647.0,
|
|
"step": 5245
|
|
},
|
|
{
|
|
"entropy": 5.2425943374633786,
|
|
"epoch": 4.393305439330544,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000352520417866222,
|
|
"loss": 4.6217,
|
|
"mean_token_accuracy": 0.23774940073490142,
|
|
"num_tokens": 9715941.0,
|
|
"step": 5250
|
|
},
|
|
{
|
|
"entropy": 5.131834554672241,
|
|
"epoch": 4.397489539748954,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003522173337256628,
|
|
"loss": 4.6844,
|
|
"mean_token_accuracy": 0.22665181905031204,
|
|
"num_tokens": 9724693.0,
|
|
"step": 5255
|
|
},
|
|
{
|
|
"entropy": 5.126883935928345,
|
|
"epoch": 4.401673640167364,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003519140906842455,
|
|
"loss": 4.6668,
|
|
"mean_token_accuracy": 0.2337368756532669,
|
|
"num_tokens": 9733314.0,
|
|
"step": 5260
|
|
},
|
|
{
|
|
"entropy": 5.288318538665772,
|
|
"epoch": 4.405857740585774,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003516106893659953,
|
|
"loss": 4.7205,
|
|
"mean_token_accuracy": 0.22745369225740433,
|
|
"num_tokens": 9742844.0,
|
|
"step": 5265
|
|
},
|
|
{
|
|
"entropy": 5.149840354919434,
|
|
"epoch": 4.410041841004184,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003513071303952635,
|
|
"loss": 4.5691,
|
|
"mean_token_accuracy": 0.23913919180631638,
|
|
"num_tokens": 9752164.0,
|
|
"step": 5270
|
|
},
|
|
{
|
|
"entropy": 5.089300584793091,
|
|
"epoch": 4.414225941422594,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003510034143967256,
|
|
"loss": 4.6119,
|
|
"mean_token_accuracy": 0.2335230827331543,
|
|
"num_tokens": 9761483.0,
|
|
"step": 5275
|
|
},
|
|
{
|
|
"entropy": 5.1148268699646,
|
|
"epoch": 4.418410041841004,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00035069954199538027,
|
|
"loss": 4.5546,
|
|
"mean_token_accuracy": 0.2350556656718254,
|
|
"num_tokens": 9769998.0,
|
|
"step": 5280
|
|
},
|
|
{
|
|
"entropy": 5.19239296913147,
|
|
"epoch": 4.422594142259414,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00035039551381654816,
|
|
"loss": 4.6277,
|
|
"mean_token_accuracy": 0.23258234262466432,
|
|
"num_tokens": 9778965.0,
|
|
"step": 5285
|
|
},
|
|
{
|
|
"entropy": 5.174081420898437,
|
|
"epoch": 4.426778242677824,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00035009133048587024,
|
|
"loss": 4.6594,
|
|
"mean_token_accuracy": 0.22826888114213945,
|
|
"num_tokens": 9788543.0,
|
|
"step": 5290
|
|
},
|
|
{
|
|
"entropy": 5.124127197265625,
|
|
"epoch": 4.430962343096234,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003497869926293069,
|
|
"loss": 4.5452,
|
|
"mean_token_accuracy": 0.2363219752907753,
|
|
"num_tokens": 9797744.0,
|
|
"step": 5295
|
|
},
|
|
{
|
|
"entropy": 5.050665187835693,
|
|
"epoch": 4.435146443514644,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003494825008731367,
|
|
"loss": 4.603,
|
|
"mean_token_accuracy": 0.23799787014722823,
|
|
"num_tokens": 9807528.0,
|
|
"step": 5300
|
|
},
|
|
{
|
|
"entropy": 5.073057174682617,
|
|
"epoch": 4.439330543933054,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00034917785584395457,
|
|
"loss": 4.6167,
|
|
"mean_token_accuracy": 0.2303186133503914,
|
|
"num_tokens": 9816078.0,
|
|
"step": 5305
|
|
},
|
|
{
|
|
"entropy": 5.1732430934906,
|
|
"epoch": 4.443514644351464,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000348873058168671,
|
|
"loss": 4.5669,
|
|
"mean_token_accuracy": 0.23577651381492615,
|
|
"num_tokens": 9825199.0,
|
|
"step": 5310
|
|
},
|
|
{
|
|
"entropy": 5.22291784286499,
|
|
"epoch": 4.447698744769874,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003485681084745106,
|
|
"loss": 4.7317,
|
|
"mean_token_accuracy": 0.23174676299095154,
|
|
"num_tokens": 9835294.0,
|
|
"step": 5315
|
|
},
|
|
{
|
|
"entropy": 5.141643428802491,
|
|
"epoch": 4.451882845188284,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000348263007389011,
|
|
"loss": 4.6007,
|
|
"mean_token_accuracy": 0.22639082223176957,
|
|
"num_tokens": 9844449.0,
|
|
"step": 5320
|
|
},
|
|
{
|
|
"entropy": 5.10732069015503,
|
|
"epoch": 4.456066945606695,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00034795775554002093,
|
|
"loss": 4.6213,
|
|
"mean_token_accuracy": 0.2389781132340431,
|
|
"num_tokens": 9852930.0,
|
|
"step": 5325
|
|
},
|
|
{
|
|
"entropy": 5.184037065505981,
|
|
"epoch": 4.460251046025105,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00034765235355569974,
|
|
"loss": 4.5611,
|
|
"mean_token_accuracy": 0.23982613235712053,
|
|
"num_tokens": 9862168.0,
|
|
"step": 5330
|
|
},
|
|
{
|
|
"entropy": 5.175882148742676,
|
|
"epoch": 4.464435146443515,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003473468020645155,
|
|
"loss": 4.6986,
|
|
"mean_token_accuracy": 0.22665689289569854,
|
|
"num_tokens": 9871702.0,
|
|
"step": 5335
|
|
},
|
|
{
|
|
"entropy": 5.242555713653564,
|
|
"epoch": 4.468619246861925,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003470411016952442,
|
|
"loss": 4.7111,
|
|
"mean_token_accuracy": 0.22500183135271073,
|
|
"num_tokens": 9880502.0,
|
|
"step": 5340
|
|
},
|
|
{
|
|
"entropy": 5.212836122512817,
|
|
"epoch": 4.472803347280335,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000346735253076968,
|
|
"loss": 4.6236,
|
|
"mean_token_accuracy": 0.23511965125799178,
|
|
"num_tokens": 9889093.0,
|
|
"step": 5345
|
|
},
|
|
{
|
|
"entropy": 5.075055360794067,
|
|
"epoch": 4.476987447698745,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003464292568390743,
|
|
"loss": 4.5185,
|
|
"mean_token_accuracy": 0.23809715807437898,
|
|
"num_tokens": 9897113.0,
|
|
"step": 5350
|
|
},
|
|
{
|
|
"entropy": 5.07142825126648,
|
|
"epoch": 4.481171548117155,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003461231136112539,
|
|
"loss": 4.5944,
|
|
"mean_token_accuracy": 0.23317514657974242,
|
|
"num_tokens": 9906478.0,
|
|
"step": 5355
|
|
},
|
|
{
|
|
"entropy": 5.180588960647583,
|
|
"epoch": 4.485355648535565,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00034581682402350057,
|
|
"loss": 4.6315,
|
|
"mean_token_accuracy": 0.22680198401212692,
|
|
"num_tokens": 9916143.0,
|
|
"step": 5360
|
|
},
|
|
{
|
|
"entropy": 5.189208269119263,
|
|
"epoch": 4.489539748953975,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00034551038870610885,
|
|
"loss": 4.629,
|
|
"mean_token_accuracy": 0.23351462036371232,
|
|
"num_tokens": 9925675.0,
|
|
"step": 5365
|
|
},
|
|
{
|
|
"entropy": 5.151645994186401,
|
|
"epoch": 4.493723849372385,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00034520380828967346,
|
|
"loss": 4.5432,
|
|
"mean_token_accuracy": 0.23736103475093842,
|
|
"num_tokens": 9934398.0,
|
|
"step": 5370
|
|
},
|
|
{
|
|
"entropy": 5.072487640380859,
|
|
"epoch": 4.497907949790795,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003448970834050877,
|
|
"loss": 4.5967,
|
|
"mean_token_accuracy": 0.23938234597444535,
|
|
"num_tokens": 9943066.0,
|
|
"step": 5375
|
|
},
|
|
{
|
|
"entropy": 5.039367198944092,
|
|
"epoch": 4.502092050209205,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003445902146835419,
|
|
"loss": 4.4998,
|
|
"mean_token_accuracy": 0.24156050235033036,
|
|
"num_tokens": 9951565.0,
|
|
"step": 5380
|
|
},
|
|
{
|
|
"entropy": 5.162084341049194,
|
|
"epoch": 4.506276150627615,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003442832027565227,
|
|
"loss": 4.6595,
|
|
"mean_token_accuracy": 0.22815261781215668,
|
|
"num_tokens": 9960997.0,
|
|
"step": 5385
|
|
},
|
|
{
|
|
"entropy": 5.203349637985229,
|
|
"epoch": 4.510460251046025,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00034397604825581115,
|
|
"loss": 4.6473,
|
|
"mean_token_accuracy": 0.23459515273571013,
|
|
"num_tokens": 9970188.0,
|
|
"step": 5390
|
|
},
|
|
{
|
|
"entropy": 5.183899784088135,
|
|
"epoch": 4.514644351464435,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003436687518134819,
|
|
"loss": 4.6387,
|
|
"mean_token_accuracy": 0.23821180015802385,
|
|
"num_tokens": 9978749.0,
|
|
"step": 5395
|
|
},
|
|
{
|
|
"entropy": 5.203292274475098,
|
|
"epoch": 4.518828451882845,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003433613140619016,
|
|
"loss": 4.7105,
|
|
"mean_token_accuracy": 0.23374868035316468,
|
|
"num_tokens": 9988107.0,
|
|
"step": 5400
|
|
},
|
|
{
|
|
"entropy": 5.116972160339356,
|
|
"epoch": 4.523012552301255,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003430537356337277,
|
|
"loss": 4.5328,
|
|
"mean_token_accuracy": 0.23596750050783158,
|
|
"num_tokens": 9996127.0,
|
|
"step": 5405
|
|
},
|
|
{
|
|
"entropy": 5.149571561813355,
|
|
"epoch": 4.527196652719665,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003427460171619073,
|
|
"loss": 4.6554,
|
|
"mean_token_accuracy": 0.23126720935106276,
|
|
"num_tokens": 10005482.0,
|
|
"step": 5410
|
|
},
|
|
{
|
|
"entropy": 5.23683910369873,
|
|
"epoch": 4.531380753138075,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00034243815927967536,
|
|
"loss": 4.6791,
|
|
"mean_token_accuracy": 0.22387361973524095,
|
|
"num_tokens": 10014765.0,
|
|
"step": 5415
|
|
},
|
|
{
|
|
"entropy": 5.1605527877807615,
|
|
"epoch": 4.535564853556485,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00034213016262055396,
|
|
"loss": 4.6376,
|
|
"mean_token_accuracy": 0.22425616085529326,
|
|
"num_tokens": 10022897.0,
|
|
"step": 5420
|
|
},
|
|
{
|
|
"entropy": 5.177553939819336,
|
|
"epoch": 4.539748953974895,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003418220278183507,
|
|
"loss": 4.7055,
|
|
"mean_token_accuracy": 0.22182696908712388,
|
|
"num_tokens": 10032281.0,
|
|
"step": 5425
|
|
},
|
|
{
|
|
"entropy": 5.105132961273194,
|
|
"epoch": 4.543933054393305,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00034151375550715753,
|
|
"loss": 4.5553,
|
|
"mean_token_accuracy": 0.24526498913764955,
|
|
"num_tokens": 10041194.0,
|
|
"step": 5430
|
|
},
|
|
{
|
|
"entropy": 5.0449440479278564,
|
|
"epoch": 4.548117154811716,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00034120534632134917,
|
|
"loss": 4.6074,
|
|
"mean_token_accuracy": 0.22482048273086547,
|
|
"num_tokens": 10050062.0,
|
|
"step": 5435
|
|
},
|
|
{
|
|
"entropy": 5.125740385055542,
|
|
"epoch": 4.552301255230126,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00034089680089558226,
|
|
"loss": 4.6979,
|
|
"mean_token_accuracy": 0.23846469968557357,
|
|
"num_tokens": 10059747.0,
|
|
"step": 5440
|
|
},
|
|
{
|
|
"entropy": 5.208222150802612,
|
|
"epoch": 4.556485355648536,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00034058811986479367,
|
|
"loss": 4.7051,
|
|
"mean_token_accuracy": 0.22276211380958558,
|
|
"num_tokens": 10069351.0,
|
|
"step": 5445
|
|
},
|
|
{
|
|
"entropy": 5.170335483551026,
|
|
"epoch": 4.560669456066946,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00034027930386419935,
|
|
"loss": 4.6303,
|
|
"mean_token_accuracy": 0.2280181735754013,
|
|
"num_tokens": 10078727.0,
|
|
"step": 5450
|
|
},
|
|
{
|
|
"entropy": 5.111565446853637,
|
|
"epoch": 4.564853556485356,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00033997035352929285,
|
|
"loss": 4.6067,
|
|
"mean_token_accuracy": 0.22922297716140747,
|
|
"num_tokens": 10088151.0,
|
|
"step": 5455
|
|
},
|
|
{
|
|
"entropy": 5.10242075920105,
|
|
"epoch": 4.569037656903766,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003396612694958444,
|
|
"loss": 4.563,
|
|
"mean_token_accuracy": 0.23516346663236617,
|
|
"num_tokens": 10097311.0,
|
|
"step": 5460
|
|
},
|
|
{
|
|
"entropy": 5.153835582733154,
|
|
"epoch": 4.573221757322176,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00033935205239989927,
|
|
"loss": 4.6906,
|
|
"mean_token_accuracy": 0.22511650621891022,
|
|
"num_tokens": 10107474.0,
|
|
"step": 5465
|
|
},
|
|
{
|
|
"entropy": 5.151067686080933,
|
|
"epoch": 4.577405857740586,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00033904270287777645,
|
|
"loss": 4.6313,
|
|
"mean_token_accuracy": 0.2434908017516136,
|
|
"num_tokens": 10116305.0,
|
|
"step": 5470
|
|
},
|
|
{
|
|
"entropy": 5.1845605850219725,
|
|
"epoch": 4.581589958158996,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003387332215660675,
|
|
"loss": 4.6603,
|
|
"mean_token_accuracy": 0.2315655156970024,
|
|
"num_tokens": 10125947.0,
|
|
"step": 5475
|
|
},
|
|
{
|
|
"entropy": 5.112628030776977,
|
|
"epoch": 4.585774058577406,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00033842360910163536,
|
|
"loss": 4.6507,
|
|
"mean_token_accuracy": 0.23253672271966935,
|
|
"num_tokens": 10134510.0,
|
|
"step": 5480
|
|
},
|
|
{
|
|
"entropy": 5.203630113601685,
|
|
"epoch": 4.589958158995816,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003381138661216125,
|
|
"loss": 4.7096,
|
|
"mean_token_accuracy": 0.22690674811601638,
|
|
"num_tokens": 10143429.0,
|
|
"step": 5485
|
|
},
|
|
{
|
|
"entropy": 5.226968288421631,
|
|
"epoch": 4.594142259414226,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00033780399326340026,
|
|
"loss": 4.6888,
|
|
"mean_token_accuracy": 0.22486475706100464,
|
|
"num_tokens": 10153405.0,
|
|
"step": 5490
|
|
},
|
|
{
|
|
"entropy": 5.219865083694458,
|
|
"epoch": 4.598326359832636,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00033749399116466723,
|
|
"loss": 4.6403,
|
|
"mean_token_accuracy": 0.23650808930397033,
|
|
"num_tokens": 10162334.0,
|
|
"step": 5495
|
|
},
|
|
{
|
|
"entropy": 5.161908149719238,
|
|
"epoch": 4.602510460251046,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003371838604633479,
|
|
"loss": 4.6394,
|
|
"mean_token_accuracy": 0.23301680088043214,
|
|
"num_tokens": 10171604.0,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"epoch": 4.602510460251046,
|
|
"eval_entropy": 4.863926047927762,
|
|
"eval_loss": 5.31691837310791,
|
|
"eval_mean_token_accuracy": 0.20018291464246854,
|
|
"eval_num_tokens": 10171604.0,
|
|
"eval_runtime": 2.0732,
|
|
"eval_samples_per_second": 1884.503,
|
|
"eval_steps_per_second": 235.864,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"entropy": 5.1064619541168215,
|
|
"epoch": 4.606694560669456,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00033687360179764125,
|
|
"loss": 4.5929,
|
|
"mean_token_accuracy": 0.23554231971502304,
|
|
"num_tokens": 10180659.0,
|
|
"step": 5505
|
|
},
|
|
{
|
|
"entropy": 5.205877780914307,
|
|
"epoch": 4.610878661087866,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003365632158060098,
|
|
"loss": 4.6724,
|
|
"mean_token_accuracy": 0.22620372623205184,
|
|
"num_tokens": 10190226.0,
|
|
"step": 5510
|
|
},
|
|
{
|
|
"entropy": 5.129668855667115,
|
|
"epoch": 4.615062761506276,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00033625270312717805,
|
|
"loss": 4.5699,
|
|
"mean_token_accuracy": 0.23736884593963622,
|
|
"num_tokens": 10198806.0,
|
|
"step": 5515
|
|
},
|
|
{
|
|
"entropy": 5.092312622070312,
|
|
"epoch": 4.619246861924686,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003359420644001311,
|
|
"loss": 4.5823,
|
|
"mean_token_accuracy": 0.2372826889157295,
|
|
"num_tokens": 10207927.0,
|
|
"step": 5520
|
|
},
|
|
{
|
|
"entropy": 5.041866493225098,
|
|
"epoch": 4.623430962343097,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00033563130026411357,
|
|
"loss": 4.5209,
|
|
"mean_token_accuracy": 0.2455090433359146,
|
|
"num_tokens": 10216241.0,
|
|
"step": 5525
|
|
},
|
|
{
|
|
"entropy": 5.1193431377410885,
|
|
"epoch": 4.627615062761507,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00033532041135862816,
|
|
"loss": 4.7036,
|
|
"mean_token_accuracy": 0.22740448713302613,
|
|
"num_tokens": 10225241.0,
|
|
"step": 5530
|
|
},
|
|
{
|
|
"entropy": 5.162849235534668,
|
|
"epoch": 4.631799163179917,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00033500939832343407,
|
|
"loss": 4.6556,
|
|
"mean_token_accuracy": 0.23305455297231675,
|
|
"num_tokens": 10233968.0,
|
|
"step": 5535
|
|
},
|
|
{
|
|
"entropy": 5.3041990280151365,
|
|
"epoch": 4.635983263598327,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003346982617985463,
|
|
"loss": 4.7341,
|
|
"mean_token_accuracy": 0.2286246508359909,
|
|
"num_tokens": 10243589.0,
|
|
"step": 5540
|
|
},
|
|
{
|
|
"entropy": 5.033699798583984,
|
|
"epoch": 4.640167364016737,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003343870024242337,
|
|
"loss": 4.5684,
|
|
"mean_token_accuracy": 0.23400027453899383,
|
|
"num_tokens": 10252416.0,
|
|
"step": 5545
|
|
},
|
|
{
|
|
"entropy": 5.101994323730469,
|
|
"epoch": 4.644351464435147,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00033407562084101795,
|
|
"loss": 4.6842,
|
|
"mean_token_accuracy": 0.22345473915338515,
|
|
"num_tokens": 10262213.0,
|
|
"step": 5550
|
|
},
|
|
{
|
|
"entropy": 5.236071681976318,
|
|
"epoch": 4.648535564853557,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00033376411768967247,
|
|
"loss": 4.7703,
|
|
"mean_token_accuracy": 0.22288744449615477,
|
|
"num_tokens": 10272774.0,
|
|
"step": 5555
|
|
},
|
|
{
|
|
"entropy": 5.15610146522522,
|
|
"epoch": 4.652719665271967,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00033345249361122056,
|
|
"loss": 4.5894,
|
|
"mean_token_accuracy": 0.2345762699842453,
|
|
"num_tokens": 10281286.0,
|
|
"step": 5560
|
|
},
|
|
{
|
|
"entropy": 5.099101305007935,
|
|
"epoch": 4.656903765690377,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00033314074924693446,
|
|
"loss": 4.6022,
|
|
"mean_token_accuracy": 0.23509458899497987,
|
|
"num_tokens": 10290101.0,
|
|
"step": 5565
|
|
},
|
|
{
|
|
"entropy": 5.090194845199585,
|
|
"epoch": 4.661087866108787,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00033282888523833394,
|
|
"loss": 4.614,
|
|
"mean_token_accuracy": 0.23831456154584885,
|
|
"num_tokens": 10299580.0,
|
|
"step": 5570
|
|
},
|
|
{
|
|
"entropy": 5.1654315948486325,
|
|
"epoch": 4.665271966527197,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00033251690222718504,
|
|
"loss": 4.6218,
|
|
"mean_token_accuracy": 0.24284600466489792,
|
|
"num_tokens": 10307860.0,
|
|
"step": 5575
|
|
},
|
|
{
|
|
"entropy": 5.151300430297852,
|
|
"epoch": 4.669456066945607,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003322048008554986,
|
|
"loss": 4.6309,
|
|
"mean_token_accuracy": 0.23785845190286636,
|
|
"num_tokens": 10316546.0,
|
|
"step": 5580
|
|
},
|
|
{
|
|
"entropy": 5.196021461486817,
|
|
"epoch": 4.673640167364017,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003318925817655291,
|
|
"loss": 4.7397,
|
|
"mean_token_accuracy": 0.23192634284496308,
|
|
"num_tokens": 10326012.0,
|
|
"step": 5585
|
|
},
|
|
{
|
|
"entropy": 5.203530311584473,
|
|
"epoch": 4.677824267782427,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003315802455997732,
|
|
"loss": 4.6788,
|
|
"mean_token_accuracy": 0.22818757593631744,
|
|
"num_tokens": 10334771.0,
|
|
"step": 5590
|
|
},
|
|
{
|
|
"entropy": 5.210732555389404,
|
|
"epoch": 4.682008368200837,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00033126779300096855,
|
|
"loss": 4.677,
|
|
"mean_token_accuracy": 0.22540452033281327,
|
|
"num_tokens": 10343416.0,
|
|
"step": 5595
|
|
},
|
|
{
|
|
"entropy": 5.252113342285156,
|
|
"epoch": 4.686192468619247,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003309552246120924,
|
|
"loss": 4.7725,
|
|
"mean_token_accuracy": 0.22517641633749008,
|
|
"num_tokens": 10353539.0,
|
|
"step": 5600
|
|
},
|
|
{
|
|
"entropy": 5.130997180938721,
|
|
"epoch": 4.690376569037657,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003306425410763601,
|
|
"loss": 4.5948,
|
|
"mean_token_accuracy": 0.2333295837044716,
|
|
"num_tokens": 10362020.0,
|
|
"step": 5605
|
|
},
|
|
{
|
|
"entropy": 5.1160571575164795,
|
|
"epoch": 4.694560669456067,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003303297430372241,
|
|
"loss": 4.5592,
|
|
"mean_token_accuracy": 0.23600176572799683,
|
|
"num_tokens": 10371277.0,
|
|
"step": 5610
|
|
},
|
|
{
|
|
"entropy": 5.103133773803711,
|
|
"epoch": 4.698744769874477,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003300168311383727,
|
|
"loss": 4.6212,
|
|
"mean_token_accuracy": 0.2366684779524803,
|
|
"num_tokens": 10380680.0,
|
|
"step": 5615
|
|
},
|
|
{
|
|
"entropy": 5.116869497299194,
|
|
"epoch": 4.702928870292887,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00032970380602372805,
|
|
"loss": 4.5578,
|
|
"mean_token_accuracy": 0.24462981671094894,
|
|
"num_tokens": 10389576.0,
|
|
"step": 5620
|
|
},
|
|
{
|
|
"entropy": 5.151834774017334,
|
|
"epoch": 4.707112970711297,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003293906683374456,
|
|
"loss": 4.6678,
|
|
"mean_token_accuracy": 0.22870336920022966,
|
|
"num_tokens": 10399182.0,
|
|
"step": 5625
|
|
},
|
|
{
|
|
"entropy": 5.239958906173706,
|
|
"epoch": 4.711297071129707,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003290774187239124,
|
|
"loss": 4.7979,
|
|
"mean_token_accuracy": 0.2265862852334976,
|
|
"num_tokens": 10409595.0,
|
|
"step": 5630
|
|
},
|
|
{
|
|
"entropy": 5.219075727462768,
|
|
"epoch": 4.7154811715481175,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00032876405782774574,
|
|
"loss": 4.7508,
|
|
"mean_token_accuracy": 0.21667829602956773,
|
|
"num_tokens": 10419057.0,
|
|
"step": 5635
|
|
},
|
|
{
|
|
"entropy": 5.129452180862427,
|
|
"epoch": 4.7196652719665275,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003284505862937919,
|
|
"loss": 4.5389,
|
|
"mean_token_accuracy": 0.2415228322148323,
|
|
"num_tokens": 10428090.0,
|
|
"step": 5640
|
|
},
|
|
{
|
|
"entropy": 5.204846668243408,
|
|
"epoch": 4.7238493723849375,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000328137004767125,
|
|
"loss": 4.6847,
|
|
"mean_token_accuracy": 0.22422901242971421,
|
|
"num_tokens": 10438088.0,
|
|
"step": 5645
|
|
},
|
|
{
|
|
"entropy": 5.096403789520264,
|
|
"epoch": 4.7280334728033475,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003278233138930454,
|
|
"loss": 4.6813,
|
|
"mean_token_accuracy": 0.2310093730688095,
|
|
"num_tokens": 10447934.0,
|
|
"step": 5650
|
|
},
|
|
{
|
|
"entropy": 5.108078479766846,
|
|
"epoch": 4.7322175732217575,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003275095143170784,
|
|
"loss": 4.5414,
|
|
"mean_token_accuracy": 0.2413547784090042,
|
|
"num_tokens": 10456088.0,
|
|
"step": 5655
|
|
},
|
|
{
|
|
"entropy": 5.126994180679321,
|
|
"epoch": 4.7364016736401675,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00032719560668497316,
|
|
"loss": 4.5577,
|
|
"mean_token_accuracy": 0.24296634197235106,
|
|
"num_tokens": 10465096.0,
|
|
"step": 5660
|
|
},
|
|
{
|
|
"entropy": 5.093886804580689,
|
|
"epoch": 4.7405857740585775,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00032688159164270114,
|
|
"loss": 4.6489,
|
|
"mean_token_accuracy": 0.23613524734973906,
|
|
"num_tokens": 10473516.0,
|
|
"step": 5665
|
|
},
|
|
{
|
|
"entropy": 5.127963638305664,
|
|
"epoch": 4.7447698744769875,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00032656746983645473,
|
|
"loss": 4.6322,
|
|
"mean_token_accuracy": 0.23411721736192703,
|
|
"num_tokens": 10482651.0,
|
|
"step": 5670
|
|
},
|
|
{
|
|
"entropy": 5.199563455581665,
|
|
"epoch": 4.7489539748953975,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003262532419126461,
|
|
"loss": 4.6585,
|
|
"mean_token_accuracy": 0.22149005085229873,
|
|
"num_tokens": 10491831.0,
|
|
"step": 5675
|
|
},
|
|
{
|
|
"entropy": 5.121698951721191,
|
|
"epoch": 4.7531380753138075,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00032593890851790593,
|
|
"loss": 4.6464,
|
|
"mean_token_accuracy": 0.2321746364235878,
|
|
"num_tokens": 10502070.0,
|
|
"step": 5680
|
|
},
|
|
{
|
|
"entropy": 5.13078031539917,
|
|
"epoch": 4.7573221757322175,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003256244702990817,
|
|
"loss": 4.642,
|
|
"mean_token_accuracy": 0.22281866818666457,
|
|
"num_tokens": 10510638.0,
|
|
"step": 5685
|
|
},
|
|
{
|
|
"entropy": 5.065270042419433,
|
|
"epoch": 4.7615062761506275,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00032530992790323664,
|
|
"loss": 4.586,
|
|
"mean_token_accuracy": 0.23764345347881316,
|
|
"num_tokens": 10519305.0,
|
|
"step": 5690
|
|
},
|
|
{
|
|
"entropy": 5.128038740158081,
|
|
"epoch": 4.7656903765690375,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00032499528197764855,
|
|
"loss": 4.6257,
|
|
"mean_token_accuracy": 0.23217536509037018,
|
|
"num_tokens": 10528844.0,
|
|
"step": 5695
|
|
},
|
|
{
|
|
"entropy": 5.157577228546143,
|
|
"epoch": 4.7698744769874475,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00032468053316980815,
|
|
"loss": 4.7285,
|
|
"mean_token_accuracy": 0.22778344452381133,
|
|
"num_tokens": 10538260.0,
|
|
"step": 5700
|
|
},
|
|
{
|
|
"entropy": 5.14331579208374,
|
|
"epoch": 4.7740585774058575,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003243656821274178,
|
|
"loss": 4.6717,
|
|
"mean_token_accuracy": 0.23287709802389145,
|
|
"num_tokens": 10547733.0,
|
|
"step": 5705
|
|
},
|
|
{
|
|
"entropy": 5.176742362976074,
|
|
"epoch": 4.7782426778242675,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003240507294983905,
|
|
"loss": 4.6306,
|
|
"mean_token_accuracy": 0.23628492206335067,
|
|
"num_tokens": 10558337.0,
|
|
"step": 5710
|
|
},
|
|
{
|
|
"entropy": 5.140197849273681,
|
|
"epoch": 4.7824267782426775,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00032373567593084804,
|
|
"loss": 4.6305,
|
|
"mean_token_accuracy": 0.22883946746587752,
|
|
"num_tokens": 10568374.0,
|
|
"step": 5715
|
|
},
|
|
{
|
|
"entropy": 5.113533449172974,
|
|
"epoch": 4.786610878661088,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00032342052207311986,
|
|
"loss": 4.698,
|
|
"mean_token_accuracy": 0.23221004456281663,
|
|
"num_tokens": 10578448.0,
|
|
"step": 5720
|
|
},
|
|
{
|
|
"entropy": 5.203852796554566,
|
|
"epoch": 4.790794979079498,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00032310526857374204,
|
|
"loss": 4.7305,
|
|
"mean_token_accuracy": 0.22515777498483658,
|
|
"num_tokens": 10586803.0,
|
|
"step": 5725
|
|
},
|
|
{
|
|
"entropy": 5.233892393112183,
|
|
"epoch": 4.794979079497908,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003227899160814556,
|
|
"loss": 4.7037,
|
|
"mean_token_accuracy": 0.23291639983654022,
|
|
"num_tokens": 10596596.0,
|
|
"step": 5730
|
|
},
|
|
{
|
|
"entropy": 5.167309713363648,
|
|
"epoch": 4.799163179916318,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003224744652452054,
|
|
"loss": 4.5909,
|
|
"mean_token_accuracy": 0.23328935205936432,
|
|
"num_tokens": 10605462.0,
|
|
"step": 5735
|
|
},
|
|
{
|
|
"entropy": 5.219909763336181,
|
|
"epoch": 4.803347280334728,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003221589167141382,
|
|
"loss": 4.7383,
|
|
"mean_token_accuracy": 0.2255614846944809,
|
|
"num_tokens": 10614908.0,
|
|
"step": 5740
|
|
},
|
|
{
|
|
"entropy": 5.182645988464356,
|
|
"epoch": 4.807531380753138,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003218432711376025,
|
|
"loss": 4.6956,
|
|
"mean_token_accuracy": 0.23108094334602355,
|
|
"num_tokens": 10624389.0,
|
|
"step": 5745
|
|
},
|
|
{
|
|
"entropy": 5.098988819122314,
|
|
"epoch": 4.811715481171548,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00032152752916514605,
|
|
"loss": 4.6508,
|
|
"mean_token_accuracy": 0.23277100026607514,
|
|
"num_tokens": 10633228.0,
|
|
"step": 5750
|
|
},
|
|
{
|
|
"entropy": 5.126742649078369,
|
|
"epoch": 4.815899581589958,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00032121169144651496,
|
|
"loss": 4.7049,
|
|
"mean_token_accuracy": 0.23087301403284072,
|
|
"num_tokens": 10641996.0,
|
|
"step": 5755
|
|
},
|
|
{
|
|
"entropy": 5.143943786621094,
|
|
"epoch": 4.820083682008368,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003208957586316525,
|
|
"loss": 4.6147,
|
|
"mean_token_accuracy": 0.23408320546150208,
|
|
"num_tokens": 10651724.0,
|
|
"step": 5760
|
|
},
|
|
{
|
|
"entropy": 5.169669914245605,
|
|
"epoch": 4.824267782426778,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003205797313706977,
|
|
"loss": 4.6714,
|
|
"mean_token_accuracy": 0.2290802299976349,
|
|
"num_tokens": 10661150.0,
|
|
"step": 5765
|
|
},
|
|
{
|
|
"entropy": 5.1223126411437985,
|
|
"epoch": 4.828451882845188,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00032026361031398386,
|
|
"loss": 4.687,
|
|
"mean_token_accuracy": 0.22545844167470933,
|
|
"num_tokens": 10669890.0,
|
|
"step": 5770
|
|
},
|
|
{
|
|
"entropy": 5.178488111495971,
|
|
"epoch": 4.832635983263598,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00031994739611203716,
|
|
"loss": 4.604,
|
|
"mean_token_accuracy": 0.23099617063999175,
|
|
"num_tokens": 10678939.0,
|
|
"step": 5775
|
|
},
|
|
{
|
|
"entropy": 5.098205709457398,
|
|
"epoch": 4.836820083682008,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00031963108941557576,
|
|
"loss": 4.5821,
|
|
"mean_token_accuracy": 0.23433270752429963,
|
|
"num_tokens": 10687465.0,
|
|
"step": 5780
|
|
},
|
|
{
|
|
"entropy": 5.1570251941680905,
|
|
"epoch": 4.841004184100418,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003193146908755078,
|
|
"loss": 4.6704,
|
|
"mean_token_accuracy": 0.2329047366976738,
|
|
"num_tokens": 10696930.0,
|
|
"step": 5785
|
|
},
|
|
{
|
|
"entropy": 5.151654958724976,
|
|
"epoch": 4.845188284518828,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00031899820114293073,
|
|
"loss": 4.5932,
|
|
"mean_token_accuracy": 0.23911323696374892,
|
|
"num_tokens": 10706820.0,
|
|
"step": 5790
|
|
},
|
|
{
|
|
"entropy": 5.146882581710815,
|
|
"epoch": 4.849372384937238,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00031868162086912946,
|
|
"loss": 4.589,
|
|
"mean_token_accuracy": 0.23868007957935333,
|
|
"num_tokens": 10715528.0,
|
|
"step": 5795
|
|
},
|
|
{
|
|
"entropy": 5.057752990722657,
|
|
"epoch": 4.853556485355648,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003183649507055754,
|
|
"loss": 4.5882,
|
|
"mean_token_accuracy": 0.2337643325328827,
|
|
"num_tokens": 10724807.0,
|
|
"step": 5800
|
|
},
|
|
{
|
|
"entropy": 5.1919629096984865,
|
|
"epoch": 4.857740585774058,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00031804819130392477,
|
|
"loss": 4.6711,
|
|
"mean_token_accuracy": 0.22639439404010772,
|
|
"num_tokens": 10733754.0,
|
|
"step": 5805
|
|
},
|
|
{
|
|
"entropy": 5.11691312789917,
|
|
"epoch": 4.861924686192468,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003177313433160175,
|
|
"loss": 4.6512,
|
|
"mean_token_accuracy": 0.23355279862880707,
|
|
"num_tokens": 10742256.0,
|
|
"step": 5810
|
|
},
|
|
{
|
|
"entropy": 5.104293012619019,
|
|
"epoch": 4.866108786610878,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003174144073938757,
|
|
"loss": 4.6142,
|
|
"mean_token_accuracy": 0.24992673397064208,
|
|
"num_tokens": 10751025.0,
|
|
"step": 5815
|
|
},
|
|
{
|
|
"entropy": 5.165078115463257,
|
|
"epoch": 4.870292887029288,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00031709738418970275,
|
|
"loss": 4.5947,
|
|
"mean_token_accuracy": 0.23556801080703735,
|
|
"num_tokens": 10759970.0,
|
|
"step": 5820
|
|
},
|
|
{
|
|
"entropy": 5.159179592132569,
|
|
"epoch": 4.874476987447698,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003167802743558813,
|
|
"loss": 4.6164,
|
|
"mean_token_accuracy": 0.2327914297580719,
|
|
"num_tokens": 10768531.0,
|
|
"step": 5825
|
|
},
|
|
{
|
|
"entropy": 5.183176517486572,
|
|
"epoch": 4.878661087866108,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003164630785449724,
|
|
"loss": 4.6965,
|
|
"mean_token_accuracy": 0.22682576030492782,
|
|
"num_tokens": 10778802.0,
|
|
"step": 5830
|
|
},
|
|
{
|
|
"entropy": 5.140844249725342,
|
|
"epoch": 4.882845188284519,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00031614579740971415,
|
|
"loss": 4.6267,
|
|
"mean_token_accuracy": 0.2291090488433838,
|
|
"num_tokens": 10788121.0,
|
|
"step": 5835
|
|
},
|
|
{
|
|
"entropy": 5.200862407684326,
|
|
"epoch": 4.887029288702929,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00031582843160302,
|
|
"loss": 4.7435,
|
|
"mean_token_accuracy": 0.22744652181863784,
|
|
"num_tokens": 10797652.0,
|
|
"step": 5840
|
|
},
|
|
{
|
|
"entropy": 5.187236499786377,
|
|
"epoch": 4.891213389121339,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00031551098177797784,
|
|
"loss": 4.6519,
|
|
"mean_token_accuracy": 0.2307753324508667,
|
|
"num_tokens": 10807564.0,
|
|
"step": 5845
|
|
},
|
|
{
|
|
"entropy": 5.176135873794555,
|
|
"epoch": 4.895397489539749,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003151934485878482,
|
|
"loss": 4.6233,
|
|
"mean_token_accuracy": 0.23188139498233795,
|
|
"num_tokens": 10816666.0,
|
|
"step": 5850
|
|
},
|
|
{
|
|
"entropy": 5.1069159507751465,
|
|
"epoch": 4.899581589958159,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00031487583268606354,
|
|
"loss": 4.6316,
|
|
"mean_token_accuracy": 0.23608867228031158,
|
|
"num_tokens": 10825639.0,
|
|
"step": 5855
|
|
},
|
|
{
|
|
"entropy": 5.109216499328613,
|
|
"epoch": 4.903765690376569,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003145581347262262,
|
|
"loss": 4.6666,
|
|
"mean_token_accuracy": 0.22962073981761932,
|
|
"num_tokens": 10834681.0,
|
|
"step": 5860
|
|
},
|
|
{
|
|
"entropy": 5.173245763778686,
|
|
"epoch": 4.907949790794979,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00031424035536210755,
|
|
"loss": 4.678,
|
|
"mean_token_accuracy": 0.23020405918359757,
|
|
"num_tokens": 10844465.0,
|
|
"step": 5865
|
|
},
|
|
{
|
|
"entropy": 5.138600635528564,
|
|
"epoch": 4.912133891213389,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00031392249524764645,
|
|
"loss": 4.6928,
|
|
"mean_token_accuracy": 0.23055480271577836,
|
|
"num_tokens": 10853327.0,
|
|
"step": 5870
|
|
},
|
|
{
|
|
"entropy": 5.091273784637451,
|
|
"epoch": 4.916317991631799,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00031360455503694787,
|
|
"loss": 4.5868,
|
|
"mean_token_accuracy": 0.2408416301012039,
|
|
"num_tokens": 10861865.0,
|
|
"step": 5875
|
|
},
|
|
{
|
|
"entropy": 5.219302558898926,
|
|
"epoch": 4.920502092050209,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003132865353842816,
|
|
"loss": 4.6845,
|
|
"mean_token_accuracy": 0.2291072890162468,
|
|
"num_tokens": 10870794.0,
|
|
"step": 5880
|
|
},
|
|
{
|
|
"entropy": 5.179766988754272,
|
|
"epoch": 4.924686192468619,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00031296843694408107,
|
|
"loss": 4.5945,
|
|
"mean_token_accuracy": 0.23731759041547776,
|
|
"num_tokens": 10879626.0,
|
|
"step": 5885
|
|
},
|
|
{
|
|
"entropy": 5.088613557815552,
|
|
"epoch": 4.928870292887029,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003126502603709417,
|
|
"loss": 4.5613,
|
|
"mean_token_accuracy": 0.23593639731407165,
|
|
"num_tokens": 10888359.0,
|
|
"step": 5890
|
|
},
|
|
{
|
|
"entropy": 5.13113923072815,
|
|
"epoch": 4.933054393305439,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003123320063196196,
|
|
"loss": 4.6122,
|
|
"mean_token_accuracy": 0.23082680106163025,
|
|
"num_tokens": 10897174.0,
|
|
"step": 5895
|
|
},
|
|
{
|
|
"entropy": 5.083344030380249,
|
|
"epoch": 4.937238493723849,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003120136754450306,
|
|
"loss": 4.6537,
|
|
"mean_token_accuracy": 0.2222206711769104,
|
|
"num_tokens": 10906660.0,
|
|
"step": 5900
|
|
},
|
|
{
|
|
"entropy": 5.16254940032959,
|
|
"epoch": 4.941422594142259,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00031169526840224836,
|
|
"loss": 4.6906,
|
|
"mean_token_accuracy": 0.22295212745666504,
|
|
"num_tokens": 10915990.0,
|
|
"step": 5905
|
|
},
|
|
{
|
|
"entropy": 5.3010660171508786,
