12167 lines
332 KiB
JSON
12167 lines
332 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 6.486749594375338,
|
|
"eval_steps": 500,
|
|
"global_step": 6000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.692022037506103,
|
|
"epoch": 0.005408328826392645,
|
|
"grad_norm": 14.3125,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.8462,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 10942.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.691925811767579,
|
|
"epoch": 0.01081665765278529,
|
|
"grad_norm": 12.5625,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7663,
|
|
"mean_token_accuracy": 0.00011025358689948916,
|
|
"num_tokens": 21887.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.69113998413086,
|
|
"epoch": 0.016224986479177934,
|
|
"grad_norm": 9.9375,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.515,
|
|
"mean_token_accuracy": 0.014385890925768762,
|
|
"num_tokens": 34345.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.684020137786865,
|
|
"epoch": 0.02163331530557058,
|
|
"grad_norm": 6.46875,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.1559,
|
|
"mean_token_accuracy": 0.0367370892316103,
|
|
"num_tokens": 45445.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.650034141540527,
|
|
"epoch": 0.02704164413196322,
|
|
"grad_norm": 3.828125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 9.8592,
|
|
"mean_token_accuracy": 0.039543523266911504,
|
|
"num_tokens": 58237.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.601274299621583,
|
|
"epoch": 0.03244997295835587,
|
|
"grad_norm": 3.34375,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.6444,
|
|
"mean_token_accuracy": 0.04297022223472595,
|
|
"num_tokens": 68098.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.589245128631593,
|
|
"epoch": 0.03785830178474851,
|
|
"grad_norm": 2.75,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.597,
|
|
"mean_token_accuracy": 0.040016197971999644,
|
|
"num_tokens": 79867.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.5987380027771,
|
|
"epoch": 0.04326663061114116,
|
|
"grad_norm": 2.71875,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.5915,
|
|
"mean_token_accuracy": 0.04129943773150444,
|
|
"num_tokens": 91257.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.589694404602051,
|
|
"epoch": 0.048674959437533805,
|
|
"grad_norm": 2.40625,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.5887,
|
|
"mean_token_accuracy": 0.04088671933859587,
|
|
"num_tokens": 103474.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.584661674499511,
|
|
"epoch": 0.05408328826392644,
|
|
"grad_norm": 2.65625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.4849,
|
|
"mean_token_accuracy": 0.039065463840961455,
|
|
"num_tokens": 114830.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.57274866104126,
|
|
"epoch": 0.05949161709031909,
|
|
"grad_norm": 2.375,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.4147,
|
|
"mean_token_accuracy": 0.042727841436862944,
|
|
"num_tokens": 126017.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.554320049285888,
|
|
"epoch": 0.06489994591671173,
|
|
"grad_norm": 2.453125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.3348,
|
|
"mean_token_accuracy": 0.048937696032226086,
|
|
"num_tokens": 138509.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.518039512634278,
|
|
"epoch": 0.07030827474310439,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.2417,
|
|
"mean_token_accuracy": 0.06088445819914341,
|
|
"num_tokens": 148353.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.38691234588623,
|
|
"epoch": 0.07571660356949703,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.2043,
|
|
"mean_token_accuracy": 0.05567694678902626,
|
|
"num_tokens": 160269.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.452953910827636,
|
|
"epoch": 0.08112493239588967,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.0443,
|
|
"mean_token_accuracy": 0.059686141461133956,
|
|
"num_tokens": 171568.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.435920333862304,
|
|
"epoch": 0.08653326122228232,
|
|
"grad_norm": 2.375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 8.9471,
|
|
"mean_token_accuracy": 0.06421410515904427,
|
|
"num_tokens": 184243.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.38515739440918,
|
|
"epoch": 0.09194159004867496,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 8.849,
|
|
"mean_token_accuracy": 0.06825026907026768,
|
|
"num_tokens": 194887.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.27032928466797,
|
|
"epoch": 0.09734991887506761,
|
|
"grad_norm": 2.53125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 8.7532,
|
|
"mean_token_accuracy": 0.07215963676571846,
|
|
"num_tokens": 206476.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.199292469024659,
|
|
"epoch": 0.10275824770146025,
|
|
"grad_norm": 2.578125,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.6005,
|
|
"mean_token_accuracy": 0.07982434406876564,
|
|
"num_tokens": 216578.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.170300388336182,
|
|
"epoch": 0.10816657652785289,
|
|
"grad_norm": 2.078125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.506,
|
|
"mean_token_accuracy": 0.0810505922883749,
|
|
"num_tokens": 226941.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.120428371429444,
|
|
"epoch": 0.11357490535424554,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.4082,
|
|
"mean_token_accuracy": 0.08101015165448189,
|
|
"num_tokens": 238060.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.04278860092163,
|
|
"epoch": 0.11898323418063818,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.2557,
|
|
