3068 lines
82 KiB
JSON
3068 lines
82 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.2552301255230125,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.742597675323486,
|
|
"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
|
|
}
|
|
],
|
|
"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": 601388123258880.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|