|
|
"epoch": 4.945606694560669,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003113767858465034,
|
|
"loss": 4.8039,
|
|
"mean_token_accuracy": 0.210269033908844,
|
|
"num_tokens": 10926355.0,
|
|
"step": 5910
|
|
},
|
|
{
|
|
"entropy": 5.113926029205322,
|
|
"epoch": 4.949790794979079,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00031105822843318173,
|
|
"loss": 4.6606,
|
|
"mean_token_accuracy": 0.2306258425116539,
|
|
"num_tokens": 10935854.0,
|
|
"step": 5915
|
|
},
|
|
{
|
|
"entropy": 5.105067920684815,
|
|
"epoch": 4.95397489539749,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003107395968178232,
|
|
"loss": 4.5984,
|
|
"mean_token_accuracy": 0.24083320200443267,
|
|
"num_tokens": 10944903.0,
|
|
"step": 5920
|
|
},
|
|
{
|
|
"entropy": 5.1270866870880125,
|
|
"epoch": 4.9581589958159,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003104208916561203,
|
|
"loss": 4.6253,
|
|
"mean_token_accuracy": 0.22781944274902344,
|
|
"num_tokens": 10954009.0,
|
|
"step": 5925
|
|
},
|
|
{
|
|
"entropy": 5.249894618988037,
|
|
"epoch": 4.96234309623431,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00031010211360391727,
|
|
"loss": 4.778,
|
|
"mean_token_accuracy": 0.21555300652980805,
|
|
"num_tokens": 10963174.0,
|
|
"step": 5930
|
|
},
|
|
{
|
|
"entropy": 5.237838649749756,
|
|
"epoch": 4.96652719665272,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000309783263317208,
|
|
"loss": 4.6953,
|
|
"mean_token_accuracy": 0.22311269193887712,
|
|
"num_tokens": 10972822.0,
|
|
"step": 5935
|
|
},
|
|
{
|
|
"entropy": 5.096681737899781,
|
|
"epoch": 4.97071129707113,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003094643414521349,
|
|
"loss": 4.5767,
|
|
"mean_token_accuracy": 0.23556727916002274,
|
|
"num_tokens": 10981758.0,
|
|
"step": 5940
|
|
},
|
|
{
|
|
"entropy": 5.115640020370483,
|
|
"epoch": 4.97489539748954,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003091453486649883,
|
|
"loss": 4.6421,
|
|
"mean_token_accuracy": 0.2290348470211029,
|
|
"num_tokens": 10991730.0,
|
|
"step": 5945
|
|
},
|
|
{
|
|
"entropy": 5.192991399765015,
|
|
"epoch": 4.97907949790795,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00030882628561220376,
|
|
"loss": 4.6601,
|
|
"mean_token_accuracy": 0.23275590538978577,
|
|
"num_tokens": 11000827.0,
|
|
"step": 5950
|
|
},
|
|
{
|
|
"entropy": 5.193814277648926,
|
|
"epoch": 4.98326359832636,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00030850715295036187,
|
|
"loss": 4.6387,
|
|
"mean_token_accuracy": 0.2392078921198845,
|
|
"num_tokens": 11010544.0,
|
|
"step": 5955
|
|
},
|
|
{
|
|
"entropy": 5.121958446502686,
|
|
"epoch": 4.98744769874477,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003081879513361863,
|
|
"loss": 4.6614,
|
|
"mean_token_accuracy": 0.22900831699371338,
|
|
"num_tokens": 11019826.0,
|
|
"step": 5960
|
|
},
|
|
{
|
|
"entropy": 5.158448791503906,
|
|
"epoch": 4.99163179916318,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003078686814265426,
|
|
"loss": 4.6584,
|
|
"mean_token_accuracy": 0.23952475786209107,
|
|
"num_tokens": 11028408.0,
|
|
"step": 5965
|
|
},
|
|
{
|
|
"entropy": 5.208160924911499,
|
|
"epoch": 4.99581589958159,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000307549343878437,
|
|
"loss": 4.7049,
|
|
"mean_token_accuracy": 0.2292460933327675,
|
|
"num_tokens": 11038564.0,
|
|
"step": 5970
|
|
},
|
|
{
|
|
"entropy": 5.19297080039978,
|
|
"epoch": 5.0,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00030722993934901487,
|
|
"loss": 4.5734,
|
|
"mean_token_accuracy": 0.24420374482870102,
|
|
"num_tokens": 11046610.0,
|
|
"step": 5975
|
|
},
|
|
{
|
|
"entropy": 4.990713357925415,
|
|
"epoch": 5.00418410041841,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00030691046849555927,
|
|
"loss": 4.3111,
|
|
"mean_token_accuracy": 0.2607599079608917,
|
|
"num_tokens": 11055929.0,
|
|
"step": 5980
|
|
},
|
|
{
|
|
"entropy": 5.122709941864014,
|
|
"epoch": 5.00836820083682,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00030659093197549,
|
|
"loss": 4.4033,
|
|
"mean_token_accuracy": 0.2586848050355911,
|
|
"num_tokens": 11064748.0,
|
|
"step": 5985
|
|
},
|
|
{
|
|
"entropy": 5.055655527114868,
|
|
"epoch": 5.01255230125523,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003062713304463616,
|
|
"loss": 4.3649,
|
|
"mean_token_accuracy": 0.2594725891947746,
|
|
"num_tokens": 11073707.0,
|
|
"step": 5990
|
|
},
|
|
{
|
|
"entropy": 5.129342174530029,
|
|
"epoch": 5.01673640167364,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003059516645658629,
|
|
"loss": 4.4069,
|
|
"mean_token_accuracy": 0.26009887307882307,
|
|
"num_tokens": 11082967.0,
|
|
"step": 5995
|
|
},
|
|
{
|
|
"entropy": 5.104907703399658,
|
|
"epoch": 5.02092050209205,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003056319349918146,
|
|
"loss": 4.401,
|
|
"mean_token_accuracy": 0.25492548644542695,
|
|
"num_tokens": 11092105.0,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"epoch": 5.02092050209205,
|
|
"eval_entropy": 4.9126372771272875,
|
|
"eval_loss": 5.304167747497559,
|
|
"eval_mean_token_accuracy": 0.20116637529831966,
|
|
"eval_num_tokens": 11092105.0,
|
|
"eval_runtime": 2.2857,
|
|
"eval_samples_per_second": 1709.3,
|
|
"eval_steps_per_second": 213.936,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"entropy": 5.168640947341919,
|
|
"epoch": 5.02510460251046,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00030531214238216906,
|
|
"loss": 4.4854,
|
|
"mean_token_accuracy": 0.2428906574845314,
|
|
"num_tokens": 11100806.0,
|
|
"step": 6005
|
|
},
|
|
{
|
|
"entropy": 5.026292228698731,
|
|
"epoch": 5.02928870292887,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00030499228739500793,
|
|
"loss": 4.3448,
|
|
"mean_token_accuracy": 0.2627560392022133,
|
|
"num_tokens": 11109779.0,
|
|
"step": 6010
|
|
},
|
|
{
|
|
"entropy": 5.079067373275757,
|
|
"epoch": 5.03347280334728,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00030467237068854136,
|
|
"loss": 4.43,
|
|
"mean_token_accuracy": 0.24878764599561692,
|
|
"num_tokens": 11118838.0,
|
|
"step": 6015
|
|
},
|
|
{
|
|
"entropy": 5.070307922363281,
|
|
"epoch": 5.03765690376569,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003043523929211064,
|
|
"loss": 4.3369,
|
|
"mean_token_accuracy": 0.2662981912493706,
|
|
"num_tokens": 11127029.0,
|
|
"step": 6020
|
|
},
|
|
{
|
|
"entropy": 5.099070835113525,
|
|
"epoch": 5.0418410041841,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000304032354751166,
|
|
"loss": 4.4019,
|
|
"mean_token_accuracy": 0.25687565952539443,
|
|
"num_tokens": 11136073.0,
|
|
"step": 6025
|
|
},
|
|
{
|
|
"entropy": 5.117370128631592,
|
|
"epoch": 5.046025104602511,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003037122568373071,
|
|
"loss": 4.4692,
|
|
"mean_token_accuracy": 0.24470825791358947,
|
|
"num_tokens": 11144807.0,
|
|
"step": 6030
|
|
},
|
|
{
|
|
"entropy": 5.136521625518799,
|
|
"epoch": 5.050209205020921,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003033920998382399,
|
|
"loss": 4.4375,
|
|
"mean_token_accuracy": 0.24109227508306502,
|
|
"num_tokens": 11154308.0,
|
|
"step": 6035
|
|
},
|
|
{
|
|
"entropy": 5.1296220302581785,
|
|
"epoch": 5.054393305439331,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003030718844127959,
|
|
"loss": 4.487,
|
|
"mean_token_accuracy": 0.24212776124477386,
|
|
"num_tokens": 11163752.0,
|
|
"step": 6040
|
|
},
|
|
{
|
|
"entropy": 5.083835744857788,
|
|
"epoch": 5.058577405857741,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003027516112199271,
|
|
"loss": 4.4262,
|
|
"mean_token_accuracy": 0.2497353583574295,
|
|
"num_tokens": 11173763.0,
|
|
"step": 6045
|
|
},
|
|
{
|
|
"entropy": 5.129863166809082,
|
|
"epoch": 5.062761506276151,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00030243128091870413,
|
|
"loss": 4.4729,
|
|
"mean_token_accuracy": 0.24305574148893355,
|
|
"num_tokens": 11183577.0,
|
|
"step": 6050
|
|
},
|
|
{
|
|
"entropy": 5.080097579956055,
|
|
"epoch": 5.066945606694561,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003021108941683152,
|
|
"loss": 4.4239,
|
|
"mean_token_accuracy": 0.25050615817308425,
|
|
"num_tokens": 11192708.0,
|
|
"step": 6055
|
|
},
|
|
{
|
|
"entropy": 5.086006736755371,
|
|
"epoch": 5.071129707112971,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00030179045162806494,
|
|
"loss": 4.476,
|
|
"mean_token_accuracy": 0.24437903016805648,
|
|
"num_tokens": 11201803.0,
|
|
"step": 6060
|
|
},
|
|
{
|
|
"entropy": 4.989393138885498,
|
|
"epoch": 5.075313807531381,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00030146995395737247,
|
|
"loss": 4.3528,
|
|
"mean_token_accuracy": 0.25479735285043714,
|
|
"num_tokens": 11211208.0,
|
|
"step": 6065
|
|
},
|
|
{
|
|
"entropy": 5.101287460327148,
|
|
"epoch": 5.079497907949791,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003011494018157704,
|
|
"loss": 4.4125,
|
|
"mean_token_accuracy": 0.2610247850418091,
|
|
"num_tokens": 11221091.0,
|
|
"step": 6070
|
|
},
|
|
{
|
|
"entropy": 5.122677612304687,
|
|
"epoch": 5.083682008368201,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003008287958629036,
|
|
"loss": 4.4787,
|
|
"mean_token_accuracy": 0.2495693415403366,
|
|
"num_tokens": 11230084.0,
|
|
"step": 6075
|
|
},
|
|
{
|
|
"entropy": 5.143477582931519,
|
|
"epoch": 5.087866108786611,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00030050813675852747,
|
|
"loss": 4.512,
|
|
"mean_token_accuracy": 0.2495734676718712,
|
|
"num_tokens": 11239310.0,
|
|
"step": 6080
|
|
},
|
|
{
|
|
"entropy": 5.103065061569214,
|
|
"epoch": 5.092050209205021,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00030018742516250697,
|
|
"loss": 4.4143,
|
|
"mean_token_accuracy": 0.2519488990306854,
|
|
"num_tokens": 11248738.0,
|
|
"step": 6085
|
|
},
|
|
{
|
|
"entropy": 5.13110122680664,
|
|
"epoch": 5.096234309623431,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000299866661734815,
|
|
"loss": 4.3907,
|
|
"mean_token_accuracy": 0.259561188519001,
|
|
"num_tokens": 11257813.0,
|
|
"step": 6090
|
|
},
|
|
{
|
|
"entropy": 5.12505202293396,
|
|
"epoch": 5.100418410041841,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00029954584713553107,
|
|
"loss": 4.4637,
|
|
"mean_token_accuracy": 0.244238044321537,
|
|
"num_tokens": 11266475.0,
|
|
"step": 6095
|
|
},
|
|
{
|
|
"entropy": 5.050375747680664,
|
|
"epoch": 5.104602510460251,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00029922498202484014,
|
|
"loss": 4.4363,
|
|
"mean_token_accuracy": 0.250096532702446,
|
|
"num_tokens": 11276011.0,
|
|
"step": 6100
|
|
},
|
|
{
|
|
"entropy": 5.116611576080322,
|
|
"epoch": 5.108786610878661,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000298904067063031,
|
|
"loss": 4.5116,
|
|
"mean_token_accuracy": 0.24519597738981247,
|
|
"num_tokens": 11284619.0,
|
|
"step": 6105
|
|
},
|
|
{
|
|
"entropy": 5.133109855651855,
|
|
"epoch": 5.112970711297071,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002985831029104949,
|
|
"loss": 4.4377,
|
|
"mean_token_accuracy": 0.2593027129769325,
|
|
"num_tokens": 11293105.0,
|
|
"step": 6110
|
|
},
|
|
{
|
|
"entropy": 5.089435720443726,
|
|
"epoch": 5.117154811715481,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002982620902277246,
|
|
"loss": 4.4036,
|
|
"mean_token_accuracy": 0.2509296879172325,
|
|
"num_tokens": 11302717.0,
|
|
"step": 6115
|
|
},
|
|
{
|
|
"entropy": 5.168311071395874,
|
|
"epoch": 5.121338912133891,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00029794102967531263,
|
|
"loss": 4.5106,
|
|
"mean_token_accuracy": 0.25531783849000933,
|
|
"num_tokens": 11312709.0,
|
|
"step": 6120
|
|
},
|
|
{
|
|
"entropy": 5.072576999664307,
|
|
"epoch": 5.125523012552302,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00029761992191394985,
|
|
"loss": 4.4269,
|
|
"mean_token_accuracy": 0.2568598002195358,
|
|
"num_tokens": 11322040.0,
|
|
"step": 6125
|
|
},
|
|
{
|
|
"entropy": 5.126050615310669,
|
|
"epoch": 5.129707112970712,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00029729876760442456,
|
|
"loss": 4.5258,
|
|
"mean_token_accuracy": 0.24892989546060562,
|
|
"num_tokens": 11332029.0,
|
|
"step": 6130
|
|
},
|
|
{
|
|
"entropy": 5.0832263946533205,
|
|
"epoch": 5.133891213389122,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002969775674076206,
|
|
"loss": 4.4803,
|
|
"mean_token_accuracy": 0.24350361078977584,
|
|
"num_tokens": 11340960.0,
|
|
"step": 6135
|
|
},
|
|
{
|
|
"entropy": 5.166423177719116,
|
|
"epoch": 5.138075313807532,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002966563219845164,
|
|
"loss": 4.509,
|
|
"mean_token_accuracy": 0.24638283550739287,
|
|
"num_tokens": 11349497.0,
|
|
"step": 6140
|
|
},
|
|
{
|
|
"entropy": 5.141126441955566,
|
|
"epoch": 5.142259414225942,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002963350319961832,
|
|
"loss": 4.4525,
|
|
"mean_token_accuracy": 0.25133324712514876,
|
|
"num_tokens": 11358477.0,
|
|
"step": 6145
|
|
},
|
|
{
|
|
"entropy": 5.117936515808106,
|
|
"epoch": 5.146443514644352,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002960136981037844,
|
|
"loss": 4.4447,
|
|
"mean_token_accuracy": 0.24857595115900039,
|
|
"num_tokens": 11367635.0,
|
|
"step": 6150
|
|
},
|
|
{
|
|
"entropy": 5.104688882827759,
|
|
"epoch": 5.150627615062762,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002956923209685733,
|
|
"loss": 4.4128,
|
|
"mean_token_accuracy": 0.2564963772892952,
|
|
"num_tokens": 11376737.0,
|
|
"step": 6155
|
|
},
|
|
{
|
|
"entropy": 5.0598899841308596,
|
|
"epoch": 5.154811715481172,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002953709012518925,
|
|
"loss": 4.4809,
|
|
"mean_token_accuracy": 0.248466332256794,
|
|
"num_tokens": 11386756.0,
|
|
"step": 6160
|
|
},
|
|
{
|
|
"entropy": 5.050072050094604,
|
|
"epoch": 5.158995815899582,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00029504943961517204,
|
|
"loss": 4.3754,
|
|
"mean_token_accuracy": 0.25987738072872163,
|
|
"num_tokens": 11394889.0,
|
|
"step": 6165
|
|
},
|
|
{
|
|
"entropy": 5.161651754379273,
|
|
"epoch": 5.163179916317992,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00029472793671992824,
|
|
"loss": 4.5483,
|
|
"mean_token_accuracy": 0.2477656528353691,
|
|
"num_tokens": 11404759.0,
|
|
"step": 6170
|
|
},
|
|
{
|
|
"entropy": 5.1545439720153805,
|
|
"epoch": 5.167364016736402,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00029440639322776246,
|
|
"loss": 4.4786,
|
|
"mean_token_accuracy": 0.24309803992509843,
|
|
"num_tokens": 11413739.0,
|
|
"step": 6175
|
|
},
|
|
{
|
|
"entropy": 5.07716064453125,
|
|
"epoch": 5.171548117154812,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00029408480980035943,
|
|
"loss": 4.498,
|
|
"mean_token_accuracy": 0.24956365525722504,
|
|
"num_tokens": 11422835.0,
|
|
"step": 6180
|
|
},
|
|
{
|
|
"entropy": 5.144747591018676,
|
|
"epoch": 5.175732217573222,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002937631870994863,
|
|
"loss": 4.4684,
|
|
"mean_token_accuracy": 0.2600972667336464,
|
|
"num_tokens": 11432613.0,
|
|
"step": 6185
|
|
},
|
|
{
|
|
"entropy": 5.053437280654907,
|
|
"epoch": 5.179916317991632,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00029344152578699074,
|
|
"loss": 4.4374,
|
|
"mean_token_accuracy": 0.24530546814203263,
|
|
"num_tokens": 11441297.0,
|
|
"step": 6190
|
|
},
|
|
{
|
|
"entropy": 5.052908086776734,
|
|
"epoch": 5.184100418410042,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00029311982652479993,
|
|
"loss": 4.4026,
|
|
"mean_token_accuracy": 0.24695399552583694,
|
|
"num_tokens": 11450409.0,
|
|
"step": 6195
|
|
},
|
|
{
|
|
"entropy": 5.1531538486480715,
|
|
"epoch": 5.188284518828452,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002927980899749195,
|
|
"loss": 4.5294,
|
|
"mean_token_accuracy": 0.23643595725297928,
|
|
"num_tokens": 11460132.0,
|
|
"step": 6200
|
|
},
|
|
{
|
|
"entropy": 5.107728958129883,
|
|
"epoch": 5.192468619246862,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002924763167994312,
|
|
"loss": 4.4848,
|
|
"mean_token_accuracy": 0.24909121692180633,
|
|
"num_tokens": 11468984.0,
|
|
"step": 6205
|
|
},
|
|
{
|
|
"entropy": 5.176926612854004,
|
|
"epoch": 5.196652719665272,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002921545076604928,
|
|
"loss": 4.5082,
|
|
"mean_token_accuracy": 0.2516917183995247,
|
|
"num_tokens": 11478371.0,
|
|
"step": 6210
|
|
},
|
|
{
|
|
"entropy": 5.079139375686646,
|
|
"epoch": 5.200836820083682,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00029183266322033565,
|
|
"loss": 4.4935,
|
|
"mean_token_accuracy": 0.24820893704891206,
|
|
"num_tokens": 11487924.0,
|
|
"step": 6215
|
|
},
|
|
{
|
|
"entropy": 5.12629017829895,
|
|
"epoch": 5.205020920502092,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002915107841412638,
|
|
"loss": 4.4778,
|
|
"mean_token_accuracy": 0.24564517438411712,
|
|
"num_tokens": 11497892.0,
|
|
"step": 6220
|
|
},
|
|
{
|
|
"entropy": 5.1255455493927,
|
|
"epoch": 5.209205020920502,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002911888710856528,
|
|
"loss": 4.3701,
|
|
"mean_token_accuracy": 0.2515479430556297,
|
|
"num_tokens": 11507266.0,
|
|
"step": 6225
|
|
},
|
|
{
|
|
"entropy": 5.093895292282104,
|
|
"epoch": 5.2133891213389125,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002908669247159477,
|
|
"loss": 4.3686,
|
|
"mean_token_accuracy": 0.25735197216272354,
|
|
"num_tokens": 11516672.0,
|
|
"step": 6230
|
|
},
|
|
{
|
|
"entropy": 5.089012575149536,
|
|
"epoch": 5.2175732217573225,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00029054494569466266,
|
|
"loss": 4.4734,
|
|
"mean_token_accuracy": 0.2501315072178841,
|
|
"num_tokens": 11525865.0,
|
|
"step": 6235
|
|
},
|
|
{
|
|
"entropy": 5.171494626998902,
|
|
"epoch": 5.2217573221757325,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00029022293468437853,
|
|
"loss": 4.5165,
|
|
"mean_token_accuracy": 0.25490754246711733,
|
|
"num_tokens": 11534910.0,
|
|
"step": 6240
|
|
},
|
|
{
|
|
"entropy": 4.994580507278442,
|
|
"epoch": 5.2259414225941425,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00028990089234774236,
|
|
"loss": 4.3695,
|
|
"mean_token_accuracy": 0.26490993052721024,
|
|
"num_tokens": 11543507.0,
|
|
"step": 6245
|
|
},
|
|
{
|
|
"entropy": 5.066041421890259,
|
|
"epoch": 5.2301255230125525,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00028957881934746544,
|
|
"loss": 4.3931,
|
|
"mean_token_accuracy": 0.2560276448726654,
|
|
"num_tokens": 11552562.0,
|
|
"step": 6250
|
|
},
|
|
{
|
|
"entropy": 5.141301441192627,
|
|
"epoch": 5.2343096234309625,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00028925671634632216,
|
|
"loss": 4.4611,
|
|
"mean_token_accuracy": 0.25285003930330274,
|
|
"num_tokens": 11561931.0,
|
|
"step": 6255
|
|
},
|
|
{
|
|
"entropy": 5.110105085372925,
|
|
"epoch": 5.2384937238493725,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00028893458400714876,
|
|
"loss": 4.4848,
|
|
"mean_token_accuracy": 0.2518938586115837,
|
|
"num_tokens": 11570639.0,
|
|
"step": 6260
|
|
},
|
|
{
|
|
"entropy": 5.1124378681182865,
|
|
"epoch": 5.2426778242677825,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002886124229928418,
|
|
"loss": 4.4177,
|
|
"mean_token_accuracy": 0.24677827060222626,
|
|
"num_tokens": 11579796.0,
|
|
"step": 6265
|
|
},
|
|
{
|
|
"entropy": 5.143540477752685,
|
|
"epoch": 5.2468619246861925,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00028829023396635683,
|
|
"loss": 4.5102,
|
|
"mean_token_accuracy": 0.2503318011760712,
|
|
"num_tokens": 11589343.0,
|
|
"step": 6270
|
|
},
|
|
{
|
|
"entropy": 5.136748600006103,
|
|
"epoch": 5.2510460251046025,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002879680175907071,
|
|
"loss": 4.5043,
|
|
"mean_token_accuracy": 0.24949095994234086,
|
|
"num_tokens": 11598633.0,
|
|
"step": 6275
|
|
},
|
|
{
|
|
"entropy": 5.0850992679595945,
|
|
"epoch": 5.2552301255230125,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002876457745289621,
|
|
"loss": 4.3806,
|
|
"mean_token_accuracy": 0.2603974059224129,
|
|
"num_tokens": 11607746.0,
|
|
"step": 6280
|
|
},
|
|
{
|
|
"entropy": 5.120009326934815,
|
|
"epoch": 5.2594142259414225,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002873235054442463,
|
|
"loss": 4.4893,
|
|
"mean_token_accuracy": 0.2473870560526848,
|
|
"num_tokens": 11616574.0,
|
|
"step": 6285
|
|
},
|
|
{
|
|
"entropy": 5.10708270072937,
|
|
"epoch": 5.2635983263598325,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00028700121099973765,
|
|
"loss": 4.4964,
|
|
"mean_token_accuracy": 0.24739756286144257,
|
|
"num_tokens": 11625918.0,
|
|
"step": 6290
|
|
},
|
|
{
|
|
"entropy": 5.157765340805054,
|
|
"epoch": 5.2677824267782425,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002866788918586661,
|
|
"loss": 4.4826,
|
|
"mean_token_accuracy": 0.247305990755558,
|
|
"num_tokens": 11634255.0,
|
|
"step": 6295
|
|
},
|
|
{
|
|
"entropy": 5.093867444992066,
|
|
"epoch": 5.2719665271966525,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00028635654868431286,
|
|
"loss": 4.4733,
|
|
"mean_token_accuracy": 0.24342973232269288,
|
|
"num_tokens": 11643532.0,
|
|
"step": 6300
|
|
},
|
|
{
|
|
"entropy": 5.087070560455322,
|
|
"epoch": 5.2761506276150625,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00028603418214000827,
|
|
"loss": 4.4297,
|
|
"mean_token_accuracy": 0.24676128923892976,
|
|
"num_tokens": 11653104.0,
|
|
"step": 6305
|
|
},
|
|
{
|
|
"entropy": 5.136249732971192,
|
|
"epoch": 5.2803347280334725,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002857117928891308,
|
|
"loss": 4.5192,
|
|
"mean_token_accuracy": 0.24575028866529464,
|
|
"num_tokens": 11662752.0,
|
|
"step": 6310
|
|
},
|
|
{
|
|
"entropy": 5.02214617729187,
|
|
"epoch": 5.2845188284518825,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00028538938159510587,
|
|
"loss": 4.3827,
|
|
"mean_token_accuracy": 0.2589569121599197,
|
|
"num_tokens": 11671566.0,
|
|
"step": 6315
|
|
},
|
|
{
|
|
"entropy": 5.120374488830566,
|
|
"epoch": 5.2887029288702925,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00028506694892140385,
|
|
"loss": 4.4706,
|
|
"mean_token_accuracy": 0.25485328435897825,
|
|
"num_tokens": 11681770.0,
|
|
"step": 6320
|
|
},
|
|
{
|
|
"entropy": 5.067798185348511,
|
|
"epoch": 5.292887029288703,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00028474449553153944,
|
|
"loss": 4.4776,
|
|
"mean_token_accuracy": 0.24201791286468505,
|
|
"num_tokens": 11691177.0,
|
|
"step": 6325
|
|
},
|
|
{
|
|
"entropy": 5.006298351287842,
|
|
"epoch": 5.297071129707113,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00028442202208906993,
|
|
"loss": 4.3972,
|
|
"mean_token_accuracy": 0.25996510237455367,
|
|
"num_tokens": 11700461.0,
|
|
"step": 6330
|
|
},
|
|
{
|
|
"entropy": 5.06848726272583,
|
|
"epoch": 5.301255230125523,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002840995292575938,
|
|
"loss": 4.4689,
|
|
"mean_token_accuracy": 0.24115028530359267,
|
|
"num_tokens": 11708783.0,
|
|
"step": 6335
|
|
},
|
|
{
|
|
"entropy": 5.127911043167114,
|
|
"epoch": 5.305439330543933,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0002837770177007494,
|
|
"loss": 4.4967,
|
|
"mean_token_accuracy": 0.24294487833976747,
|
|
"num_tokens": 11717419.0,
|
|
"step": 6340
|
|
},
|
|
{
|
|
"entropy": 5.142444086074829,
|
|
"epoch": 5.309623430962343,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00028345448808221374,
|
|
"loss": 4.5398,
|
|
"mean_token_accuracy": 0.24216974377632142,
|
|
"num_tokens": 11727014.0,
|
|
"step": 6345
|
|
},
|
|
{
|
|
"entropy": 5.103272247314453,
|
|
"epoch": 5.313807531380753,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00028313194106570095,
|
|
"loss": 4.4795,
|
|
"mean_token_accuracy": 0.24722935855388642,
|
|
"num_tokens": 11736718.0,
|
|
"step": 6350
|
|
},
|
|
{
|
|
"entropy": 5.207290840148926,
|
|
"epoch": 5.317991631799163,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002828093773149607,
|
|
"loss": 4.5332,
|
|
"mean_token_accuracy": 0.23772156089544297,
|
|
"num_tokens": 11747236.0,
|
|
"step": 6355
|
|
},
|
|
{
|
|
"entropy": 5.1225443363189695,
|
|
"epoch": 5.322175732217573,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00028248679749377754,
|
|
"loss": 4.4476,
|
|
"mean_token_accuracy": 0.24965260028839112,
|
|
"num_tokens": 11756912.0,
|
|
"step": 6360
|
|
},
|
|
{
|
|
"entropy": 5.0315266132354735,
|
|
"epoch": 5.326359832635983,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00028216420226596884,
|
|
"loss": 4.4165,
|
|
"mean_token_accuracy": 0.25815355181694033,
|
|
"num_tokens": 11766035.0,
|
|
"step": 6365
|
|
},
|
|
{
|
|
"entropy": 5.010232734680176,
|
|
"epoch": 5.330543933054393,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002818415922953837,
|
|
"loss": 4.3621,
|
|
"mean_token_accuracy": 0.2595366328954697,
|
|
"num_tokens": 11775934.0,
|
|
"step": 6370
|
|
},
|
|
{
|
|
"entropy": 5.0897420883178714,
|
|
"epoch": 5.334728033472803,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00028151896824590156,
|
|
"loss": 4.4505,
|
|
"mean_token_accuracy": 0.24970170855522156,
|
|
"num_tokens": 11784549.0,
|
|
"step": 6375
|
|
},
|
|
{
|
|
"entropy": 5.1013298511505125,
|
|
"epoch": 5.338912133891213,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002811963307814309,
|
|
"loss": 4.5072,
|
|
"mean_token_accuracy": 0.24299495071172714,
|
|
"num_tokens": 11794369.0,
|
|
"step": 6380
|
|
},
|
|
{
|
|
"entropy": 5.078360652923584,
|
|
"epoch": 5.343096234309623,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00028087368056590767,
|
|
"loss": 4.4777,
|
|
"mean_token_accuracy": 0.2452381432056427,
|
|
"num_tokens": 11803221.0,
|
|
"step": 6385
|
|
},
|
|
{
|
|
"entropy": 5.161740112304687,
|
|
"epoch": 5.347280334728033,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002805510182632942,
|
|
"loss": 4.4807,
|
|
"mean_token_accuracy": 0.24622632414102555,
|
|
"num_tokens": 11812811.0,
|
|
"step": 6390
|
|
},
|
|
{
|
|
"entropy": 5.14595136642456,
|
|
"epoch": 5.351464435146443,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002802283445375775,
|
|
"loss": 4.4989,
|
|
"mean_token_accuracy": 0.2507223039865494,
|
|
"num_tokens": 11821522.0,
|
|
"step": 6395
|
|
},
|
|
{
|
|
"entropy": 5.054875373840332,
|
|
"epoch": 5.355648535564853,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00027990566005276846,
|
|
"loss": 4.4299,
|
|
"mean_token_accuracy": 0.25411662459373474,
|
|
"num_tokens": 11830021.0,
|
|
"step": 6400
|
|
},
|
|
{
|
|
"entropy": 5.047816944122315,
|
|
"epoch": 5.359832635983263,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00027958296547289966,
|
|
"loss": 4.4111,
|
|
"mean_token_accuracy": 0.2576426908373833,
|
|
"num_tokens": 11838612.0,
|
|
"step": 6405
|
|
},
|
|
{
|
|
"entropy": 5.0519373416900635,
|
|
"epoch": 5.364016736401673,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00027926026146202475,
|
|
"loss": 4.3712,
|
|
"mean_token_accuracy": 0.26127527207136153,
|
|
"num_tokens": 11847648.0,
|
|
"step": 6410
|
|
},
|
|
{
|
|
"entropy": 5.012980556488037,
|
|
"epoch": 5.368200836820083,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00027893754868421665,
|
|
"loss": 4.4102,
|
|
"mean_token_accuracy": 0.25309607684612273,
|
|
"num_tokens": 11857008.0,
|
|
"step": 6415
|
|
},
|
|
{
|
|
"entropy": 5.073302412033081,
|
|
"epoch": 5.372384937238493,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002786148278035665,
|
|
"loss": 4.4374,
|
|
"mean_token_accuracy": 0.25854425728321073,
|
|
"num_tokens": 11866055.0,
|
|
"step": 6420
|
|
},
|
|
{
|
|
"entropy": 5.1710968017578125,
|
|
"epoch": 5.376569037656903,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002782920994841819,
|
|
"loss": 4.516,
|
|
"mean_token_accuracy": 0.24146082550287246,
|
|
"num_tokens": 11875496.0,
|
|
"step": 6425
|
|
},
|
|
{
|
|
"entropy": 5.107016611099243,
|
|
"epoch": 5.380753138075314,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002779693643901858,
|
|
"loss": 4.4311,
|
|
"mean_token_accuracy": 0.2498142972588539,
|
|
"num_tokens": 11884115.0,
|
|
"step": 6430
|
|
},
|
|
{
|
|
"entropy": 5.086424493789673,
|
|
"epoch": 5.384937238493724,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00027764662318571526,
|
|
"loss": 4.5956,
|
|
"mean_token_accuracy": 0.2409585788846016,
|
|
"num_tokens": 11893930.0,
|
|
"step": 6435
|
|
},
|
|
{
|
|
"entropy": 5.040985631942749,
|
|
"epoch": 5.389121338912134,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002773238765349196,
|
|
"loss": 4.4561,
|
|
"mean_token_accuracy": 0.2477280393242836,
|
|
"num_tokens": 11902702.0,
|
|
"step": 6440
|
|
},
|
|
{
|
|
"entropy": 5.163437128067017,
|
|
"epoch": 5.393305439330544,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002770011251019598,
|
|
"loss": 4.4913,
|
|
"mean_token_accuracy": 0.24457105100154877,
|
|
"num_tokens": 11911915.0,
|
|
"step": 6445
|
|
},
|
|
{
|
|
"entropy": 5.067444705963135,
|
|
"epoch": 5.397489539748954,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00027667836955100615,
|
|
"loss": 4.4465,
|
|
"mean_token_accuracy": 0.25000817477703097,
|
|
"num_tokens": 11920737.0,
|
|
"step": 6450
|
|
},
|
|
{
|
|
"entropy": 5.085964632034302,
|
|
"epoch": 5.401673640167364,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002763556105462378,
|
|
"loss": 4.4605,
|
|
"mean_token_accuracy": 0.2499578818678856,
|
|
"num_tokens": 11929548.0,
|
|
"step": 6455
|
|
},
|
|
{
|
|
"entropy": 5.150844764709473,
|
|
"epoch": 5.405857740585774,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00027603284875184095,
|
|
"loss": 4.4316,
|
|
"mean_token_accuracy": 0.25477868765592576,
|
|
"num_tokens": 11938459.0,
|
|
"step": 6460
|
|
},
|
|
{
|
|
"entropy": 5.162665891647339,
|
|
"epoch": 5.410041841004184,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00027571008483200744,
|
|
"loss": 4.4482,
|
|
"mean_token_accuracy": 0.25221012681722643,
|
|
"num_tokens": 11947282.0,
|
|
"step": 6465
|
|
},
|
|
{
|
|
"entropy": 5.057677221298218,
|
|
"epoch": 5.414225941422594,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00027538731945093346,
|
|
"loss": 4.417,
|
|
"mean_token_accuracy": 0.24799353182315825,
|
|
"num_tokens": 11956209.0,
|
|
"step": 6470
|
|
},
|
|
{
|
|
"entropy": 5.0593235969543455,
|
|
"epoch": 5.418410041841004,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00027506455327281834,
|
|
"loss": 4.4664,
|
|
"mean_token_accuracy": 0.24558320343494416,
|
|
"num_tokens": 11965689.0,
|
|
"step": 6475
|
|
},
|
|
{
|
|
"entropy": 5.099319648742676,
|
|
"epoch": 5.422594142259414,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000274741786961863,
|
|
"loss": 4.484,
|
|
"mean_token_accuracy": 0.24559568166732787,
|
|
"num_tokens": 11975622.0,
|
|
"step": 6480
|
|
},
|
|
{
|
|
"entropy": 5.105125427246094,
|
|
"epoch": 5.426778242677824,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002744190211822685,
|
|
"loss": 4.5647,
|
|
"mean_token_accuracy": 0.2358685702085495,
|
|
"num_tokens": 11984904.0,
|
|
"step": 6485
|
|
},
|
|
{
|
|
"entropy": 5.136060285568237,
|
|
"epoch": 5.430962343096234,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000274096256598235,
|
|
"loss": 4.5482,
|
|
"mean_token_accuracy": 0.23940227478742598,
|
|
"num_tokens": 11995456.0,
|
|
"step": 6490
|
|
},
|
|
{
|
|
"entropy": 5.099023723602295,
|
|
"epoch": 5.435146443514644,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00027377349387396017,
|
|
"loss": 4.4504,
|
|
"mean_token_accuracy": 0.2372775763273239,
|
|
"num_tokens": 12004566.0,
|
|
"step": 6495
|
|
},
|
|
{
|
|
"entropy": 5.0777482986450195,
|
|
"epoch": 5.439330543933054,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00027345073367363775,
|
|
"loss": 4.3975,
|
|
"mean_token_accuracy": 0.25032069981098176,
|
|
"num_tokens": 12013304.0,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"epoch": 5.439330543933054,
|
|
"eval_entropy": 4.899447086642368,
|
|
"eval_loss": 5.299992084503174,
|
|
"eval_mean_token_accuracy": 0.2025544162999633,
|
|
"eval_num_tokens": 12013304.0,
|
|
"eval_runtime": 2.0866,
|
|
"eval_samples_per_second": 1872.385,
|
|
"eval_steps_per_second": 234.348,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"entropy": 5.019331979751587,
|
|
"epoch": 5.443514644351464,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002731279766614563,
|
|
"loss": 4.397,
|
|
"mean_token_accuracy": 0.250000961124897,
|
|
"num_tokens": 12022007.0,
|
|
"step": 6505
|
|
},
|
|
{
|
|
"entropy": 5.067392110824585,
|
|
"epoch": 5.447698744769874,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000272805223501598,
|
|
"loss": 4.4781,
|
|
"mean_token_accuracy": 0.23938673436641694,
|
|
"num_tokens": 12031493.0,
|
|
"step": 6510
|
|
},
|
|
{
|
|
"entropy": 5.055736303329468,
|
|
"epoch": 5.451882845188284,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002724824748582369,
|
|
"loss": 4.4404,
|
|
"mean_token_accuracy": 0.25732064843177793,
|
|
"num_tokens": 12040143.0,
|
|
"step": 6515
|
|
},
|
|
{
|
|
"entropy": 5.235954427719117,
|
|
"epoch": 5.456066945606695,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002721597313955376,
|
|
"loss": 4.6086,
|
|
"mean_token_accuracy": 0.2400415360927582,
|
|
"num_tokens": 12050136.0,
|
|
"step": 6520
|
|
},
|
|
{
|
|
"entropy": 5.133109474182129,
|
|
"epoch": 5.460251046025105,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002718369937776546,
|
|
"loss": 4.4926,
|
|
"mean_token_accuracy": 0.24647392630577086,
|
|
"num_tokens": 12060065.0,
|
|
"step": 6525
|
|
},
|
|
{
|
|
"entropy": 5.149432420730591,
|
|
"epoch": 5.464435146443515,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002715142626687298,
|
|
"loss": 4.5347,
|
|
"mean_token_accuracy": 0.234070485830307,
|
|
"num_tokens": 12069175.0,
|
|
"step": 6530
|
|
},
|
|
{
|
|
"entropy": 5.113833856582642,
|
|
"epoch": 5.468619246861925,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000271191538732892,
|
|
"loss": 4.4355,
|
|
"mean_token_accuracy": 0.2553034946322441,
|
|
"num_tokens": 12077308.0,
|
|
"step": 6535
|
|
},
|
|
{
|
|
"entropy": 5.034334421157837,
|
|
"epoch": 5.472803347280335,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00027086882263425527,
|
|
"loss": 4.3929,
|
|
"mean_token_accuracy": 0.2552476391196251,
|
|
"num_tokens": 12085678.0,
|
|
"step": 6540
|
|
},
|
|
{
|
|
"entropy": 5.079328203201294,
|
|
"epoch": 5.476987447698745,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00027054611503691736,
|
|
"loss": 4.496,
|
|
"mean_token_accuracy": 0.24969979077577592,
|
|
"num_tokens": 12095163.0,
|
|
"step": 6545
|
|
},
|
|
{
|
|
"entropy": 5.116035223007202,
|
|
"epoch": 5.481171548117155,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00027022341660495853,
|
|
"loss": 4.4908,
|
|
"mean_token_accuracy": 0.24912148118019103,
|
|
"num_tokens": 12104185.0,
|
|
"step": 6550
|
|
},
|
|
{
|
|
"entropy": 5.1298102855682375,
|
|
"epoch": 5.485355648535565,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00026990072800244045,
|
|
"loss": 4.5115,
|
|
"mean_token_accuracy": 0.2372227892279625,
|
|
"num_tokens": 12114298.0,
|
|
"step": 6555
|
|
},
|
|
{
|
|
"entropy": 5.095721435546875,
|
|
"epoch": 5.489539748953975,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00026957804989340427,
|
|
"loss": 4.4654,
|
|
"mean_token_accuracy": 0.25308566838502883,
|
|
"num_tokens": 12122512.0,
|
|
"step": 6560
|
|
},
|
|
{
|
|
"entropy": 5.029066371917724,