"mean_token_accuracy": 0.08279627189040184,
|
|
"num_tokens": 249137.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 9.902883434295655,
|
|
"epoch": 0.12439156300703083,
|
|
"grad_norm": 1.8984375,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.1556,
|
|
"mean_token_accuracy": 0.08368096873164177,
|
|
"num_tokens": 260712.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.842740917205811,
|
|
"epoch": 0.12979989183342347,
|
|
"grad_norm": 2.109375,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 7.9778,
|
|
"mean_token_accuracy": 0.08551804721355438,
|
|
"num_tokens": 271039.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.667721843719482,
|
|
"epoch": 0.1352082206598161,
|
|
"grad_norm": 1.8203125,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 7.8296,
|
|
"mean_token_accuracy": 0.088394845277071,
|
|
"num_tokens": 281353.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.477235412597656,
|
|
"epoch": 0.14061654948620878,
|
|
"grad_norm": 1.96875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 7.8005,
|
|
"mean_token_accuracy": 0.08695972561836243,
|
|
"num_tokens": 292221.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.307325172424317,
|
|
"epoch": 0.1460248783126014,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.6751,
|
|
"mean_token_accuracy": 0.08949057161808013,
|
|
"num_tokens": 303353.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.106245803833009,
|
|
"epoch": 0.15143320713899405,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.6111,
|
|
"mean_token_accuracy": 0.09538959562778473,
|
|
"num_tokens": 314165.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 8.896690082550048,
|
|
"epoch": 0.1568415359653867,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.4219,
|
|
"mean_token_accuracy": 0.09479932934045791,
|
|
"num_tokens": 326387.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.698641109466553,
|
|
"epoch": 0.16224986479177933,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.4523,
|
|
"mean_token_accuracy": 0.08837239667773247,
|
|
"num_tokens": 339155.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.527585792541505,
|
|
"epoch": 0.167658193618172,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.3469,
|
|
"mean_token_accuracy": 0.09304244369268418,
|
|
"num_tokens": 350643.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.301559352874756,
|
|
"epoch": 0.17306652244456464,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.257,
|
|
"mean_token_accuracy": 0.09016378521919251,
|
|
"num_tokens": 361977.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.236385345458984,
|
|
"epoch": 0.17847485127095727,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.2952,
|
|
"mean_token_accuracy": 0.09531197249889374,
|
|
"num_tokens": 373604.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.092022514343261,
|
|
"epoch": 0.1838831800973499,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.2732,
|
|
"mean_token_accuracy": 0.09200240671634674,
|
|
"num_tokens": 384306.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 7.969397687911988,
|
|
"epoch": 0.18929150892374255,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.169,
|
|
"mean_token_accuracy": 0.1011082261800766,
|
|
"num_tokens": 396479.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 7.91082444190979,
|
|
"epoch": 0.19469983775013522,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.1033,
|
|
"mean_token_accuracy": 0.10111359134316444,
|
|
"num_tokens": 408489.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 7.861333703994751,
|
|
"epoch": 0.20010816657652786,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.1182,
|
|
"mean_token_accuracy": 0.10262049958109856,
|
|
"num_tokens": 419339.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.818029165267944,
|
|
"epoch": 0.2055164954029205,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.0883,
|
|
"mean_token_accuracy": 0.10314406752586365,
|
|
"num_tokens": 430486.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.71181092262268,
|
|
"epoch": 0.21092482422931313,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.1069,
|
|
"mean_token_accuracy": 0.10316698178648949,
|
|
"num_tokens": 440414.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.695619297027588,
|
|
"epoch": 0.21633315305570577,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.0605,
|
|
"mean_token_accuracy": 0.1009691059589386,
|
|
"num_tokens": 451475.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.618443393707276,
|
|
"epoch": 0.22174148188209844,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.126,
|
|
"mean_token_accuracy": 0.1053870052099228,
|
|
"num_tokens": 462053.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.729849576950073,
|
|
"epoch": 0.22714981070849108,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.0694,
|
|
"mean_token_accuracy": 0.10431502312421799,
|
|
"num_tokens": 473207.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.642491388320923,
|
|
"epoch": 0.23255813953488372,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.0593,
|
|
"mean_token_accuracy": 0.10497235506772995,
|
|
"num_tokens": 484999.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.669218826293945,
|
|
"epoch": 0.23796646836127636,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.0536,
|
|
"mean_token_accuracy": 0.10244816839694977,
|
|
"num_tokens": 496772.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.611874437332153,
|
|
"epoch": 0.24337479718766902,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.0209,
|
|
"mean_token_accuracy": 0.10049235001206398,
|
|
"num_tokens": 507782.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.446110391616822,
|
|
"epoch": 0.24878312601406166,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 6.9319,
|
|
"mean_token_accuracy": 0.10533562302589417,
|
|
"num_tokens": 518950.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.5693611145019535,
|
|
"epoch": 0.2541914548404543,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.0201,
|
|
"mean_token_accuracy": 0.1056381680071354,
|
|
"num_tokens": 530795.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.5523681640625,
|
|
"epoch": 0.25959978366684694,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.027,
|
|
"mean_token_accuracy": 0.10181101113557815,
|
|
"num_tokens": 542878.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.477112722396851,
|
|
"epoch": 0.2650081124932396,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000122,
|
|
"loss": 6.9276,
|
|
"mean_token_accuracy": 0.10972334071993828,
|
|
"num_tokens": 553682.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.378021717071533,
|
|
"epoch": 0.2704164413196322,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 6.9296,
|
|
"mean_token_accuracy": 0.11474497318267822,
|
|
"num_tokens": 564433.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.477735948562622,
|
|
"epoch": 0.27582477014602486,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000127,
|
|
"loss": 6.9645,
|
|
"mean_token_accuracy": 0.10399928912520409,
|
|
"num_tokens": 576001.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.369494915008545,
|
|
"epoch": 0.28123309897241755,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 6.8768,
|
|
"mean_token_accuracy": 0.11646146923303605,
|
|
"num_tokens": 587138.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.286924839019775,
|
|
"epoch": 0.2866414277988102,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000132,
|
|
"loss": 6.851,
|
|
"mean_token_accuracy": 0.1115494042634964,
|
|
"num_tokens": 597572.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.285667753219604,
|
|
"epoch": 0.2920497566252028,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 6.8139,
|
|
"mean_token_accuracy": 0.11849615946412087,
|
|
"num_tokens": 608879.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.397619819641113,
|
|
"epoch": 0.29745808545159547,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 6.8427,
|
|
"mean_token_accuracy": 0.10948013365268708,
|
|
"num_tokens": 620439.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.3241359233856205,
|
|
"epoch": 0.3028664142779881,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 6.9064,
|
|
"mean_token_accuracy": 0.10746880844235421,
|
|
"num_tokens": 632543.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.305241250991822,
|
|
"epoch": 0.30827474310438074,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 6.9523,
|
|
"mean_token_accuracy": 0.11276887506246566,
|
|
"num_tokens": 643369.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.290642261505127,
|
|
"epoch": 0.3136830719307734,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.8175,
|
|
"mean_token_accuracy": 0.11630758419632911,
|
|
"num_tokens": 655328.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.304603433609008,
|
|
"epoch": 0.319091400757166,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000147,
|
|
"loss": 6.8837,
|
|
"mean_token_accuracy": 0.1105578251183033,
|
|
"num_tokens": 665780.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.252665376663208,
|
|
"epoch": 0.32449972958355866,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.7811,
|
|
"mean_token_accuracy": 0.11430338621139527,
|
|
"num_tokens": 676566.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.213497543334961,
|
|
"epoch": 0.3299080584099513,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.7701,
|
|
"mean_token_accuracy": 0.115247243642807,
|
|
"num_tokens": 687407.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.21800651550293,
|
|
"epoch": 0.335316387236344,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.7086,
|
|
"mean_token_accuracy": 0.11728021055459976,
|
|
"num_tokens": 698162.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.1137049198150635,
|
|
"epoch": 0.34072471606273663,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.6813,
|
|
"mean_token_accuracy": 0.11424923315644264,
|
|
"num_tokens": 708743.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.287932205200195,
|
|
"epoch": 0.34613304488912927,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.9224,
|
|
"mean_token_accuracy": 0.11441284045577049,
|
|
"num_tokens": 719828.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.240033435821533,
|
|
"epoch": 0.3515413737155219,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.8683,
|
|
"mean_token_accuracy": 0.11263540089130401,
|
|
"num_tokens": 732514.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.1317566394805905,
|
|
"epoch": 0.35694970254191455,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.6928,
|
|
"mean_token_accuracy": 0.11701238229870796,
|
|
"num_tokens": 742391.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.036797666549683,
|
|
"epoch": 0.3623580313683072,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.6964,
|
|
"mean_token_accuracy": 0.11464602425694466,
|
|
"num_tokens": 753638.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.286917877197266,
|
|
"epoch": 0.3677663601946998,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.9188,
|
|
"mean_token_accuracy": 0.11068255305290223,
|
|
"num_tokens": 764850.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.178950262069702,
|
|
"epoch": 0.37317468902109246,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.6992,
|
|
"mean_token_accuracy": 0.1204465426504612,
|
|