|
|
"epoch": 5.493723849372385,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002692553829418697,
|
|
"loss": 4.4275,
|
|
"mean_token_accuracy": 0.2512552201747894,
|
|
"num_tokens": 12132133.0,
|
|
"step": 6565
|
|
},
|
|
{
|
|
"entropy": 5.077983617782593,
|
|
"epoch": 5.497907949790795,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002689327278118334,
|
|
"loss": 4.4892,
|
|
"mean_token_accuracy": 0.2465217798948288,
|
|
"num_tokens": 12141299.0,
|
|
"step": 6570
|
|
},
|
|
{
|
|
"entropy": 5.173110818862915,
|
|
"epoch": 5.502092050209205,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000268610085167268,
|
|
"loss": 4.5672,
|
|
"mean_token_accuracy": 0.24299231469631194,
|
|
"num_tokens": 12150412.0,
|
|
"step": 6575
|
|
},
|
|
{
|
|
"entropy": 5.099612092971801,
|
|
"epoch": 5.506276150627615,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002682874556721198,
|
|
"loss": 4.4963,
|
|
"mean_token_accuracy": 0.2501836374402046,
|
|
"num_tokens": 12159754.0,
|
|
"step": 6580
|
|
},
|
|
{
|
|
"entropy": 5.1209635734558105,
|
|
"epoch": 5.510460251046025,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002679648399903087,
|
|
"loss": 4.4761,
|
|
"mean_token_accuracy": 0.25393230766057967,
|
|
"num_tokens": 12169098.0,
|
|
"step": 6585
|
|
},
|
|
{
|
|
"entropy": 5.099884510040283,
|
|
"epoch": 5.514644351464435,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002676422387857258,
|
|
"loss": 4.4401,
|
|
"mean_token_accuracy": 0.2535048007965088,
|
|
"num_tokens": 12178304.0,
|
|
"step": 6590
|
|
},
|
|
{
|
|
"entropy": 5.028957271575928,
|
|
"epoch": 5.518828451882845,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00026731965272223246,
|
|
"loss": 4.4098,
|
|
"mean_token_accuracy": 0.25376600176095965,
|
|
"num_tokens": 12188029.0,
|
|
"step": 6595
|
|
},
|
|
{
|
|
"entropy": 5.053542709350586,
|
|
"epoch": 5.523012552301255,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002669970824636591,
|
|
"loss": 4.4037,
|
|
"mean_token_accuracy": 0.24876107573509215,
|
|
"num_tokens": 12196663.0,
|
|
"step": 6600
|
|
},
|
|
{
|
|
"entropy": 5.093118667602539,
|
|
"epoch": 5.527196652719665,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002666745286738033,
|
|
"loss": 4.517,
|
|
"mean_token_accuracy": 0.2403578817844391,
|
|
"num_tokens": 12206496.0,
|
|
"step": 6605
|
|
},
|
|
{
|
|
"entropy": 5.186746644973755,
|
|
"epoch": 5.531380753138075,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00026635199201642914,
|
|
"loss": 4.5411,
|
|
"mean_token_accuracy": 0.24637376815080642,
|
|
"num_tokens": 12216597.0,
|
|
"step": 6610
|
|
},
|
|
{
|
|
"entropy": 5.065533828735352,
|
|
"epoch": 5.535564853556485,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002660294731552649,
|
|
"loss": 4.4497,
|
|
"mean_token_accuracy": 0.2502744674682617,
|
|
"num_tokens": 12225125.0,
|
|
"step": 6615
|
|
},
|
|
{
|
|
"entropy": 5.052310514450073,
|
|
"epoch": 5.539748953974895,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00026570697275400267,
|
|
"loss": 4.4215,
|
|
"mean_token_accuracy": 0.2520707845687866,
|
|
"num_tokens": 12234402.0,
|
|
"step": 6620
|
|
},
|
|
{
|
|
"entropy": 5.052184343338013,
|
|
"epoch": 5.543933054393305,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002653844914762964,
|
|
"loss": 4.4086,
|
|
"mean_token_accuracy": 0.2480204701423645,
|
|
"num_tokens": 12244136.0,
|
|
"step": 6625
|
|
},
|
|
{
|
|
"entropy": 5.124208259582519,
|
|
"epoch": 5.548117154811716,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002650620299857607,
|
|
"loss": 4.4381,
|
|
"mean_token_accuracy": 0.25527730882167815,
|
|
"num_tokens": 12253039.0,
|
|
"step": 6630
|
|
},
|
|
{
|
|
"entropy": 5.149023008346558,
|
|
"epoch": 5.552301255230126,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00026473958894596963,
|
|
"loss": 4.4927,
|
|
"mean_token_accuracy": 0.24831282794475557,
|
|
"num_tokens": 12263172.0,
|
|
"step": 6635
|
|
},
|
|
{
|
|
"entropy": 5.1168745994567875,
|
|
"epoch": 5.556485355648536,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000264417169020455,
|
|
"loss": 4.5149,
|
|
"mean_token_accuracy": 0.24542383551597596,
|
|
"num_tokens": 12273324.0,
|
|
"step": 6640
|
|
},
|
|
{
|
|
"entropy": 5.156293106079102,
|
|
"epoch": 5.560669456066946,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002640947708727049,
|
|
"loss": 4.5655,
|
|
"mean_token_accuracy": 0.23956019133329393,
|
|
"num_tokens": 12283719.0,
|
|
"step": 6645
|
|
},
|
|
{
|
|
"entropy": 5.051907300949097,
|
|
"epoch": 5.564853556485356,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002637723951661631,
|
|
"loss": 4.496,
|
|
"mean_token_accuracy": 0.23797832578420638,
|
|
"num_tokens": 12293451.0,
|
|
"step": 6650
|
|
},
|
|
{
|
|
"entropy": 5.1931952953338625,
|
|
"epoch": 5.569037656903766,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002634500425642268,
|
|
"loss": 4.5688,
|
|
"mean_token_accuracy": 0.24010612815618515,
|
|
"num_tokens": 12302560.0,
|
|
"step": 6655
|
|
},
|
|
{
|
|
"entropy": 5.191447973251343,
|
|
"epoch": 5.573221757322176,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002631277137302461,
|
|
"loss": 4.5258,
|
|
"mean_token_accuracy": 0.23543630987405778,
|
|
"num_tokens": 12312115.0,
|
|
"step": 6660
|
|
},
|
|
{
|
|
"entropy": 5.019646310806275,
|
|
"epoch": 5.577405857740586,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00026280540932752176,
|
|
"loss": 4.3931,
|
|
"mean_token_accuracy": 0.25971721261739733,
|
|
"num_tokens": 12320689.0,
|
|
"step": 6665
|
|
},
|
|
{
|
|
"entropy": 4.991688299179077,
|
|
"epoch": 5.581589958158996,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00026248313001930445,
|
|
"loss": 4.3859,
|
|
"mean_token_accuracy": 0.25562838912010194,
|
|
"num_tokens": 12329643.0,
|
|
"step": 6670
|
|
},
|
|
{
|
|
"entropy": 5.108033180236816,
|
|
"epoch": 5.585774058577406,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00026216087646879326,
|
|
"loss": 4.5046,
|
|
"mean_token_accuracy": 0.24654557853937148,
|
|
"num_tokens": 12339267.0,
|
|
"step": 6675
|
|
},
|
|
{
|
|
"entropy": 5.210212707519531,
|
|
"epoch": 5.589958158995816,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002618386493391339,
|
|
"loss": 4.5849,
|
|
"mean_token_accuracy": 0.23922317177057267,
|
|
"num_tokens": 12348515.0,
|
|
"step": 6680
|
|
},
|
|
{
|
|
"entropy": 5.11072850227356,
|
|
"epoch": 5.594142259414226,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00026151644929341833,
|
|
"loss": 4.5467,
|
|
"mean_token_accuracy": 0.24149613976478576,
|
|
"num_tokens": 12358023.0,
|
|
"step": 6685
|
|
},
|
|
{
|
|
"entropy": 5.013644504547119,
|
|
"epoch": 5.598326359832636,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00026119427699468226,
|
|
"loss": 4.432,
|
|
"mean_token_accuracy": 0.24544403851032257,
|
|
"num_tokens": 12367129.0,
|
|
"step": 6690
|
|
},
|
|
{
|
|
"entropy": 5.061335325241089,
|
|
"epoch": 5.602510460251046,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00026087213310590455,
|
|
"loss": 4.4651,
|
|
"mean_token_accuracy": 0.2562482938170433,
|
|
"num_tokens": 12375933.0,
|
|
"step": 6695
|
|
},
|
|
{
|
|
"entropy": 5.021448850631714,
|
|
"epoch": 5.606694560669456,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00026055001829000543,
|
|
"loss": 4.4614,
|
|
"mean_token_accuracy": 0.2524800539016724,
|
|
"num_tokens": 12384784.0,
|
|
"step": 6700
|
|
},
|
|
{
|
|
"entropy": 5.072281217575073,
|
|
"epoch": 5.610878661087866,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00026022793320984553,
|
|
"loss": 4.4814,
|
|
"mean_token_accuracy": 0.24833172857761382,
|
|
"num_tokens": 12393933.0,
|
|
"step": 6705
|
|
},
|
|
{
|
|
"entropy": 5.128530788421631,
|
|
"epoch": 5.615062761506276,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002599058785282241,
|
|
"loss": 4.5016,
|
|
"mean_token_accuracy": 0.2365942567586899,
|
|
"num_tokens": 12403442.0,
|
|
"step": 6710
|
|
},
|
|
{
|
|
"entropy": 5.065907907485962,
|
|
"epoch": 5.619246861924686,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00025958385490787797,
|
|
"loss": 4.408,
|
|
"mean_token_accuracy": 0.2596138596534729,
|
|
"num_tokens": 12412317.0,
|
|
"step": 6715
|
|
},
|
|
{
|
|
"entropy": 5.039059591293335,
|
|
"epoch": 5.623430962343097,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002592618630114798,
|
|
"loss": 4.4486,
|
|
"mean_token_accuracy": 0.25201161950826645,
|
|
"num_tokens": 12421459.0,
|
|
"step": 6720
|
|
},
|
|
{
|
|
"entropy": 5.07504620552063,
|
|
"epoch": 5.627615062761507,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00025893990350163724,
|
|
"loss": 4.4907,
|
|
"mean_token_accuracy": 0.24800920039415358,
|
|
"num_tokens": 12431224.0,
|
|
"step": 6725
|
|
},
|
|
{
|
|
"entropy": 5.034825468063355,
|
|
"epoch": 5.631799163179917,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00025861797704089133,
|
|
"loss": 4.3862,
|
|
"mean_token_accuracy": 0.2583120226860046,
|
|
"num_tokens": 12439891.0,
|
|
"step": 6730
|
|
},
|
|
{
|
|
"entropy": 5.046555852890014,
|
|
"epoch": 5.635983263598327,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002582960842917148,
|
|
"loss": 4.3944,
|
|
"mean_token_accuracy": 0.2507047638297081,
|
|
"num_tokens": 12449852.0,
|
|
"step": 6735
|
|
},
|
|
{
|
|
"entropy": 5.1154985427856445,
|
|
"epoch": 5.640167364016737,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002579742259165113,
|
|
"loss": 4.4852,
|
|
"mean_token_accuracy": 0.2537687852978706,
|
|
"num_tokens": 12459058.0,
|
|
"step": 6740
|
|
},
|
|
{
|
|
"entropy": 5.156938219070435,
|
|
"epoch": 5.644351464435147,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002576524025776135,
|
|
"loss": 4.5236,
|
|
"mean_token_accuracy": 0.24604057669639587,
|
|
"num_tokens": 12468684.0,
|
|
"step": 6745
|
|
},
|
|
{
|
|
"entropy": 5.093432474136352,
|
|
"epoch": 5.648535564853557,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002573306149372821,
|
|
"loss": 4.4974,
|
|
"mean_token_accuracy": 0.25132050216197965,
|
|
"num_tokens": 12478102.0,
|
|
"step": 6750
|
|
},
|
|
{
|
|
"entropy": 5.121048927307129,
|
|
"epoch": 5.652719665271967,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002570088636577044,
|
|
"loss": 4.5128,
|
|
"mean_token_accuracy": 0.24179583340883254,
|
|
"num_tokens": 12488036.0,
|
|
"step": 6755
|
|
},
|
|
{
|
|
"entropy": 5.132516765594483,
|
|
"epoch": 5.656903765690377,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002566871494009928,
|
|
"loss": 4.4782,
|
|
"mean_token_accuracy": 0.24441462904214858,
|
|
"num_tokens": 12496593.0,
|
|
"step": 6760
|
|
},
|
|
{
|
|
"entropy": 5.1246960163116455,
|
|
"epoch": 5.661087866108787,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002563654728291835,
|
|
"loss": 4.5584,
|
|
"mean_token_accuracy": 0.24160375744104384,
|
|
"num_tokens": 12505752.0,
|
|
"step": 6765
|
|
},
|
|
{
|
|
"entropy": 5.080409336090088,
|
|
"epoch": 5.665271966527197,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002560438346042351,
|
|
"loss": 4.4413,
|
|
"mean_token_accuracy": 0.25703504383563996,
|
|
"num_tokens": 12515192.0,
|
|
"step": 6770
|
|
},
|
|
{
|
|
"entropy": 5.165335750579834,
|
|
"epoch": 5.669456066945607,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002557222353880275,
|
|
"loss": 4.5207,
|
|
"mean_token_accuracy": 0.24085431843996047,
|
|
"num_tokens": 12525613.0,
|
|
"step": 6775
|
|
},
|
|
{
|
|
"entropy": 5.133619737625122,
|
|
"epoch": 5.673640167364017,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00025540067584235997,
|
|
"loss": 4.5048,
|
|
"mean_token_accuracy": 0.24708575308322905,
|
|
"num_tokens": 12535076.0,
|
|
"step": 6780
|
|
},
|
|
{
|
|
"entropy": 5.064434814453125,
|
|
"epoch": 5.677824267782427,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00025507915662895055,
|
|
"loss": 4.4983,
|
|
"mean_token_accuracy": 0.24683099687099458,
|
|
"num_tokens": 12544333.0,
|
|
"step": 6785
|
|
},
|
|
{
|
|
"entropy": 5.122288417816162,
|
|
"epoch": 5.682008368200837,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00025475767840943403,
|
|
"loss": 4.4641,
|
|
"mean_token_accuracy": 0.2520321413874626,
|
|
"num_tokens": 12554292.0,
|
|
"step": 6790
|
|
},
|
|
{
|
|
"entropy": 5.102587747573852,
|
|
"epoch": 5.686192468619247,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002544362418453608,
|
|
"loss": 4.5205,
|
|
"mean_token_accuracy": 0.2413293331861496,
|
|
"num_tokens": 12564240.0,
|
|
"step": 6795
|
|
},
|
|
{
|
|
"entropy": 5.095020532608032,
|
|
"epoch": 5.690376569037657,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00025411484759819584,
|
|
"loss": 4.4459,
|
|
"mean_token_accuracy": 0.2601528763771057,
|
|
"num_tokens": 12573280.0,
|
|
"step": 6800
|
|
},
|
|
{
|
|
"entropy": 5.117878723144531,
|
|
"epoch": 5.694560669456067,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00025379349632931666,
|
|
"loss": 4.5399,
|
|
"mean_token_accuracy": 0.24549717009067534,
|
|
"num_tokens": 12583133.0,
|
|
"step": 6805
|
|
},
|
|
{
|
|
"entropy": 5.105498361587524,
|
|
"epoch": 5.698744769874477,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002534721887000127,
|
|
"loss": 4.5039,
|
|
"mean_token_accuracy": 0.24512191265821456,
|
|
"num_tokens": 12591709.0,
|
|
"step": 6810
|
|
},
|
|
{
|
|
"entropy": 5.134109115600586,
|
|
"epoch": 5.702928870292887,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002531509253714831,
|
|
"loss": 4.4389,
|
|
"mean_token_accuracy": 0.25467728525400163,
|
|
"num_tokens": 12600187.0,
|
|
"step": 6815
|
|
},
|
|
{
|
|
"entropy": 5.031455421447754,
|
|
"epoch": 5.707112970711297,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002528297070048364,
|
|
"loss": 4.4403,
|
|
"mean_token_accuracy": 0.24836181700229645,
|
|
"num_tokens": 12608910.0,
|
|
"step": 6820
|
|
},
|
|
{
|
|
"entropy": 5.048375368118286,
|
|
"epoch": 5.711297071129707,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00025250853426108814,
|
|
"loss": 4.4108,
|
|
"mean_token_accuracy": 0.2574331611394882,
|
|
"num_tokens": 12617370.0,
|
|
"step": 6825
|
|
},
|
|
{
|
|
"entropy": 5.116286277770996,
|
|
"epoch": 5.7154811715481175,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00025218740780116037,
|
|
"loss": 4.4886,
|
|
"mean_token_accuracy": 0.24766309410333634,
|
|
"num_tokens": 12626146.0,
|
|
"step": 6830
|
|
},
|
|
{
|
|
"entropy": 5.143683767318725,
|
|
"epoch": 5.7196652719665275,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00025186632828587977,
|
|
"loss": 4.4568,
|
|
"mean_token_accuracy": 0.250521057844162,
|
|
"num_tokens": 12635565.0,
|
|
"step": 6835
|
|
},
|
|
{
|
|
"entropy": 5.137791633605957,
|
|
"epoch": 5.7238493723849375,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002515452963759761,
|
|
"loss": 4.5212,
|
|
"mean_token_accuracy": 0.2418299749493599,
|
|
"num_tokens": 12644869.0,
|
|
"step": 6840
|
|
},
|
|
{
|
|
"entropy": 5.084618282318115,
|
|
"epoch": 5.7280334728033475,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002512243127320817,
|
|
"loss": 4.455,
|
|
"mean_token_accuracy": 0.2505542978644371,
|
|
"num_tokens": 12655215.0,
|
|
"step": 6845
|
|
},
|
|
{
|
|
"entropy": 5.080519151687622,
|
|
"epoch": 5.7322175732217575,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002509033780147291,
|
|
"loss": 4.4078,
|
|
"mean_token_accuracy": 0.25304082334041594,
|
|
"num_tokens": 12663684.0,
|
|
"step": 6850
|
|
},
|
|
{
|
|
"entropy": 5.070897245407105,
|
|
"epoch": 5.7364016736401675,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00025058249288435037,
|
|
"loss": 4.4904,
|
|
"mean_token_accuracy": 0.24235712140798568,
|
|
"num_tokens": 12673367.0,
|
|
"step": 6855
|
|
},
|
|
{
|
|
"entropy": 5.082643938064575,
|
|
"epoch": 5.7405857740585775,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002502616580012756,
|
|
"loss": 4.4414,
|
|
"mean_token_accuracy": 0.25600948333740237,
|
|
"num_tokens": 12682472.0,
|
|
"step": 6860
|
|
},
|
|
{
|
|
"entropy": 5.022229766845703,
|
|
"epoch": 5.7447698744769875,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00024994087402573133,
|
|
"loss": 4.4455,
|
|
"mean_token_accuracy": 0.2510752141475677,
|
|
"num_tokens": 12691215.0,
|
|
"step": 6865
|
|
},
|
|
{
|
|
"entropy": 5.0094822406768795,
|
|
"epoch": 5.7489539748953975,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00024962014161783946,
|
|
"loss": 4.3899,
|
|
"mean_token_accuracy": 0.24387340247631073,
|
|
"num_tokens": 12700332.0,
|
|
"step": 6870
|
|
},
|
|
{
|
|
"entropy": 5.098571920394898,
|
|
"epoch": 5.7531380753138075,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00024929946143761585,
|
|
"loss": 4.4505,
|
|
"mean_token_accuracy": 0.24581302851438522,
|
|
"num_tokens": 12708886.0,
|
|
"step": 6875
|
|
},
|
|
{
|
|
"entropy": 5.0760712146759035,
|
|
"epoch": 5.7573221757322175,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00024897883414496844,
|
|
"loss": 4.4283,
|
|
"mean_token_accuracy": 0.2575951188802719,
|
|
"num_tokens": 12717476.0,
|
|
"step": 6880
|
|
},
|
|
{
|
|
"entropy": 5.18282561302185,
|
|
"epoch": 5.7615062761506275,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00024865826039969683,
|
|
"loss": 4.5713,
|
|
"mean_token_accuracy": 0.24546981155872344,
|
|
"num_tokens": 12726737.0,
|
|
"step": 6885
|
|
},
|
|
{
|
|
"entropy": 5.091027450561524,
|
|
"epoch": 5.7656903765690375,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002483377408614903,
|
|
"loss": 4.4661,
|
|
"mean_token_accuracy": 0.24852315634489058,
|
|
"num_tokens": 12735750.0,
|
|
"step": 6890
|
|
},
|
|
{
|
|
"entropy": 5.044934701919556,
|
|
"epoch": 5.7698744769874475,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00024801727618992635,
|
|
"loss": 4.4381,
|
|
"mean_token_accuracy": 0.2477365255355835,
|
|
"num_tokens": 12745143.0,
|
|
"step": 6895
|
|
},
|
|
{
|
|
"entropy": 5.032395696640014,
|
|
"epoch": 5.7740585774058575,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00024769686704446996,
|
|
"loss": 4.4501,
|
|
"mean_token_accuracy": 0.24524296671152115,
|
|
"num_tokens": 12754623.0,
|
|
"step": 6900
|
|
},
|
|
{
|
|
"entropy": 5.133699607849121,
|
|
"epoch": 5.7782426778242675,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00024737651408447155,
|
|
"loss": 4.487,
|
|
"mean_token_accuracy": 0.24803201258182525,
|
|
"num_tokens": 12762921.0,
|
|
"step": 6905
|
|
},
|
|
{
|
|
"entropy": 5.114299249649048,
|
|
"epoch": 5.7824267782426775,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000247056217969166,
|
|
"loss": 4.535,
|
|
"mean_token_accuracy": 0.24999246150255203,
|
|
"num_tokens": 12772159.0,
|
|
"step": 6910
|
|
},
|
|
{
|
|
"entropy": 5.062676811218262,
|
|
"epoch": 5.786610878661088,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00024673597935767123,
|
|
"loss": 4.4979,
|
|
"mean_token_accuracy": 0.24298299551010133,
|
|
"num_tokens": 12781916.0,
|
|
"step": 6915
|
|
},
|
|
{
|
|
"entropy": 5.061027526855469,
|
|
"epoch": 5.790794979079498,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002464157989089867,
|
|
"loss": 4.4866,
|
|
"mean_token_accuracy": 0.246560762822628,
|
|
"num_tokens": 12791275.0,
|
|
"step": 6920
|
|
},
|
|
{
|
|
"entropy": 5.104857730865478,
|
|
"epoch": 5.794979079497908,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002460956772819925,
|
|
"loss": 4.47,
|
|
"mean_token_accuracy": 0.24929350912570952,
|
|
"num_tokens": 12800667.0,
|
|
"step": 6925
|
|
},
|
|
{
|
|
"entropy": 5.072886800765991,
|
|
"epoch": 5.799163179916318,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00024577561513544724,
|
|
"loss": 4.4934,
|
|
"mean_token_accuracy": 0.24380537420511245,
|
|
"num_tokens": 12808986.0,
|
|
"step": 6930
|
|
},
|
|
{
|
|
"entropy": 5.106848430633545,
|
|
"epoch": 5.803347280334728,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00024545561312798756,
|
|
"loss": 4.5023,
|
|
"mean_token_accuracy": 0.24585724025964736,
|
|
"num_tokens": 12817658.0,
|
|
"step": 6935
|
|
},
|
|
{
|
|
"entropy": 5.117634105682373,
|
|
"epoch": 5.807531380753138,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00024513567191812607,
|
|
"loss": 4.4927,
|
|
"mean_token_accuracy": 0.248333877325058,
|
|
"num_tokens": 12826301.0,
|
|
"step": 6940
|
|
},
|
|
{
|
|
"entropy": 5.204719161987304,
|
|
"epoch": 5.811715481171548,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002448157921642505,
|
|
"loss": 4.5184,
|
|
"mean_token_accuracy": 0.23951416015625,
|
|
"num_tokens": 12836295.0,
|
|
"step": 6945
|
|
},
|
|
{
|
|
"entropy": 5.106586456298828,
|
|
"epoch": 5.815899581589958,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00024449597452462165,
|
|
"loss": 4.4767,
|
|
"mean_token_accuracy": 0.24392015486955643,
|
|
"num_tokens": 12847128.0,
|
|
"step": 6950
|
|
},
|
|
{
|
|
"entropy": 5.054434537887573,
|
|
"epoch": 5.820083682008368,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002441762196573729,
|
|
"loss": 4.5189,
|
|
"mean_token_accuracy": 0.24588286131620407,
|
|
"num_tokens": 12856198.0,
|
|
"step": 6955
|
|
},
|
|
{
|
|
"entropy": 5.121337175369263,
|
|
"epoch": 5.824267782426778,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00024385652822050846,
|
|
"loss": 4.4405,
|
|
"mean_token_accuracy": 0.2535675525665283,
|
|
"num_tokens": 12865513.0,
|
|
"step": 6960
|
|
},
|
|
{
|
|
"entropy": 5.133240747451782,
|
|
"epoch": 5.828451882845188,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00024353690087190188,
|
|
"loss": 4.4761,
|
|
"mean_token_accuracy": 0.24401120841503143,
|
|
"num_tokens": 12874086.0,
|
|
"step": 6965
|
|
},
|
|
{
|
|
"entropy": 5.04276967048645,
|
|
"epoch": 5.832635983263598,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00024321733826929487,
|
|
"loss": 4.4715,
|
|
"mean_token_accuracy": 0.2528166607022285,
|
|
"num_tokens": 12883382.0,
|
|
"step": 6970
|
|
},
|
|
{
|
|
"entropy": 5.188482189178467,
|
|
"epoch": 5.836820083682008,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00024289784107029588,
|
|
"loss": 4.5266,
|
|
"mean_token_accuracy": 0.2440932422876358,
|
|
"num_tokens": 12893510.0,
|
|
"step": 6975
|
|
},
|
|
{
|
|
"entropy": 5.20506739616394,
|
|
"epoch": 5.841004184100418,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00024257840993237862,
|
|
"loss": 4.5531,
|
|
"mean_token_accuracy": 0.2400622308254242,
|
|
"num_tokens": 12902719.0,
|
|
"step": 6980
|
|
},
|
|
{
|
|
"entropy": 5.124751901626587,
|
|
"epoch": 5.845188284518828,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00024225904551288124,
|
|
"loss": 4.5583,
|
|
"mean_token_accuracy": 0.2348041996359825,
|
|
"num_tokens": 12912889.0,
|
|
"step": 6985
|
|
},
|
|
{
|
|
"entropy": 5.095812654495239,
|
|
"epoch": 5.849372384937238,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00024193974846900424,
|
|
"loss": 4.4259,
|
|
"mean_token_accuracy": 0.24815673083066941,
|
|
"num_tokens": 12922454.0,
|
|
"step": 6990
|
|
},
|
|
{
|
|
"entropy": 5.081589031219482,
|
|
"epoch": 5.853556485355648,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00024162051945780967,
|
|
"loss": 4.4401,
|
|
"mean_token_accuracy": 0.2495010554790497,
|
|
"num_tokens": 12931038.0,
|
|
"step": 6995
|
|
},
|
|
{
|
|
"entropy": 5.079547786712647,
|
|
"epoch": 5.857740585774058,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002413013591362195,
|
|
"loss": 4.5131,
|
|
"mean_token_accuracy": 0.24270990937948228,
|
|
"num_tokens": 12940830.0,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"epoch": 5.857740585774058,
|
|
"eval_entropy": 4.886750562546931,
|
|
"eval_loss": 5.28425931930542,
|
|
"eval_mean_token_accuracy": 0.20318048753621387,
|
|
"eval_num_tokens": 12940830.0,
|
|
"eval_runtime": 2.2671,
|
|
"eval_samples_per_second": 1723.368,
|
|
"eval_steps_per_second": 215.697,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"entropy": 5.138467693328858,
|
|
"epoch": 5.861924686192468,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00024098226816101427,
|
|
"loss": 4.5417,
|
|
"mean_token_accuracy": 0.23605234622955323,
|
|
"num_tokens": 12949159.0,
|
|
"step": 7005
|
|
},
|
|
{
|
|
"entropy": 5.170948457717896,
|
|
"epoch": 5.866108786610878,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00024066324718883205,
|
|
"loss": 4.5475,
|
|
"mean_token_accuracy": 0.24482942074537278,
|
|
"num_tokens": 12957818.0,
|
|
"step": 7010
|
|
},
|
|
{
|
|
"entropy": 5.141408634185791,
|
|
"epoch": 5.870292887029288,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002403442968761666,
|
|
"loss": 4.5627,
|
|
"mean_token_accuracy": 0.24214372783899307,
|
|
"num_tokens": 12966667.0,
|
|
"step": 7015
|
|
},
|
|
{
|
|
"entropy": 5.127597236633301,
|
|
"epoch": 5.874476987447698,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00024002541787936644,
|
|
"loss": 4.4495,
|
|
"mean_token_accuracy": 0.2393207758665085,
|
|
"num_tokens": 12976081.0,
|
|
"step": 7020
|
|
},
|
|
{
|
|
"entropy": 5.078018045425415,
|
|
"epoch": 5.878661087866108,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002397066108546332,
|
|
"loss": 4.5393,
|
|
"mean_token_accuracy": 0.24502900689840318,
|
|
"num_tokens": 12985006.0,
|
|
"step": 7025
|
|
},
|
|
{
|
|
"entropy": 5.090643882751465,
|
|
"epoch": 5.882845188284519,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00023938787645802056,
|
|
"loss": 4.4778,
|
|
"mean_token_accuracy": 0.2502941772341728,
|
|
"num_tokens": 12993862.0,
|
|
"step": 7030
|
|
},
|
|
{
|
|
"entropy": 5.0531378269195555,
|
|
"epoch": 5.887029288702929,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00023906921534543269,
|
|
"loss": 4.4815,
|
|
"mean_token_accuracy": 0.2494940459728241,
|
|
"num_tokens": 13003637.0,
|
|
"step": 7035
|
|
},
|
|
{
|
|
"entropy": 5.142329502105713,
|
|
"epoch": 5.891213389121339,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00023875062817262284,
|
|
"loss": 4.5243,
|
|
"mean_token_accuracy": 0.24456612020730972,
|
|
"num_tokens": 13013364.0,
|
|
"step": 7040
|
|
},
|
|
{
|
|
"entropy": 5.056112432479859,
|
|
"epoch": 5.895397489539749,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00023843211559519228,
|
|
"loss": 4.4855,
|
|
"mean_token_accuracy": 0.2474406346678734,
|
|
"num_tokens": 13023432.0,
|
|
"step": 7045
|
|
},
|
|
{
|
|
"entropy": 5.1363420486450195,
|
|
"epoch": 5.899581589958159,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00023811367826858854,
|
|
"loss": 4.5155,
|
|
"mean_token_accuracy": 0.24509378522634506,
|
|
"num_tokens": 13034293.0,
|
|
"step": 7050
|
|
},
|
|
{
|
|
"entropy": 5.1593788146972654,
|
|
"epoch": 5.903765690376569,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00023779531684810462,
|
|
"loss": 4.5208,
|
|
"mean_token_accuracy": 0.24647510945796966,
|
|
"num_tokens": 13043985.0,
|
|
"step": 7055
|
|
},
|
|
{
|
|
"entropy": 5.015690040588379,
|
|
"epoch": 5.907949790794979,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00023747703198887698,
|
|
"loss": 4.3764,
|
|
"mean_token_accuracy": 0.2601540356874466,
|
|
"num_tokens": 13052339.0,
|
|
"step": 7060
|
|
},
|
|
{
|
|
"entropy": 5.077298498153686,
|
|
"epoch": 5.912133891213389,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00023715882434588483,
|
|
"loss": 4.4321,
|
|
"mean_token_accuracy": 0.24911000281572343,
|
|
"num_tokens": 13060858.0,
|
|
"step": 7065
|
|
},
|
|
{
|
|
"entropy": 5.102647972106934,
|
|
"epoch": 5.916317991631799,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00023684069457394826,
|
|
"loss": 4.4031,
|
|
"mean_token_accuracy": 0.261798283457756,
|
|
"num_tokens": 13069542.0,
|
|
"step": 7070
|
|
},
|
|
{
|
|
"entropy": 5.056877851486206,
|
|
"epoch": 5.920502092050209,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00023652264332772721,
|
|
"loss": 4.3791,
|
|
"mean_token_accuracy": 0.2659183740615845,
|
|
"num_tokens": 13078868.0,
|
|
"step": 7075
|
|
},
|
|
{
|
|
"entropy": 5.113799047470093,
|
|
"epoch": 5.924686192468619,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00023620467126172001,
|
|
"loss": 4.5585,
|
|
"mean_token_accuracy": 0.2380124509334564,
|
|
"num_tokens": 13088573.0,
|
|
"step": 7080
|
|
},
|
|
{
|
|
"entropy": 5.080211591720581,
|
|
"epoch": 5.928870292887029,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00023588677903026202,
|
|
"loss": 4.4729,
|
|
"mean_token_accuracy": 0.24575835168361665,
|
|
"num_tokens": 13097723.0,
|
|
"step": 7085
|
|
},
|
|
{
|
|
"entropy": 5.1612372398376465,
|
|
"epoch": 5.933054393305439,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00023556896728752435,
|
|
"loss": 4.4262,
|
|
"mean_token_accuracy": 0.24828552901744844,
|
|
"num_tokens": 13106071.0,
|
|
"step": 7090
|
|
},
|
|
{
|
|
"entropy": 5.0523108959198,
|
|
"epoch": 5.937238493723849,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002352512366875125,
|
|
"loss": 4.4611,
|
|
"mean_token_accuracy": 0.24787884950637817,
|
|
"num_tokens": 13115150.0,
|
|
"step": 7095
|
|
},
|
|
{
|
|
"entropy": 5.0561525344848635,
|
|
"epoch": 5.941422594142259,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000234933587884065,
|
|
"loss": 4.5192,
|
|
"mean_token_accuracy": 0.24879903346300125,
|
|
"num_tokens": 13124833.0,
|
|
"step": 7100
|
|
},
|
|
{
|
|
"entropy": 5.140002775192261,
|
|
"epoch": 5.945606694560669,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000234616021530852,
|
|
"loss": 4.5456,
|
|
"mean_token_accuracy": 0.2430487707257271,
|
|
"num_tokens": 13135751.0,
|
|
"step": 7105
|
|
},
|
|
{
|
|
"entropy": 5.070494604110718,
|
|
"epoch": 5.949790794979079,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00023429853828137393,
|
|
"loss": 4.4,
|
|
"mean_token_accuracy": 0.25536018759012225,
|
|
"num_tokens": 13144285.0,
|
|
"step": 7110
|
|
},
|
|
{
|
|
"entropy": 5.087052249908448,
|
|
"epoch": 5.95397489539749,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00023398113878896033,
|
|
"loss": 4.4873,
|
|
"mean_token_accuracy": 0.24541891664266585,
|
|
"num_tokens": 13152856.0,
|
|
"step": 7115
|
|
},
|
|
{
|
|
"entropy": 5.150256967544555,
|
|
"epoch": 5.9581589958159,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00023366382370676824,
|
|
"loss": 4.5149,
|
|
"mean_token_accuracy": 0.2472623333334923,
|
|
"num_tokens": 13162665.0,
|
|
"step": 7120
|
|
},
|
|
{
|
|
"entropy": 5.075791072845459,
|
|
"epoch": 5.96234309623431,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00023334659368778117,
|
|
"loss": 4.4384,
|
|
"mean_token_accuracy": 0.2565875440835953,
|
|
"num_tokens": 13171110.0,
|
|
"step": 7125
|
|
},
|
|
{
|
|
"entropy": 5.073829984664917,
|
|
"epoch": 5.96652719665272,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00023302944938480752,
|
|
"loss": 4.5033,
|
|
"mean_token_accuracy": 0.24355719834566117,
|
|
"num_tokens": 13179948.0,
|
|
"step": 7130
|
|
},
|
|
{
|
|
"entropy": 5.160698795318604,
|
|
"epoch": 5.97071129707113,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00023271239145047917,
|
|
"loss": 4.5055,
|
|
"mean_token_accuracy": 0.2471671834588051,
|
|
"num_tokens": 13190214.0,
|
|
"step": 7135
|
|
},
|
|
{
|
|
"entropy": 4.986247396469116,
|
|
"epoch": 5.97489539748954,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00023239542053725044,
|
|
"loss": 4.3631,
|
|
"mean_token_accuracy": 0.2576016321778297,
|
|
"num_tokens": 13199048.0,
|
|
"step": 7140
|
|
},
|
|
{
|
|
"entropy": 5.080405044555664,
|
|
"epoch": 5.97907949790795,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00023207853729739663,
|
|
"loss": 4.457,
|
|
"mean_token_accuracy": 0.2480512022972107,
|
|
"num_tokens": 13208553.0,
|
|
"step": 7145
|
|
},
|
|
{
|
|
"entropy": 5.184884929656983,
|
|
"epoch": 5.98326359832636,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00023176174238301228,
|
|
"loss": 4.5494,
|
|
"mean_token_accuracy": 0.24254859387874603,
|
|
"num_tokens": 13218151.0,
|
|
"step": 7150
|
|
},
|
|
{
|
|
"entropy": 5.081519651412964,
|
|
"epoch": 5.98744769874477,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002314450364460105,
|
|
"loss": 4.4514,
|
|
"mean_token_accuracy": 0.24984516501426696,
|
|
"num_tokens": 13227730.0,
|
|
"step": 7155
|
|
},
|
|
{
|
|
"entropy": 5.098627805709839,
|
|
"epoch": 5.99163179916318,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002311284201381212,
|
|
"loss": 4.5465,
|
|
"mean_token_accuracy": 0.24166882336139678,
|
|
"num_tokens": 13237449.0,
|
|
"step": 7160
|
|
},
|
|
{
|
|
"entropy": 5.082867622375488,
|
|
"epoch": 5.99581589958159,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00023081189411088972,
|
|
"loss": 4.4537,
|
|
"mean_token_accuracy": 0.24508282244205476,
|
|
"num_tokens": 13246662.0,
|
|
"step": 7165
|
|
},
|
|
{
|
|
"entropy": 5.138063621520996,
|
|
"epoch": 6.0,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00023049545901567586,
|
|
"loss": 4.5036,
|
|
"mean_token_accuracy": 0.24773088544607164,
|
|
"num_tokens": 13255932.0,
|
|
"step": 7170
|
|
},
|
|
{
|
|
"entropy": 5.074365472793579,
|
|
"epoch": 6.00418410041841,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00023017911550365233,
|
|
"loss": 4.3163,
|
|
"mean_token_accuracy": 0.2662795126438141,
|
|
"num_tokens": 13264600.0,
|
|
"step": 7175
|
|
},
|
|
{
|
|
"entropy": 5.180688190460205,
|
|
"epoch": 6.00836820083682,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00022986286422580288,
|
|
"loss": 4.3803,
|
|
"mean_token_accuracy": 0.2599422439932823,
|
|
"num_tokens": 13274261.0,
|
|
"step": 7180
|
|
},
|
|
{
|
|
"entropy": 5.106724405288697,
|
|
"epoch": 6.01255230125523,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00022954670583292196,
|
|
"loss": 4.34,
|
|
"mean_token_accuracy": 0.26268582940101626,
|
|
"num_tokens": 13284150.0,
|
|
"step": 7185
|
|
},
|
|
{
|
|
"entropy": 5.139708662033081,
|
|
"epoch": 6.01673640167364,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00022923064097561265,
|
|
"loss": 4.3205,
|
|
"mean_token_accuracy": 0.25518807023763657,
|
|
"num_tokens": 13293114.0,
|
|
"step": 7190
|
|
},
|
|
{
|
|
"entropy": 5.060097408294678,
|
|
"epoch": 6.02092050209205,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002289146703042857,
|
|
"loss": 4.3655,
|
|
"mean_token_accuracy": 0.2617349445819855,
|
|
"num_tokens": 13303391.0,
|
|
"step": 7195
|
|
},
|
|
{
|
|
"entropy": 5.110640811920166,
|
|
"epoch": 6.02510460251046,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00022859879446915788,
|
|
"loss": 4.4307,
|
|
"mean_token_accuracy": 0.2584112733602524,
|
|
"num_tokens": 13312372.0,
|
|
"step": 7200
|
|
},
|
|
{
|
|
"entropy": 5.13575930595398,
|
|
"epoch": 6.02928870292887,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00022828301412025094,
|
|
"loss": 4.376,
|
|
"mean_token_accuracy": 0.2626110836863518,
|
|
"num_tokens": 13321543.0,
|
|
"step": 7205
|
|
},
|
|
{
|
|
"entropy": 5.090541839599609,
|
|
"epoch": 6.03347280334728,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002279673299073901,
|
|
"loss": 4.327,
|
|
"mean_token_accuracy": 0.264054936170578,
|
|
"num_tokens": 13330179.0,
|
|
"step": 7210
|
|
},
|
|
{
|
|
"entropy": 5.0801926612854,
|
|
"epoch": 6.03765690376569,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00022765174248020258,
|
|
"loss": 4.2979,
|
|
"mean_token_accuracy": 0.26571533679962156,
|
|
"num_tokens": 13339116.0,
|
|
"step": 7215
|
|
},
|
|
{
|
|
"entropy": 4.99830322265625,
|
|
"epoch": 6.0418410041841,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00022733625248811656,
|
|
"loss": 4.3041,
|
|
"mean_token_accuracy": 0.2684531956911087,
|
|
"num_tokens": 13347908.0,
|
|
"step": 7220
|
|
},
|
|
{
|
|
"entropy": 4.996017789840698,
|
|
"epoch": 6.046025104602511,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00022702086058035985,