"num_tokens": 776192.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 6.9970947265625,
|
|
"epoch": 0.3785830178474851,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.6463,
|
|
"mean_token_accuracy": 0.12572311908006667,
|
|
"num_tokens": 787051.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.054988002777099,
|
|
"epoch": 0.3839913466738778,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.6818,
|
|
"mean_token_accuracy": 0.1151166208088398,
|
|
"num_tokens": 799281.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.076799201965332,
|
|
"epoch": 0.38939967550027044,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.7153,
|
|
"mean_token_accuracy": 0.11469916179776192,
|
|
"num_tokens": 810003.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 6.988044166564942,
|
|
"epoch": 0.3948080043266631,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.6715,
|
|
"mean_token_accuracy": 0.11684603095054627,
|
|
"num_tokens": 819749.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.086754989624024,
|
|
"epoch": 0.4002163331530557,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.669,
|
|
"mean_token_accuracy": 0.1218308962881565,
|
|
"num_tokens": 830552.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.108803462982178,
|
|
"epoch": 0.40562466197944835,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.7743,
|
|
"mean_token_accuracy": 0.12103541642427444,
|
|
"num_tokens": 840495.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 6.959353446960449,
|
|
"epoch": 0.411032990805841,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.7243,
|
|
"mean_token_accuracy": 0.11957930251955987,
|
|
"num_tokens": 852915.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.139857006072998,
|
|
"epoch": 0.41644131963223363,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.6824,
|
|
"mean_token_accuracy": 0.11836005300283432,
|
|
"num_tokens": 862762.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 6.925649166107178,
|
|
"epoch": 0.42184964845862627,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.6169,
|
|
"mean_token_accuracy": 0.11832203418016433,
|
|
"num_tokens": 873023.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.009290456771851,
|
|
"epoch": 0.4272579772850189,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.7184,
|
|
"mean_token_accuracy": 0.11268399432301521,
|
|
"num_tokens": 885281.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.027568912506103,
|
|
"epoch": 0.43266630611141155,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.7191,
|
|
"mean_token_accuracy": 0.11810961216688157,
|
|
"num_tokens": 896210.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 6.915582704544067,
|
|
"epoch": 0.43807463493780424,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.6367,
|
|
"mean_token_accuracy": 0.12111475020647049,
|
|
"num_tokens": 909028.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 6.9255085468292235,
|
|
"epoch": 0.4434829637641969,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.5633,
|
|
"mean_token_accuracy": 0.12149589061737061,
|
|
"num_tokens": 919710.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.012827968597412,
|
|
"epoch": 0.4488912925905895,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.633,
|
|
"mean_token_accuracy": 0.1167762778699398,
|
|
"num_tokens": 930019.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 6.988396644592285,
|
|
"epoch": 0.45429962141698216,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.5543,
|
|
"mean_token_accuracy": 0.12245730757713318,
|
|
"num_tokens": 940535.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 6.710291481018066,
|
|
"epoch": 0.4597079502433748,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.5134,
|
|
"mean_token_accuracy": 0.12176155820488929,
|
|
"num_tokens": 951390.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 6.93554916381836,
|
|
"epoch": 0.46511627906976744,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.5595,
|
|
"mean_token_accuracy": 0.12662623450160027,
|
|
"num_tokens": 963010.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 6.83835015296936,
|
|
"epoch": 0.4705246078961601,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.5177,
|
|
"mean_token_accuracy": 0.12881121560931205,
|
|
"num_tokens": 974032.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 6.867484664916992,
|
|
"epoch": 0.4759329367225527,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.5658,
|
|
"mean_token_accuracy": 0.12559181153774263,
|
|
"num_tokens": 983954.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 6.875189256668091,
|
|
"epoch": 0.48134126554894535,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.6071,
|
|
"mean_token_accuracy": 0.12600242644548415,
|
|
"num_tokens": 994135.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 6.954481601715088,
|
|
"epoch": 0.48674959437533805,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.5621,
|
|
"mean_token_accuracy": 0.12190107628703117,
|
|
"num_tokens": 1006275.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 6.6844987869262695,
|
|
"epoch": 0.4921579232017307,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.3918,
|
|
"mean_token_accuracy": 0.13092931881546974,
|
|
"num_tokens": 1017767.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 6.859748840332031,
|
|
"epoch": 0.4975662520281233,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.5152,
|
|
"mean_token_accuracy": 0.1274128295481205,
|
|
"num_tokens": 1028286.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 6.7749512672424315,
|
|
"epoch": 0.502974580854516,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.5196,
|
|
"mean_token_accuracy": 0.12368692606687545,
|
|