|
|
"loss": 4.2637,
|
|
"mean_token_accuracy": 0.27434307932853697,
|
|
"num_tokens": 13356840.0,
|
|
"step": 7225
|
|
},
|
|
{
|
|
"entropy": 5.039406394958496,
|
|
"epoch": 6.050209205020921,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00022670556740595827,
|
|
"loss": 4.3136,
|
|
"mean_token_accuracy": 0.26353214830160143,
|
|
"num_tokens": 13365288.0,
|
|
"step": 7230
|
|
},
|
|
{
|
|
"entropy": 5.0219217300415036,
|
|
"epoch": 6.054393305439331,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00022639037361373444,
|
|
"loss": 4.2927,
|
|
"mean_token_accuracy": 0.27515466809272765,
|
|
"num_tokens": 13374612.0,
|
|
"step": 7235
|
|
},
|
|
{
|
|
"entropy": 5.089159154891968,
|
|
"epoch": 6.058577405857741,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00022607527985230648,
|
|
"loss": 4.3192,
|
|
"mean_token_accuracy": 0.2659605577588081,
|
|
"num_tokens": 13384307.0,
|
|
"step": 7240
|
|
},
|
|
{
|
|
"entropy": 5.073921632766724,
|
|
"epoch": 6.062761506276151,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00022576028677008688,
|
|
"loss": 4.3141,
|
|
"mean_token_accuracy": 0.2657345920801163,
|
|
"num_tokens": 13392626.0,
|
|
"step": 7245
|
|
},
|
|
{
|
|
"entropy": 5.100242376327515,
|
|
"epoch": 6.066945606694561,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002254453950152804,
|
|
"loss": 4.3207,
|
|
"mean_token_accuracy": 0.26862914562225343,
|
|
"num_tokens": 13401694.0,
|
|
"step": 7250
|
|
},
|
|
{
|
|
"entropy": 5.073161363601685,
|
|
"epoch": 6.071129707112971,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002251306052358839,
|
|
"loss": 4.375,
|
|
"mean_token_accuracy": 0.26331732273101804,
|
|
"num_tokens": 13410330.0,
|
|
"step": 7255
|
|
},
|
|
{
|
|
"entropy": 5.033152055740357,
|
|
"epoch": 6.075313807531381,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00022481591807968417,
|
|
"loss": 4.2646,
|
|
"mean_token_accuracy": 0.2685937911272049,
|
|
"num_tokens": 13418988.0,
|
|
"step": 7260
|
|
},
|
|
{
|
|
"entropy": 5.06645393371582,
|
|
"epoch": 6.079497907949791,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002245013341942566,
|
|
"loss": 4.3781,
|
|
"mean_token_accuracy": 0.25999888479709626,
|
|
"num_tokens": 13427487.0,
|
|
"step": 7265
|
|
},
|
|
{
|
|
"entropy": 5.075822448730468,
|
|
"epoch": 6.083682008368201,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00022418685422696444,
|
|
"loss": 4.3134,
|
|
"mean_token_accuracy": 0.2724656596779823,
|
|
"num_tokens": 13436450.0,
|
|
"step": 7270
|
|
},
|
|
{
|
|
"entropy": 5.039012813568116,
|
|
"epoch": 6.087866108786611,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002238724788249568,
|
|
"loss": 4.3066,
|
|
"mean_token_accuracy": 0.26200431436300275,
|
|
"num_tokens": 13445363.0,
|
|
"step": 7275
|
|
},
|
|
{
|
|
"entropy": 5.061498737335205,
|
|
"epoch": 6.092050209205021,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00022355820863516774,
|
|
"loss": 4.2484,
|
|
"mean_token_accuracy": 0.27218090295791625,
|
|
"num_tokens": 13453960.0,
|
|
"step": 7280
|
|
},
|
|
{
|
|
"entropy": 5.106298112869263,
|
|
"epoch": 6.096234309623431,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002232440443043147,
|
|
"loss": 4.3766,
|
|
"mean_token_accuracy": 0.2682182937860489,
|
|
"num_tokens": 13463619.0,
|
|
"step": 7285
|
|
},
|
|
{
|
|
"entropy": 5.079867172241211,
|
|
"epoch": 6.100418410041841,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002229299864788975,
|
|
"loss": 4.3269,
|
|
"mean_token_accuracy": 0.26354132443666456,
|
|
"num_tokens": 13473566.0,
|
|
"step": 7290
|
|
},
|
|
{
|
|
"entropy": 4.999058437347412,
|
|
"epoch": 6.104602510460251,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00022261603580519656,
|
|
"loss": 4.2678,
|
|
"mean_token_accuracy": 0.2678779110312462,
|
|
"num_tokens": 13481848.0,
|
|
"step": 7295
|
|
},
|
|
{
|
|
"entropy": 5.1179498672485355,
|
|
"epoch": 6.108786610878661,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00022230219292927188,
|
|
"loss": 4.4353,
|
|
"mean_token_accuracy": 0.2544319868087769,
|
|
"num_tokens": 13491600.0,
|
|
"step": 7300
|
|
},
|
|
{
|
|
"entropy": 5.11618595123291,
|
|
"epoch": 6.112970711297071,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00022198845849696175,
|
|
"loss": 4.4035,
|
|
"mean_token_accuracy": 0.25373679101467134,
|
|
"num_tokens": 13500888.0,
|
|
"step": 7305
|
|
},
|
|
{
|
|
"entropy": 5.105974578857422,
|
|
"epoch": 6.117154811715481,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00022167483315388108,
|
|
"loss": 4.3566,
|
|
"mean_token_accuracy": 0.26570618003606794,
|
|
"num_tokens": 13511031.0,
|
|
"step": 7310
|
|
},
|
|
{
|
|
"entropy": 5.02967791557312,
|
|
"epoch": 6.121338912133891,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00022136131754542048,
|
|
"loss": 4.3788,
|
|
"mean_token_accuracy": 0.2533226668834686,
|
|
"num_tokens": 13520704.0,
|
|
"step": 7315
|
|
},
|
|
{
|
|
"entropy": 5.050959968566895,
|
|
"epoch": 6.125523012552302,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00022104791231674465,
|
|
"loss": 4.2645,
|
|
"mean_token_accuracy": 0.2682411178946495,
|
|
"num_tokens": 13530297.0,
|
|
"step": 7320
|
|
},
|
|
{
|
|
"entropy": 5.040257501602173,
|
|
"epoch": 6.129707112970712,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00022073461811279106,
|
|
"loss": 4.2841,
|
|
"mean_token_accuracy": 0.26461021900177,
|
|
"num_tokens": 13538796.0,
|
|
"step": 7325
|
|
},
|
|
{
|
|
"entropy": 4.952008295059204,
|
|
"epoch": 6.133891213389122,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00022042143557826892,
|
|
"loss": 4.2277,
|
|
"mean_token_accuracy": 0.27523902952671053,
|
|
"num_tokens": 13548145.0,
|
|
"step": 7330
|
|
},
|
|
{
|
|
"entropy": 4.993928670883179,
|
|
"epoch": 6.138075313807532,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00022010836535765749,
|
|
"loss": 4.277,
|
|
"mean_token_accuracy": 0.26273745894432066,
|
|
"num_tokens": 13557025.0,
|
|
"step": 7335
|
|
},
|
|
{
|
|
"entropy": 4.972987937927246,
|
|
"epoch": 6.142259414225942,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00021979540809520504,
|
|
"loss": 4.3329,
|
|
"mean_token_accuracy": 0.26788567155599596,
|
|
"num_tokens": 13566485.0,
|
|
"step": 7340
|
|
},
|
|
{
|
|
"entropy": 5.0744242668151855,
|
|
"epoch": 6.146443514644352,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002194825644349272,
|
|
"loss": 4.3341,
|
|
"mean_token_accuracy": 0.26522617489099504,
|
|
"num_tokens": 13575167.0,
|
|
"step": 7345
|
|
},
|
|
{
|
|
"entropy": 5.1914349555969235,
|
|
"epoch": 6.150627615062762,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00021916983502060595,
|
|
"loss": 4.3679,
|
|
"mean_token_accuracy": 0.2540760785341263,
|
|
"num_tokens": 13585757.0,
|
|
"step": 7350
|
|
},
|
|
{
|
|
"entropy": 5.079670476913452,
|
|
"epoch": 6.154811715481172,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00021885722049578817,
|
|
"loss": 4.368,
|
|
"mean_token_accuracy": 0.2585496664047241,
|
|
"num_tokens": 13595862.0,
|
|
"step": 7355
|
|
},
|
|
{
|
|
"entropy": 5.085423994064331,
|
|
"epoch": 6.158995815899582,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002185447215037842,
|
|
"loss": 4.3764,
|
|
"mean_token_accuracy": 0.25471221357584,
|
|
"num_tokens": 13605106.0,
|
|
"step": 7360
|
|
},
|
|
{
|
|
"entropy": 5.031488561630249,
|
|
"epoch": 6.163179916317992,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00021823233868766684,
|
|
"loss": 4.2959,
|
|
"mean_token_accuracy": 0.2658937364816666,
|
|
"num_tokens": 13614317.0,
|
|
"step": 7365
|
|
},
|
|
{
|
|
"entropy": 5.010358381271362,
|
|
"epoch": 6.167364016736402,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002179200726902697,
|
|
"loss": 4.2605,
|
|
"mean_token_accuracy": 0.27398955821990967,
|
|
"num_tokens": 13623196.0,
|
|
"step": 7370
|
|
},
|
|
{
|
|
"entropy": 5.048473930358886,
|
|
"epoch": 6.171548117154812,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002176079241541859,
|
|
"loss": 4.3052,
|
|
"mean_token_accuracy": 0.2672801211476326,
|
|
"num_tokens": 13631872.0,
|
|
"step": 7375
|
|
},
|
|
{
|
|
"entropy": 5.081488704681396,
|
|
"epoch": 6.175732217573222,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000217295893721767,
|
|
"loss": 4.3315,
|
|
"mean_token_accuracy": 0.26303794234991074,
|
|
"num_tokens": 13641240.0,
|
|
"step": 7380
|
|
},
|
|
{
|
|
"entropy": 4.989028882980347,
|
|
"epoch": 6.179916317991632,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00021698398203512148,
|
|
"loss": 4.2902,
|
|
"mean_token_accuracy": 0.2644805312156677,
|
|
"num_tokens": 13650495.0,
|
|
"step": 7385
|
|
},
|
|
{
|
|
"entropy": 4.986727571487426,
|
|
"epoch": 6.184100418410042,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00021667218973611346,
|
|
"loss": 4.2481,
|
|
"mean_token_accuracy": 0.2774860769510269,
|
|
"num_tokens": 13660234.0,
|
|
"step": 7390
|
|
},
|
|
{
|
|
"entropy": 5.09214334487915,
|
|
"epoch": 6.188284518828452,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00021636051746636126,
|
|
"loss": 4.4269,
|
|
"mean_token_accuracy": 0.26026774793863294,
|
|
"num_tokens": 13670060.0,
|
|
"step": 7395
|
|
},
|
|
{
|
|
"entropy": 5.073004198074341,
|
|
"epoch": 6.192468619246862,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00021604896586723644,
|
|
"loss": 4.3648,
|
|
"mean_token_accuracy": 0.2661386698484421,
|
|
"num_tokens": 13679551.0,
|
|
"step": 7400
|
|
},
|
|
{
|
|
"entropy": 5.095697593688965,
|
|
"epoch": 6.196652719665272,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00021573753557986198,
|
|
"loss": 4.3658,
|
|
"mean_token_accuracy": 0.26226079016923903,
|
|
"num_tokens": 13688772.0,
|
|
"step": 7405
|
|
},
|
|
{
|
|
"entropy": 5.182313919067383,
|
|
"epoch": 6.200836820083682,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00021542622724511146,
|
|
"loss": 4.4025,
|
|
"mean_token_accuracy": 0.26056388914585116,
|
|
"num_tokens": 13698110.0,
|
|
"step": 7410
|
|
},
|
|
{
|
|
"entropy": 5.114210414886474,
|
|
"epoch": 6.205020920502092,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00021511504150360724,
|
|
"loss": 4.318,
|
|
"mean_token_accuracy": 0.26869817078113556,
|
|
"num_tokens": 13707459.0,
|
|
"step": 7415
|
|
},
|
|
{
|
|
"entropy": 5.061584901809693,
|
|
"epoch": 6.209205020920502,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00021480397899571963,
|
|
"loss": 4.3986,
|
|
"mean_token_accuracy": 0.2572581544518471,
|
|
"num_tokens": 13717596.0,
|
|
"step": 7420
|
|
},
|
|
{
|
|
"entropy": 5.007045650482178,
|
|
"epoch": 6.2133891213389125,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00021449304036156525,
|
|
"loss": 4.312,
|
|
"mean_token_accuracy": 0.26988656967878344,
|
|
"num_tokens": 13726871.0,
|
|
"step": 7425
|
|
},
|
|
{
|
|
"entropy": 5.028404331207275,
|
|
"epoch": 6.2175732217573225,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00021418222624100587,
|
|
"loss": 4.2902,
|
|
"mean_token_accuracy": 0.26517807245254515,
|
|
"num_tokens": 13734960.0,
|
|
"step": 7430
|
|
},
|
|
{
|
|
"entropy": 5.092596435546875,
|
|
"epoch": 6.2217573221757325,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00021387153727364677,
|
|
"loss": 4.3267,
|
|
"mean_token_accuracy": 0.26866778284311293,
|
|
"num_tokens": 13743771.0,
|
|
"step": 7435
|
|
},
|
|
{
|
|
"entropy": 5.011627292633056,
|
|
"epoch": 6.2259414225941425,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0002135609740988362,
|
|
"loss": 4.2926,
|
|
"mean_token_accuracy": 0.2614522695541382,
|
|
"num_tokens": 13753080.0,
|
|
"step": 7440
|
|
},
|
|
{
|
|
"entropy": 5.067791080474853,
|
|
"epoch": 6.2301255230125525,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0002132505373556632,
|
|
"loss": 4.3861,
|
|
"mean_token_accuracy": 0.2613463640213013,
|
|
"num_tokens": 13762698.0,
|
|
"step": 7445
|
|
},
|
|
{
|
|
"entropy": 5.043681478500366,
|
|
"epoch": 6.2343096234309625,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002129402276829564,
|
|
"loss": 4.3109,
|
|
"mean_token_accuracy": 0.264802049100399,
|
|
"num_tokens": 13771693.0,
|
|
"step": 7450
|
|
},
|
|
{
|
|
"entropy": 5.07982873916626,
|
|
"epoch": 6.2384937238493725,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00021263004571928345,
|
|
"loss": 4.425,
|
|
"mean_token_accuracy": 0.25666316896677016,
|
|
"num_tokens": 13780939.0,
|
|
"step": 7455
|
|
},
|
|
{
|
|
"entropy": 5.03602786064148,
|
|
"epoch": 6.2426778242677825,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00021231999210294882,
|
|
"loss": 4.277,
|
|
"mean_token_accuracy": 0.27305288016796114,
|
|
"num_tokens": 13790001.0,
|
|
"step": 7460
|
|
},
|
|
{
|
|
"entropy": 5.124095630645752,
|
|
"epoch": 6.2468619246861925,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00021201006747199313,
|
|
"loss": 4.3613,
|
|
"mean_token_accuracy": 0.2635511845350266,
|
|
"num_tokens": 13799360.0,
|
|
"step": 7465
|
|
},
|
|
{
|
|
"entropy": 5.05175633430481,
|
|
"epoch": 6.2510460251046025,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00021170027246419137,
|
|
"loss": 4.3371,
|
|
"mean_token_accuracy": 0.26772892475128174,
|
|
"num_tokens": 13808518.0,
|
|
"step": 7470
|
|
},
|
|
{
|
|
"entropy": 5.138366413116455,
|
|
"epoch": 6.2552301255230125,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00021139060771705205,
|
|
"loss": 4.4191,
|
|
"mean_token_accuracy": 0.2561225429177284,
|
|
"num_tokens": 13818144.0,
|
|
"step": 7475
|
|
},
|
|
{
|
|
"entropy": 5.035146760940552,
|
|
"epoch": 6.2594142259414225,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00021108107386781517,
|
|
"loss": 4.3097,
|
|
"mean_token_accuracy": 0.26861542761325835,
|
|
"num_tokens": 13828063.0,
|
|
"step": 7480
|
|
},
|
|
{
|
|
"entropy": 4.975880336761475,
|
|
"epoch": 6.2635983263598325,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00021077167155345166,
|
|
"loss": 4.267,
|
|
"mean_token_accuracy": 0.2627613991498947,
|
|
"num_tokens": 13837014.0,
|
|
"step": 7485
|
|
},
|
|
{
|
|
"entropy": 5.07652440071106,
|
|
"epoch": 6.2677824267782425,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002104624014106618,
|
|
"loss": 4.3879,
|
|
"mean_token_accuracy": 0.26356396079063416,
|
|
"num_tokens": 13845518.0,
|
|
"step": 7490
|
|
},
|
|
{
|
|
"entropy": 5.065155029296875,
|
|
"epoch": 6.2719665271966525,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00021015326407587389,
|
|
"loss": 4.375,
|
|
"mean_token_accuracy": 0.260299389064312,
|
|
"num_tokens": 13854154.0,
|
|
"step": 7495
|
|
},
|
|
{
|
|
"entropy": 5.087391567230225,
|
|
"epoch": 6.2761506276150625,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00020984426018524278,
|
|
"loss": 4.3635,
|
|
"mean_token_accuracy": 0.26884748935699465,
|
|
"num_tokens": 13863404.0,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"epoch": 6.2761506276150625,
|
|
"eval_entropy": 4.848500711542453,
|
|
"eval_loss": 5.298111915588379,
|
|
"eval_mean_token_accuracy": 0.20408154117488178,
|
|
"eval_num_tokens": 13863404.0,
|
|
"eval_runtime": 2.0812,
|
|
"eval_samples_per_second": 1877.327,
|
|
"eval_steps_per_second": 234.966,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"entropy": 4.978644895553589,
|
|
"epoch": 6.2803347280334725,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00020953539037464882,
|
|
"loss": 4.2291,
|
|
"mean_token_accuracy": 0.2701076939702034,
|
|
"num_tokens": 13872554.0,
|
|
"step": 7505
|
|
},
|
|
{
|
|
"entropy": 5.0080237865448,
|
|
"epoch": 6.2845188284518825,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00020922665527969646,
|
|
"loss": 4.3225,
|
|
"mean_token_accuracy": 0.26744169294834136,
|
|
"num_tokens": 13881740.0,
|
|
"step": 7510
|
|
},
|
|
{
|
|
"entropy": 5.098605966567993,
|
|
"epoch": 6.2887029288702925,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0002089180555357128,
|
|
"loss": 4.3923,
|
|
"mean_token_accuracy": 0.26935034096240995,
|
|
"num_tokens": 13890114.0,
|
|
"step": 7515
|
|
},
|
|
{
|
|
"entropy": 5.082124519348144,
|
|
"epoch": 6.292887029288703,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00020860959177774652,
|
|
"loss": 4.458,
|
|
"mean_token_accuracy": 0.24792285412549972,
|
|
"num_tokens": 13899744.0,
|
|
"step": 7520
|
|
},
|
|
{
|
|
"entropy": 5.159330892562866,
|
|
"epoch": 6.297071129707113,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00020830126464056647,
|
|
"loss": 4.371,
|
|
"mean_token_accuracy": 0.25871034413576127,
|
|
"num_tokens": 13909011.0,
|
|
"step": 7525
|
|
},
|
|
{
|
|
"entropy": 5.066532897949219,
|
|
"epoch": 6.301255230125523,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00020799307475866035,
|
|
"loss": 4.3363,
|
|
"mean_token_accuracy": 0.2609132409095764,
|
|
"num_tokens": 13918435.0,
|
|
"step": 7530
|
|
},
|
|
{
|
|
"entropy": 5.04642391204834,
|
|
"epoch": 6.305439330543933,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00020768502276623336,
|
|
"loss": 4.3499,
|
|
"mean_token_accuracy": 0.2623241916298866,
|
|
"num_tokens": 13927164.0,
|
|
"step": 7535
|
|
},
|
|
{
|
|
"entropy": 5.052899789810181,
|
|
"epoch": 6.309623430962343,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00020737710929720694,
|
|
"loss": 4.2647,
|
|
"mean_token_accuracy": 0.2732947036623955,
|
|
"num_tokens": 13935694.0,
|
|
"step": 7540
|
|
},
|
|
{
|
|
"entropy": 5.140844678878784,
|
|
"epoch": 6.313807531380753,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00020706933498521767,
|
|
"loss": 4.4525,
|
|
"mean_token_accuracy": 0.24935196191072465,
|
|
"num_tokens": 13946068.0,
|
|
"step": 7545
|
|
},
|
|
{
|
|
"entropy": 5.135636520385742,
|
|
"epoch": 6.317991631799163,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002067617004636153,
|
|
"loss": 4.4251,
|
|
"mean_token_accuracy": 0.2586972013115883,
|
|
"num_tokens": 13955238.0,
|
|
"step": 7550
|
|
},
|
|
{
|
|
"entropy": 5.076630115509033,
|
|
"epoch": 6.322175732217573,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00020645420636546246,
|
|
"loss": 4.3098,
|
|
"mean_token_accuracy": 0.2750520408153534,
|
|
"num_tokens": 13964251.0,
|
|
"step": 7555
|
|
},
|
|
{
|
|
"entropy": 5.1363832473754885,
|
|
"epoch": 6.326359832635983,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00020614685332353246,
|
|
"loss": 4.4559,
|
|
"mean_token_accuracy": 0.24637118428945542,
|
|
"num_tokens": 13974111.0,
|
|
"step": 7560
|
|
},
|
|
{
|
|
"entropy": 5.055293893814087,
|
|
"epoch": 6.330543933054393,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00020583964197030852,
|
|
"loss": 4.3069,
|
|
"mean_token_accuracy": 0.26215541660785674,
|
|
"num_tokens": 13982842.0,
|
|
"step": 7565
|
|
},
|
|
{
|
|
"entropy": 4.982450819015503,
|
|
"epoch": 6.334728033472803,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00020553257293798216,
|
|
"loss": 4.2739,
|
|
"mean_token_accuracy": 0.2685685306787491,
|
|
"num_tokens": 13992010.0,
|
|
"step": 7570
|
|
},
|
|
{
|
|
"entropy": 5.054913902282715,
|
|
"epoch": 6.338912133891213,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00020522564685845207,
|
|
"loss": 4.264,
|
|
"mean_token_accuracy": 0.2730960175395012,
|
|
"num_tokens": 14000746.0,
|
|
"step": 7575
|
|
},
|
|
{
|
|
"entropy": 5.055222606658935,
|
|
"epoch": 6.343096234309623,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00020491886436332287,
|
|
"loss": 4.3219,
|
|
"mean_token_accuracy": 0.2686514586210251,
|
|
"num_tokens": 14010664.0,
|
|
"step": 7580
|
|
},
|
|
{
|
|
"entropy": 5.074668312072754,
|
|
"epoch": 6.347280334728033,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00020461222608390346,
|
|
"loss": 4.368,
|
|
"mean_token_accuracy": 0.2645489886403084,
|
|
"num_tokens": 14019473.0,
|
|
"step": 7585
|
|
},
|
|
{
|
|
"entropy": 5.035215091705322,
|
|
"epoch": 6.351464435146443,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00020430573265120623,
|
|
"loss": 4.3769,
|
|
"mean_token_accuracy": 0.27022518068552015,
|
|
"num_tokens": 14027971.0,
|
|
"step": 7590
|
|
},
|
|
{
|
|
"entropy": 4.92142333984375,
|
|
"epoch": 6.355648535564853,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0002039993846959453,
|
|
"loss": 4.1948,
|
|
"mean_token_accuracy": 0.27719499468803405,
|
|
"num_tokens": 14036972.0,
|
|
"step": 7595
|
|
},
|
|
{
|
|
"entropy": 5.050300979614258,
|
|
"epoch": 6.359832635983263,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00020369318284853562,
|
|
"loss": 4.3128,
|
|
"mean_token_accuracy": 0.2577369913458824,
|
|
"num_tokens": 14046486.0,
|
|
"step": 7600
|
|
},
|
|
{
|
|
"entropy": 5.024500942230224,
|
|
"epoch": 6.364016736401673,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00020338712773909125,
|
|
"loss": 4.2983,
|
|
"mean_token_accuracy": 0.2696241304278374,
|
|
"num_tokens": 14056398.0,
|
|
"step": 7605
|
|
},
|
|
{
|
|
"entropy": 5.179713249206543,
|
|
"epoch": 6.368200836820083,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00020308121999742458,
|
|
"loss": 4.4745,
|
|
"mean_token_accuracy": 0.24856538623571395,
|
|
"num_tokens": 14065517.0,
|
|
"step": 7610
|
|
},
|
|
{
|
|
"entropy": 5.057542133331299,
|
|
"epoch": 6.372384937238493,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00020277546025304432,
|
|
"loss": 4.3526,
|
|
"mean_token_accuracy": 0.26338842511177063,
|
|
"num_tokens": 14074563.0,
|
|
"step": 7615
|
|
},
|
|
{
|
|
"entropy": 5.078920984268189,
|
|
"epoch": 6.376569037656903,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00020246984913515503,
|
|
"loss": 4.3815,
|
|
"mean_token_accuracy": 0.2627051919698715,
|
|
"num_tokens": 14083125.0,
|
|
"step": 7620
|
|
},
|
|
{
|
|
"entropy": 5.055297327041626,
|
|
"epoch": 6.380753138075314,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00020216438727265517,
|
|
"loss": 4.3195,
|
|
"mean_token_accuracy": 0.2680700168013573,
|
|
"num_tokens": 14091965.0,
|
|
"step": 7625
|
|
},
|
|
{
|
|
"entropy": 5.00025782585144,
|
|
"epoch": 6.384937238493724,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00020185907529413622,
|
|
"loss": 4.3125,
|
|
"mean_token_accuracy": 0.26454683393239975,
|
|
"num_tokens": 14100872.0,
|
|
"step": 7630
|
|
},
|
|
{
|
|
"entropy": 5.1153617858886715,
|
|
"epoch": 6.389121338912134,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00020155391382788108,
|
|
"loss": 4.3836,
|
|
"mean_token_accuracy": 0.2641046553850174,
|
|
"num_tokens": 14109860.0,
|
|
"step": 7635
|
|
},
|
|
{
|
|
"entropy": 5.115455579757691,
|
|
"epoch": 6.393305439330544,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00020124890350186304,
|
|
"loss": 4.4098,
|
|
"mean_token_accuracy": 0.2558906123042107,
|
|
"num_tokens": 14119256.0,
|
|
"step": 7640
|
|
},
|
|
{
|
|
"entropy": 5.173307704925537,
|
|
"epoch": 6.397489539748954,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0002009440449437443,
|
|
"loss": 4.4972,
|
|
"mean_token_accuracy": 0.24952636957168578,
|
|
"num_tokens": 14128160.0,
|
|
"step": 7645
|
|
},
|
|
{
|
|
"entropy": 5.128319883346558,
|
|
"epoch": 6.401673640167364,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002006393387808747,
|
|
"loss": 4.4319,
|
|
"mean_token_accuracy": 0.25553305000066756,
|
|
"num_tokens": 14137810.0,
|
|
"step": 7650
|
|
},
|
|
{
|
|
"entropy": 5.089184284210205,
|
|
"epoch": 6.405857740585774,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00020033478564029063,
|
|
"loss": 4.2627,
|
|
"mean_token_accuracy": 0.27127518355846403,
|
|
"num_tokens": 14145914.0,
|
|
"step": 7655
|
|
},
|
|
{
|
|
"entropy": 5.139502286911011,
|
|
"epoch": 6.410041841004184,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002000303861487134,
|
|
"loss": 4.4545,
|
|
"mean_token_accuracy": 0.2523098334670067,
|
|
"num_tokens": 14156348.0,
|
|
"step": 7660
|
|
},
|
|
{
|
|
"entropy": 5.181828927993775,
|
|
"epoch": 6.414225941422594,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00019972614093254822,
|
|
"loss": 4.4639,
|
|
"mean_token_accuracy": 0.2589636266231537,
|
|
"num_tokens": 14166079.0,
|
|
"step": 7665
|
|
},
|
|
{
|
|
"entropy": 5.077084064483643,
|
|
"epoch": 6.418410041841004,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00019942205061788292,
|
|
"loss": 4.3601,
|
|
"mean_token_accuracy": 0.2625689893960953,
|
|
"num_tokens": 14175732.0,
|
|
"step": 7670
|
|
},
|
|
{
|
|
"entropy": 5.08632264137268,
|
|
"epoch": 6.422594142259414,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00019911811583048645,
|
|
"loss": 4.3629,
|
|
"mean_token_accuracy": 0.2561234936118126,
|
|
"num_tokens": 14184935.0,
|
|
"step": 7675
|
|
},
|
|
{
|
|
"entropy": 5.086982154846192,
|
|
"epoch": 6.426778242677824,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00019881433719580766,
|
|
"loss": 4.4284,
|
|
"mean_token_accuracy": 0.260427688062191,
|
|
"num_tokens": 14193693.0,
|
|
"step": 7680
|
|
},
|
|
{
|
|
"entropy": 5.056566286087036,
|
|
"epoch": 6.430962343096234,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00019851071533897425,
|
|
"loss": 4.3478,
|
|
"mean_token_accuracy": 0.2657705545425415,
|
|
"num_tokens": 14203323.0,
|
|
"step": 7685
|
|
},
|
|
{
|
|
"entropy": 5.025413656234742,
|
|
"epoch": 6.435146443514644,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00019820725088479107,
|
|
"loss": 4.2969,
|
|
"mean_token_accuracy": 0.26010903567075727,
|
|
"num_tokens": 14211992.0,
|
|
"step": 7690
|
|
},
|
|
{
|
|
"entropy": 5.050085639953613,
|
|
"epoch": 6.439330543933054,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001979039444577393,
|
|
"loss": 4.3435,
|
|
"mean_token_accuracy": 0.2624731048941612,
|
|
"num_tokens": 14221578.0,
|
|
"step": 7695
|
|
},
|
|
{
|
|
"entropy": 5.095527458190918,
|
|
"epoch": 6.443514644351464,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00019760079668197474,
|
|
"loss": 4.3633,
|
|
"mean_token_accuracy": 0.26120885759592055,
|
|
"num_tokens": 14231411.0,
|
|
"step": 7700
|
|
},
|
|
{
|
|
"entropy": 5.075884628295898,
|
|
"epoch": 6.447698744769874,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00019729780818132664,
|
|
"loss": 4.3287,
|
|
"mean_token_accuracy": 0.2689286217093468,
|
|
"num_tokens": 14241317.0,
|
|
"step": 7705
|
|
},
|
|
{
|
|
"entropy": 5.134666538238525,
|
|
"epoch": 6.451882845188284,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00019699497957929702,
|
|
"loss": 4.394,
|
|
"mean_token_accuracy": 0.2579468905925751,
|
|
"num_tokens": 14251248.0,
|
|
"step": 7710
|
|
},
|
|
{
|
|
"entropy": 5.014488077163696,
|
|
"epoch": 6.456066945606695,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00019669231149905804,
|
|
"loss": 4.3295,
|
|
"mean_token_accuracy": 0.2599170312285423,
|
|
"num_tokens": 14261487.0,
|
|
"step": 7715
|
|
},
|
|
{
|
|
"entropy": 5.062679767608643,
|
|
"epoch": 6.460251046025105,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00019638980456345207,
|
|
"loss": 4.3904,
|
|
"mean_token_accuracy": 0.2560228779911995,
|
|
"num_tokens": 14271438.0,
|
|
"step": 7720
|
|
},
|
|
{
|
|
"entropy": 5.072897005081177,
|
|
"epoch": 6.464435146443515,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00019608745939498983,
|
|
"loss": 4.3422,
|
|
"mean_token_accuracy": 0.25823580771684645,
|
|
"num_tokens": 14280683.0,
|
|
"step": 7725
|
|
},
|
|
{
|
|
"entropy": 5.054940271377563,
|
|
"epoch": 6.468619246861925,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00019578527661584898,
|
|
"loss": 4.4003,
|
|
"mean_token_accuracy": 0.25856180638074877,
|
|
"num_tokens": 14290041.0,
|
|
"step": 7730
|
|
},
|
|
{
|
|
"entropy": 5.087418746948242,
|
|
"epoch": 6.472803347280335,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00019548325684787311,
|
|
"loss": 4.3904,
|
|
"mean_token_accuracy": 0.25272819548845293,
|
|
"num_tokens": 14298914.0,
|
|
"step": 7735
|
|
},
|
|
{
|
|
"entropy": 5.082686948776245,
|
|
"epoch": 6.476987447698745,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001951814007125704,
|
|
"loss": 4.3332,
|
|
"mean_token_accuracy": 0.25361494719982147,
|
|
"num_tokens": 14308347.0,
|
|
"step": 7740
|
|
},
|
|
{
|
|
"entropy": 5.043616867065429,
|
|
"epoch": 6.481171548117155,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00019487970883111234,
|
|
"loss": 4.3541,
|
|
"mean_token_accuracy": 0.25611075311899184,
|
|
"num_tokens": 14317839.0,
|
|
"step": 7745
|
|
},
|
|
{
|
|
"entropy": 5.039049959182739,
|
|
"epoch": 6.485355648535565,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00019457818182433206,
|
|
"loss": 4.3489,
|
|
"mean_token_accuracy": 0.2597661390900612,
|
|
"num_tokens": 14327037.0,
|
|
"step": 7750
|
|
},
|
|
{
|
|
"entropy": 5.060854244232178,
|
|
"epoch": 6.489539748953975,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00019427682031272382,
|
|
"loss": 4.358,
|
|
"mean_token_accuracy": 0.2636822506785393,
|
|
"num_tokens": 14337417.0,
|
|
"step": 7755
|
|
},
|
|
{
|
|
"entropy": 5.189823007583618,
|
|
"epoch": 6.493723849372385,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00019397562491644112,
|
|
"loss": 4.4573,
|
|
"mean_token_accuracy": 0.25790254324674605,
|
|
"num_tokens": 14347037.0,
|
|
"step": 7760
|
|
},
|
|
{
|
|
"entropy": 5.090803003311157,
|
|
"epoch": 6.497907949790795,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00019367459625529564,
|
|
"loss": 4.4021,
|
|
"mean_token_accuracy": 0.2566482648253441,
|
|
"num_tokens": 14356509.0,
|
|
"step": 7765
|
|
},
|
|
{
|
|
"entropy": 5.047531652450561,
|
|
"epoch": 6.502092050209205,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00019337373494875604,
|
|
"loss": 4.2851,
|
|
"mean_token_accuracy": 0.2700349003076553,
|
|
"num_tokens": 14365276.0,
|
|
"step": 7770
|
|
},
|
|
{
|
|
"entropy": 5.123957014083862,
|
|
"epoch": 6.506276150627615,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00019307304161594657,
|
|
"loss": 4.4022,
|
|
"mean_token_accuracy": 0.2643188700079918,
|
|
"num_tokens": 14373546.0,
|
|
"step": 7775
|
|
},
|
|
{
|
|
"entropy": 5.062945795059204,
|
|
"epoch": 6.510460251046025,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00019277251687564557,
|
|
"loss": 4.3293,
|
|
"mean_token_accuracy": 0.2669775947928429,
|
|
"num_tokens": 14382767.0,
|
|
"step": 7780
|
|
},
|
|
{
|
|
"entropy": 5.0304361343383786,
|
|
"epoch": 6.514644351464435,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00019247216134628476,
|
|
"loss": 4.3377,
|
|
"mean_token_accuracy": 0.25244544446468353,
|
|
"num_tokens": 14392602.0,
|
|
"step": 7785
|
|
},
|
|
{
|
|
"entropy": 5.03928861618042,
|
|
"epoch": 6.518828451882845,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00019217197564594746,
|
|
"loss": 4.4267,
|
|
"mean_token_accuracy": 0.25861469060182574,
|
|
"num_tokens": 14402191.0,
|
|
"step": 7790
|
|
},
|
|
{
|
|
"entropy": 5.1294880390167235,
|
|
"epoch": 6.523012552301255,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00019187196039236772,
|
|
"loss": 4.3592,
|
|
"mean_token_accuracy": 0.26094249188899993,
|
|
"num_tokens": 14411116.0,
|
|
"step": 7795
|
|
},
|
|
{
|
|
"entropy": 5.031586933135986,
|
|
"epoch": 6.527196652719665,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00019157211620292853,
|
|
"loss": 4.2935,
|
|
"mean_token_accuracy": 0.26494442820549013,
|
|
"num_tokens": 14419725.0,
|
|
"step": 7800
|
|
},
|
|
{
|
|
"entropy": 5.0798896789550785,
|
|
"epoch": 6.531380753138075,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00019127244369466118,
|
|
"loss": 4.4182,
|
|
"mean_token_accuracy": 0.2552990406751633,
|
|
"num_tokens": 14429904.0,
|
|
"step": 7805
|
|
},
|
|
{
|
|
"entropy": 5.030403137207031,
|
|
"epoch": 6.535564853556485,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00019097294348424348,
|
|
"loss": 4.3067,
|
|
"mean_token_accuracy": 0.2629008337855339,
|
|
"num_tokens": 14439384.0,
|
|
"step": 7810
|
|
},
|
|
{
|
|
"entropy": 5.049721431732178,
|
|
"epoch": 6.539748953974895,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00019067361618799856,
|
|
"loss": 4.31,
|
|
"mean_token_accuracy": 0.2663272351026535,
|
|
"num_tokens": 14448002.0,
|
|
"step": 7815
|
|
},
|
|
{
|
|
"entropy": 5.056630039215088,
|
|
"epoch": 6.543933054393305,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00019037446242189398,
|
|
"loss": 4.3231,
|
|
"mean_token_accuracy": 0.2571879461407661,
|
|
"num_tokens": 14456887.0,
|
|
"step": 7820
|
|
},
|
|
{
|
|
"entropy": 5.115304565429687,
|
|
"epoch": 6.548117154811716,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00019007548280154007,
|
|
"loss": 4.482,
|
|
"mean_token_accuracy": 0.24738727360963822,
|
|
"num_tokens": 14466615.0,
|
|
"step": 7825
|
|
},
|
|
{
|
|
"entropy": 5.050726079940796,
|
|
"epoch": 6.552301255230126,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00018977667794218894,
|
|
"loss": 4.3053,
|
|
"mean_token_accuracy": 0.2661483645439148,
|
|
"num_tokens": 14476684.0,
|
|
"step": 7830
|
|
},
|
|
{
|
|
"entropy": 5.035591077804566,
|
|
"epoch": 6.556485355648536,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0001894780484587328,
|
|
"loss": 4.3019,
|
|
"mean_token_accuracy": 0.268118879199028,
|
|
"num_tokens": 14485364.0,
|
|
"step": 7835
|
|
},
|
|
{
|
|
"entropy": 5.050217723846435,
|
|
"epoch": 6.560669456066946,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00018917959496570318,
|
|
"loss": 4.381,
|
|
"mean_token_accuracy": 0.26267995983362197,
|
|
"num_tokens": 14495099.0,
|
|
"step": 7840
|
|
},
|
|
{
|
|
"entropy": 5.099884128570556,
|
|
"epoch": 6.564853556485356,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001888813180772694,
|
|
"loss": 4.3364,
|
|
"mean_token_accuracy": 0.2596940204501152,
|
|
"num_tokens": 14504051.0,
|
|
"step": 7845
|
|
},
|
|
{
|
|
"entropy": 5.10122537612915,
|
|
"epoch": 6.569037656903766,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00018858321840723724,
|
|
"loss": 4.3399,
|
|
"mean_token_accuracy": 0.2580750986933708,
|
|
"num_tokens": 14513517.0,
|
|
"step": 7850
|
|
},
|
|
{
|
|
"entropy": 5.009996509552002,
|
|
"epoch": 6.573221757322176,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00018828529656904802,
|
|
"loss": 4.4023,
|
|
"mean_token_accuracy": 0.25650571286678314,
|
|
"num_tokens": 14522504.0,
|
|
"step": 7855
|
|
},
|
|
{
|
|
"entropy": 5.1013140201568605,
|
|
"epoch": 6.577405857740586,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00018798755317577691,
|
|
"loss": 4.3965,
|
|
"mean_token_accuracy": 0.26109466403722764,
|
|
"num_tokens": 14532215.0,
|
|
"step": 7860
|
|
},
|
|
{
|
|
"entropy": 4.97636661529541,
|
|
"epoch": 6.581589958158996,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.000187689988840132,
|
|
"loss": 4.249,
|
|
"mean_token_accuracy": 0.27210971266031264,
|
|
"num_tokens": 14540489.0,
|
|
"step": 7865
|
|
},
|
|
{
|
|
"entropy": 5.041528129577637,
|
|
"epoch": 6.585774058577406,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001873926041744528,
|
|
"loss": 4.3429,
|
|
"mean_token_accuracy": 0.2642879575490952,
|
|
"num_tokens": 14550476.0,
|
|
"step": 7870
|
|
},
|
|
{
|
|
"entropy": 5.08097243309021,
|
|
"epoch": 6.589958158995816,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00018709539979070922,
|
|
"loss": 4.3014,
|
|
"mean_token_accuracy": 0.266031388938427,
|
|
"num_tokens": 14559349.0,
|
|
"step": 7875
|
|
},
|
|
{
|
|
"entropy": 5.0223753452301025,
|
|
"epoch": 6.594142259414226,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00018679837630049994,
|
|
"loss": 4.3472,
|
|
"mean_token_accuracy": 0.27019475400447845,
|
|
"num_tokens": 14568480.0,
|
|
"step": 7880
|
|
},
|
|
{
|
|
"entropy": 5.082499361038208,
|
|
"epoch": 6.598326359832636,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00018650153431505167,