"num_tokens": 1040555.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 6.847295665740967,
|
|
"epoch": 0.5083829096809086,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.6368,
|
|
"mean_token_accuracy": 0.12430229112505913,
|
|
"num_tokens": 1053668.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 6.735847425460816,
|
|
"epoch": 0.5137912385073012,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.5764,
|
|
"mean_token_accuracy": 0.12110921442508697,
|
|
"num_tokens": 1066042.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 6.780033254623413,
|
|
"epoch": 0.5191995673336939,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.5933,
|
|
"mean_token_accuracy": 0.12213548719882965,
|
|
"num_tokens": 1077843.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 6.815031719207764,
|
|
"epoch": 0.5246078961600865,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.479,
|
|
"mean_token_accuracy": 0.12793946266174316,
|
|
"num_tokens": 1089448.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 6.753958320617675,
|
|
"epoch": 0.5300162249864792,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.5871,
|
|
"mean_token_accuracy": 0.12737660706043244,
|
|
"num_tokens": 1100989.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 6.929710721969604,
|
|
"epoch": 0.5354245538128718,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.6087,
|
|
"mean_token_accuracy": 0.12573689594864845,
|
|
"num_tokens": 1113012.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 6.811409521102905,
|
|
"epoch": 0.5408328826392644,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.5551,
|
|
"mean_token_accuracy": 0.12299831435084344,
|
|
"num_tokens": 1123551.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.5408328826392644,
|
|
"eval_entropy": 6.899274277687073,
|
|
"eval_loss": 6.698337554931641,
|
|
"eval_mean_token_accuracy": 0.12483972003683448,
|
|
"eval_num_tokens": 1123551.0,
|
|
"eval_runtime": 2.6565,
|
|
"eval_samples_per_second": 1323.152,
|
|
"eval_steps_per_second": 165.629,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 6.823844623565674,
|
|
"epoch": 0.5462412114656571,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.4921,
|
|
"mean_token_accuracy": 0.1252983532845974,
|
|
"num_tokens": 1133838.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 6.688885736465454,
|
|
"epoch": 0.5516495402920497,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.524,
|
|
"mean_token_accuracy": 0.12103727906942367,
|
|
"num_tokens": 1146005.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 6.737024593353271,
|
|
"epoch": 0.5570578691184424,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.6172,
|
|
"mean_token_accuracy": 0.12063505500555038,
|
|
"num_tokens": 1157328.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 6.776409530639649,
|
|
"epoch": 0.5624661979448351,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.4227,
|
|
"mean_token_accuracy": 0.13085018917918206,
|
|
"num_tokens": 1168592.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 6.724823951721191,
|
|
"epoch": 0.5678745267712277,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.5439,
|
|
"mean_token_accuracy": 0.12476739659905434,
|
|
"num_tokens": 1178680.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.7444133281707765,
|
|
"epoch": 0.5732828555976204,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.4865,
|
|
"mean_token_accuracy": 0.12942970767617226,
|
|
"num_tokens": 1188825.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 6.775806856155396,
|
|
"epoch": 0.578691184424013,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.5289,
|
|
"mean_token_accuracy": 0.12784937024116516,
|
|
"num_tokens": 1201375.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.79164457321167,
|
|
"epoch": 0.5840995132504057,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.6684,
|
|
"mean_token_accuracy": 0.12346827238798141,
|
|
"num_tokens": 1211915.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.693835020065308,
|
|
"epoch": 0.5895078420767983,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.5662,
|
|
"mean_token_accuracy": 0.13108632192015648,
|
|
"num_tokens": 1223151.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.779698181152344,
|
|
"epoch": 0.5949161709031909,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.4848,
|
|
"mean_token_accuracy": 0.1311143383383751,
|
|
"num_tokens": 1234229.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.77201509475708,
|
|
"epoch": 0.6003244997295836,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.4983,
|
|
"mean_token_accuracy": 0.12925008907914162,
|
|
"num_tokens": 1245298.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.639382028579712,
|
|
"epoch": 0.6057328285559762,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.4937,
|
|
"mean_token_accuracy": 0.12552526965737343,
|
|
"num_tokens": 1256154.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.685818862915039,
|
|
"epoch": 0.6111411573823688,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.4008,
|
|
"mean_token_accuracy": 0.13165971264243126,
|
|
"num_tokens": 1267404.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.706040239334106,
|
|
"epoch": 0.6165494862087615,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.4547,
|
|
"mean_token_accuracy": 0.12809761390089988,
|
|
"num_tokens": 1278894.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.580960464477539,
|
|
"epoch": 0.6219578150351541,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.4485,
|
|
"mean_token_accuracy": 0.12942247837781906,
|
|
"num_tokens": 1290408.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.720835113525391,
|
|
"epoch": 0.6273661438615468,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.5157,
|
|
"mean_token_accuracy": 0.1271141864359379,
|
|
"num_tokens": 1301508.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.751352787017822,
|
|
"epoch": 0.6327744726879394,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.4879,
|
|
"mean_token_accuracy": 0.1288011610507965,
|
|
"num_tokens": 1312551.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.7323637962341305,
|
|
"epoch": 0.638182801514332,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.5873,
|
|
"mean_token_accuracy": 0.1275964416563511,
|
|
"num_tokens": 1324876.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.695128726959228,
|
|
"epoch": 0.6435911303407247,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.4719,
|
|
"mean_token_accuracy": 0.12831967324018478,
|
|
"num_tokens": 1335401.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.631118583679199,
|
|
"epoch": 0.6489994591671173,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.4032,
|
|
"mean_token_accuracy": 0.12815723195672035,
|
|
"num_tokens": 1348265.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.5179417610168455,
|
|
"epoch": 0.65440778799351,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.3546,
|
|
"mean_token_accuracy": 0.1407169371843338,
|
|
"num_tokens": 1359078.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.48005108833313,
|
|
"epoch": 0.6598161168199026,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.2681,
|
|
"mean_token_accuracy": 0.1330178216099739,
|
|
"num_tokens": 1369499.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.575146198272705,
|
|
"epoch": 0.6652244456462953,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.3726,
|
|
"mean_token_accuracy": 0.1281535267829895,
|
|
"num_tokens": 1380329.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.63279767036438,
|
|
"epoch": 0.670632774472688,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.4672,
|
|
"mean_token_accuracy": 0.12824212312698363,
|
|
"num_tokens": 1393047.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.590985298156738,
|
|
"epoch": 0.6760411032990806,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.4586,
|
|
"mean_token_accuracy": 0.12962278202176095,
|
|
"num_tokens": 1405600.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.63199200630188,
|
|
"epoch": 0.6814494321254733,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.3874,
|
|
"mean_token_accuracy": 0.13001631796360016,
|
|
"num_tokens": 1416670.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.5218299388885494,
|
|
"epoch": 0.6868577609518659,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.4098,
|
|
"mean_token_accuracy": 0.13109819069504738,
|
|
"num_tokens": 1428929.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.6227617263793945,
|
|
"epoch": 0.6922660897782585,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.329,
|
|
"mean_token_accuracy": 0.1342690460383892,
|
|
"num_tokens": 1439298.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.598624324798584,
|
|
"epoch": 0.6976744186046512,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.4277,
|
|
"mean_token_accuracy": 0.13435010835528374,
|
|
"num_tokens": 1449853.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.6222474575042725,
|
|
"epoch": 0.7030827474310438,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.5087,
|
|
"mean_token_accuracy": 0.1277529202401638,
|
|
"num_tokens": 1460816.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.46167483329773,
|
|
"epoch": 0.7084910762574365,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.3511,
|
|
"mean_token_accuracy": 0.1322698675096035,
|
|
"num_tokens": 1473003.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.622555923461914,
|
|
"epoch": 0.7138994050838291,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.4762,
|
|
"mean_token_accuracy": 0.1253452330827713,
|
|
"num_tokens": 1484745.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.60373821258545,
|
|
"epoch": 0.7193077339102217,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.3472,
|
|
"mean_token_accuracy": 0.13652867451310158,
|
|
"num_tokens": 1495803.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.512233686447144,
|
|
"epoch": 0.7247160627366144,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.3497,
|
|
"mean_token_accuracy": 0.14011847078800202,
|
|
"num_tokens": 1507504.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.610670757293701,
|
|
"epoch": 0.730124391563007,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.4504,
|
|
"mean_token_accuracy": 0.12672319188714026,
|
|
"num_tokens": 1518765.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.40890736579895,
|
|
"epoch": 0.7355327203893997,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.3244,
|
|
"mean_token_accuracy": 0.13610832542181014,
|
|
"num_tokens": 1529481.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.598431587219238,
|
|
"epoch": 0.7409410492157923,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.3494,
|
|
"mean_token_accuracy": 0.13574561551213266,
|
|
"num_tokens": 1539612.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.5662445545196535,
|
|
"epoch": 0.7463493780421849,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.5187,
|
|
"mean_token_accuracy": 0.1300526537001133,
|
|
"num_tokens": 1551370.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.570234394073486,
|
|
"epoch": 0.7517577068685776,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.3954,
|
|
"mean_token_accuracy": 0.14097690880298613,
|
|
"num_tokens": 1561662.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.546575880050659,
|
|
"epoch": 0.7571660356949702,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.3739,
|
|