|
|
"loss": 4.3449,
|
|
"mean_token_accuracy": 0.26862629055976867,
|
|
"num_tokens": 14578071.0,
|
|
"step": 7885
|
|
},
|
|
{
|
|
"entropy": 5.045102882385254,
|
|
"epoch": 6.602510460251046,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001862048744452175,
|
|
"loss": 4.3417,
|
|
"mean_token_accuracy": 0.2667803481221199,
|
|
"num_tokens": 14586211.0,
|
|
"step": 7890
|
|
},
|
|
{
|
|
"entropy": 5.047694683074951,
|
|
"epoch": 6.606694560669456,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001859083973014757,
|
|
"loss": 4.3497,
|
|
"mean_token_accuracy": 0.2631842628121376,
|
|
"num_tokens": 14595206.0,
|
|
"step": 7895
|
|
},
|
|
{
|
|
"entropy": 5.073220729827881,
|
|
"epoch": 6.610878661087866,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00018561210349392855,
|
|
"loss": 4.3548,
|
|
"mean_token_accuracy": 0.26026915311813353,
|
|
"num_tokens": 14604569.0,
|
|
"step": 7900
|
|
},
|
|
{
|
|
"entropy": 5.135022068023682,
|
|
"epoch": 6.615062761506276,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001853159936323011,
|
|
"loss": 4.3666,
|
|
"mean_token_accuracy": 0.258091539144516,
|
|
"num_tokens": 14614040.0,
|
|
"step": 7905
|
|
},
|
|
{
|
|
"entropy": 5.061950826644898,
|
|
"epoch": 6.619246861924686,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00018502006832593992,
|
|
"loss": 4.3449,
|
|
"mean_token_accuracy": 0.2690091371536255,
|
|
"num_tokens": 14624461.0,
|
|
"step": 7910
|
|
},
|
|
{
|
|
"entropy": 5.039607381820678,
|
|
"epoch": 6.623430962343097,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00018472432818381158,
|
|
"loss": 4.3801,
|
|
"mean_token_accuracy": 0.2627553537487984,
|
|
"num_tokens": 14633417.0,
|
|
"step": 7915
|
|
},
|
|
{
|
|
"entropy": 5.015586042404175,
|
|
"epoch": 6.627615062761507,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00018442877381450175,
|
|
"loss": 4.2823,
|
|
"mean_token_accuracy": 0.2706668317317963,
|
|
"num_tokens": 14642387.0,
|
|
"step": 7920
|
|
},
|
|
{
|
|
"entropy": 4.969128894805908,
|
|
"epoch": 6.631799163179917,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0001841334058262139,
|
|
"loss": 4.2784,
|
|
"mean_token_accuracy": 0.26777905523777007,
|
|
"num_tokens": 14650613.0,
|
|
"step": 7925
|
|
},
|
|
{
|
|
"entropy": 4.996392107009887,
|
|
"epoch": 6.635983263598327,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00018383822482676777,
|
|
"loss": 4.3811,
|
|
"mean_token_accuracy": 0.2690281540155411,
|
|
"num_tokens": 14659411.0,
|
|
"step": 7930
|
|
},
|
|
{
|
|
"entropy": 4.997714281082153,
|
|
"epoch": 6.640167364016737,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00018354323142359848,
|
|
"loss": 4.3382,
|
|
"mean_token_accuracy": 0.2615195870399475,
|
|
"num_tokens": 14668375.0,
|
|
"step": 7935
|
|
},
|
|
{
|
|
"entropy": 5.045203638076782,
|
|
"epoch": 6.644351464435147,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00018324842622375497,
|
|
"loss": 4.3104,
|
|
"mean_token_accuracy": 0.26712699234485626,
|
|
"num_tokens": 14677608.0,
|
|
"step": 7940
|
|
},
|
|
{
|
|
"entropy": 5.111136436462402,
|
|
"epoch": 6.648535564853557,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00018295380983389905,
|
|
"loss": 4.3661,
|
|
"mean_token_accuracy": 0.2637924775481224,
|
|
"num_tokens": 14686646.0,
|
|
"step": 7945
|
|
},
|
|
{
|
|
"entropy": 5.013623476028442,
|
|
"epoch": 6.652719665271967,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001826593828603037,
|
|
"loss": 4.3206,
|
|
"mean_token_accuracy": 0.26068522930145266,
|
|
"num_tokens": 14695703.0,
|
|
"step": 7950
|
|
},
|
|
{
|
|
"entropy": 5.006418704986572,
|
|
"epoch": 6.656903765690377,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001823651459088524,
|
|
"loss": 4.3273,
|
|
"mean_token_accuracy": 0.2670476630330086,
|
|
"num_tokens": 14704493.0,
|
|
"step": 7955
|
|
},
|
|
{
|
|
"entropy": 5.08234281539917,
|
|
"epoch": 6.661087866108787,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00018207109958503747,
|
|
"loss": 4.4096,
|
|
"mean_token_accuracy": 0.2644865497946739,
|
|
"num_tokens": 14713119.0,
|
|
"step": 7960
|
|
},
|
|
{
|
|
"entropy": 5.053924322128296,
|
|
"epoch": 6.665271966527197,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000181777244493959,
|
|
"loss": 4.3613,
|
|
"mean_token_accuracy": 0.2604943573474884,
|
|
"num_tokens": 14722360.0,
|
|
"step": 7965
|
|
},
|
|
{
|
|
"entropy": 5.107659912109375,
|
|
"epoch": 6.669456066945607,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001814835812403235,
|
|
"loss": 4.3812,
|
|
"mean_token_accuracy": 0.2630063146352768,
|
|
"num_tokens": 14732066.0,
|
|
"step": 7970
|
|
},
|
|
{
|
|
"entropy": 5.0399716854095455,
|
|
"epoch": 6.673640167364017,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00018119011042844286,
|
|
"loss": 4.3112,
|
|
"mean_token_accuracy": 0.2605651393532753,
|
|
"num_tokens": 14740968.0,
|
|
"step": 7975
|
|
},
|
|
{
|
|
"entropy": 5.036748647689819,
|
|
"epoch": 6.677824267782427,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001808968326622325,
|
|
"loss": 4.3455,
|
|
"mean_token_accuracy": 0.2557371199131012,
|
|
"num_tokens": 14750574.0,
|
|
"step": 7980
|
|
},
|
|
{
|
|
"entropy": 5.053411293029785,
|
|
"epoch": 6.682008368200837,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00018060374854521116,
|
|
"loss": 4.3111,
|
|
"mean_token_accuracy": 0.2627292677760124,
|
|
"num_tokens": 14760064.0,
|
|
"step": 7985
|
|
},
|
|
{
|
|
"entropy": 5.041105031967163,
|
|
"epoch": 6.686192468619247,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00018031085868049884,
|
|
"loss": 4.3765,
|
|
"mean_token_accuracy": 0.26246311515569687,
|
|
"num_tokens": 14769721.0,
|
|
"step": 7990
|
|
},
|
|
{
|
|
"entropy": 5.1072190284729,
|
|
"epoch": 6.690376569037657,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001800181636708157,
|
|
"loss": 4.4017,
|
|
"mean_token_accuracy": 0.25853475630283357,
|
|
"num_tokens": 14778895.0,
|
|
"step": 7995
|
|
},
|
|
{
|
|
"entropy": 5.042085027694702,
|
|
"epoch": 6.694560669456067,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00017972566411848112,
|
|
"loss": 4.3617,
|
|
"mean_token_accuracy": 0.26257790476083753,
|
|
"num_tokens": 14787541.0,
|
|
"step": 8000
|
|
},
|
|
{
|
|
"epoch": 6.694560669456067,
|
|
"eval_entropy": 4.859297962520264,
|
|
"eval_loss": 5.287324905395508,
|
|
"eval_mean_token_accuracy": 0.20587582493120907,
|
|
"eval_num_tokens": 14787541.0,
|
|
"eval_runtime": 2.2797,
|
|
"eval_samples_per_second": 1713.824,
|
|
"eval_steps_per_second": 214.502,
|
|
"step": 8000
|
|
},
|
|
{
|
|
"entropy": 5.062149381637573,
|
|
"epoch": 6.698744769874477,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000179433360625412,
|
|
"loss": 4.3461,
|
|
"mean_token_accuracy": 0.26085254848003386,
|
|
"num_tokens": 14795669.0,
|
|
"step": 8005
|
|
},
|
|
{
|
|
"entropy": 5.070108842849732,
|
|
"epoch": 6.702928870292887,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00017914125379312208,
|
|
"loss": 4.3263,
|
|
"mean_token_accuracy": 0.2619004756212234,
|
|
"num_tokens": 14804587.0,
|
|
"step": 8010
|
|
},
|
|
{
|
|
"entropy": 5.030421400070191,
|
|
"epoch": 6.707112970711297,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00017884934422272026,
|
|
"loss": 4.3093,
|
|
"mean_token_accuracy": 0.26812771111726763,
|
|
"num_tokens": 14813746.0,
|
|
"step": 8015
|
|
},
|
|
{
|
|
"entropy": 5.039276361465454,
|
|
"epoch": 6.711297071129707,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00017855763251490924,
|
|
"loss": 4.3791,
|
|
"mean_token_accuracy": 0.2563133776187897,
|
|
"num_tokens": 14822906.0,
|
|
"step": 8020
|
|
},
|
|
{
|
|
"entropy": 5.0845130443572994,
|
|
"epoch": 6.7154811715481175,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00017826611926998514,
|
|
"loss": 4.3857,
|
|
"mean_token_accuracy": 0.2646468535065651,
|
|
"num_tokens": 14832212.0,
|
|
"step": 8025
|
|
},
|
|
{
|
|
"entropy": 5.051948499679566,
|
|
"epoch": 6.7196652719665275,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001779748050878352,
|
|
"loss": 4.3603,
|
|
"mean_token_accuracy": 0.2570932567119598,
|
|
"num_tokens": 14841381.0,
|
|
"step": 8030
|
|
},
|
|
{
|
|
"entropy": 4.984414672851562,
|
|
"epoch": 6.7238493723849375,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00017768369056793727,
|
|
"loss": 4.2382,
|
|
"mean_token_accuracy": 0.2748552232980728,
|
|
"num_tokens": 14850090.0,
|
|
"step": 8035
|
|
},
|
|
{
|
|
"entropy": 5.044227457046508,
|
|
"epoch": 6.7280334728033475,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.000177392776309358,
|
|
"loss": 4.3526,
|
|
"mean_token_accuracy": 0.2622598364949226,
|
|
"num_tokens": 14858885.0,
|
|
"step": 8040
|
|
},
|
|
{
|
|
"entropy": 5.018987989425659,
|
|
"epoch": 6.7322175732217575,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00017710206291075248,
|
|
"loss": 4.331,
|
|
"mean_token_accuracy": 0.2611741289496422,
|
|
"num_tokens": 14868173.0,
|
|
"step": 8045
|
|
},
|
|
{
|
|
"entropy": 5.030536651611328,
|
|
"epoch": 6.7364016736401675,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00017681155097036194,
|
|
"loss": 4.3337,
|
|
"mean_token_accuracy": 0.2665359631180763,
|
|
"num_tokens": 14877164.0,
|
|
"step": 8050
|
|
},
|
|
{
|
|
"entropy": 5.046485042572021,
|
|
"epoch": 6.7405857740585775,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001765212410860132,
|
|
"loss": 4.3516,
|
|
"mean_token_accuracy": 0.26321945041418077,
|
|
"num_tokens": 14886285.0,
|
|
"step": 8055
|
|
},
|
|
{
|
|
"entropy": 4.977289390563965,
|
|
"epoch": 6.7447698744769875,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00017623113385511753,
|
|
"loss": 4.3117,
|
|
"mean_token_accuracy": 0.26183947920799255,
|
|
"num_tokens": 14895079.0,
|
|
"step": 8060
|
|
},
|
|
{
|
|
"entropy": 5.0467628002166744,
|
|
"epoch": 6.7489539748953975,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001759412298746687,
|
|
"loss": 4.3437,
|
|
"mean_token_accuracy": 0.25639995783567426,
|
|
"num_tokens": 14904418.0,
|
|
"step": 8065
|
|
},
|
|
{
|
|
"entropy": 5.050811719894409,
|
|
"epoch": 6.7531380753138075,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00017565152974124276,
|
|
"loss": 4.3954,
|
|
"mean_token_accuracy": 0.24853759855031968,
|
|
"num_tokens": 14913621.0,
|
|
"step": 8070
|
|
},
|
|
{
|
|
"entropy": 5.018222379684448,
|
|
"epoch": 6.7573221757322175,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.0001753620340509959,
|
|
"loss": 4.3059,
|
|
"mean_token_accuracy": 0.26400296241044996,
|
|
"num_tokens": 14921621.0,
|
|
"step": 8075
|
|
},
|
|
{
|
|
"entropy": 4.981066036224365,
|
|
"epoch": 6.7615062761506275,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00017507274339966368,
|
|
"loss": 4.2866,
|
|
"mean_token_accuracy": 0.2669717788696289,
|
|
"num_tokens": 14930732.0,
|
|
"step": 8080
|
|
},
|
|
{
|
|
"entropy": 5.001305103302002,
|
|
"epoch": 6.7656903765690375,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00017478365838255978,
|
|
"loss": 4.3812,
|
|
"mean_token_accuracy": 0.2603939563035965,
|
|
"num_tokens": 14939626.0,
|
|
"step": 8085
|
|
},
|
|
{
|
|
"entropy": 5.081197595596313,
|
|
"epoch": 6.7698744769874475,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001744947795945746,
|
|
"loss": 4.3533,
|
|
"mean_token_accuracy": 0.2690828189253807,
|
|
"num_tokens": 14948718.0,
|
|
"step": 8090
|
|
},
|
|
{
|
|
"entropy": 5.051125764846802,
|
|
"epoch": 6.7740585774058575,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001742061076301744,
|
|
"loss": 4.1991,
|
|
"mean_token_accuracy": 0.2752571314573288,
|
|
"num_tokens": 14957703.0,
|
|
"step": 8095
|
|
},
|
|
{
|
|
"entropy": 4.997343826293945,
|
|
"epoch": 6.7782426778242675,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00017391764308339957,
|
|
"loss": 4.2845,
|
|
"mean_token_accuracy": 0.2664765894412994,
|
|
"num_tokens": 14966693.0,
|
|
"step": 8100
|
|
},
|
|
{
|
|
"entropy": 5.086082029342651,
|
|
"epoch": 6.7824267782426775,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00017362938654786383,
|
|
"loss": 4.4034,
|
|
"mean_token_accuracy": 0.26028103828430177,
|
|
"num_tokens": 14977267.0,
|
|
"step": 8105
|
|
},
|
|
{
|
|
"entropy": 5.095875263214111,
|
|
"epoch": 6.786610878661088,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00017334133861675278,
|
|
"loss": 4.3871,
|
|
"mean_token_accuracy": 0.25572931617498396,
|
|
"num_tokens": 14985648.0,
|
|
"step": 8110
|
|
},
|
|
{
|
|
"entropy": 5.047555160522461,
|
|
"epoch": 6.790794979079498,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0001730534998828226,
|
|
"loss": 4.3422,
|
|
"mean_token_accuracy": 0.26074492037296293,
|
|
"num_tokens": 14994915.0,
|
|
"step": 8115
|
|
},
|
|
{
|
|
"entropy": 5.043225383758545,
|
|
"epoch": 6.794979079497908,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001727658709383993,
|
|
"loss": 4.3153,
|
|
"mean_token_accuracy": 0.26998747885227203,
|
|
"num_tokens": 15004288.0,
|
|
"step": 8120
|
|
},
|
|
{
|
|
"entropy": 5.085497331619263,
|
|
"epoch": 6.799163179916318,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00017247845237537693,
|
|
"loss": 4.3795,
|
|
"mean_token_accuracy": 0.2614174589514732,
|
|
"num_tokens": 15012970.0,
|
|
"step": 8125
|
|
},
|
|
{
|
|
"entropy": 5.072023820877075,
|
|
"epoch": 6.803347280334728,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00017219124478521676,
|
|
"loss": 4.4198,
|
|
"mean_token_accuracy": 0.2502362594008446,
|
|
"num_tokens": 15022989.0,
|
|
"step": 8130
|
|
},
|
|
{
|
|
"entropy": 5.116892671585083,
|
|
"epoch": 6.807531380753138,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00017190424875894567,
|
|
"loss": 4.3835,
|
|
"mean_token_accuracy": 0.26622540652751925,
|
|
"num_tokens": 15032665.0,
|
|
"step": 8135
|
|
},
|
|
{
|
|
"entropy": 5.112889289855957,
|
|
"epoch": 6.811715481171548,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00017161746488715546,
|
|
"loss": 4.356,
|
|
"mean_token_accuracy": 0.2618714287877083,
|
|
"num_tokens": 15041109.0,
|
|
"step": 8140
|
|
},
|
|
{
|
|
"entropy": 4.965719079971313,
|
|
"epoch": 6.815899581589958,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00017133089376000142,
|
|
"loss": 4.2781,
|
|
"mean_token_accuracy": 0.273608335852623,
|
|
"num_tokens": 15050080.0,
|
|
"step": 8145
|
|
},
|
|
{
|
|
"entropy": 5.003106307983399,
|
|
"epoch": 6.820083682008368,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0001710445359672005,
|
|
"loss": 4.3439,
|
|
"mean_token_accuracy": 0.2660954982042313,
|
|
"num_tokens": 15058429.0,
|
|
"step": 8150
|
|
},
|
|
{
|
|
"entropy": 5.039329910278321,
|
|
"epoch": 6.824267782426778,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00017075839209803117,
|
|
"loss": 4.3045,
|
|
"mean_token_accuracy": 0.26352973133325575,
|
|
"num_tokens": 15068652.0,
|
|
"step": 8155
|
|
},
|
|
{
|
|
"entropy": 5.084462451934814,
|
|
"epoch": 6.828451882845188,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00017047246274133144,
|
|
"loss": 4.3112,
|
|
"mean_token_accuracy": 0.25799543857574464,
|
|
"num_tokens": 15077643.0,
|
|
"step": 8160
|
|
},
|
|
{
|
|
"entropy": 5.088665294647217,
|
|
"epoch": 6.832635983263598,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001701867484854981,
|
|
"loss": 4.3515,
|
|
"mean_token_accuracy": 0.2615647569298744,
|
|
"num_tokens": 15087471.0,
|
|
"step": 8165
|
|
},
|
|
{
|
|
"entropy": 5.157473182678222,
|
|
"epoch": 6.836820083682008,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00016990124991848497,
|
|
"loss": 4.4588,
|
|
"mean_token_accuracy": 0.2577340483665466,
|
|
"num_tokens": 15096357.0,
|
|
"step": 8170
|
|
},
|
|
{
|
|
"entropy": 5.057214164733887,
|
|
"epoch": 6.841004184100418,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001696159676278023,
|
|
"loss": 4.3153,
|
|
"mean_token_accuracy": 0.2582933992147446,
|
|
"num_tokens": 15105424.0,
|
|
"step": 8175
|
|
},
|
|
{
|
|
"entropy": 5.047346687316894,
|
|
"epoch": 6.845188284518828,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001693309022005153,
|
|
"loss": 4.3667,
|
|
"mean_token_accuracy": 0.26533435136079786,
|
|
"num_tokens": 15115641.0,
|
|
"step": 8180
|
|
},
|
|
{
|
|
"entropy": 5.064873218536377,
|
|
"epoch": 6.849372384937238,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00016904605422324255,
|
|
"loss": 4.3292,
|
|
"mean_token_accuracy": 0.26459945142269137,
|
|
"num_tokens": 15124765.0,
|
|
"step": 8185
|
|
},
|
|
{
|
|
"entropy": 4.963242769241333,
|
|
"epoch": 6.853556485355648,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00016876142428215552,
|
|
"loss": 4.2832,
|
|
"mean_token_accuracy": 0.2670376360416412,
|
|
"num_tokens": 15133591.0,
|
|
"step": 8190
|
|
},
|
|
{
|
|
"entropy": 5.059290266036987,
|
|
"epoch": 6.857740585774058,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0001684770129629768,
|
|
"loss": 4.3973,
|
|
"mean_token_accuracy": 0.25397323071956635,
|
|
"num_tokens": 15143291.0,
|
|
"step": 8195
|
|
},
|
|
{
|
|
"entropy": 5.122475957870483,
|
|
"epoch": 6.861924686192468,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00016819282085097918,
|
|
"loss": 4.4571,
|
|
"mean_token_accuracy": 0.24787489473819732,
|
|
"num_tokens": 15152806.0,
|
|
"step": 8200
|
|
},
|
|
{
|
|
"entropy": 5.075448036193848,
|
|
"epoch": 6.866108786610878,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00016790884853098441,
|
|
"loss": 4.3622,
|
|
"mean_token_accuracy": 0.25432624071836474,
|
|
"num_tokens": 15163116.0,
|
|
"step": 8205
|
|
},
|
|
{
|
|
"entropy": 5.043636989593506,
|
|
"epoch": 6.870292887029288,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001676250965873619,
|
|
"loss": 4.3589,
|
|
"mean_token_accuracy": 0.2569292441010475,
|
|
"num_tokens": 15172916.0,
|
|
"step": 8210
|
|
},
|
|
{
|
|
"entropy": 5.065410232543945,
|
|
"epoch": 6.874476987447698,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00016734156560402742,
|
|
"loss": 4.3399,
|
|
"mean_token_accuracy": 0.26283870488405225,
|
|
"num_tokens": 15181715.0,
|
|
"step": 8215
|
|
},
|
|
{
|
|
"entropy": 5.03677544593811,
|
|
"epoch": 6.878661087866108,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001670582561644421,
|
|
"loss": 4.3202,
|
|
"mean_token_accuracy": 0.26895684748888016,
|
|
"num_tokens": 15191290.0,
|
|
"step": 8220
|
|
},
|
|
{
|
|
"entropy": 5.026115036010742,
|
|
"epoch": 6.882845188284519,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00016677516885161127,
|
|
"loss": 4.3249,
|
|
"mean_token_accuracy": 0.26135091185569764,
|
|
"num_tokens": 15200614.0,
|
|
"step": 8225
|
|
},
|
|
{
|
|
"entropy": 5.14013524055481,
|
|
"epoch": 6.887029288702929,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001664923042480832,
|
|
"loss": 4.4482,
|
|
"mean_token_accuracy": 0.2467765063047409,
|
|
"num_tokens": 15210793.0,
|
|
"step": 8230
|
|
},
|
|
{
|
|
"entropy": 5.040735244750977,
|
|
"epoch": 6.891213389121339,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00016620966293594763,
|
|
"loss": 4.2784,
|
|
"mean_token_accuracy": 0.26931155025959014,
|
|
"num_tokens": 15219968.0,
|
|
"step": 8235
|
|
},
|
|
{
|
|
"entropy": 5.0545717716217045,
|
|
"epoch": 6.895397489539749,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00016592724549683514,
|
|
"loss": 4.3356,
|
|
"mean_token_accuracy": 0.26684281080961225,
|
|
"num_tokens": 15229434.0,
|
|
"step": 8240
|
|
},
|
|
{
|
|
"entropy": 4.9897627353668215,
|
|
"epoch": 6.899581589958159,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001656450525119152,
|
|
"loss": 4.2604,
|
|
"mean_token_accuracy": 0.27228819131851195,
|
|
"num_tokens": 15238784.0,
|
|
"step": 8245
|
|
},
|
|
{
|
|
"entropy": 5.071241426467895,
|
|
"epoch": 6.903765690376569,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00016536308456189568,
|
|
"loss": 4.3957,
|
|
"mean_token_accuracy": 0.2549637109041214,
|
|
"num_tokens": 15247926.0,
|
|
"step": 8250
|
|
},
|
|
{
|
|
"entropy": 5.127165699005127,
|
|
"epoch": 6.907949790794979,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00016508134222702133,
|
|
"loss": 4.4268,
|
|
"mean_token_accuracy": 0.25532266944646836,
|
|
"num_tokens": 15257158.0,
|
|
"step": 8255
|
|
},
|
|
{
|
|
"entropy": 5.11683406829834,
|
|
"epoch": 6.912133891213389,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00016479982608707245,
|
|
"loss": 4.3875,
|
|
"mean_token_accuracy": 0.258647920191288,
|
|
"num_tokens": 15266639.0,
|
|
"step": 8260
|
|
},
|
|
{
|
|
"entropy": 5.0732522964477536,
|
|
"epoch": 6.916317991631799,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00016451853672136418,
|
|
"loss": 4.324,
|
|
"mean_token_accuracy": 0.2630228266119957,
|
|
"num_tokens": 15275446.0,
|
|
"step": 8265
|
|
},
|
|
{
|
|
"entropy": 5.063825845718384,
|
|
"epoch": 6.920502092050209,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001642374747087446,
|
|
"loss": 4.3951,
|
|
"mean_token_accuracy": 0.2583335295319557,
|
|
"num_tokens": 15285011.0,
|
|
"step": 8270
|
|
},
|
|
{
|
|
"entropy": 5.081723260879516,
|
|
"epoch": 6.924686192468619,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00016395664062759442,
|
|
"loss": 4.519,
|
|
"mean_token_accuracy": 0.2532519116997719,
|
|
"num_tokens": 15295066.0,
|
|
"step": 8275
|
|
},
|
|
{
|
|
"entropy": 5.118585014343262,
|
|
"epoch": 6.928870292887029,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00016367603505582483,
|
|
"loss": 4.3408,
|
|
"mean_token_accuracy": 0.2631371781229973,
|
|
"num_tokens": 15304660.0,
|
|
"step": 8280
|
|
},
|
|
{
|
|
"entropy": 5.041181135177612,
|
|
"epoch": 6.933054393305439,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001633956585708769,
|
|
"loss": 4.3515,
|
|
"mean_token_accuracy": 0.26335434168577193,
|
|
"num_tokens": 15314376.0,
|
|
"step": 8285
|
|
},
|
|
{
|
|
"entropy": 4.969250154495239,
|
|
"epoch": 6.937238493723849,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00016311551174972062,
|
|
"loss": 4.2613,
|
|
"mean_token_accuracy": 0.27426518201828004,
|
|
"num_tokens": 15323414.0,
|
|
"step": 8290
|
|
},
|
|
{
|
|
"entropy": 5.042141532897949,
|
|
"epoch": 6.941422594142259,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00016283559516885285,
|
|
"loss": 4.3742,
|
|
"mean_token_accuracy": 0.25510680228471755,
|
|
"num_tokens": 15332266.0,
|
|
"step": 8295
|
|
},
|
|
{
|
|
"entropy": 5.049210834503174,
|
|
"epoch": 6.945606694560669,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00016255590940429717,
|
|
"loss": 4.3642,
|
|
"mean_token_accuracy": 0.25772922188043595,
|
|
"num_tokens": 15342238.0,
|
|
"step": 8300
|
|
},
|
|
{
|
|
"entropy": 5.126092624664307,
|
|
"epoch": 6.949790794979079,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001622764550316016,
|
|
"loss": 4.4452,
|
|
"mean_token_accuracy": 0.25809556394815447,
|
|
"num_tokens": 15351565.0,
|
|
"step": 8305
|
|
},
|
|
{
|
|
"entropy": 5.076201343536377,
|
|
"epoch": 6.95397489539749,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001619972326258386,
|
|
"loss": 4.3694,
|
|
"mean_token_accuracy": 0.2544846460223198,
|
|
"num_tokens": 15361064.0,
|
|
"step": 8310
|
|
},
|
|
{
|
|
"entropy": 5.008136510848999,
|
|
"epoch": 6.9581589958159,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00016171824276160288,
|
|
"loss": 4.3775,
|
|
"mean_token_accuracy": 0.253261524438858,
|
|
"num_tokens": 15371152.0,
|
|
"step": 8315
|
|
},
|
|
{
|
|
"entropy": 5.074177265167236,
|
|
"epoch": 6.96234309623431,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00016143948601301064,
|
|
"loss": 4.3919,
|
|
"mean_token_accuracy": 0.25521156638860704,
|
|
"num_tokens": 15381546.0,
|
|
"step": 8320
|
|
},
|
|
{
|
|
"entropy": 5.021331930160523,
|
|
"epoch": 6.96652719665272,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001611609629536986,
|
|
"loss": 4.297,
|
|
"mean_token_accuracy": 0.2707701876759529,
|
|
"num_tokens": 15390581.0,
|
|
"step": 8325
|
|
},
|
|
{
|
|
"entropy": 5.0649947166442875,
|
|
"epoch": 6.97071129707113,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001608826741568223,
|
|
"loss": 4.3422,
|
|
"mean_token_accuracy": 0.26897594034671785,
|
|
"num_tokens": 15400035.0,
|
|
"step": 8330
|
|
},
|
|
{
|
|
"entropy": 5.0840735912323,
|
|
"epoch": 6.97489539748954,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001606046201950555,
|
|
"loss": 4.392,
|
|
"mean_token_accuracy": 0.2564506381750107,
|
|
"num_tokens": 15409199.0,
|
|
"step": 8335
|
|
},
|
|
{
|
|
"entropy": 5.020877361297607,
|
|
"epoch": 6.97907949790795,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00016032680164058832,
|
|
"loss": 4.2716,
|
|
"mean_token_accuracy": 0.2738534063100815,
|
|
"num_tokens": 15417851.0,
|
|
"step": 8340
|
|
},
|
|
{
|
|
"entropy": 5.055433368682861,
|
|
"epoch": 6.98326359832636,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00016004921906512699,
|
|
"loss": 4.2813,
|
|
"mean_token_accuracy": 0.27073655277490616,
|
|
"num_tokens": 15427072.0,
|
|
"step": 8345
|
|
},
|
|
{
|
|
"entropy": 5.075017404556275,
|
|
"epoch": 6.98744769874477,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001597718730398916,
|
|
"loss": 4.3865,
|
|
"mean_token_accuracy": 0.25538610816001894,
|
|
"num_tokens": 15437486.0,
|
|
"step": 8350
|
|
},
|
|
{
|
|
"entropy": 5.036880588531494,
|
|
"epoch": 6.99163179916318,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00015949476413561564,
|
|
"loss": 4.3712,
|
|
"mean_token_accuracy": 0.26474476903676986,
|
|
"num_tokens": 15447650.0,
|
|
"step": 8355
|
|
},
|
|
{
|
|
"entropy": 4.9965174198150635,
|
|
"epoch": 6.99581589958159,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00015921789292254484,
|
|
"loss": 4.294,
|
|
"mean_token_accuracy": 0.26825576424598696,
|
|
"num_tokens": 15456650.0,
|
|
"step": 8360
|
|
},
|
|
{
|
|
"entropy": 5.040670156478882,
|
|
"epoch": 7.0,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00015894125997043551,
|
|
"loss": 4.3139,
|
|
"mean_token_accuracy": 0.2720404386520386,
|
|
"num_tokens": 15465254.0,
|
|
"step": 8365
|
|
},
|
|
{
|
|
"entropy": 5.053529453277588,
|
|
"epoch": 7.00418410041841,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00015866486584855397,
|
|
"loss": 4.2176,
|
|
"mean_token_accuracy": 0.27785159945487975,
|
|
"num_tokens": 15474018.0,
|
|
"step": 8370
|
|
},
|
|
{
|
|
"entropy": 5.071825742721558,
|
|
"epoch": 7.00836820083682,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00015838871112567465,
|
|
"loss": 4.2976,
|
|
"mean_token_accuracy": 0.2650803431868553,
|
|
"num_tokens": 15483871.0,
|
|
"step": 8375
|
|
},
|
|
{
|
|
"entropy": 4.992147922515869,
|
|
"epoch": 7.01255230125523,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001581127963700798,
|
|
"loss": 4.2221,
|
|
"mean_token_accuracy": 0.2777676463127136,
|
|
"num_tokens": 15492996.0,
|
|
"step": 8380
|
|
},
|
|
{
|
|
"entropy": 5.120091438293457,
|
|
"epoch": 7.01673640167364,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0001578371221495575,
|
|
"loss": 4.3136,
|
|
"mean_token_accuracy": 0.2664042204618454,
|
|
"num_tokens": 15501703.0,
|
|
"step": 8385
|
|
},
|
|
{
|
|
"entropy": 5.092587900161743,
|
|
"epoch": 7.02092050209205,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0001575616890314009,
|
|
"loss": 4.2719,
|
|
"mean_token_accuracy": 0.2778256595134735,
|
|
"num_tokens": 15510110.0,
|
|
"step": 8390
|
|
},
|
|
{
|
|
"entropy": 4.948746252059936,
|
|
"epoch": 7.02510460251046,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00015728649758240727,
|
|
"loss": 4.1961,
|
|
"mean_token_accuracy": 0.2842837870121002,
|
|
"num_tokens": 15518352.0,
|
|
"step": 8395
|
|
},
|
|
{
|
|
"entropy": 5.0083235740661625,
|
|
"epoch": 7.02928870292887,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001570115483688762,
|
|
"loss": 4.2024,
|
|
"mean_token_accuracy": 0.2811959758400917,
|
|
"num_tokens": 15527133.0,
|
|
"step": 8400
|
|
},
|
|
{
|
|
"entropy": 5.022755193710327,
|
|
"epoch": 7.03347280334728,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0001567368419566091,
|
|
"loss": 4.2013,
|
|
"mean_token_accuracy": 0.27631216645240786,
|
|
"num_tokens": 15536326.0,
|
|
"step": 8405
|
|
},
|
|
{
|
|
"entropy": 5.089591217041016,
|
|
"epoch": 7.03765690376569,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00015646237891090749,
|
|
"loss": 4.2945,
|
|
"mean_token_accuracy": 0.26310173869132997,
|
|
"num_tokens": 15546408.0,
|
|
"step": 8410
|
|
},
|
|
{
|
|
"entropy": 5.035111045837402,
|
|
"epoch": 7.0418410041841,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0001561881597965722,
|
|
"loss": 4.2913,
|
|
"mean_token_accuracy": 0.27444498240947723,
|
|
"num_tokens": 15554988.0,
|
|
"step": 8415
|
|
},
|
|
{
|
|
"entropy": 5.013755750656128,
|
|
"epoch": 7.046025104602511,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00015591418517790217,
|
|
"loss": 4.1681,
|
|
"mean_token_accuracy": 0.28711867332458496,
|
|
"num_tokens": 15564127.0,
|
|
"step": 8420
|
|
},
|
|
{
|
|
"entropy": 5.0585071563720705,
|
|
"epoch": 7.050209205020921,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.000155640455618693,
|
|
"loss": 4.2678,
|
|
"mean_token_accuracy": 0.27647695541381834,
|
|
"num_tokens": 15573658.0,
|
|
"step": 8425
|
|
},
|
|
{
|
|
"entropy": 5.117292881011963,
|
|
"epoch": 7.054393305439331,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00015536697168223626,
|
|
"loss": 4.3235,
|
|
"mean_token_accuracy": 0.26910134553909304,
|
|
"num_tokens": 15583728.0,
|
|
"step": 8430
|
|
},
|
|
{
|
|
"entropy": 4.945322561264038,
|
|
"epoch": 7.058577405857741,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0001550937339313177,
|
|
"loss": 4.1956,
|
|
"mean_token_accuracy": 0.28086566030979154,
|
|
"num_tokens": 15593335.0,
|
|
"step": 8435
|
|
},
|
|
{
|
|
"entropy": 4.964973115921021,
|
|
"epoch": 7.062761506276151,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00015482074292821687,
|
|
"loss": 4.1373,
|
|
"mean_token_accuracy": 0.28245575726032257,
|
|
"num_tokens": 15603133.0,
|
|
"step": 8440
|
|
},
|
|
{
|
|
"entropy": 5.039173793792725,
|
|
"epoch": 7.066945606694561,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001545479992347055,
|
|
"loss": 4.2591,
|
|
"mean_token_accuracy": 0.26864075362682344,
|
|
"num_tokens": 15612509.0,
|
|
"step": 8445
|
|
},
|
|
{
|
|
"entropy": 5.0185717105865475,
|
|
"epoch": 7.071129707112971,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00015427550341204585,
|
|
"loss": 4.2401,
|
|
"mean_token_accuracy": 0.2785619229078293,
|
|
"num_tokens": 15622160.0,
|
|
"step": 8450
|
|
},
|
|
{
|
|
"entropy": 5.115309143066407,
|
|
"epoch": 7.075313807531381,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00015400325602099086,
|
|
"loss": 4.371,
|
|
"mean_token_accuracy": 0.25720364302396775,
|
|
"num_tokens": 15631268.0,
|
|
"step": 8455
|
|
},
|
|
{
|
|
"entropy": 5.1116204261779785,
|
|
"epoch": 7.079497907949791,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00015373125762178168,
|
|
"loss": 4.3238,
|
|
"mean_token_accuracy": 0.2686139330267906,
|
|
"num_tokens": 15640153.0,
|
|
"step": 8460
|
|
},
|
|
{
|
|
"entropy": 5.008873176574707,
|
|
"epoch": 7.083682008368201,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00015345950877414732,
|
|
"loss": 4.1539,
|
|
"mean_token_accuracy": 0.2791593551635742,
|
|
"num_tokens": 15647752.0,
|
|
"step": 8465
|
|
},
|
|
{
|
|
"entropy": 5.088695621490478,
|
|
"epoch": 7.087866108786611,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00015318801003730344,
|
|
"loss": 4.3161,
|
|
"mean_token_accuracy": 0.26208445578813555,
|
|
"num_tokens": 15656822.0,
|
|
"step": 8470
|
|
},
|
|
{
|
|
"entropy": 4.966014671325683,
|
|
"epoch": 7.092050209205021,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00015291676196995047,
|
|
"loss": 4.1663,
|
|
"mean_token_accuracy": 0.2835097014904022,
|
|
"num_tokens": 15665045.0,
|
|
"step": 8475
|
|
},
|
|
{
|
|
"entropy": 5.036664867401123,
|
|
"epoch": 7.096234309623431,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00015264576513027367,
|
|
"loss": 4.2511,
|
|
"mean_token_accuracy": 0.27102883011102674,
|
|
"num_tokens": 15674206.0,
|
|
"step": 8480
|
|
},
|
|
{
|
|
"entropy": 5.128562593460083,
|
|
"epoch": 7.100418410041841,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0001523750200759406,
|
|
"loss": 4.3158,
|
|
"mean_token_accuracy": 0.26051937490701677,
|
|
"num_tokens": 15683075.0,
|
|
"step": 8485
|
|
},
|
|
{
|
|
"entropy": 5.044271993637085,
|
|
"epoch": 7.104602510460251,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001521045273641013,
|
|
"loss": 4.1565,
|
|
"mean_token_accuracy": 0.2831128314137459,
|
|
"num_tokens": 15691698.0,
|
|
"step": 8490
|
|
},
|
|
{
|
|
"entropy": 5.0921674251556395,
|
|
"epoch": 7.108786610878661,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00015183428755138613,
|
|
"loss": 4.3157,
|
|
"mean_token_accuracy": 0.2629671946167946,
|
|
"num_tokens": 15701927.0,
|
|
"step": 8495
|
|
},
|
|
{
|
|
"entropy": 5.046346998214721,
|
|
"epoch": 7.112970711297071,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00015156430119390524,
|
|
"loss": 4.27,
|
|
"mean_token_accuracy": 0.2713933870196342,
|
|
"num_tokens": 15710813.0,
|
|
"step": 8500
|
|
},
|
|
{
|
|
"epoch": 7.112970711297071,
|
|
"eval_entropy": 4.87325340692251,
|
|
"eval_loss": 5.297507286071777,
|
|
"eval_mean_token_accuracy": 0.2046695337430831,
|
|
"eval_num_tokens": 15710813.0,
|
|
"eval_runtime": 2.0821,
|
|
"eval_samples_per_second": 1876.472,
|
|
"eval_steps_per_second": 234.859,
|
|
"step": 8500
|
|
},
|
|
{
|
|
"entropy": 5.137725734710694,
|
|
"epoch": 7.117154811715481,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00015129456884724723,
|
|
"loss": 4.2821,
|
|
"mean_token_accuracy": 0.26927792429924013,
|
|
"num_tokens": 15720559.0,
|
|
"step": 8505
|
|
},
|
|
{
|
|
"entropy": 5.014296817779541,
|
|
"epoch": 7.121338912133891,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00015102509106647782,
|
|
"loss": 4.2608,
|
|
"mean_token_accuracy": 0.2737458646297455,
|
|
"num_tokens": 15730649.0,
|
|
"step": 8510
|
|
},
|
|
{
|
|
"entropy": 5.015346908569336,
|
|
"epoch": 7.125523012552302,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00015075586840613902,
|
|
"loss": 4.1875,
|
|
"mean_token_accuracy": 0.2838317513465881,
|
|
"num_tokens": 15739619.0,
|
|
"step": 8515
|
|
},
|
|
{
|
|
"entropy": 5.057281255722046,
|
|
"epoch": 7.129707112970712,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00015048690142024763,
|
|
"loss": 4.2753,
|
|
"mean_token_accuracy": 0.26924674212932587,
|
|
"num_tokens": 15748457.0,
|
|
"step": 8520
|
|
},
|
|
{
|
|
"entropy": 5.001475191116333,
|
|
"epoch": 7.133891213389122,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00015021819066229452,
|
|
"loss": 4.1997,
|
|
"mean_token_accuracy": 0.2717295467853546,
|
|
"num_tokens": 15758527.0,
|
|
"step": 8525
|
|
},
|
|
{
|
|
"entropy": 5.045412826538086,
|
|
"epoch": 7.138075313807532,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00014994973668524343,
|
|
"loss": 4.3298,
|
|
"mean_token_accuracy": 0.26308275759220123,
|
|
"num_tokens": 15767980.0,
|
|
"step": 8530
|
|
},
|
|
{
|
|
"entropy": 4.981498336791992,
|
|
"epoch": 7.142259414225942,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00014968154004152922,
|
|
"loss": 4.2236,
|
|
"mean_token_accuracy": 0.27652217745780944,
|
|
"num_tokens": 15777323.0,
|
|
"step": 8535
|
|
},
|
|
{
|
|
"entropy": 5.014065313339233,
|
|
"epoch": 7.146443514644352,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00014941360128305765,
|
|
"loss": 4.2412,
|
|
"mean_token_accuracy": 0.2810399204492569,
|
|
"num_tokens": 15787589.0,
|
|
"step": 8540
|
|
},
|
|
{
|
|
"entropy": 5.050902652740478,
|
|
"epoch": 7.150627615062762,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00014914592096120348,
|
|
"loss": 4.2858,
|
|
"mean_token_accuracy": 0.2709465831518173,
|
|
"num_tokens": 15797248.0,
|
|
"step": 8545
|
|
},
|
|
{
|
|
"entropy": 5.052913570404053,
|
|
"epoch": 7.154811715481172,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00014887849962680978,
|
|
"loss": 4.2992,
|
|
"mean_token_accuracy": 0.2598874017596245,
|
|
"num_tokens": 15806389.0,
|
|
"step": 8550
|
|
},
|
|
{
|
|
"entropy": 5.0877069473266605,
|
|
"epoch": 7.158995815899582,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00014861133783018678,
|
|
"loss": 4.2605,
|
|
"mean_token_accuracy": 0.27973571717739104,
|
|
"num_tokens": 15814719.0,
|
|
"step": 8555
|
|
},
|
|
{
|
|
"entropy": 5.057853651046753,
|
|
"epoch": 7.163179916317992,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00014834443612111035,
|
|
"loss": 4.3173,
|
|
"mean_token_accuracy": 0.2663189485669136,
|
|
"num_tokens": 15823751.0,
|
|
"step": 8560
|
|
},
|
|
{
|
|
"entropy": 5.060284852981567,
|
|
"epoch": 7.167364016736402,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00014807779504882147,
|
|
"loss": 4.3054,
|
|
"mean_token_accuracy": 0.2679447799921036,
|
|
"num_tokens": 15832717.0,
|
|
"step": 8565
|
|
},
|
|
{
|
|
"entropy": 5.033549785614014,
|
|
"epoch": 7.171548117154812,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00014781141516202439,
|
|
"loss": 4.2484,
|
|
"mean_token_accuracy": 0.274546617269516,
|
|
"num_tokens": 15841494.0,
|
|
"step": 8570
|
|
},
|
|
{
|
|
"entropy": 5.061661195755005,
|
|
"epoch": 7.175732217573222,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001475452970088863,
|
|
"loss": 4.3151,
|
|
"mean_token_accuracy": 0.2674101710319519,
|
|
"num_tokens": 15851619.0,
|
|
"step": 8575
|
|
},
|
|
{
|
|
"entropy": 5.044713973999023,
|
|
"epoch": 7.179916317991632,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0001472794411370355,
|
|
"loss": 4.296,
|
|
"mean_token_accuracy": 0.26447591930627823,
|
|
"num_tokens": 15861563.0,
|
|
"step": 8580
|
|
},
|
|
{
|
|
"entropy": 5.04661111831665,
|
|
"epoch": 7.184100418410042,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00014701384809356053,
|
|
"loss": 4.283,
|
|
"mean_token_accuracy": 0.27113932818174363,
|
|
"num_tokens": 15870554.0,
|
|
"step": 8585
|
|
},
|
|
{
|
|
"entropy": 5.062147760391236,
|
|
"epoch": 7.188284518828452,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00014674851842500931,
|
|
"loss": 4.2343,
|
|
"mean_token_accuracy": 0.28210595846176145,
|
|
"num_tokens": 15879679.0,
|
|
"step": 8590
|
|
},
|
|
{
|
|
"entropy": 4.985602807998657,
|
|
"epoch": 7.192468619246862,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001464834526773875,
|
|
"loss": 4.2204,
|
|
"mean_token_accuracy": 0.2676385298371315,
|
|
"num_tokens": 15888533.0,
|
|
"step": 8595
|
|
},
|
|
{
|
|
"entropy": 5.05936450958252,
|
|
"epoch": 7.196652719665272,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00014621865139615795,
|
|
"loss": 4.2333,
|
|
"mean_token_accuracy": 0.27673305571079254,
|
|
"num_tokens": 15896888.0,
|
|
"step": 8600
|
|
},
|
|
{
|
|
"entropy": 5.080636310577392,
|
|
"epoch": 7.200836820083682,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00014595411512623895,
|
|
"loss": 4.3234,
|
|
"mean_token_accuracy": 0.26925985515117645,
|
|
"num_tokens": 15906843.0,
|
|
"step": 8605
|
|
},
|
|
{
|
|
"entropy": 4.927389287948609,
|
|
"epoch": 7.205020920502092,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00014568984441200384,
|
|
"loss": 4.1987,
|
|
"mean_token_accuracy": 0.27644574344158174,
|
|
"num_tokens": 15915333.0,
|
|
"step": 8610
|
|
},
|
|
{
|
|
"entropy": 5.048716115951538,
|
|
"epoch": 7.209205020920502,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00014542583979727908,
|
|
"loss": 4.3007,
|
|
"mean_token_accuracy": 0.2728378862142563,
|
|
"num_tokens": 15924187.0,
|
|
"step": 8615
|
|
},
|
|
{
|
|
"entropy": 5.056351852416992,
|
|
"epoch": 7.2133891213389125,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00014516210182534373,
|
|
"loss": 4.2786,
|
|
"mean_token_accuracy": 0.27571345269680025,
|
|
"num_tokens": 15932763.0,
|
|
"step": 8620
|
|
},
|
|
{
|
|
"entropy": 5.075104236602783,
|
|
"epoch": 7.2175732217573225,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00014489863103892825,
|
|
"loss": 4.2265,
|
|
"mean_token_accuracy": 0.2737642928957939,
|
|
"num_tokens": 15942272.0,
|
|
"step": 8625
|
|
},
|
|
{
|
|
"entropy": 5.037778425216675,
|
|
"epoch": 7.2217573221757325,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000144635427980213,
|
|
"loss": 4.3277,
|
|
"mean_token_accuracy": 0.2651097521185875,
|
|
"num_tokens": 15952099.0,
|
|
"step": 8630
|
|
},
|
|
{
|
|
"entropy": 4.982480573654175,
|
|
"epoch": 7.2259414225941425,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00014437249319082773,
|
|
"loss": 4.1832,
|
|
"mean_token_accuracy": 0.27846368551254275,
|
|
"num_tokens": 15961206.0,
|
|
"step": 8635
|
|
},
|
|
{
|
|
"entropy": 5.013817358016968,
|
|
"epoch": 7.2301255230125525,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00014410982721184984,
|
|
"loss": 4.2353,
|
|
"mean_token_accuracy": 0.27771357744932174,
|
|
"num_tokens": 15970694.0,
|
|
"step": 8640
|
|
},
|
|
{
|
|
"entropy": 5.0025615215301515,
|
|
"epoch": 7.2343096234309625,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00014384743058380373,
|
|
"loss": 4.2195,
|
|
"mean_token_accuracy": 0.2759588658809662,
|
|
"num_tokens": 15979271.0,
|
|
"step": 8645
|
|
},
|
|
{
|
|
"entropy": 5.064763593673706,
|
|
"epoch": 7.2384937238493725,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00014358530384665953,
|
|
"loss": 4.2777,
|
|
"mean_token_accuracy": 0.27130317091941836,
|
|
"num_tokens": 15988124.0,
|
|
"step": 8650
|
|
},
|
|
{
|
|
"entropy": 5.092198753356934,
|
|
"epoch": 7.2426778242677825,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00014332344753983174,
|
|
"loss": 4.2976,
|
|
"mean_token_accuracy": 0.27071579098701476,
|
|
"num_tokens": 15998912.0,
|
|
"step": 8655
|
|
},
|
|
{
|
|
"entropy": 5.056737565994263,
|
|
"epoch": 7.2468619246861925,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00014306186220217873,
|
|
"loss": 4.2813,
|
|
"mean_token_accuracy": 0.2685153901576996,
|
|
"num_tokens": 16008295.0,
|
|
"step": 8660
|
|
},
|
|
{
|
|
"entropy": 5.051494216918945,
|
|
"epoch": 7.2510460251046025,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00014280054837200094,
|
|
"loss": 4.2595,
|
|
"mean_token_accuracy": 0.2847561925649643,
|
|
"num_tokens": 16017120.0,
|
|
"step": 8665
|
|
},
|
|
{
|
|
"entropy": 5.077377700805664,
|
|
"epoch": 7.2552301255230125,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00014253950658704034,
|
|
"loss": 4.3333,
|
|
"mean_token_accuracy": 0.26760763078927996,
|
|
"num_tokens": 16026630.0,
|
|
"step": 8670
|
|
},
|
|
{
|
|
"entropy": 5.037253570556641,
|
|
"epoch": 7.2594142259414225,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00014227873738447883,
|
|
"loss": 4.2784,
|
|
"mean_token_accuracy": 0.27751528322696684,
|
|
"num_tokens": 16034897.0,
|
|
"step": 8675
|
|
},
|
|
{
|
|
"entropy": 4.977588033676147,
|
|
"epoch": 7.2635983263598325,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00014201824130093764,
|
|
"loss": 4.1917,
|
|
"mean_token_accuracy": 0.27370653450489046,
|
|
"num_tokens": 16044387.0,
|
|
"step": 8680
|
|
},
|
|
{
|
|
"entropy": 5.012709331512451,
|
|
"epoch": 7.2677824267782425,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00014175801887247576,
|
|
"loss": 4.2395,
|
|
"mean_token_accuracy": 0.27253492176532745,
|
|
"num_tokens": 16053659.0,
|
|
"step": 8685
|
|
},
|
|
{
|
|
"entropy": 5.019836711883545,
|
|
"epoch": 7.2719665271966525,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00014149807063458896,
|
|
"loss": 4.2432,
|
|
"mean_token_accuracy": 0.27708993256092074,
|
|
"num_tokens": 16062997.0,
|
|
"step": 8690
|
|
},
|
|
{
|
|
"entropy": 5.09762864112854,
|
|
"epoch": 7.2761506276150625,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00014123839712220921,
|
|
"loss": 4.2545,
|
|
"mean_token_accuracy": 0.2704265475273132,
|
|
"num_tokens": 16071491.0,
|
|
"step": 8695
|
|
},
|
|
{
|
|
"entropy": 5.096727991104126,
|
|
"epoch": 7.2803347280334725,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001409789988697026,
|
|
"loss": 4.3249,
|
|
"mean_token_accuracy": 0.2706155315041542,
|
|
"num_tokens": 16081008.0,
|
|
"step": 8700
|
|
},
|
|
{
|
|
"entropy": 5.004104709625244,
|
|
"epoch": 7.2845188284518825,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00014071987641086926,
|
|
"loss": 4.2235,
|
|
"mean_token_accuracy": 0.27506672739982607,
|
|
"num_tokens": 16090516.0,
|
|
"step": 8705
|
|
},
|
|
{
|
|
"entropy": 5.071073293685913,
|
|
"epoch": 7.2887029288702925,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001404610302789414,
|
|
"loss": 4.3175,
|
|
"mean_token_accuracy": 0.2676409766077995,
|
|
"num_tokens": 16100952.0,
|
|
"step": 8710
|
|
},
|
|
{
|
|
"entropy": 5.044872188568116,
|
|
"epoch": 7.292887029288703,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00014020246100658268,
|
|
"loss": 4.279,
|
|
"mean_token_accuracy": 0.27356911152601243,
|
|
"num_tokens": 16110493.0,
|
|
"step": 8715
|
|
},
|
|
{
|
|
"entropy": 5.029181385040284,
|
|
"epoch": 7.297071129707113,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00013994416912588726,
|
|
"loss": 4.2542,
|
|
"mean_token_accuracy": 0.27263363301753996,
|
|
"num_tokens": 16119227.0,
|
|
"step": 8720
|
|
},
|
|
{
|
|
"entropy": 5.068349027633667,
|
|
"epoch": 7.301255230125523,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001396861551683781,
|
|
"loss": 4.343,
|
|
"mean_token_accuracy": 0.2722461983561516,
|
|
"num_tokens": 16129268.0,
|
|
"step": 8725
|
|
},
|
|
{
|
|
"entropy": 5.076955270767212,
|
|
"epoch": 7.305439330543933,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001394284196650067,
|
|
"loss": 4.295,
|
|
"mean_token_accuracy": 0.2715486019849777,
|
|
"num_tokens": 16137523.0,
|
|
"step": 8730
|
|
},
|
|
{
|
|
"entropy": 5.050295305252075,
|
|
"epoch": 7.309623430962343,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001391709631461511,
|
|
"loss": 4.3316,
|
|
"mean_token_accuracy": 0.2641110524535179,
|
|
"num_tokens": 16146535.0,
|
|
"step": 8735
|
|
},
|
|
{
|
|
"entropy": 5.047996711730957,
|
|
"epoch": 7.313807531380753,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00013891378614161538,
|
|
"loss": 4.2682,
|
|
"mean_token_accuracy": 0.27303022146224976,
|
|
"num_tokens": 16155892.0,
|
|
"step": 8740
|
|
},
|
|
{
|
|
"entropy": 4.987729215621949,
|
|
"epoch": 7.317991631799163,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00013865688918062875,
|
|
"loss": 4.2285,
|
|
"mean_token_accuracy": 0.2712548956274986,
|
|
"num_tokens": 16164697.0,
|
|
"step": 8745
|
|
},
|
|
{
|
|
"entropy": 5.0376465797424315,
|
|
"epoch": 7.322175732217573,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00013840027279184353,
|
|
"loss": 4.2546,
|
|
"mean_token_accuracy": 0.2734776079654694,
|
|
"num_tokens": 16173387.0,
|
|
"step": 8750
|
|
},
|
|
{
|
|
"entropy": 5.031920051574707,
|
|
"epoch": 7.326359832635983,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00013814393750333524,
|
|
"loss": 4.2197,
|
|
"mean_token_accuracy": 0.2799023240804672,
|
|
"num_tokens": 16182438.0,
|
|
"step": 8755
|
|
},
|
|
{
|
|
"entropy": 5.024758052825928,
|
|
"epoch": 7.330543933054393,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00013788788384260043,
|
|
"loss": 4.1919,
|
|
"mean_token_accuracy": 0.27620892226696014,
|
|
"num_tokens": 16193149.0,
|
|
"step": 8760
|
|
},
|
|
{
|
|
"entropy": 5.064318466186523,
|
|
"epoch": 7.334728033472803,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00013763211233655656,
|
|
"loss": 4.3543,
|
|
"mean_token_accuracy": 0.2649631813168526,
|
|
"num_tokens": 16202354.0,
|
|
"step": 8765
|
|
},
|
|
{
|
|
"entropy": 4.999471664428711,
|
|
"epoch": 7.338912133891213,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00013737662351154028,
|
|
"loss": 4.2712,
|
|
"mean_token_accuracy": 0.26226536482572554,
|
|
"num_tokens": 16210608.0,
|
|
"step": 8770
|
|
},
|
|
{
|
|
"entropy": 5.075949859619141,
|
|
"epoch": 7.343096234309623,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00013712141789330642,
|
|
"loss": 4.3312,
|
|
"mean_token_accuracy": 0.27122950553894043,
|
|
"num_tokens": 16219544.0,
|
|
"step": 8775
|
|
},
|
|
{
|
|
"entropy": 5.021850061416626,
|
|
"epoch": 7.347280334728033,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00013686649600702728,
|
|
"loss": 4.2378,
|
|
"mean_token_accuracy": 0.27088230550289155,
|
|
"num_tokens": 16228820.0,
|
|
"step": 8780
|
|
},
|
|
{
|
|
"entropy": 4.9606890201568605,
|
|
"epoch": 7.351464435146443,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00013661185837729085,
|
|
"loss": 4.148,
|
|
"mean_token_accuracy": 0.2787727937102318,
|
|
"num_tokens": 16237751.0,
|
|
"step": 8785
|
|
},
|
|
{
|
|
"entropy": 5.049793767929077,
|
|
"epoch": 7.355648535564853,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00013635750552810065,
|
|
"loss": 4.2619,
|
|
"mean_token_accuracy": 0.27321857213974,
|
|
"num_tokens": 16246944.0,
|
|
"step": 8790
|
|
},
|
|
{
|
|
"entropy": 5.07344536781311,
|
|
"epoch": 7.359832635983263,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00013610343798287376,
|
|
"loss": 4.3382,
|
|
"mean_token_accuracy": 0.26902366876602174,
|
|
"num_tokens": 16256565.0,
|
|
"step": 8795
|
|
},
|
|
{
|
|
"entropy": 4.998224115371704,
|
|
"epoch": 7.364016736401673,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00013584965626444045,
|
|
"loss": 4.1991,
|
|
"mean_token_accuracy": 0.2762854859232903,
|
|
"num_tokens": 16264886.0,
|
|
"step": 8800
|
|
},
|
|
{
|
|
"entropy": 5.1031632900238035,
|
|
"epoch": 7.368200836820083,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.00013559616089504285,
|
|
"loss": 4.3857,
|
|
"mean_token_accuracy": 0.2609135702252388,
|
|
"num_tokens": 16274039.0,
|
|
"step": 8805
|
|
},
|
|
{
|
|
"entropy": 4.998230600357056,
|
|
"epoch": 7.372384937238493,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00013534295239633342,
|
|
"loss": 4.2491,
|
|
"mean_token_accuracy": 0.2725122168660164,
|
|
"num_tokens": 16283274.0,
|
|
"step": 8810
|
|
},
|
|
{
|
|
"entropy": 5.0272191047668455,
|
|
"epoch": 7.376569037656903,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0001350900312893747,
|
|
"loss": 4.244,
|
|
"mean_token_accuracy": 0.27896369695663453,
|
|
"num_tokens": 16292048.0,
|
|
"step": 8815
|
|
},
|
|
{
|
|
"entropy": 5.044795179367066,
|
|
"epoch": 7.380753138075314,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00013483739809463753,
|
|
"loss": 4.2415,
|
|
"mean_token_accuracy": 0.2776633337140083,
|
|
"num_tokens": 16301910.0,
|
|
"step": 8820
|
|
},
|
|
{
|
|
"entropy": 5.031274652481079,
|
|
"epoch": 7.384937238493724,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00013458505333200044,
|
|
"loss": 4.2374,
|
|
"mean_token_accuracy": 0.2756684094667435,
|
|
"num_tokens": 16311221.0,
|
|
"step": 8825
|
|
},
|
|
{
|
|
"entropy": 4.982961559295655,
|
|
"epoch": 7.389121338912134,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00013433299752074857,
|
|
"loss": 4.2056,
|
|
"mean_token_accuracy": 0.2794200211763382,
|
|
"num_tokens": 16319530.0,
|
|
"step": 8830
|
|
},
|
|
{
|
|
"entropy": 4.972958707809449,
|
|
"epoch": 7.393305439330544,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.000134081231179572,
|
|
"loss": 4.1679,
|
|
"mean_token_accuracy": 0.2839868158102036,
|
|
"num_tokens": 16327927.0,
|
|
"step": 8835
|
|
},
|
|
{
|
|
"entropy": 5.094141292572021,
|
|
"epoch": 7.397489539748954,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00013382975482656556,
|
|
"loss": 4.3068,
|
|
"mean_token_accuracy": 0.2647473633289337,
|
|
"num_tokens": 16337551.0,
|
|
"step": 8840
|
|
},
|
|
{
|
|
"entropy": 5.026058053970337,
|
|
"epoch": 7.401673640167364,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001335785689792271,
|
|
"loss": 4.2582,
|
|
"mean_token_accuracy": 0.27456394881010054,
|
|
"num_tokens": 16345937.0,
|
|
"step": 8845
|
|
},
|
|
{
|
|
"entropy": 5.059644746780395,
|
|
"epoch": 7.405857740585774,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001333276741544566,
|
|
"loss": 4.3181,
|
|
"mean_token_accuracy": 0.26773334443569186,
|
|
"num_tokens": 16355797.0,
|
|
"step": 8850
|
|
},
|
|
{
|
|
"entropy": 5.020276117324829,
|
|
"epoch": 7.410041841004184,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00013307707086855546,
|
|
"loss": 4.2613,
|
|
"mean_token_accuracy": 0.27164504677057266,
|
|
"num_tokens": 16364547.0,
|
|
"step": 8855
|
|
},
|
|
{
|
|
"entropy": 4.964390611648559,
|
|
"epoch": 7.414225941422594,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001328267596372248,
|
|
"loss": 4.1684,
|
|
"mean_token_accuracy": 0.29008165597915647,
|
|
"num_tokens": 16373506.0,
|
|
"step": 8860
|
|
},
|
|
{
|
|
"entropy": 5.037272787094116,
|
|
"epoch": 7.418410041841004,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000132576740975565,
|
|
"loss": 4.2374,
|
|
"mean_token_accuracy": 0.2810764268040657,
|
|
"num_tokens": 16383006.0,
|
|
"step": 8865
|
|
},
|
|
{
|
|
"entropy": 5.048643779754639,
|
|
"epoch": 7.422594142259414,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00013232701539807415,
|
|
"loss": 4.2933,
|
|
"mean_token_accuracy": 0.27292863428592684,
|
|
"num_tokens": 16392721.0,
|
|
"step": 8870
|
|
},
|
|
{
|
|
"entropy": 4.976942777633667,
|
|
"epoch": 7.426778242677824,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001320775834186475,
|
|
"loss": 4.2582,
|
|
"mean_token_accuracy": 0.2710895583033562,
|
|
"num_tokens": 16402404.0,
|
|
"step": 8875
|
|
},
|
|
{
|
|
"entropy": 5.0821326732635494,
|
|
"epoch": 7.430962343096234,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00013182844555057583,
|
|
"loss": 4.2923,
|
|
"mean_token_accuracy": 0.263348913192749,
|
|
"num_tokens": 16411386.0,
|
|
"step": 8880
|
|
},
|
|
{
|
|
"entropy": 4.989440965652466,
|
|
"epoch": 7.435146443514644,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001315796023065448,
|
|
"loss": 4.2383,
|
|
"mean_token_accuracy": 0.26915232092142105,
|
|
"num_tokens": 16420407.0,
|
|
"step": 8885
|
|
},
|
|
{
|
|
"entropy": 5.005561113357544,
|
|
"epoch": 7.439330543933054,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00013133105419863394,
|
|
"loss": 4.1577,
|
|
"mean_token_accuracy": 0.2873361647129059,
|
|
"num_tokens": 16429197.0,
|
|
"step": 8890
|
|
},
|
|
{
|
|
"entropy": 5.040199327468872,
|
|
"epoch": 7.443514644351464,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00013108280173831517,
|
|
"loss": 4.2972,
|
|
"mean_token_accuracy": 0.2679921954870224,
|
|
"num_tokens": 16438144.0,
|
|
"step": 8895
|
|
},
|
|
{
|
|
"entropy": 5.002023077011108,
|
|
"epoch": 7.447698744769874,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0001308348454364522,
|
|
"loss": 4.3016,
|
|
"mean_token_accuracy": 0.26979140341281893,
|
|
"num_tokens": 16447861.0,
|
|
"step": 8900
|
|
},
|
|
{
|
|
"entropy": 4.972557449340821,
|
|
"epoch": 7.451882845188284,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00013058718580329917,
|
|
"loss": 4.1472,
|
|
"mean_token_accuracy": 0.2920081168413162,
|
|
"num_tokens": 16456937.0,
|
|
"step": 8905
|
|
},
|
|
{
|
|
"entropy": 5.045724534988404,
|
|
"epoch": 7.456066945606695,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001303398233484999,
|
|
"loss": 4.2816,
|
|
"mean_token_accuracy": 0.27083266973495485,
|
|
"num_tokens": 16466714.0,
|
|
"step": 8910
|
|
},
|
|
{
|
|
"entropy": 5.033455562591553,
|
|
"epoch": 7.460251046025105,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.00013009275858108648,
|
|
"loss": 4.2087,
|
|
"mean_token_accuracy": 0.282195907831192,
|
|
"num_tokens": 16475219.0,
|
|
"step": 8915
|
|
},
|
|
{
|
|
"entropy": 5.017781686782837,
|
|
"epoch": 7.464435146443515,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0001298459920094784,
|
|
"loss": 4.1817,
|
|
"mean_token_accuracy": 0.2830035999417305,
|
|
"num_tokens": 16484169.0,
|
|
"step": 8920
|
|
},
|
|
{
|
|
"entropy": 5.049612045288086,
|
|
"epoch": 7.468619246861925,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001295995241414818,
|
|
"loss": 4.2696,
|
|
"mean_token_accuracy": 0.2670405611395836,
|
|
"num_tokens": 16493413.0,
|
|
"step": 8925
|
|
},
|
|
{
|
|
"entropy": 4.967992162704467,
|
|
"epoch": 7.472803347280335,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00012935335548428767,
|
|
"loss": 4.2334,
|
|
"mean_token_accuracy": 0.273973685503006,
|
|
"num_tokens": 16503195.0,
|
|
"step": 8930
|
|
},
|
|
{
|
|
"entropy": 4.9723328113555905,
|
|
"epoch": 7.476987447698745,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001291074865444718,
|
|
"loss": 4.1386,
|
|
"mean_token_accuracy": 0.2769825026392937,
|
|
"num_tokens": 16511168.0,
|
|
"step": 8935
|
|
},
|
|
{
|
|
"entropy": 5.0018514633178714,
|
|
"epoch": 7.481171548117155,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00012886191782799267,
|
|
"loss": 4.2479,
|
|
"mean_token_accuracy": 0.2801429361104965,
|
|
"num_tokens": 16520125.0,
|
|
"step": 8940
|
|
},
|
|
{
|
|
"entropy": 4.963721418380738,
|
|
"epoch": 7.485355648535565,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00012861664984019144,
|
|
"loss": 4.1641,
|
|
"mean_token_accuracy": 0.28001548498868944,
|
|
"num_tokens": 16528121.0,
|
|
"step": 8945
|
|
},
|
|
{
|
|
"entropy": 4.976732873916626,
|
|
"epoch": 7.489539748953975,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00012837168308579012,
|
|
"loss": 4.1944,
|
|
"mean_token_accuracy": 0.27863424718379975,
|
|
"num_tokens": 16537741.0,
|
|
"step": 8950
|
|
},
|
|
{
|
|
"entropy": 5.01279354095459,
|
|
"epoch": 7.493723849372385,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00012812701806889077,
|
|
"loss": 4.2861,
|
|
"mean_token_accuracy": 0.2727734223008156,
|
|
"num_tokens": 16548091.0,
|
|
"step": 8955
|
|
},
|
|
{
|
|
"entropy": 5.038434982299805,
|
|
"epoch": 7.497907949790795,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001278826552929749,
|
|
"loss": 4.2428,
|
|
"mean_token_accuracy": 0.2747714564204216,
|
|
"num_tokens": 16557011.0,
|
|
"step": 8960
|
|
},
|
|
{
|
|
"entropy": 5.073702573776245,
|
|
"epoch": 7.502092050209205,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.0001276385952609016,
|
|
"loss": 4.3047,
|
|
"mean_token_accuracy": 0.2638867825269699,
|
|
"num_tokens": 16565868.0,
|
|
"step": 8965
|
|
},
|
|
{
|
|
"entropy": 5.075294160842896,
|
|
"epoch": 7.506276150627615,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00012739483847490734,
|
|
"loss": 4.2889,
|
|
"mean_token_accuracy": 0.27467488646507265,
|
|
"num_tokens": 16575131.0,
|
|
"step": 8970
|
|
},
|
|
{
|
|
"entropy": 5.047591209411621,
|
|
"epoch": 7.510460251046025,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00012715138543660425,
|
|
"loss": 4.2992,
|
|
"mean_token_accuracy": 0.2682076901197433,
|
|
"num_tokens": 16584789.0,
|
|
"step": 8975
|
|
},
|
|
{
|
|
"entropy": 5.107950067520141,
|
|
"epoch": 7.514644351464435,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001269082366469797,
|
|
"loss": 4.287,
|
|
"mean_token_accuracy": 0.271279938519001,
|
|
"num_tokens": 16594488.0,
|
|
"step": 8980
|
|
},
|
|
{
|
|
"entropy": 5.003856134414673,
|
|
"epoch": 7.518828451882845,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.00012666539260639475,
|
|
"loss": 4.1852,
|
|
"mean_token_accuracy": 0.27887404561042783,
|
|
"num_tokens": 16602691.0,
|
|
"step": 8985
|
|
},
|
|
{
|
|
"entropy": 5.097051572799683,
|
|
"epoch": 7.523012552301255,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001264228538145833,
|
|
"loss": 4.3255,
|
|
"mean_token_accuracy": 0.2749505043029785,
|
|
"num_tokens": 16611843.0,
|
|
"step": 8990
|
|
},
|
|
{
|
|
"entropy": 5.078509426116943,
|
|
"epoch": 7.527196652719665,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00012618062077065126,
|
|
"loss": 4.2874,
|
|
"mean_token_accuracy": 0.2615717425942421,
|
|
"num_tokens": 16620587.0,
|
|
"step": 8995
|
|
},
|
|
{
|
|
"entropy": 5.067453289031983,
|
|
"epoch": 7.531380753138075,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00012593869397307527,
|
|
"loss": 4.3512,
|
|
"mean_token_accuracy": 0.2601457521319389,
|
|
"num_tokens": 16631113.0,
|
|
"step": 9000
|
|
},
|
|
{
|
|
"epoch": 7.531380753138075,
|
|
"eval_entropy": 4.849638395270443,
|
|
"eval_loss": 5.294634819030762,
|
|
"eval_mean_token_accuracy": 0.20558827676656055,
|
|
"eval_num_tokens": 16631113.0,
|
|
"eval_runtime": 2.2645,
|
|
"eval_samples_per_second": 1725.318,
|
|
"eval_steps_per_second": 215.941,
|
|
"step": 9000
|
|
},
|
|
{
|
|
"entropy": 5.068883848190308,
|
|
"epoch": 7.535564853556485,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00012569707391970187,
|
|
"loss": 4.1982,
|
|
"mean_token_accuracy": 0.27720783054828646,
|
|
"num_tokens": 16639747.0,
|
|
"step": 9005
|
|
},
|
|
{
|
|
"entropy": 5.067685508728028,
|
|
"epoch": 7.539748953974895,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00012545576110774617,
|
|
"loss": 4.2383,
|
|
"mean_token_accuracy": 0.2758551150560379,
|
|
"num_tokens": 16648619.0,
|
|
"step": 9010
|
|
},
|
|
{
|
|
"entropy": 5.0643151760101315,
|
|
"epoch": 7.543933054393305,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00012521475603379117,
|
|
"loss": 4.3234,
|
|
"mean_token_accuracy": 0.2736960932612419,
|
|
"num_tokens": 16656860.0,
|
|
"step": 9015
|
|
},
|
|
{
|
|
"entropy": 5.098158359527588,
|
|
"epoch": 7.548117154811716,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00012497405919378662,
|
|
"loss": 4.3737,
|
|
"mean_token_accuracy": 0.26648738235235214,
|
|
"num_tokens": 16666121.0,
|
|
"step": 9020
|
|
},
|
|
{
|
|
"entropy": 5.027002239227295,
|
|
"epoch": 7.552301255230126,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0001247336710830478,
|
|
"loss": 4.2329,
|
|
"mean_token_accuracy": 0.27224652618169787,
|
|
"num_tokens": 16675296.0,
|
|
"step": 9025
|
|
},
|
|
{
|
|
"entropy": 4.95601806640625,
|
|
"epoch": 7.556485355648536,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00012449359219625494,
|
|
"loss": 4.1859,
|
|
"mean_token_accuracy": 0.28932357430458067,
|
|
"num_tokens": 16683554.0,
|
|
"step": 9030
|
|
},
|
|
{
|
|
"entropy": 5.079730081558227,
|
|
"epoch": 7.560669456066946,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00012425382302745167,
|
|
"loss": 4.3025,
|
|
"mean_token_accuracy": 0.2714738458395004,
|
|
"num_tokens": 16692081.0,
|
|
"step": 9035
|
|
},
|
|
{
|
|
"entropy": 5.070031785964966,
|
|
"epoch": 7.564853556485356,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00012401436407004444,
|
|
"loss": 4.3263,
|
|
"mean_token_accuracy": 0.2680123969912529,
|
|
"num_tokens": 16701543.0,
|
|
"step": 9040
|
|
},
|
|
{
|
|
"entropy": 5.060573625564575,
|
|
"epoch": 7.569037656903766,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00012377521581680146,
|
|
"loss": 4.2769,
|
|
"mean_token_accuracy": 0.2722619652748108,
|
|
"num_tokens": 16710044.0,
|
|
"step": 9045
|
|
},
|
|
{
|
|
"entropy": 5.057760238647461,
|
|
"epoch": 7.573221757322176,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000123536378759851,
|
|
"loss": 4.2725,
|
|
"mean_token_accuracy": 0.28056506961584093,
|
|
"num_tokens": 16718765.0,
|
|
"step": 9050
|
|
},
|
|
{
|
|
"entropy": 5.015079402923584,
|
|
"epoch": 7.577405857740586,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00012329785339068163,
|
|
"loss": 4.1615,
|
|
"mean_token_accuracy": 0.27901374995708467,
|
|
"num_tokens": 16727470.0,
|
|
"step": 9055
|
|
},
|
|
{
|
|
"entropy": 5.091164779663086,
|
|
"epoch": 7.581589958158996,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001230596402001401,
|
|
"loss": 4.3221,
|
|
"mean_token_accuracy": 0.26617172062397004,
|
|
"num_tokens": 16736557.0,
|
|
"step": 9060
|
|
},
|
|
{
|
|
"entropy": 5.054933738708496,
|
|
"epoch": 7.585774058577406,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00012282173967843085,
|
|
"loss": 4.2543,
|
|
"mean_token_accuracy": 0.2751393675804138,
|
|
"num_tokens": 16745601.0,
|
|
"step": 9065
|
|
},
|
|
{
|
|
"entropy": 5.0628868579864506,
|
|
"epoch": 7.589958158995816,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00012258415231511497,
|
|
"loss": 4.2919,
|
|
"mean_token_accuracy": 0.26894889026880264,
|
|
"num_tokens": 16756141.0,
|
|
"step": 9070
|
|
},
|
|
{
|
|
"entropy": 5.014832878112793,
|
|
"epoch": 7.594142259414226,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00012234687859910895,
|
|
"loss": 4.3037,
|
|
"mean_token_accuracy": 0.2745773047208786,
|
|
"num_tokens": 16766080.0,
|
|
"step": 9075
|
|
},
|
|
{
|
|
"entropy": 5.055583333969116,
|
|
"epoch": 7.598326359832636,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0001221099190186842,
|
|
"loss": 4.2423,
|
|
"mean_token_accuracy": 0.27104195654392244,
|
|
"num_tokens": 16775826.0,
|
|
"step": 9080
|
|
},
|
|
{
|
|
"entropy": 5.004641056060791,
|
|
"epoch": 7.602510460251046,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0001218732740614651,
|
|
"loss": 4.3,
|
|
"mean_token_accuracy": 0.27334684431552886,
|
|
"num_tokens": 16785139.0,
|
|
"step": 9085
|
|
},
|
|
{
|
|
"entropy": 5.026444387435913,
|
|
"epoch": 7.606694560669456,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00012163694421442905,
|
|
"loss": 4.2686,
|
|
"mean_token_accuracy": 0.27255396395921705,
|
|
"num_tokens": 16793935.0,
|
|
"step": 9090
|
|
},
|
|
{
|
|
"entropy": 5.042926597595215,
|
|
"epoch": 7.610878661087866,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00012140092996390502,
|
|
"loss": 4.2188,
|
|
"mean_token_accuracy": 0.275225305557251,
|
|
"num_tokens": 16803451.0,
|
|
"step": 9095
|
|
},
|
|
{
|
|
"entropy": 5.093796443939209,
|
|
"epoch": 7.615062761506276,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00012116523179557221,
|
|
"loss": 4.2656,
|
|
"mean_token_accuracy": 0.27019921094179156,
|
|
"num_tokens": 16813229.0,
|
|
"step": 9100
|
|
},
|
|
{
|
|
"entropy": 5.009964942932129,
|
|
"epoch": 7.619246861924686,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00012092985019445968,
|
|
"loss": 4.2397,
|
|
"mean_token_accuracy": 0.2722703918814659,
|
|
"num_tokens": 16822088.0,
|
|
"step": 9105
|
|
},
|
|
{
|
|
"entropy": 5.019072866439819,
|
|
"epoch": 7.623430962343097,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001206947856449448,
|
|
"loss": 4.2586,
|
|
"mean_token_accuracy": 0.27781941592693327,
|
|
"num_tokens": 16831172.0,
|
|
"step": 9110
|
|
},
|
|
{
|
|
"entropy": 5.072237300872803,
|
|
"epoch": 7.627615062761507,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00012046003863075285,
|
|
"loss": 4.3216,
|
|
"mean_token_accuracy": 0.2608758002519608,
|
|
"num_tokens": 16840482.0,
|
|
"step": 9115
|
|
},
|
|
{
|
|
"entropy": 5.0504148483276365,
|
|
"epoch": 7.631799163179917,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00012022560963495516,
|
|
"loss": 4.2661,
|
|
"mean_token_accuracy": 0.2628014534711838,
|
|
"num_tokens": 16849205.0,
|
|
"step": 9120
|
|
},
|
|
{
|
|
"entropy": 5.108313035964966,
|
|
"epoch": 7.635983263598327,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00011999149913996902,
|
|
"loss": 4.2837,
|
|
"mean_token_accuracy": 0.2715302675962448,
|
|
"num_tokens": 16857500.0,
|
|
"step": 9125
|
|
},
|
|
{
|
|
"entropy": 5.055491876602173,
|
|
"epoch": 7.640167364016737,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00011975770762755627,
|
|
"loss": 4.1902,
|
|
"mean_token_accuracy": 0.2780496314167976,
|
|
"num_tokens": 16866329.0,
|
|
"step": 9130
|
|
},
|
|
{
|
|
"entropy": 5.046244335174561,
|
|
"epoch": 7.644351464435147,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001195242355788221,
|
|
"loss": 4.264,
|
|
"mean_token_accuracy": 0.27194504737854003,
|
|
"num_tokens": 16875408.0,
|
|
"step": 9135
|
|
},
|
|
{
|
|
"entropy": 5.056604719161987,
|
|
"epoch": 7.648535564853557,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00011929108347421468,
|
|
"loss": 4.2969,
|
|
"mean_token_accuracy": 0.2683531165122986,
|
|
"num_tokens": 16885272.0,
|
|
"step": 9140
|
|
},
|
|
{
|
|
"entropy": 4.96450982093811,
|
|
"epoch": 7.652719665271967,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00011905825179352337,
|
|
"loss": 4.2538,
|
|
"mean_token_accuracy": 0.27425817102193834,
|
|
"num_tokens": 16894375.0,
|
|
"step": 9145
|
|
},
|
|
{
|
|
"entropy": 4.9829323291778564,
|
|
"epoch": 7.656903765690377,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00011882574101587834,
|
|
"loss": 4.2079,
|
|
"mean_token_accuracy": 0.27803242206573486,
|
|
"num_tokens": 16903790.0,
|
|
"step": 9150
|
|
},
|
|
{
|
|
"entropy": 4.98489236831665,
|
|
"epoch": 7.661087866108787,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00011859355161974948,
|
|
"loss": 4.2506,
|
|
"mean_token_accuracy": 0.2752090528607368,
|
|
"num_tokens": 16913453.0,
|
|
"step": 9155
|
|
},
|
|
{
|
|
"entropy": 4.9712934494018555,
|
|
"epoch": 7.665271966527197,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00011836168408294513,
|
|
"loss": 4.2418,
|
|
"mean_token_accuracy": 0.27315788716077805,
|
|
"num_tokens": 16922392.0,
|
|
"step": 9160
|
|
},
|
|
{
|
|
"entropy": 5.062557601928711,
|
|
"epoch": 7.669456066945607,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00011813013888261154,
|
|
"loss": 4.3301,
|
|
"mean_token_accuracy": 0.26658158004283905,
|
|
"num_tokens": 16931945.0,
|
|
"step": 9165
|
|
},
|
|
{
|
|
"entropy": 5.040381956100464,
|
|
"epoch": 7.673640167364017,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00011789891649523137,
|
|
"loss": 4.2662,
|
|
"mean_token_accuracy": 0.2667974576354027,
|
|
"num_tokens": 16941285.0,
|
|
"step": 9170
|
|
},
|
|
{
|
|
"entropy": 5.074206590652466,
|
|
"epoch": 7.677824267782427,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001176680173966232,
|
|
"loss": 4.2678,
|
|
"mean_token_accuracy": 0.2669205844402313,
|
|
"num_tokens": 16950155.0,
|
|
"step": 9175
|
|
},
|
|
{
|
|
"entropy": 5.044518280029297,
|
|
"epoch": 7.682008368200837,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00011743744206194028,
|
|
"loss": 4.3635,
|
|
"mean_token_accuracy": 0.26337283849716187,
|
|
"num_tokens": 16959994.0,
|
|
"step": 9180
|
|
},
|
|
{
|
|
"entropy": 5.0134725093841555,
|
|
"epoch": 7.686192468619247,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00011720719096566946,
|
|
"loss": 4.2649,
|
|
"mean_token_accuracy": 0.27254819720983503,
|
|
"num_tokens": 16969095.0,
|
|
"step": 9185
|
|
},
|
|
{
|
|
"entropy": 5.086143207550049,
|
|
"epoch": 7.690376569037657,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00011697726458163056,
|
|
"loss": 4.3302,
|
|
"mean_token_accuracy": 0.264897446334362,
|
|
"num_tokens": 16978574.0,
|
|
"step": 9190
|
|
},
|
|
{
|
|
"entropy": 5.082922267913818,
|
|
"epoch": 7.694560669456067,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00011674766338297508,
|
|
"loss": 4.3462,
|
|
"mean_token_accuracy": 0.2686086013913155,
|
|
"num_tokens": 16988165.0,
|
|
"step": 9195
|
|
},
|
|
{
|
|
"entropy": 5.013553524017334,
|
|
"epoch": 7.698744769874477,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00011651838784218545,
|
|
"loss": 4.2241,