"mean_token_accuracy": 0.1312471553683281,
|
|
"num_tokens": 1573479.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.5348869323730465,
|
|
"epoch": 0.7625743645213628,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.3272,
|
|
"mean_token_accuracy": 0.13626974672079087,
|
|
"num_tokens": 1583684.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.479220628738403,
|
|
"epoch": 0.7679826933477556,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.3699,
|
|
"mean_token_accuracy": 0.13402081429958343,
|
|
"num_tokens": 1595694.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.457759761810303,
|
|
"epoch": 0.7733910221741482,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.3701,
|
|
"mean_token_accuracy": 0.13271448239684105,
|
|
"num_tokens": 1606104.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.550243330001831,
|
|
"epoch": 0.7787993510005409,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.4455,
|
|
"mean_token_accuracy": 0.13614369183778763,
|
|
"num_tokens": 1617368.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.569984006881714,
|
|
"epoch": 0.7842076798269335,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.4495,
|
|
"mean_token_accuracy": 0.12869683876633645,
|
|
"num_tokens": 1630485.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.535257720947266,
|
|
"epoch": 0.7896160086533262,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.2776,
|
|
"mean_token_accuracy": 0.1384786419570446,
|
|
"num_tokens": 1641411.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.42301983833313,
|
|
"epoch": 0.7950243374797188,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.29,
|
|
"mean_token_accuracy": 0.138799499720335,
|
|
"num_tokens": 1653737.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.551464939117432,
|
|
"epoch": 0.8004326663061114,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.4953,
|
|
"mean_token_accuracy": 0.12869839891791343,
|
|
"num_tokens": 1665910.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.504175233840942,
|
|
"epoch": 0.8058409951325041,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.2587,
|
|
"mean_token_accuracy": 0.13608988225460053,
|
|
"num_tokens": 1676927.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.426490259170532,
|
|
"epoch": 0.8112493239588967,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.3135,
|
|
"mean_token_accuracy": 0.13178193494677543,
|
|
"num_tokens": 1688167.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.408829879760742,
|
|
"epoch": 0.8166576527852893,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.3907,
|
|
"mean_token_accuracy": 0.13767326027154922,
|
|
"num_tokens": 1699899.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.4508825778961185,
|
|
"epoch": 0.822065981611682,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.3248,
|
|
"mean_token_accuracy": 0.13979970589280127,
|
|
"num_tokens": 1709862.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.433129453659058,
|
|
"epoch": 0.8274743104380746,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.3001,
|
|
"mean_token_accuracy": 0.13528692349791527,
|
|
"num_tokens": 1719902.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.432600688934326,
|
|
"epoch": 0.8328826392644673,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.3136,
|
|
"mean_token_accuracy": 0.13330746367573737,
|
|
"num_tokens": 1730351.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.397243452072144,
|
|
"epoch": 0.8382909680908599,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.2879,
|
|
"mean_token_accuracy": 0.13299915418028832,
|
|
"num_tokens": 1741783.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.329720830917358,
|
|
"epoch": 0.8436992969172525,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.3301,
|
|
"mean_token_accuracy": 0.13444845601916314,
|
|
"num_tokens": 1752952.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.591363334655762,
|
|
"epoch": 0.8491076257436452,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.3485,
|
|
"mean_token_accuracy": 0.13427395597100258,
|
|
"num_tokens": 1764014.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.388641452789306,
|
|
"epoch": 0.8545159545700378,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.3036,
|
|
"mean_token_accuracy": 0.1394558310508728,
|
|
"num_tokens": 1776154.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.477089548110962,
|
|
"epoch": 0.8599242833964305,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.3705,
|
|
"mean_token_accuracy": 0.13453256264328955,
|
|
"num_tokens": 1787292.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.41263780593872,
|
|
"epoch": 0.8653326122228231,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.3337,
|
|
"mean_token_accuracy": 0.13896974176168442,
|
|
"num_tokens": 1799469.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.326815843582153,
|
|
"epoch": 0.8707409410492158,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.2143,
|
|
"mean_token_accuracy": 0.14395588785409927,
|
|
"num_tokens": 1810091.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.444188117980957,
|
|
"epoch": 0.8761492698756085,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.3674,
|
|
"mean_token_accuracy": 0.13579124063253403,
|
|
"num_tokens": 1820835.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.424579000473022,
|
|
"epoch": 0.8815575987020011,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.3276,
|
|
"mean_token_accuracy": 0.1343837335705757,
|
|
"num_tokens": 1831868.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.34673490524292,
|
|
"epoch": 0.8869659275283938,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.2949,
|
|
"mean_token_accuracy": 0.13721646144986152,