|
|
"mean_token_accuracy": 0.27429974675178526,
|
|
"num_tokens": 16998419.0,
|
|
"step": 9200
|
|
},
|
|
{
|
|
"entropy": 5.017548894882202,
|
|
"epoch": 7.702928870292887,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00011628943843107372,
|
|
"loss": 4.2728,
|
|
"mean_token_accuracy": 0.2614956647157669,
|
|
"num_tokens": 17008276.0,
|
|
"step": 9205
|
|
},
|
|
{
|
|
"entropy": 5.028089427947998,
|
|
"epoch": 7.707112970711297,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00011606081562078113,
|
|
"loss": 4.3134,
|
|
"mean_token_accuracy": 0.2672402963042259,
|
|
"num_tokens": 17017157.0,
|
|
"step": 9210
|
|
},
|
|
{
|
|
"entropy": 4.9897887229919435,
|
|
"epoch": 7.711297071129707,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00011583251988177664,
|
|
"loss": 4.2512,
|
|
"mean_token_accuracy": 0.27037367075681684,
|
|
"num_tokens": 17027100.0,
|
|
"step": 9215
|
|
},
|
|
{
|
|
"entropy": 4.9647791385650635,
|
|
"epoch": 7.7154811715481175,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00011560455168385601,
|
|
"loss": 4.2033,
|
|
"mean_token_accuracy": 0.2734866887331009,
|
|
"num_tokens": 17036094.0,
|
|
"step": 9220
|
|
},
|
|
{
|
|
"entropy": 5.003016662597656,
|
|
"epoch": 7.7196652719665275,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00011537691149614136,
|
|
"loss": 4.2195,
|
|
"mean_token_accuracy": 0.2857777878642082,
|
|
"num_tokens": 17044720.0,
|
|
"step": 9225
|
|
},
|
|
{
|
|
"entropy": 4.996941137313843,
|
|
"epoch": 7.7238493723849375,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00011514959978707944,
|
|
"loss": 4.2904,
|
|
"mean_token_accuracy": 0.27402451634407043,
|
|
"num_tokens": 17053074.0,
|
|
"step": 9230
|
|
},
|
|
{
|
|
"entropy": 5.080314254760742,
|
|
"epoch": 7.7280334728033475,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001149226170244413,
|
|
"loss": 4.3185,
|
|
"mean_token_accuracy": 0.2646454766392708,
|
|
"num_tokens": 17063036.0,
|
|
"step": 9235
|
|
},
|
|
{
|
|
"entropy": 5.049683618545532,
|
|
"epoch": 7.7322175732217575,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0001146959636753209,
|
|
"loss": 4.269,
|
|
"mean_token_accuracy": 0.2721634715795517,
|
|
"num_tokens": 17072722.0,
|
|
"step": 9240
|
|
},
|
|
{
|
|
"entropy": 5.065503406524658,
|
|
"epoch": 7.7364016736401675,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00011446964020613448,
|
|
"loss": 4.2758,
|
|
"mean_token_accuracy": 0.2736674875020981,
|
|
"num_tokens": 17081620.0,
|
|
"step": 9245
|
|
},
|
|
{
|
|
"entropy": 4.9678871631622314,
|
|
"epoch": 7.7405857740585775,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00011424364708261931,
|
|
"loss": 4.1573,
|
|
"mean_token_accuracy": 0.2903356432914734,
|
|
"num_tokens": 17090628.0,
|
|
"step": 9250
|
|
},
|
|
{
|
|
"entropy": 5.07299017906189,
|
|
"epoch": 7.7447698744769875,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00011401798476983281,
|
|
"loss": 4.3196,
|
|
"mean_token_accuracy": 0.26951802521944046,
|
|
"num_tokens": 17100089.0,
|
|
"step": 9255
|
|
},
|
|
{
|
|
"entropy": 5.029959058761596,
|
|
"epoch": 7.7489539748953975,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00011379265373215189,
|
|
"loss": 4.2707,
|
|
"mean_token_accuracy": 0.2690222695469856,
|
|
"num_tokens": 17108455.0,
|
|
"step": 9260
|
|
},
|
|
{
|
|
"entropy": 5.028404617309571,
|
|
"epoch": 7.7531380753138075,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00011356765443327146,
|
|
"loss": 4.2724,
|
|
"mean_token_accuracy": 0.2748915910720825,
|
|
"num_tokens": 17117803.0,
|
|
"step": 9265
|
|
},
|
|
{
|
|
"entropy": 5.062588024139404,
|
|
"epoch": 7.7573221757322175,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.00011334298733620405,
|
|
"loss": 4.2497,
|
|
"mean_token_accuracy": 0.2838040068745613,
|
|
"num_tokens": 17126436.0,
|
|
"step": 9270
|
|
},
|
|
{
|
|
"entropy": 5.087263488769532,
|
|
"epoch": 7.7615062761506275,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00011311865290327827,
|
|
"loss": 4.3476,
|
|
"mean_token_accuracy": 0.2598974838852882,
|
|
"num_tokens": 17135467.0,
|
|
"step": 9275
|
|
},
|
|
{
|
|
"entropy": 5.062054061889649,
|
|
"epoch": 7.7656903765690375,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00011289465159613849,
|
|
"loss": 4.3035,
|
|
"mean_token_accuracy": 0.27165281027555466,
|
|
"num_tokens": 17145282.0,
|
|
"step": 9280
|
|
},
|
|
{
|
|
"entropy": 5.050103712081909,
|
|
"epoch": 7.7698744769874475,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00011267098387574328,
|
|
"loss": 4.3364,
|
|
"mean_token_accuracy": 0.27117717564105986,
|
|
"num_tokens": 17155003.0,
|
|
"step": 9285
|
|
},
|
|
{
|
|
"entropy": 5.030095386505127,
|
|
"epoch": 7.7740585774058575,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00011244765020236477,
|
|
"loss": 4.2478,
|
|
"mean_token_accuracy": 0.27389675080776216,
|
|
"num_tokens": 17163784.0,
|
|
"step": 9290
|
|
},
|
|
{
|
|
"entropy": 5.013273143768311,
|
|
"epoch": 7.7782426778242675,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00011222465103558794,
|
|
"loss": 4.2462,
|
|
"mean_token_accuracy": 0.27200443744659425,
|
|
"num_tokens": 17172991.0,
|
|
"step": 9295
|
|
},
|
|
{
|
|
"entropy": 5.040381288528442,
|
|
"epoch": 7.7824267782426775,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00011200198683430904,
|
|
"loss": 4.2305,
|
|
"mean_token_accuracy": 0.27734278589487077,
|
|
"num_tokens": 17182035.0,
|
|
"step": 9300
|
|
},
|
|
{
|
|
"entropy": 5.021309089660645,
|
|
"epoch": 7.786610878661088,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00011177965805673528,
|
|
"loss": 4.2076,
|
|
"mean_token_accuracy": 0.2759990572929382,
|
|
"num_tokens": 17191257.0,
|
|
"step": 9305
|
|
},
|
|
{
|
|
"entropy": 5.0517433166503904,
|
|
"epoch": 7.790794979079498,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00011155766516038363,
|
|
"loss": 4.2416,
|
|
"mean_token_accuracy": 0.27116229832172395,
|
|
"num_tokens": 17200792.0,
|
|
"step": 9310
|
|
},
|
|
{
|
|
"entropy": 4.9979713439941404,
|
|
"epoch": 7.794979079497908,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00011133600860207965,
|
|
"loss": 4.2764,
|
|
"mean_token_accuracy": 0.26407433301210403,
|
|
"num_tokens": 17209833.0,
|
|
"step": 9315
|
|
},
|
|
{
|
|
"entropy": 5.007322978973389,
|
|
"epoch": 7.799163179916318,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00011111468883795698,
|
|
"loss": 4.2394,
|
|
"mean_token_accuracy": 0.2688806042075157,
|
|
"num_tokens": 17218982.0,
|
|
"step": 9320
|
|
},
|
|
{
|
|
"entropy": 5.058834838867187,
|
|
"epoch": 7.803347280334728,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00011089370632345599,
|
|
"loss": 4.2889,
|
|
"mean_token_accuracy": 0.2670976594090462,
|
|
"num_tokens": 17228557.0,
|
|
"step": 9325
|
|
},
|
|
{
|
|
"entropy": 5.040580892562867,
|
|
"epoch": 7.807531380753138,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00011067306151332331,
|
|
"loss": 4.2821,
|
|
"mean_token_accuracy": 0.27458608746528623,
|
|
"num_tokens": 17238081.0,
|
|
"step": 9330
|
|
},
|
|
{
|
|
"entropy": 4.982531547546387,
|
|
"epoch": 7.811715481171548,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00011045275486161039,
|
|
"loss": 4.2306,
|
|
"mean_token_accuracy": 0.2768711239099503,
|
|
"num_tokens": 17249000.0,
|
|
"step": 9335
|
|
},
|
|
{
|
|
"entropy": 5.02601990699768,
|
|
"epoch": 7.815899581589958,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001102327868216729,
|
|
"loss": 4.3042,
|
|
"mean_token_accuracy": 0.2723618522286415,
|
|
"num_tokens": 17259374.0,
|
|
"step": 9340
|
|
},
|
|
{
|
|
"entropy": 4.976259040832519,
|
|
"epoch": 7.820083682008368,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00011001315784616988,
|
|
"loss": 4.2355,
|
|
"mean_token_accuracy": 0.27565877288579943,
|
|
"num_tokens": 17268334.0,
|
|
"step": 9345
|
|
},
|
|
{
|
|
"entropy": 5.0442976474761965,
|
|
"epoch": 7.824267782426778,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0001097938683870622,
|
|
"loss": 4.2117,
|
|
"mean_token_accuracy": 0.27425878793001174,
|
|
"num_tokens": 17277364.0,
|
|
"step": 9350
|
|
},
|
|
{
|
|
"entropy": 5.029729652404785,
|
|
"epoch": 7.828451882845188,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00010957491889561244,
|
|
"loss": 4.2509,
|
|
"mean_token_accuracy": 0.2655960828065872,
|
|
"num_tokens": 17286552.0,
|
|
"step": 9355
|
|
},
|
|
{
|
|
"entropy": 5.0138932228088375,
|
|
"epoch": 7.832635983263598,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001093563098223834,
|
|
"loss": 4.2866,
|
|
"mean_token_accuracy": 0.2666684627532959,
|
|
"num_tokens": 17297114.0,
|
|
"step": 9360
|
|
},
|
|
{
|
|
"entropy": 5.0435230255126955,
|
|
"epoch": 7.836820083682008,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00010913804161723742,
|
|
"loss": 4.3057,
|
|
"mean_token_accuracy": 0.2676535725593567,
|
|
"num_tokens": 17306469.0,
|
|
"step": 9365
|
|
},
|
|
{
|
|
"entropy": 5.07969388961792,
|
|
"epoch": 7.841004184100418,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00010892011472933547,
|
|
"loss": 4.3156,
|
|
"mean_token_accuracy": 0.2768995717167854,
|
|
"num_tokens": 17315790.0,
|
|
"step": 9370
|
|
},
|
|
{
|
|
"entropy": 5.04618821144104,
|
|
"epoch": 7.845188284518828,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00010870252960713587,
|
|
"loss": 4.3253,
|
|
"mean_token_accuracy": 0.26163008660078046,
|
|
"num_tokens": 17324803.0,
|
|
"step": 9375
|
|
},
|
|
{
|
|
"entropy": 5.044091844558716,
|
|
"epoch": 7.849372384937238,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00010848528669839409,
|
|
"loss": 4.3269,
|
|
"mean_token_accuracy": 0.271440315246582,
|
|
"num_tokens": 17333856.0,
|
|
"step": 9380
|
|
},
|
|
{
|
|
"entropy": 4.999814748764038,
|
|
"epoch": 7.853556485355648,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00010826838645016077,
|
|
"loss": 4.1853,
|
|
"mean_token_accuracy": 0.2770646199584007,
|
|
"num_tokens": 17343399.0,
|
|
"step": 9385
|
|
},
|
|
{
|
|
"entropy": 5.0035968780517575,
|
|
"epoch": 7.857740585774058,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00010805182930878191,
|
|
"loss": 4.1788,
|
|
"mean_token_accuracy": 0.2797183275222778,
|
|
"num_tokens": 17352547.0,
|
|
"step": 9390
|
|
},
|
|
{
|
|
"entropy": 5.067364692687988,
|
|
"epoch": 7.861924686192468,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00010783561571989739,
|
|
"loss": 4.3115,
|
|
"mean_token_accuracy": 0.2673242509365082,
|
|
"num_tokens": 17362507.0,
|
|
"step": 9395
|
|
},
|
|
{
|
|
"entropy": 4.994563817977905,
|
|
"epoch": 7.866108786610878,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00010761974612843984,
|
|
"loss": 4.1803,
|
|
"mean_token_accuracy": 0.2784202516078949,
|
|
"num_tokens": 17372133.0,
|
|
"step": 9400
|
|
},
|
|
{
|
|
"entropy": 5.074862813949585,
|
|
"epoch": 7.870292887029288,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00010740422097863431,
|
|
"loss": 4.3228,
|
|
"mean_token_accuracy": 0.2619020164012909,
|
|
"num_tokens": 17383178.0,
|
|
"step": 9405
|
|
},
|
|
{
|
|
"entropy": 5.048515701293946,
|
|
"epoch": 7.874476987447698,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001071890407139968,
|
|
"loss": 4.3216,
|
|
"mean_token_accuracy": 0.2625690117478371,
|
|
"num_tokens": 17392357.0,
|
|
"step": 9410
|
|
},
|
|
{
|
|
"entropy": 5.071331310272217,
|
|
"epoch": 7.878661087866108,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00010697420577733385,
|
|
"loss": 4.3052,
|
|
"mean_token_accuracy": 0.27173558473587034,
|
|
"num_tokens": 17401101.0,
|
|
"step": 9415
|
|
},
|
|
{
|
|
"entropy": 4.981981086730957,
|
|
"epoch": 7.882845188284519,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00010675971661074102,
|
|
"loss": 4.1685,
|
|
"mean_token_accuracy": 0.2866115659475327,
|
|
"num_tokens": 17409909.0,
|
|
"step": 9420
|
|
},
|
|
{
|
|
"entropy": 5.0866528987884525,
|
|
"epoch": 7.887029288702929,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00010654557365560261,
|
|
"loss": 4.3326,
|
|
"mean_token_accuracy": 0.27177735418081284,
|
|
"num_tokens": 17419631.0,
|
|
"step": 9425
|
|
},
|
|
{
|
|
"entropy": 5.026048040390014,
|
|
"epoch": 7.891213389121339,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00010633177735259047,
|
|
"loss": 4.3103,
|
|
"mean_token_accuracy": 0.2631794348359108,
|
|
"num_tokens": 17428557.0,
|
|
"step": 9430
|
|
},
|
|
{
|
|
"entropy": 5.007184410095215,
|
|
"epoch": 7.895397489539749,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00010611832814166292,
|
|
"loss": 4.2677,
|
|
"mean_token_accuracy": 0.2721746176481247,
|
|
"num_tokens": 17438442.0,
|
|
"step": 9435
|
|
},
|
|
{
|
|
"entropy": 5.070800542831421,
|
|
"epoch": 7.899581589958159,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00010590522646206419,
|
|
"loss": 4.2964,
|
|
"mean_token_accuracy": 0.27106854170560835,
|
|
"num_tokens": 17448116.0,
|
|
"step": 9440
|
|
},
|
|
{
|
|
"entropy": 5.005953550338745,
|
|
"epoch": 7.903765690376569,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00010569247275232328,
|
|
"loss": 4.1989,
|
|
"mean_token_accuracy": 0.2795607328414917,
|
|
"num_tokens": 17456823.0,
|
|
"step": 9445
|
|
},
|
|
{
|
|
"entropy": 5.082686185836792,
|
|
"epoch": 7.907949790794979,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00010548006745025298,
|
|
"loss": 4.2782,
|
|
"mean_token_accuracy": 0.27925231158733366,
|
|
"num_tokens": 17466313.0,
|
|
"step": 9450
|
|
},
|
|
{
|
|
"entropy": 5.067076587677002,
|
|
"epoch": 7.912133891213389,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00010526801099294947,
|
|
"loss": 4.3669,
|
|
"mean_token_accuracy": 0.26218350380659106,
|
|
"num_tokens": 17477448.0,
|
|
"step": 9455
|
|
},
|
|
{
|
|
"entropy": 5.099254941940307,
|
|
"epoch": 7.916317991631799,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00010505630381679063,
|
|
"loss": 4.3311,
|
|
"mean_token_accuracy": 0.2724431723356247,
|
|
"num_tokens": 17487587.0,
|
|
"step": 9460
|
|
},
|
|
{
|
|
"entropy": 5.070095252990723,
|
|
"epoch": 7.920502092050209,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00010484494635743598,
|
|
"loss": 4.2672,
|
|
"mean_token_accuracy": 0.27221591025590897,
|
|
"num_tokens": 17496282.0,
|
|
"step": 9465
|
|
},
|
|
{
|
|
"entropy": 5.069140100479126,
|
|
"epoch": 7.924686192468619,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.000104633939049825,
|
|
"loss": 4.3243,
|
|
"mean_token_accuracy": 0.2629933461546898,
|
|
"num_tokens": 17505468.0,
|
|
"step": 9470
|
|
},
|
|
{
|
|
"entropy": 5.083333826065063,
|
|
"epoch": 7.928870292887029,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00010442328232817702,
|
|
"loss": 4.2513,
|
|
"mean_token_accuracy": 0.26906110942363737,
|
|
"num_tokens": 17515743.0,
|
|
"step": 9475
|
|
},
|
|
{
|
|
"entropy": 5.090427207946777,
|
|
"epoch": 7.933054393305439,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00010421297662598955,
|
|
"loss": 4.2312,
|
|
"mean_token_accuracy": 0.2801044970750809,
|
|
"num_tokens": 17524350.0,
|
|
"step": 9480
|
|
},
|
|
{
|
|
"entropy": 5.011372089385986,
|
|
"epoch": 7.937238493723849,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00010400302237603793,
|
|
"loss": 4.2157,
|
|
"mean_token_accuracy": 0.27107050120830534,
|
|
"num_tokens": 17533439.0,
|
|
"step": 9485
|
|
},
|
|
{
|
|
"entropy": 5.039130163192749,
|
|
"epoch": 7.941422594142259,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00010379342001037431,
|
|
"loss": 4.2488,
|
|
"mean_token_accuracy": 0.2697942167520523,
|
|
"num_tokens": 17542782.0,
|
|
"step": 9490
|
|
},
|
|
{
|
|
"entropy": 5.043299055099487,
|
|
"epoch": 7.945606694560669,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00010358416996032658,
|
|
"loss": 4.2731,
|
|
"mean_token_accuracy": 0.2744456797838211,
|
|
"num_tokens": 17552495.0,
|
|
"step": 9495
|
|
},
|
|
{
|
|
"entropy": 4.993641471862793,
|
|
"epoch": 7.949790794979079,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00010337527265649782,
|
|
"loss": 4.2745,
|
|
"mean_token_accuracy": 0.26963996738195417,
|
|
"num_tokens": 17560933.0,
|
|
"step": 9500
|
|
},
|
|
{
|
|
"epoch": 7.949790794979079,
|
|
"eval_entropy": 4.825869235036807,
|
|
"eval_loss": 5.289889812469482,
|
|
"eval_mean_token_accuracy": 0.2055124320805926,
|
|
"eval_num_tokens": 17560933.0,
|
|
"eval_runtime": 2.0831,
|
|
"eval_samples_per_second": 1875.6,
|
|
"eval_steps_per_second": 234.75,
|
|
"step": 9500
|
|
},
|
|
{
|
|
"entropy": 5.052238464355469,
|
|
"epoch": 7.95397489539749,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00010316672852876494,
|
|
"loss": 4.3069,
|
|
"mean_token_accuracy": 0.27297005653381345,
|
|
"num_tokens": 17571077.0,
|
|
"step": 9505
|
|
},
|
|
{
|
|
"entropy": 5.039814376831055,
|
|
"epoch": 7.9581589958159,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00010295853800627836,
|
|
"loss": 4.2481,
|
|
"mean_token_accuracy": 0.2719253093004227,
|
|
"num_tokens": 17580128.0,
|
|
"step": 9510
|
|
},
|
|
{
|
|
"entropy": 5.097062492370606,
|
|
"epoch": 7.96234309623431,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00010275070151746069,
|
|
"loss": 4.3125,
|
|
"mean_token_accuracy": 0.26955891102552415,
|
|
"num_tokens": 17589811.0,
|
|
"step": 9515
|
|
},
|
|
{
|
|
"entropy": 5.066234827041626,
|
|
"epoch": 7.96652719665272,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00010254321949000587,
|
|
"loss": 4.2853,
|
|
"mean_token_accuracy": 0.26900836527347566,
|
|
"num_tokens": 17599018.0,
|
|
"step": 9520
|
|
},
|
|
{
|
|
"entropy": 5.013735389709472,
|
|
"epoch": 7.97071129707113,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00010233609235087874,
|
|
"loss": 4.2003,
|
|
"mean_token_accuracy": 0.2722196146845818,
|
|
"num_tokens": 17607416.0,
|
|
"step": 9525
|
|
},
|
|
{
|
|
"entropy": 5.033866500854492,
|
|
"epoch": 7.97489539748954,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00010212932052631351,
|
|
"loss": 4.3162,
|
|
"mean_token_accuracy": 0.26662529110908506,
|
|
"num_tokens": 17616674.0,
|
|
"step": 9530
|
|
},
|
|
{
|
|
"entropy": 5.030138540267944,
|
|
"epoch": 7.97907949790795,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00010192290444181347,
|
|
"loss": 4.2705,
|
|
"mean_token_accuracy": 0.26915808767080307,
|
|
"num_tokens": 17626099.0,
|
|
"step": 9535
|
|
},
|
|
{
|
|
"entropy": 4.991519451141357,
|
|
"epoch": 7.98326359832636,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00010171684452214959,
|
|
"loss": 4.2514,
|
|
"mean_token_accuracy": 0.2685385629534721,
|
|
"num_tokens": 17636013.0,
|
|
"step": 9540
|
|
},
|
|
{
|
|
"entropy": 5.0862504005432125,
|
|
"epoch": 7.98744769874477,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00010151114119136024,
|
|
"loss": 4.2905,
|
|
"mean_token_accuracy": 0.2681483760476112,
|
|
"num_tokens": 17645849.0,
|
|
"step": 9545
|
|
},
|
|
{
|
|
"entropy": 5.087320137023926,
|
|
"epoch": 7.99163179916318,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00010130579487274969,
|
|
"loss": 4.3354,
|
|
"mean_token_accuracy": 0.26767594665288924,
|
|
"num_tokens": 17655557.0,
|
|
"step": 9550
|
|
},
|
|
{
|
|
"entropy": 5.056091356277466,
|
|
"epoch": 7.99581589958159,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001011008059888876,
|
|
"loss": 4.3505,
|
|
"mean_token_accuracy": 0.2669521152973175,
|
|
"num_tokens": 17665717.0,
|
|
"step": 9555
|
|
},
|
|
{
|
|
"entropy": 4.973353433609009,
|
|
"epoch": 8.0,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00010089617496160825,
|
|
"loss": 4.2041,
|
|
"mean_token_accuracy": 0.279686176776886,
|
|
"num_tokens": 17674576.0,
|
|
"step": 9560
|
|
},
|
|
{
|
|
"entropy": 5.063303422927857,
|
|
"epoch": 8.00418410041841,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00010069190221200929,
|
|
"loss": 4.1901,
|
|
"mean_token_accuracy": 0.2867306709289551,
|
|
"num_tokens": 17684147.0,
|
|
"step": 9565
|
|
},
|
|
{
|
|
"entropy": 4.962224960327148,
|
|
"epoch": 8.00836820083682,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00010048798816045132,
|
|
"loss": 4.1281,
|
|
"mean_token_accuracy": 0.2884212374687195,
|
|
"num_tokens": 17693082.0,
|
|
"step": 9570
|
|
},
|
|
{
|
|
"entropy": 4.9703936100006105,
|
|
"epoch": 8.01255230125523,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00010028443322655652,
|
|
"loss": 4.1669,
|
|
"mean_token_accuracy": 0.2806610107421875,
|
|
"num_tokens": 17702149.0,
|
|
"step": 9575
|
|
},
|
|
{
|
|
"entropy": 4.992264604568481,
|
|
"epoch": 8.01673640167364,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00010008123782920836,
|
|
"loss": 4.1529,
|
|
"mean_token_accuracy": 0.28808309733867643,
|
|
"num_tokens": 17710666.0,
|
|
"step": 9580
|
|
},
|
|
{
|
|
"entropy": 5.0076658725738525,
|
|
"epoch": 8.02092050209205,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 9.987840238655026e-05,
|
|
"loss": 4.214,
|
|
"mean_token_accuracy": 0.2777795284986496,
|
|
"num_tokens": 17719643.0,
|
|
"step": 9585
|
|
},
|
|
{
|
|
"entropy": 5.074099683761597,
|
|
"epoch": 8.02510460251046,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 9.967592731598487e-05,
|
|
"loss": 4.2532,
|
|
"mean_token_accuracy": 0.27414805740118026,
|
|
"num_tokens": 17728714.0,
|
|
"step": 9590
|
|
},
|
|
{
|
|
"entropy": 5.086824989318847,
|
|
"epoch": 8.02928870292887,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 9.947381303417346e-05,
|
|
"loss": 4.2558,
|
|
"mean_token_accuracy": 0.28147734701633453,
|
|
"num_tokens": 17736496.0,
|
|
"step": 9595
|
|
},
|
|
{
|
|
"entropy": 5.087257432937622,
|
|
"epoch": 8.03347280334728,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 9.927205995703456e-05,
|
|
"loss": 4.2963,
|
|
"mean_token_accuracy": 0.2720399841666222,
|
|
"num_tokens": 17745909.0,
|
|
"step": 9600
|
|
},
|
|
{
|
|
"entropy": 5.040671539306641,
|
|
"epoch": 8.03765690376569,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 9.907066849974372e-05,
|
|
"loss": 4.1886,
|
|
"mean_token_accuracy": 0.28223493993282317,
|
|
"num_tokens": 17755125.0,
|
|
"step": 9605
|
|
},
|
|
{
|
|
"entropy": 5.003476524353028,
|
|
"epoch": 8.0418410041841,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 9.886963907673214e-05,
|
|
"loss": 4.1669,
|
|
"mean_token_accuracy": 0.2803024411201477,
|
|
"num_tokens": 17764253.0,
|
|
"step": 9610
|
|
},
|
|
{
|
|
"entropy": 5.009642267227173,
|
|
"epoch": 8.04602510460251,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 9.866897210168608e-05,
|
|
"loss": 4.2297,
|
|
"mean_token_accuracy": 0.2759902149438858,
|
|
"num_tokens": 17773485.0,
|
|
"step": 9615
|
|
},
|
|
{
|
|
"entropy": 5.050246143341065,
|
|
"epoch": 8.05020920502092,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 9.846866798754583e-05,
|
|
"loss": 4.266,
|
|
"mean_token_accuracy": 0.27509860694408417,
|
|
"num_tokens": 17781942.0,
|
|
"step": 9620
|
|
},
|
|
{
|
|
"entropy": 5.0431334495544435,
|
|
"epoch": 8.05439330543933,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 9.826872714650509e-05,
|
|
"loss": 4.1868,
|
|
"mean_token_accuracy": 0.27666711807250977,
|
|
"num_tokens": 17791576.0,
|
|
"step": 9625
|
|
},
|
|
{
|
|
"entropy": 5.067641401290894,
|
|
"epoch": 8.05857740585774,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 9.806914999000998e-05,
|
|
"loss": 4.2388,
|
|
"mean_token_accuracy": 0.2772431164979935,
|
|
"num_tokens": 17801439.0,
|
|
"step": 9630
|
|
},
|
|
{
|
|
"entropy": 5.047262716293335,
|
|
"epoch": 8.06276150627615,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 9.786993692875814e-05,
|
|
"loss": 4.2637,
|
|
"mean_token_accuracy": 0.27281187772750853,
|
|
"num_tokens": 17810523.0,
|
|
"step": 9635
|
|
},
|
|
{
|
|
"entropy": 4.977340507507324,
|
|
"epoch": 8.06694560669456,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 9.767108837269808e-05,
|
|
"loss": 4.1598,
|
|
"mean_token_accuracy": 0.28288719356060027,
|
|
"num_tokens": 17819050.0,
|
|
"step": 9640
|
|
},
|
|
{
|
|
"entropy": 5.024219036102295,
|
|
"epoch": 8.07112970711297,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 9.747260473102829e-05,
|
|
"loss": 4.2314,
|
|
"mean_token_accuracy": 0.27856398224830625,
|
|
"num_tokens": 17828215.0,
|
|
"step": 9645
|
|
},
|
|
{
|
|
"entropy": 4.965454626083374,
|
|
"epoch": 8.07531380753138,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 9.727448641219591e-05,
|
|
"loss": 4.2085,
|
|
"mean_token_accuracy": 0.2770579934120178,
|
|
"num_tokens": 17837091.0,
|
|
"step": 9650
|
|
},
|
|
{
|
|
"entropy": 4.980905151367187,
|
|
"epoch": 8.07949790794979,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 9.707673382389678e-05,
|
|
"loss": 4.2038,
|
|
"mean_token_accuracy": 0.28658299446105956,
|
|
"num_tokens": 17845647.0,
|
|
"step": 9655
|
|
},
|
|
{
|
|
"entropy": 5.033685684204102,
|
|
"epoch": 8.0836820083682,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 9.687934737307386e-05,
|
|
"loss": 4.2033,
|
|
"mean_token_accuracy": 0.28195141404867174,
|
|
"num_tokens": 17854096.0,
|
|
"step": 9660
|
|
},
|
|
{
|
|
"entropy": 5.067026805877686,
|
|
"epoch": 8.087866108786612,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 9.668232746591671e-05,
|
|
"loss": 4.2041,
|
|
"mean_token_accuracy": 0.27937687635421754,
|
|
"num_tokens": 17863239.0,
|
|
"step": 9665
|
|
},
|
|
{
|
|
"entropy": 5.029227018356323,
|
|
"epoch": 8.092050209205022,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 9.648567450786077e-05,
|
|
"loss": 4.1939,
|
|
"mean_token_accuracy": 0.27796803414821625,
|
|
"num_tokens": 17872447.0,
|
|
"step": 9670
|
|
},
|
|
{
|
|
"entropy": 4.994064235687256,
|
|
"epoch": 8.096234309623432,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 9.628938890358604e-05,
|
|
"loss": 4.1731,
|
|
"mean_token_accuracy": 0.278631666302681,
|
|
"num_tokens": 17881862.0,
|
|
"step": 9675
|
|
},
|
|
{
|
|
"entropy": 4.978284788131714,
|
|
"epoch": 8.100418410041842,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 9.609347105701691e-05,
|
|
"loss": 4.2245,
|
|
"mean_token_accuracy": 0.2763476148247719,
|
|
"num_tokens": 17891683.0,
|
|
"step": 9680
|
|
},
|
|
{
|
|
"entropy": 5.077682542800903,
|
|
"epoch": 8.104602510460252,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 9.589792137132061e-05,
|
|
"loss": 4.2598,
|
|
"mean_token_accuracy": 0.27501572370529176,
|
|
"num_tokens": 17901105.0,
|
|
"step": 9685
|
|
},
|
|
{
|
|
"entropy": 4.925991249084473,
|
|
"epoch": 8.108786610878662,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 9.570274024890704e-05,
|
|
"loss": 4.1778,
|
|
"mean_token_accuracy": 0.28336783349514005,
|
|
"num_tokens": 17910761.0,
|
|
"step": 9690
|
|
},
|
|
{
|
|
"entropy": 5.0583244323730465,
|
|
"epoch": 8.112970711297072,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 9.550792809142764e-05,
|
|
"loss": 4.2117,
|
|
"mean_token_accuracy": 0.2813915640115738,
|
|
"num_tokens": 17919917.0,
|
|
"step": 9695
|
|
},
|
|
{
|
|
"entropy": 5.082311391830444,
|
|
"epoch": 8.117154811715482,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 9.531348529977439e-05,
|
|
"loss": 4.2846,
|
|
"mean_token_accuracy": 0.26801287978887556,
|
|
"num_tokens": 17928588.0,
|
|
"step": 9700
|
|
},
|
|
{
|
|
"entropy": 5.056627416610718,
|
|
"epoch": 8.121338912133892,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 9.51194122740794e-05,
|
|
"loss": 4.2292,
|
|
"mean_token_accuracy": 0.2816961854696274,
|
|
"num_tokens": 17938142.0,
|
|
"step": 9705
|
|
},
|
|
{
|
|
"entropy": 4.977713966369629,
|
|
"epoch": 8.125523012552302,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 9.492570941371365e-05,
|
|
"loss": 4.1402,
|
|
"mean_token_accuracy": 0.2833647862076759,
|
|
"num_tokens": 17947676.0,
|
|
"step": 9710
|
|
},
|
|
{
|
|
"entropy": 4.962055206298828,
|
|
"epoch": 8.129707112970712,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 9.47323771172866e-05,
|
|
"loss": 4.2223,
|
|
"mean_token_accuracy": 0.2822852790355682,
|
|
"num_tokens": 17957166.0,
|
|
"step": 9715
|
|
},
|
|
{
|
|
"entropy": 5.05087513923645,
|
|
"epoch": 8.133891213389122,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 9.453941578264499e-05,
|
|
"loss": 4.2388,
|
|
"mean_token_accuracy": 0.278366906940937,
|
|
"num_tokens": 17965592.0,
|
|
"step": 9720
|
|
},
|
|
{
|
|
"entropy": 4.969899225234985,
|
|
"epoch": 8.138075313807532,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 9.434682580687215e-05,
|
|
"loss": 4.182,
|
|
"mean_token_accuracy": 0.2748723983764648,
|
|
"num_tokens": 17973949.0,
|
|
"step": 9725
|
|
},
|
|
{
|
|
"entropy": 4.960282230377198,
|
|
"epoch": 8.142259414225942,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 9.415460758628739e-05,
|
|
"loss": 4.1767,
|
|
"mean_token_accuracy": 0.27881801426410674,
|
|
"num_tokens": 17982540.0,
|
|
"step": 9730
|
|
},
|
|
{
|
|
"entropy": 4.982939910888672,
|
|
"epoch": 8.146443514644352,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 9.396276151644478e-05,
|
|
"loss": 4.1671,
|
|
"mean_token_accuracy": 0.27901965379714966,
|
|
"num_tokens": 17991611.0,
|
|
"step": 9735
|
|
},
|
|
{
|
|
"entropy": 5.037381792068482,
|
|
"epoch": 8.150627615062762,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 9.377128799213286e-05,
|
|
"loss": 4.1612,
|
|
"mean_token_accuracy": 0.2865539491176605,
|
|
"num_tokens": 17999462.0,
|
|
"step": 9740
|
|
},
|
|
{
|
|
"entropy": 5.079920625686645,
|
|
"epoch": 8.154811715481172,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 9.358018740737325e-05,
|
|
"loss": 4.3062,
|
|
"mean_token_accuracy": 0.27679474502801893,
|
|
"num_tokens": 18009693.0,
|
|
"step": 9745
|
|
},
|
|
{
|
|
"entropy": 4.970609426498413,
|
|
"epoch": 8.158995815899582,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 9.338946015542018e-05,
|
|
"loss": 4.091,
|
|
"mean_token_accuracy": 0.28968202471733095,
|
|
"num_tokens": 18019101.0,
|
|
"step": 9750
|
|
},
|
|
{
|
|
"entropy": 5.0230405807495115,
|
|
"epoch": 8.163179916317992,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 9.31991066287598e-05,
|
|
"loss": 4.3203,
|
|
"mean_token_accuracy": 0.26787225604057313,
|
|
"num_tokens": 18028198.0,
|
|
"step": 9755
|
|
},
|
|
{
|
|
"entropy": 5.059514665603638,
|
|
"epoch": 8.167364016736402,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 9.300912721910894e-05,
|
|
"loss": 4.274,
|
|
"mean_token_accuracy": 0.27333677709102633,
|
|
"num_tokens": 18037592.0,
|
|
"step": 9760
|
|
},
|
|
{
|
|
"entropy": 5.024917078018189,
|
|
"epoch": 8.171548117154812,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 9.28195223174148e-05,
|
|
"loss": 4.2389,
|
|
"mean_token_accuracy": 0.27330819666385653,
|
|
"num_tokens": 18046988.0,
|
|
"step": 9765
|
|
},
|
|
{
|
|
"entropy": 4.948165607452393,
|
|
"epoch": 8.175732217573222,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 9.263029231385366e-05,
|
|
"loss": 4.135,
|
|
"mean_token_accuracy": 0.2916024178266525,
|
|
"num_tokens": 18056322.0,
|
|
"step": 9770
|
|
},
|
|
{
|
|
"entropy": 4.95274977684021,
|
|
"epoch": 8.179916317991632,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 9.244143759783058e-05,
|
|
"loss": 4.1448,
|
|
"mean_token_accuracy": 0.2786892384290695,
|
|
"num_tokens": 18065157.0,
|
|
"step": 9775
|
|
},
|
|
{
|
|
"entropy": 5.075520801544189,
|
|
"epoch": 8.184100418410042,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 9.225295855797812e-05,
|
|
"loss": 4.2752,
|
|
"mean_token_accuracy": 0.27706539034843447,
|
|
"num_tokens": 18074389.0,
|
|
"step": 9780
|
|
},
|
|
{
|
|
"entropy": 5.052859115600586,
|
|
"epoch": 8.188284518828452,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 9.20648555821558e-05,
|
|
"loss": 4.1984,
|
|
"mean_token_accuracy": 0.2828655779361725,
|
|
"num_tokens": 18083434.0,
|
|
"step": 9785
|
|
},
|
|
{
|
|
"entropy": 4.988186073303223,
|
|
"epoch": 8.192468619246862,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 9.18771290574494e-05,
|
|
"loss": 4.1783,
|
|
"mean_token_accuracy": 0.2772080272436142,
|
|
"num_tokens": 18094145.0,
|
|
"step": 9790
|
|
},
|
|
{
|
|
"entropy": 5.037233448028564,
|
|
"epoch": 8.196652719665272,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 9.168977937016981e-05,
|
|
"loss": 4.2425,
|
|
"mean_token_accuracy": 0.276852636039257,
|
|
"num_tokens": 18103707.0,
|
|
"step": 9795
|
|
},
|
|
{
|
|
"entropy": 4.984955644607544,
|
|
"epoch": 8.200836820083682,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 9.150280690585271e-05,
|
|
"loss": 4.1783,
|
|
"mean_token_accuracy": 0.28189956545829775,
|
|
"num_tokens": 18113416.0,
|
|
"step": 9800
|
|
},
|
|
{
|
|
"entropy": 4.962972640991211,
|
|
"epoch": 8.205020920502092,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 9.131621204925724e-05,
|
|
"loss": 4.1572,
|
|
"mean_token_accuracy": 0.2784869909286499,
|
|
"num_tokens": 18122809.0,
|
|
"step": 9805
|
|
},
|
|
{
|
|
"entropy": 5.078785848617554,
|
|
"epoch": 8.209205020920502,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 9.112999518436574e-05,
|
|
"loss": 4.2896,
|
|
"mean_token_accuracy": 0.264019201695919,
|
|
"num_tokens": 18132518.0,
|
|
"step": 9810
|
|
},
|
|
{
|
|
"entropy": 5.0259480476379395,
|
|
"epoch": 8.213389121338912,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 9.09441566943825e-05,
|
|
"loss": 4.2259,
|
|
"mean_token_accuracy": 0.2733228117227554,
|
|
"num_tokens": 18141499.0,
|
|
"step": 9815
|
|
},
|
|
{
|
|
"entropy": 5.054711246490479,
|
|
"epoch": 8.217573221757322,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 9.075869696173321e-05,
|
|
"loss": 4.207,
|
|
"mean_token_accuracy": 0.2811878636479378,
|
|
"num_tokens": 18150330.0,
|
|
"step": 9820
|
|
},
|
|
{
|
|
"entropy": 5.0294029712677,
|
|
"epoch": 8.221757322175732,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 9.057361636806428e-05,
|
|
"loss": 4.1646,
|
|
"mean_token_accuracy": 0.28192920982837677,
|
|
"num_tokens": 18158925.0,
|
|
"step": 9825
|
|
},
|
|
{
|
|
"entropy": 5.008471918106079,
|
|
"epoch": 8.225941422594142,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 9.038891529424173e-05,
|
|
"loss": 4.1752,
|
|
"mean_token_accuracy": 0.28283385932445526,
|
|
"num_tokens": 18167612.0,
|
|
"step": 9830
|
|
},
|
|
{
|
|
"entropy": 5.063526964187622,
|
|
"epoch": 8.230125523012552,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 9.020459412035075e-05,
|
|
"loss": 4.2167,
|
|
"mean_token_accuracy": 0.28126252591609957,
|
|
"num_tokens": 18176727.0,
|
|
"step": 9835
|
|
},
|
|
{
|
|
"entropy": 5.062085866928101,
|
|
"epoch": 8.234309623430962,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 9.002065322569464e-05,
|
|
"loss": 4.2281,
|
|
"mean_token_accuracy": 0.276458115875721,
|
|
"num_tokens": 18186542.0,
|
|
"step": 9840
|
|
},
|
|
{
|
|
"entropy": 4.978729200363159,
|
|
"epoch": 8.238493723849372,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 8.983709298879424e-05,
|
|
"loss": 4.1668,
|
|
"mean_token_accuracy": 0.27908451706171034,
|
|
"num_tokens": 18195514.0,
|
|
"step": 9845
|
|
},
|
|
{
|
|
"entropy": 4.982341670989991,
|
|
"epoch": 8.242677824267782,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 8.965391378738692e-05,
|
|
"loss": 4.1656,
|
|
"mean_token_accuracy": 0.278090363740921,
|
|
"num_tokens": 18204481.0,
|
|
"step": 9850
|
|
},
|
|
{
|
|
"entropy": 5.086486291885376,
|
|
"epoch": 8.246861924686192,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 8.947111599842606e-05,
|
|
"loss": 4.2105,
|
|
"mean_token_accuracy": 0.2768116295337677,
|
|
"num_tokens": 18213342.0,
|
|
"step": 9855
|
|
},
|
|
{
|
|
"entropy": 5.051018619537354,
|
|
"epoch": 8.251046025104603,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 8.928869999808017e-05,
|
|
"loss": 4.2536,
|
|
"mean_token_accuracy": 0.27513068318367007,
|
|
"num_tokens": 18223083.0,
|
|
"step": 9860
|
|
},
|
|
{
|
|
"entropy": 4.977971506118775,
|
|
"epoch": 8.255230125523013,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 8.910666616173196e-05,
|
|
"loss": 4.1712,
|
|
"mean_token_accuracy": 0.28243237286806105,
|
|
"num_tokens": 18231792.0,
|
|
"step": 9865
|
|
},
|
|
{
|
|
"entropy": 5.008806037902832,
|
|
"epoch": 8.259414225941423,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 8.89250148639779e-05,
|
|
"loss": 4.2029,
|
|
"mean_token_accuracy": 0.2870588004589081,
|
|
"num_tokens": 18240099.0,
|
|
"step": 9870
|
|
},
|
|
{
|
|
"entropy": 5.006771087646484,
|
|
"epoch": 8.263598326359833,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 8.874374647862704e-05,
|
|
"loss": 4.234,
|
|
"mean_token_accuracy": 0.27143135815858843,
|
|
"num_tokens": 18249237.0,
|
|
"step": 9875
|
|
},
|
|
{
|
|
"entropy": 5.053074502944947,
|
|
"epoch": 8.267782426778243,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 8.856286137870068e-05,
|
|
"loss": 4.2177,
|
|
"mean_token_accuracy": 0.28085602819919586,
|
|
"num_tokens": 18257896.0,
|
|
"step": 9880
|
|
},
|
|
{
|
|
"entropy": 5.017831182479858,
|
|
"epoch": 8.271966527196653,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 8.838235993643125e-05,
|
|
"loss": 4.1735,
|
|
"mean_token_accuracy": 0.2854412138462067,
|
|
"num_tokens": 18266251.0,
|
|
"step": 9885
|
|
},
|
|
{
|
|
"entropy": 5.002558279037475,
|
|
"epoch": 8.276150627615063,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 8.820224252326161e-05,
|
|
"loss": 4.1777,
|
|
"mean_token_accuracy": 0.2826448917388916,
|
|
"num_tokens": 18275729.0,
|
|
"step": 9890
|
|
},
|
|
{
|
|
"entropy": 5.055112981796265,
|
|
"epoch": 8.280334728033473,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 8.802250950984461e-05,
|
|
"loss": 4.2509,
|
|
"mean_token_accuracy": 0.27306572049856187,
|
|
"num_tokens": 18284625.0,
|
|
"step": 9895
|
|
},
|
|
{
|
|
"entropy": 4.966627693176269,
|
|
"epoch": 8.284518828451883,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 8.784316126604169e-05,
|
|
"loss": 4.1401,
|
|
"mean_token_accuracy": 0.2849645525217056,
|
|
"num_tokens": 18293390.0,
|
|
"step": 9900
|
|
},
|
|
{
|
|
"entropy": 4.9697594165802,
|
|
"epoch": 8.288702928870293,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 8.766419816092284e-05,
|
|
"loss": 4.1578,
|
|
"mean_token_accuracy": 0.28503442704677584,
|
|
"num_tokens": 18302955.0,
|
|
"step": 9905
|
|
},
|
|
{
|
|
"entropy": 5.038019514083862,
|
|
"epoch": 8.292887029288703,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 8.748562056276535e-05,
|
|
"loss": 4.2569,
|
|
"mean_token_accuracy": 0.2719696074724197,
|
|
"num_tokens": 18311614.0,
|
|
"step": 9910
|
|
},
|
|
{
|
|
"entropy": 5.06715350151062,
|
|
"epoch": 8.297071129707113,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 8.730742883905321e-05,
|
|
"loss": 4.2412,
|
|
"mean_token_accuracy": 0.2767865478992462,
|
|
"num_tokens": 18320588.0,
|
|
"step": 9915
|
|
},
|
|
{
|
|
"entropy": 4.951652145385742,
|
|
"epoch": 8.301255230125523,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 8.712962335647629e-05,
|
|
"loss": 4.0977,
|
|
"mean_token_accuracy": 0.2911615252494812,
|
|
"num_tokens": 18329976.0,
|
|
"step": 9920
|
|
},
|
|
{
|
|
"entropy": 5.004001712799072,
|
|
"epoch": 8.305439330543933,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 8.695220448092964e-05,
|
|
"loss": 4.2216,
|
|
"mean_token_accuracy": 0.2751340627670288,
|
|
"num_tokens": 18338495.0,
|
|
"step": 9925
|
|
},
|
|
{
|
|
"entropy": 5.060586071014404,
|
|
"epoch": 8.309623430962343,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 8.677517257751286e-05,
|
|
"loss": 4.2082,
|
|
"mean_token_accuracy": 0.2826252207159996,
|
|
"num_tokens": 18348212.0,
|
|
"step": 9930
|
|
},
|
|
{
|
|
"entropy": 4.953409814834595,
|
|
"epoch": 8.313807531380753,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 8.659852801052915e-05,
|
|
"loss": 4.1007,
|
|
"mean_token_accuracy": 0.29038176238536834,
|
|
"num_tokens": 18356803.0,
|
|
"step": 9935
|
|
},
|
|
{
|
|
"entropy": 5.122460079193115,
|
|
"epoch": 8.317991631799163,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 8.64222711434846e-05,
|
|
"loss": 4.3375,
|
|
"mean_token_accuracy": 0.27103967070579527,
|
|
"num_tokens": 18366822.0,
|
|
"step": 9940
|
|
},
|
|
{
|
|
"entropy": 5.049137926101684,
|
|
"epoch": 8.322175732217573,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 8.62464023390876e-05,
|
|
"loss": 4.2403,
|
|
"mean_token_accuracy": 0.27120221108198167,
|
|
"num_tokens": 18376495.0,
|
|
"step": 9945
|
|
},
|
|
{
|
|
"entropy": 4.961305856704712,
|
|
"epoch": 8.326359832635983,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 8.607092195924768e-05,
|
|
"loss": 4.1102,
|
|
"mean_token_accuracy": 0.2882045090198517,
|
|
"num_tokens": 18385652.0,
|
|
"step": 9950
|
|
},
|
|
{
|
|
"entropy": 4.991803550720215,
|
|
"epoch": 8.330543933054393,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 8.58958303650754e-05,
|
|
"loss": 4.1914,
|
|
"mean_token_accuracy": 0.28207722008228303,
|
|
"num_tokens": 18394604.0,
|
|
"step": 9955
|
|
},
|
|
{
|
|
"entropy": 5.060005855560303,
|
|
"epoch": 8.334728033472803,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 8.572112791688105e-05,
|
|
"loss": 4.1963,
|
|
"mean_token_accuracy": 0.28778715133666993,
|
|
"num_tokens": 18403477.0,
|
|
"step": 9960
|
|
},
|
|
{
|
|
"entropy": 5.023592138290406,
|
|
"epoch": 8.338912133891213,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 8.554681497417416e-05,
|
|
"loss": 4.1972,
|
|
"mean_token_accuracy": 0.28311238139867784,
|
|
"num_tokens": 18412866.0,
|
|
"step": 9965
|
|
},
|
|
{
|
|
"entropy": 5.030958032608032,
|
|
"epoch": 8.343096234309623,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 8.537289189566284e-05,
|
|
"loss": 4.1872,
|
|
"mean_token_accuracy": 0.2783506542444229,
|
|
"num_tokens": 18421742.0,
|
|
"step": 9970
|
|
},
|
|
{
|
|
"entropy": 5.05708327293396,
|
|
"epoch": 8.347280334728033,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 8.519935903925271e-05,
|
|
"loss": 4.2478,
|
|
"mean_token_accuracy": 0.26256821155548093,
|
|
"num_tokens": 18431331.0,
|
|
"step": 9975
|
|
},
|
|
{
|
|
"entropy": 5.0161481380462645,
|
|
"epoch": 8.351464435146443,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 8.502621676204666e-05,
|
|
"loss": 4.288,
|
|
"mean_token_accuracy": 0.2753719657659531,
|
|
"num_tokens": 18440906.0,
|
|
"step": 9980
|
|
},
|
|
{
|
|
"entropy": 5.0068212985992435,
|
|
"epoch": 8.355648535564853,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 8.485346542034343e-05,
|
|
"loss": 4.235,
|
|
"mean_token_accuracy": 0.26996591240167617,
|
|
"num_tokens": 18450093.0,
|
|
"step": 9985
|
|
},
|
|
{
|
|
"entropy": 4.943309116363525,
|
|
"epoch": 8.359832635983263,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 8.468110536963763e-05,
|
|
"loss": 4.1303,
|
|
"mean_token_accuracy": 0.2823055058717728,
|
|
"num_tokens": 18458536.0,
|
|
"step": 9990
|
|
},
|
|
{
|
|
"entropy": 5.052586412429809,
|
|
"epoch": 8.364016736401673,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 8.450913696461856e-05,
|
|
"loss": 4.1831,
|
|
"mean_token_accuracy": 0.2845236137509346,
|
|
"num_tokens": 18466910.0,
|
|
"step": 9995
|
|
},
|
|
{
|
|
"entropy": 5.054701852798462,
|
|
"epoch": 8.368200836820083,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 8.433756055916944e-05,
|
|
"loss": 4.2878,
|
|
"mean_token_accuracy": 0.27250601202249525,
|
|
"num_tokens": 18476098.0,
|
|
"step": 10000
|
|
},
|
|
{
|
|
"epoch": 8.368200836820083,
|
|
"eval_entropy": 4.824316427995335,
|
|
"eval_loss": 5.298755168914795,
|
|
"eval_mean_token_accuracy": 0.205755040095262,
|
|
"eval_num_tokens": 18476098.0,
|
|
"eval_runtime": 2.0835,
|
|
"eval_samples_per_second": 1875.241,
|
|
"eval_steps_per_second": 234.705,
|
|
"step": 10000
|
|
},
|
|
{
|
|
"entropy": 4.9570457458496096,
|
|
"epoch": 8.372384937238493,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 8.416637650636708e-05,
|
|
"loss": 4.1713,
|
|
"mean_token_accuracy": 0.28225075751543044,
|
|
"num_tokens": 18486059.0,
|
|
"step": 10005
|
|
},
|
|
{
|
|
"entropy": 5.0787766456604,
|
|
"epoch": 8.376569037656903,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 8.399558515848059e-05,
|
|
"loss": 4.2751,
|
|
"mean_token_accuracy": 0.27219097763299943,
|
|
"num_tokens": 18495373.0,
|
|
"step": 10010
|
|
},
|
|
{
|
|
"entropy": 5.007111644744873,
|
|
"epoch": 8.380753138075313,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 8.382518686697122e-05,
|
|
"loss": 4.1736,
|
|
"mean_token_accuracy": 0.2838737219572067,
|
|
"num_tokens": 18504559.0,
|
|
"step": 10015
|
|
},
|
|
{
|
|
"entropy": 5.037234210968018,
|
|
"epoch": 8.384937238493723,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 8.365518198249122e-05,
|
|
"loss": 4.203,
|
|
"mean_token_accuracy": 0.28168347775936126,
|
|
"num_tokens": 18514267.0,
|
|
"step": 10020
|
|
},
|
|
{
|
|
"entropy": 4.90367169380188,
|
|
"epoch": 8.389121338912133,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 8.348557085488323e-05,
|
|
"loss": 4.0876,
|
|
"mean_token_accuracy": 0.29268074333667754,
|
|
"num_tokens": 18523664.0,
|
|
"step": 10025
|
|
},
|
|
{
|
|
"entropy": 4.9796000003814695,
|
|
"epoch": 8.393305439330543,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 8.331635383317981e-05,
|
|
"loss": 4.2748,
|
|
"mean_token_accuracy": 0.2772753670811653,
|
|
"num_tokens": 18532731.0,
|
|
"step": 10030
|
|
},
|
|
{
|
|
"entropy": 5.002987384796143,
|
|
"epoch": 8.397489539748953,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 8.314753126560229e-05,
|
|
"loss": 4.1875,
|
|
"mean_token_accuracy": 0.2758780851960182,
|
|
"num_tokens": 18542041.0,
|
|
"step": 10035
|
|
},
|
|
{
|
|
"entropy": 5.088874101638794,
|
|
"epoch": 8.401673640167363,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 8.297910349956045e-05,
|
|
"loss": 4.2309,
|
|
"mean_token_accuracy": 0.2806305274367332,
|
|
"num_tokens": 18551006.0,
|
|
"step": 10040
|
|
},
|
|
{
|
|
"entropy": 5.018345689773559,
|
|
"epoch": 8.405857740585773,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 8.28110708816515e-05,
|
|
"loss": 4.233,
|
|
"mean_token_accuracy": 0.27696211338043214,
|
|
"num_tokens": 18560822.0,
|
|
"step": 10045
|
|
},
|
|
{
|
|
"entropy": 4.955094480514527,
|
|
"epoch": 8.410041841004183,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 8.264343375765951e-05,
|
|
"loss": 4.1889,
|
|
"mean_token_accuracy": 0.2775804489850998,
|
|
"num_tokens": 18570993.0,
|
|
"step": 10050
|
|
},
|
|
{
|
|
"entropy": 4.993603324890136,
|
|
"epoch": 8.414225941422593,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 8.247619247255486e-05,
|
|
"loss": 4.2329,
|
|
"mean_token_accuracy": 0.2754255771636963,
|
|
"num_tokens": 18580424.0,
|
|
"step": 10055
|
|
},
|
|
{
|
|
"entropy": 4.976649427413941,
|
|
"epoch": 8.418410041841003,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 8.230934737049309e-05,
|
|
"loss": 4.214,
|
|
"mean_token_accuracy": 0.2767063617706299,
|
|
"num_tokens": 18588789.0,
|
|
"step": 10060
|
|
},
|
|
{
|
|
"entropy": 5.027304315567017,
|
|
"epoch": 8.422594142259415,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 8.21428987948147e-05,
|
|
"loss": 4.2125,
|
|
"mean_token_accuracy": 0.27069309651851653,
|
|
"num_tokens": 18597553.0,
|
|
"step": 10065
|
|
},
|
|
{
|
|
"entropy": 5.062617826461792,
|
|
"epoch": 8.426778242677825,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 8.197684708804399e-05,
|
|
"loss": 4.2679,
|
|
"mean_token_accuracy": 0.2750175341963768,
|
|
"num_tokens": 18606997.0,
|
|
"step": 10070
|
|
},
|
|
{
|
|
"entropy": 5.007031869888306,
|
|
"epoch": 8.430962343096235,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 8.18111925918888e-05,
|
|
"loss": 4.1853,
|
|
"mean_token_accuracy": 0.28118045926094054,
|
|
"num_tokens": 18615807.0,
|
|
"step": 10075
|
|
},
|
|
{
|
|
"entropy": 5.05542402267456,
|
|
"epoch": 8.435146443514645,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 8.164593564723938e-05,
|
|
"loss": 4.2493,
|
|
"mean_token_accuracy": 0.27469212412834165,
|
|
"num_tokens": 18625276.0,
|
|
"step": 10080
|
|
},
|
|
{
|
|
"entropy": 5.062376689910889,
|
|
"epoch": 8.439330543933055,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 8.148107659416793e-05,
|
|
"loss": 4.2529,
|
|
"mean_token_accuracy": 0.27851465046405793,
|
|
"num_tokens": 18634269.0,
|
|
"step": 10085
|
|
},
|
|
{
|
|
"entropy": 5.145097875595093,
|
|
"epoch": 8.443514644351465,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 8.131661577192792e-05,
|
|
"loss": 4.3243,
|
|
"mean_token_accuracy": 0.2753843739628792,
|
|
"num_tokens": 18644506.0,
|
|
"step": 10090
|
|
},
|
|
{
|
|
"entropy": 5.0679045677185055,
|
|
"epoch": 8.447698744769875,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 8.115255351895325e-05,
|
|
"loss": 4.1899,
|
|
"mean_token_accuracy": 0.2766395941376686,
|
|
"num_tokens": 18654250.0,
|
|
"step": 10095
|
|
},
|
|
{
|
|
"entropy": 5.092316961288452,
|
|
"epoch": 8.451882845188285,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 8.098889017285771e-05,
|
|
"loss": 4.2393,
|
|
"mean_token_accuracy": 0.2761916607618332,
|
|
"num_tokens": 18664089.0,
|
|
"step": 10100
|
|
},
|
|
{
|
|
"entropy": 5.129371070861817,
|
|
"epoch": 8.456066945606695,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 8.082562607043407e-05,
|
|
"loss": 4.3081,
|
|
"mean_token_accuracy": 0.2711872115731239,
|
|
"num_tokens": 18674233.0,
|
|
"step": 10105
|
|
},
|
|
{
|
|
"entropy": 5.0900859355926515,
|
|
"epoch": 8.460251046025105,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 8.066276154765374e-05,
|
|
"loss": 4.2369,
|
|
"mean_token_accuracy": 0.2794917166233063,
|
|
"num_tokens": 18683960.0,
|
|
"step": 10110
|
|
},
|
|
{
|
|
"entropy": 4.990304183959961,
|
|
"epoch": 8.464435146443515,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 8.050029693966561e-05,
|
|
"loss": 4.2431,
|
|
"mean_token_accuracy": 0.27529713064432143,
|
|
"num_tokens": 18694418.0,
|
|
"step": 10115
|
|
},
|
|
{
|
|
"entropy": 5.014610004425049,
|
|
"epoch": 8.468619246861925,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 8.033823258079572e-05,
|
|
"loss": 4.2043,
|
|
"mean_token_accuracy": 0.2807725012302399,
|
|
"num_tokens": 18703986.0,
|
|
"step": 10120
|
|
},
|
|
{
|
|
"entropy": 5.047932958602905,
|
|
"epoch": 8.472803347280335,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 8.017656880454651e-05,
|
|
"loss": 4.2462,
|
|
"mean_token_accuracy": 0.27711226642131803,
|
|
"num_tokens": 18712856.0,
|
|
"step": 10125
|
|
},
|
|
{
|
|
"entropy": 5.000606298446655,
|
|
"epoch": 8.476987447698745,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 8.0015305943596e-05,
|
|
"loss": 4.2513,
|
|
"mean_token_accuracy": 0.275727154314518,
|
|
"num_tokens": 18721795.0,
|
|
"step": 10130
|
|
},
|
|
{
|
|
"entropy": 4.99550085067749,
|
|
"epoch": 8.481171548117155,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 7.985444432979733e-05,
|
|
"loss": 4.2674,
|
|
"mean_token_accuracy": 0.27308051884174345,
|
|
"num_tokens": 18730475.0,
|
|
"step": 10135
|
|
},
|
|
{
|
|
"entropy": 5.058686590194702,
|
|
"epoch": 8.485355648535565,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 7.969398429417768e-05,
|
|
"loss": 4.1792,
|
|
"mean_token_accuracy": 0.2821617811918259,
|
|
"num_tokens": 18740017.0,
|
|
"step": 10140
|
|
},
|
|
{
|
|
"entropy": 5.073039722442627,
|
|
"epoch": 8.489539748953975,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 7.953392616693814e-05,
|
|
"loss": 4.2525,
|
|
"mean_token_accuracy": 0.2722253829240799,
|
|
"num_tokens": 18749018.0,
|
|
"step": 10145
|
|
},
|
|
{
|
|
"entropy": 5.096699523925781,
|
|
"epoch": 8.493723849372385,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.937427027745259e-05,
|
|
"loss": 4.3063,
|
|
"mean_token_accuracy": 0.26944058686494826,
|
|
"num_tokens": 18758661.0,
|
|
"step": 10150
|
|
},
|
|
{
|
|
"entropy": 5.008637714385986,
|
|
"epoch": 8.497907949790795,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 7.921501695426707e-05,
|
|
"loss": 4.1898,
|
|
"mean_token_accuracy": 0.27210600674152374,
|
|
"num_tokens": 18767769.0,
|
|
"step": 10155
|
|
},
|
|
{
|
|
"entropy": 5.036049699783325,
|
|
"epoch": 8.502092050209205,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 7.905616652509945e-05,
|
|
"loss": 4.2868,
|
|
"mean_token_accuracy": 0.26859958469867706,
|
|
"num_tokens": 18777069.0,
|
|
"step": 10160
|
|
},
|
|
{
|
|
"entropy": 5.064014863967896,
|
|
"epoch": 8.506276150627615,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 7.889771931683826e-05,
|
|
"loss": 4.2638,
|
|
"mean_token_accuracy": 0.2722799763083458,
|
|
"num_tokens": 18788123.0,
|
|
"step": 10165
|
|
},
|
|
{
|
|
"entropy": 5.029590463638305,
|
|
"epoch": 8.510460251046025,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 7.873967565554247e-05,
|
|
"loss": 4.2121,
|
|
"mean_token_accuracy": 0.28239852786064146,
|
|
"num_tokens": 18797421.0,
|
|
"step": 10170
|
|
},
|
|
{
|
|
"entropy": 4.956624746322632,
|
|
"epoch": 8.514644351464435,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 7.858203586644047e-05,
|
|
"loss": 4.1666,
|
|
"mean_token_accuracy": 0.2844740390777588,
|
|
"num_tokens": 18806542.0,
|
|
"step": 10175
|
|
},
|
|
{
|
|
"entropy": 5.07797064781189,
|
|
"epoch": 8.518828451882845,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 7.842480027392966e-05,
|
|
"loss": 4.2695,
|
|
"mean_token_accuracy": 0.2714936062693596,
|
|
"num_tokens": 18815498.0,
|
|
"step": 10180
|
|
},
|
|
{
|
|
"entropy": 5.011352396011352,
|
|
"epoch": 8.523012552301255,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 7.826796920157558e-05,
|
|
"loss": 4.194,
|
|
"mean_token_accuracy": 0.27937617897987366,
|
|
"num_tokens": 18824388.0,
|
|
"step": 10185
|
|
},
|
|
{
|
|
"entropy": 5.03595871925354,
|
|
"epoch": 8.527196652719665,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 7.81115429721113e-05,
|
|
"loss": 4.1857,
|
|
"mean_token_accuracy": 0.27123529762029647,
|
|
"num_tokens": 18833180.0,
|
|
"step": 10190
|
|
},
|
|
{
|
|
"entropy": 5.030209589004516,
|
|
"epoch": 8.531380753138075,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.795552190743693e-05,
|
|
"loss": 4.2421,
|
|
"mean_token_accuracy": 0.2764876678586006,
|
|
"num_tokens": 18842306.0,
|
|
"step": 10195
|
|
},
|
|
{
|
|
"entropy": 5.007592487335205,
|
|
"epoch": 8.535564853556485,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 7.779990632861869e-05,
|
|
"loss": 4.211,
|
|
"mean_token_accuracy": 0.27322245836257936,
|
|
"num_tokens": 18851843.0,
|
|
"step": 10200
|
|
},
|
|
{
|
|
"entropy": 5.035088968276978,
|
|
"epoch": 8.539748953974895,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 7.764469655588845e-05,
|
|
"loss": 4.277,
|
|
"mean_token_accuracy": 0.27226158529520034,
|
|
"num_tokens": 18860943.0,
|
|
"step": 10205
|
|
},
|
|
{
|
|
"entropy": 4.993625020980835,
|
|
"epoch": 8.543933054393305,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 7.748989290864301e-05,
|
|
"loss": 4.1325,
|
|
"mean_token_accuracy": 0.28864702880382537,
|
|
"num_tokens": 18869925.0,
|
|
"step": 10210
|
|
},
|
|
{
|
|
"entropy": 4.9478741645812985,
|
|
"epoch": 8.548117154811715,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 7.733549570544334e-05,
|
|
"loss": 4.2949,
|
|
"mean_token_accuracy": 0.2777721583843231,
|
|
"num_tokens": 18880022.0,
|
|
"step": 10215
|
|
},
|
|
{
|
|
"entropy": 4.960660076141357,
|
|
"epoch": 8.552301255230125,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 7.718150526401405e-05,
|
|
"loss": 4.1366,
|
|
"mean_token_accuracy": 0.2883001446723938,
|
|
"num_tokens": 18889756.0,
|
|
"step": 10220
|
|
},
|
|
{
|
|
"entropy": 4.990696001052856,
|
|
"epoch": 8.556485355648535,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 7.702792190124268e-05,
|
|
"loss": 4.2146,
|
|
"mean_token_accuracy": 0.27981131374835966,
|
|
"num_tokens": 18899688.0,
|
|
"step": 10225
|
|
},
|
|
{
|
|
"entropy": 4.985267543792725,
|
|
"epoch": 8.560669456066945,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 7.687474593317915e-05,
|
|
"loss": 4.1929,
|
|
"mean_token_accuracy": 0.2868585079908371,
|
|
"num_tokens": 18908862.0,
|
|
"step": 10230
|
|
},
|
|
{
|
|
"entropy": 5.07468547821045,
|
|
"epoch": 8.564853556485355,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 7.6721977675035e-05,
|
|
"loss": 4.2321,
|
|
"mean_token_accuracy": 0.27738107442855836,
|
|
"num_tokens": 18918292.0,
|
|
"step": 10235
|
|
},
|
|
{
|
|
"entropy": 4.925774097442627,
|
|
"epoch": 8.569037656903765,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 7.656961744118266e-05,
|
|
"loss": 4.1127,
|
|
"mean_token_accuracy": 0.2795661598443985,
|
|
"num_tokens": 18927181.0,
|
|
"step": 10240
|
|
},
|
|
{
|
|
"entropy": 4.984150075912476,
|
|
"epoch": 8.573221757322175,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 7.641766554515512e-05,
|
|
"loss": 4.2112,
|
|
"mean_token_accuracy": 0.2843546062707901,
|
|
"num_tokens": 18936259.0,
|
|
"step": 10245
|
|
},
|
|
{
|
|
"entropy": 5.044968795776367,
|
|
"epoch": 8.577405857740585,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 7.626612229964486e-05,
|
|
"loss": 4.2695,
|
|
"mean_token_accuracy": 0.27446300685405733,
|
|
"num_tokens": 18946073.0,
|
|
"step": 10250
|
|
},
|
|
{
|
|
"entropy": 5.014589118957519,
|
|
"epoch": 8.581589958158997,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.611498801650353e-05,
|
|
"loss": 4.1569,
|
|
"mean_token_accuracy": 0.2869542181491852,
|
|
"num_tokens": 18955472.0,
|
|
"step": 10255
|
|
},
|
|
{
|
|
"entropy": 4.981923770904541,
|
|
"epoch": 8.585774058577407,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 7.596426300674113e-05,
|
|
"loss": 4.1986,
|
|
"mean_token_accuracy": 0.2767299622297287,
|
|
"num_tokens": 18965407.0,
|
|
"step": 10260
|
|
},
|
|
{
|
|
"entropy": 5.007739067077637,
|
|
"epoch": 8.589958158995817,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 7.581394758052559e-05,
|
|
"loss": 4.172,
|
|
"mean_token_accuracy": 0.2722089499235153,
|
|
"num_tokens": 18974505.0,
|
|
"step": 10265
|
|
},
|
|
{
|
|
"entropy": 5.077593803405762,
|
|
"epoch": 8.594142259414227,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 7.566404204718188e-05,
|
|
"loss": 4.2637,
|
|
"mean_token_accuracy": 0.27450929284095765,
|
|
"num_tokens": 18983439.0,
|
|
"step": 10270
|
|
},
|
|
{
|
|
"entropy": 4.9693304061889645,
|
|
"epoch": 8.598326359832637,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 7.551454671519142e-05,
|
|
"loss": 4.1928,
|
|
"mean_token_accuracy": 0.2757311508059502,
|
|
"num_tokens": 18992548.0,
|
|
"step": 10275
|
|
},
|
|
{
|
|
"entropy": 4.992247629165649,
|
|
"epoch": 8.602510460251047,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 7.536546189219171e-05,
|
|
"loss": 4.1989,
|
|
"mean_token_accuracy": 0.28285996317863465,
|
|
"num_tokens": 19000846.0,
|
|
"step": 10280
|
|
},
|
|
{
|
|
"entropy": 4.9671117782592775,
|
|
"epoch": 8.606694560669457,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 7.521678788497515e-05,
|
|
"loss": 4.1315,
|
|
"mean_token_accuracy": 0.29310665726661683,
|
|
"num_tokens": 19009387.0,
|
|
"step": 10285
|
|
},
|
|
{
|
|
"entropy": 4.979158401489258,
|
|
"epoch": 8.610878661087867,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 7.506852499948905e-05,
|
|
"loss": 4.2251,
|
|
"mean_token_accuracy": 0.27544511407613753,
|
|
"num_tokens": 19018039.0,
|
|
"step": 10290
|
|
},
|
|
{
|
|
"entropy": 4.998078107833862,
|
|
"epoch": 8.615062761506277,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 7.492067354083462e-05,
|
|
"loss": 4.162,
|
|
"mean_token_accuracy": 0.28275784850120544,
|
|
"num_tokens": 19026769.0,
|
|
"step": 10295
|
|
},
|
|
{
|
|
"entropy": 5.047854280471801,
|
|
"epoch": 8.619246861924687,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 7.47732338132663e-05,
|
|
"loss": 4.2926,
|
|
"mean_token_accuracy": 0.2689140230417252,
|
|
"num_tokens": 19037030.0,
|
|
"step": 10300
|
|
},
|
|
{
|
|
"entropy": 5.107204389572144,
|
|
"epoch": 8.623430962343097,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 7.462620612019146e-05,
|
|
"loss": 4.3095,
|
|
"mean_token_accuracy": 0.2656465440988541,
|
|
"num_tokens": 19046891.0,
|
|
"step": 10305
|
|
},
|
|
{
|
|
"entropy": 5.017383241653443,
|
|
"epoch": 8.627615062761507,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 7.447959076416934e-05,
|
|
"loss": 4.1936,
|
|
"mean_token_accuracy": 0.2744273319840431,
|
|
"num_tokens": 19055863.0,
|
|
"step": 10310
|
|
},
|
|
{
|
|
"entropy": 5.044226169586182,
|
|
"epoch": 8.631799163179917,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 7.433338804691087e-05,
|
|
"loss": 4.2437,
|
|
"mean_token_accuracy": 0.2786155045032501,
|
|
"num_tokens": 19065564.0,
|
|
"step": 10315
|
|
},
|
|
{
|
|
"entropy": 5.109070444107056,
|
|
"epoch": 8.635983263598327,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 7.41875982692777e-05,
|
|
"loss": 4.3169,
|
|
"mean_token_accuracy": 0.26440167874097825,
|
|
"num_tokens": 19075532.0,
|
|
"step": 10320
|
|
},
|
|
{
|
|
"entropy": 4.998457384109497,
|
|
"epoch": 8.640167364016737,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 7.40422217312817e-05,
|
|
"loss": 4.1466,
|
|
"mean_token_accuracy": 0.2848371297121048,
|
|
"num_tokens": 19083915.0,
|
|
"step": 10325
|
|
},
|
|
{
|
|
"entropy": 4.956072235107422,
|
|
"epoch": 8.644351464435147,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 7.38972587320845e-05,
|
|
"loss": 4.1886,
|
|
"mean_token_accuracy": 0.27727857530117034,
|
|
"num_tokens": 19093094.0,
|
|
"step": 10330
|
|
},
|
|
{
|
|
"entropy": 5.072630548477173,
|
|
"epoch": 8.648535564853557,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 7.375270956999652e-05,
|
|
"loss": 4.2494,
|
|
"mean_token_accuracy": 0.2835138916969299,
|
|
"num_tokens": 19103124.0,
|
|
"step": 10335
|
|
},
|
|
{
|
|
"entropy": 4.9755185604095455,
|
|
"epoch": 8.652719665271967,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 7.360857454247686e-05,
|
|
"loss": 4.175,
|
|
"mean_token_accuracy": 0.2828449532389641,
|
|
"num_tokens": 19111307.0,
|
|
"step": 10340
|
|
},
|
|
{
|
|
"entropy": 4.999535703659058,
|
|
"epoch": 8.656903765690377,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 7.34648539461321e-05,
|
|
"loss": 4.2327,
|
|
"mean_token_accuracy": 0.271081006526947,
|
|
"num_tokens": 19120756.0,
|
|
"step": 10345
|
|
},
|
|
{
|
|
"entropy": 5.010644912719727,
|
|
"epoch": 8.661087866108787,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 7.332154807671623e-05,
|
|
"loss": 4.1642,
|
|
"mean_token_accuracy": 0.28579142689704895,
|
|
"num_tokens": 19129374.0,
|
|
"step": 10350
|
|
},
|
|
{
|
|
"entropy": 5.016131496429443,
|
|
"epoch": 8.665271966527197,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 7.317865722912958e-05,
|
|
"loss": 4.2063,
|
|
"mean_token_accuracy": 0.277300500869751,
|
|
"num_tokens": 19138384.0,
|
|
"step": 10355
|
|
},
|
|
{
|
|
"entropy": 4.97669415473938,
|
|
"epoch": 8.669456066945607,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 7.30361816974186e-05,
|
|
"loss": 4.174,
|
|
"mean_token_accuracy": 0.2849442720413208,
|
|
"num_tokens": 19146926.0,
|
|
"step": 10360
|
|
},
|
|
{
|
|
"entropy": 5.067147636413575,
|
|
"epoch": 8.673640167364017,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.289412177477505e-05,
|
|
"loss": 4.2552,
|
|
"mean_token_accuracy": 0.2759119838476181,
|
|
"num_tokens": 19156181.0,
|
|
"step": 10365
|
|
},
|
|
{
|
|
"entropy": 5.058348226547241,
|
|
"epoch": 8.677824267782427,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 7.275247775353531e-05,
|
|
"loss": 4.3017,
|
|
"mean_token_accuracy": 0.2739701122045517,
|
|
"num_tokens": 19165743.0,
|
|
"step": 10370
|
|
},
|
|
{
|
|
"entropy": 5.039249753952026,
|
|
"epoch": 8.682008368200837,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 7.261124992518016e-05,
|
|
"loss": 4.2626,
|
|
"mean_token_accuracy": 0.27092191874980925,
|
|
"num_tokens": 19174498.0,
|
|
"step": 10375
|
|
},
|
|
{
|
|
"entropy": 5.01496148109436,
|
|
"epoch": 8.686192468619247,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 7.247043858033366e-05,
|
|
"loss": 4.2179,
|
|
"mean_token_accuracy": 0.2821288913488388,
|
|
"num_tokens": 19183326.0,
|
|
"step": 10380
|
|
},
|
|
{
|
|
"entropy": 5.034399318695068,
|
|
"epoch": 8.690376569037657,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.233004400876287e-05,
|
|
"loss": 4.1818,
|
|
"mean_token_accuracy": 0.28557226061820984,
|
|
"num_tokens": 19192593.0,
|
|
"step": 10385
|
|
},
|
|
{
|
|
"entropy": 4.9995386600494385,
|
|
"epoch": 8.694560669456067,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 7.21900664993774e-05,
|
|
"loss": 4.2511,
|
|
"mean_token_accuracy": 0.2762255594134331,
|
|
"num_tokens": 19201751.0,
|
|
"step": 10390
|
|
},
|
|
{
|
|
"entropy": 4.963431739807129,
|
|
"epoch": 8.698744769874477,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 7.20505063402283e-05,
|
|
"loss": 4.168,
|
|
"mean_token_accuracy": 0.2841141015291214,
|
|
"num_tokens": 19211293.0,
|
|
"step": 10395
|
|
},
|
|
{
|
|
"entropy": 4.9720453262329105,
|
|
"epoch": 8.702928870292887,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 7.191136381850806e-05,
|
|
"loss": 4.241,
|
|
"mean_token_accuracy": 0.26752633303403855,
|
|
"num_tokens": 19220555.0,
|
|
"step": 10400
|
|
},
|
|
{
|
|
"entropy": 4.9709250926971436,
|
|
"epoch": 8.707112970711297,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 7.177263922054956e-05,
|
|
"loss": 4.2086,
|
|
"mean_token_accuracy": 0.27654631733894347,
|
|
"num_tokens": 19230149.0,
|
|
"step": 10405
|
|
},
|
|
{
|
|
"entropy": 4.957654428482056,
|
|
"epoch": 8.711297071129707,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 7.163433283182571e-05,
|
|
"loss": 4.2401,
|
|
"mean_token_accuracy": 0.2788254275918007,
|
|
"num_tokens": 19238575.0,
|
|
"step": 10410
|
|
},
|
|
{
|
|
"entropy": 4.938343334197998,
|
|
"epoch": 8.715481171548117,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 7.14964449369489e-05,
|
|
"loss": 4.0665,
|
|
"mean_token_accuracy": 0.28815277516841886,
|
|
"num_tokens": 19247540.0,
|
|
"step": 10415
|
|
},
|
|
{
|
|
"entropy": 5.051509809494019,
|
|
"epoch": 8.719665271966527,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 7.135897581967013e-05,
|
|
"loss": 4.2127,
|
|
"mean_token_accuracy": 0.2796056717634201,
|
|
"num_tokens": 19257245.0,
|
|
"step": 10420
|
|
},
|
|
{
|
|
"entropy": 4.9939367294311525,
|
|
"epoch": 8.723849372384937,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 7.122192576287885e-05,
|
|
"loss": 4.2042,
|
|
"mean_token_accuracy": 0.2830051273107529,
|
|
"num_tokens": 19267580.0,
|
|
"step": 10425
|
|
},
|
|
{
|
|
"entropy": 5.013699913024903,
|
|
"epoch": 8.728033472803347,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 7.108529504860195e-05,
|
|
"loss": 4.1961,
|
|
"mean_token_accuracy": 0.2850960254669189,
|
|
"num_tokens": 19276461.0,
|
|
"step": 10430
|
|
},
|
|
{
|
|
"entropy": 5.060430479049683,
|
|
"epoch": 8.732217573221757,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 7.094908395800357e-05,
|
|
"loss": 4.2214,
|
|
"mean_token_accuracy": 0.27945521771907805,
|
|
"num_tokens": 19285184.0,
|
|
"step": 10435
|
|
},
|
|
{
|
|
"entropy": 5.022622060775757,
|
|
"epoch": 8.736401673640167,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 7.081329277138412e-05,
|
|
"loss": 4.2163,
|
|
"mean_token_accuracy": 0.2759193629026413,
|
|
"num_tokens": 19294170.0,
|
|
"step": 10440
|
|
},
|
|
{
|
|
"entropy": 5.061263370513916,
|
|
"epoch": 8.740585774058577,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 7.067792176818011e-05,
|
|
"loss": 4.3116,
|
|
"mean_token_accuracy": 0.26989395916461945,
|
|
"num_tokens": 19303786.0,
|
|
"step": 10445
|
|
},
|
|
{
|
|
"entropy": 4.986194276809693,
|
|
"epoch": 8.744769874476987,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 7.054297122696336e-05,
|
|
"loss": 4.2024,
|
|
"mean_token_accuracy": 0.27779615521430967,
|
|
"num_tokens": 19313759.0,
|
|
"step": 10450
|
|
},
|
|
{
|
|
"entropy": 5.040928173065185,
|
|
"epoch": 8.748953974895397,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 7.040844142544026e-05,
|
|
"loss": 4.3049,
|
|
"mean_token_accuracy": 0.2696812987327576,
|
|
"num_tokens": 19322462.0,
|
|
"step": 10455
|
|
},
|
|
{
|
|
"entropy": 5.060733509063721,
|
|
"epoch": 8.753138075313807,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 7.027433264045156e-05,
|
|
"loss": 4.2418,
|
|
"mean_token_accuracy": 0.27631075382232667,
|
|
"num_tokens": 19330860.0,
|
|
"step": 10460
|
|
},
|
|
{
|
|
"entropy": 5.0259006977081295,
|
|
"epoch": 8.757322175732218,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 7.014064514797157e-05,
|
|
"loss": 4.2811,
|
|
"mean_token_accuracy": 0.2739049166440964,
|
|
"num_tokens": 19340226.0,
|
|
"step": 10465
|
|
},
|
|
{
|
|
"entropy": 4.991638708114624,
|
|
"epoch": 8.761506276150628,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 7.000737922310772e-05,
|
|
"loss": 4.1749,
|
|
"mean_token_accuracy": 0.2842607259750366,
|
|
"num_tokens": 19349238.0,
|
|
"step": 10470
|
|
},
|
|
{
|
|
"entropy": 5.0447385787963865,
|
|
"epoch": 8.765690376569038,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.987453514009977e-05,
|
|
"loss": 4.2658,
|
|
"mean_token_accuracy": 0.2802996799349785,
|
|
"num_tokens": 19358887.0,
|
|
"step": 10475
|
|
},
|
|
{
|
|
"entropy": 5.007854652404785,
|
|
"epoch": 8.769874476987448,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 6.974211317231954e-05,
|
|
"loss": 4.2065,
|
|
"mean_token_accuracy": 0.2757023602724075,
|
|
"num_tokens": 19368800.0,
|
|
"step": 10480
|
|
},
|
|
{
|
|
"entropy": 4.985948085784912,
|
|
"epoch": 8.774058577405858,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.961011359227014e-05,
|
|
"loss": 4.1349,
|
|
"mean_token_accuracy": 0.2844905763864517,
|
|
"num_tokens": 19377162.0,
|
|
"step": 10485
|
|
},
|
|
{
|
|
"entropy": 5.102771425247193,
|
|
"epoch": 8.778242677824268,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 6.947853667158544e-05,
|
|
"loss": 4.2865,
|
|
"mean_token_accuracy": 0.26930278688669207,
|
|
"num_tokens": 19387236.0,
|
|
"step": 10490
|
|
},
|
|
{
|
|
"entropy": 5.017039585113525,
|
|
"epoch": 8.782426778242678,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 6.934738268102963e-05,
|
|
"loss": 4.2559,
|
|
"mean_token_accuracy": 0.2739841639995575,
|
|
"num_tokens": 19396880.0,
|
|
"step": 10495
|
|
},
|
|
{
|
|
"entropy": 5.044538259506226,
|
|
"epoch": 8.786610878661088,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 6.921665189049654e-05,
|
|
"loss": 4.2263,
|
|
"mean_token_accuracy": 0.2748436689376831,
|
|
"num_tokens": 19405796.0,
|
|
"step": 10500
|
|
},
|
|
{
|
|
"epoch": 8.786610878661088,
|
|
"eval_entropy": 4.833877158311247,
|
|
"eval_loss": 5.298664569854736,
|
|
"eval_mean_token_accuracy": 0.20602963112127073,
|
|
"eval_num_tokens": 19405796.0,
|
|
"eval_runtime": 2.082,
|
|
"eval_samples_per_second": 1876.601,
|
|
"eval_steps_per_second": 234.875,
|
|
"step": 10500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 11950,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"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": 4191700508835840.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|