|
|
"num_tokens": 1842040.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.5023425102233885,
|
|
"epoch": 0.8923742563547864,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.3664,
|
|
"mean_token_accuracy": 0.13853828459978104,
|
|
"num_tokens": 1853765.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.460512018203735,
|
|
"epoch": 0.897782585181179,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.3883,
|
|
"mean_token_accuracy": 0.1366254411637783,
|
|
"num_tokens": 1866081.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.409449529647827,
|
|
"epoch": 0.9031909140075717,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.2842,
|
|
"mean_token_accuracy": 0.13905733972787857,
|
|
"num_tokens": 1876754.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.462026214599609,
|
|
"epoch": 0.9085992428339643,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.3939,
|
|
"mean_token_accuracy": 0.13471903949975966,
|
|
"num_tokens": 1888299.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.240159559249878,
|
|
"epoch": 0.914007571660357,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.1764,
|
|
"mean_token_accuracy": 0.14900295436382294,
|
|
"num_tokens": 1899027.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.445553398132324,
|
|
"epoch": 0.9194159004867496,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.3012,
|
|
"mean_token_accuracy": 0.13794053345918655,
|
|
"num_tokens": 1911043.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.317274713516236,
|
|
"epoch": 0.9248242293131422,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.2364,
|
|
"mean_token_accuracy": 0.13861771374940873,
|
|
"num_tokens": 1921295.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.43291745185852,
|
|
"epoch": 0.9302325581395349,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.331,
|
|
"mean_token_accuracy": 0.1411274828016758,
|
|
"num_tokens": 1932944.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.272318315505982,
|
|
"epoch": 0.9356408869659275,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.2742,
|
|
"mean_token_accuracy": 0.13169358149170876,
|
|
"num_tokens": 1944072.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.327965927124024,
|
|
"epoch": 0.9410492157923201,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.2785,
|
|
"mean_token_accuracy": 0.14310004264116288,
|
|
"num_tokens": 1954798.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.351740598678589,
|
|
"epoch": 0.9464575446187128,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.2393,
|
|
"mean_token_accuracy": 0.14451056867837905,
|
|
"num_tokens": 1965304.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.361933422088623,
|
|
"epoch": 0.9518658734451054,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.2942,
|
|
"mean_token_accuracy": 0.13480084165930747,
|
|
"num_tokens": 1976797.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.353989362716675,
|
|
"epoch": 0.9572742022714981,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.3118,
|
|
"mean_token_accuracy": 0.13902182057499884,
|
|
"num_tokens": 1987448.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.1734973907470705,
|
|
"epoch": 0.9626825310978907,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.1444,
|
|
"mean_token_accuracy": 0.13558762967586518,
|
|
"num_tokens": 2000948.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.344043302536011,
|
|
"epoch": 0.9680908599242833,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.2367,
|
|
"mean_token_accuracy": 0.14082473739981652,
|
|
"num_tokens": 2011807.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.353227615356445,
|
|
"epoch": 0.9734991887506761,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.3003,
|
|
"mean_token_accuracy": 0.134646288305521,
|
|
"num_tokens": 2022835.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.265669250488282,
|
|
"epoch": 0.9789075175770687,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.2119,
|
|
"mean_token_accuracy": 0.14002675116062163,
|
|
"num_tokens": 2034331.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.342136430740356,
|
|
"epoch": 0.9843158464034614,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.2217,
|
|
"mean_token_accuracy": 0.13417831286787987,
|
|
"num_tokens": 2047002.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.174915409088134,
|
|
"epoch": 0.989724175229854,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.1613,
|
|
"mean_token_accuracy": 0.1420857109129429,
|
|
"num_tokens": 2056299.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.296426773071289,
|
|
"epoch": 0.9951325040562466,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.2942,
|
|
"mean_token_accuracy": 0.1392147921025753,
|
|
"num_tokens": 2068031.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.3065918816460504,
|
|
"epoch": 1.0,
|
|
"grad_norm": 1.8984375,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.2526,
|
|
"mean_token_accuracy": 0.14379818571938408,
|
|
"num_tokens": 2076889.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.321426105499268,
|
|
"epoch": 1.0054083288263926,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.0714,
|
|
"mean_token_accuracy": 0.14781692028045654,
|
|
"num_tokens": 2087316.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.282159948348999,
|
|
"epoch": 1.0108166576527853,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.0678,
|
|
"mean_token_accuracy": 0.1447027936577797,
|
|
"num_tokens": 2098522.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.112043714523315,
|
|
"epoch": 1.016224986479178,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.0422,
|
|
"mean_token_accuracy": 0.15174026042222977,
|
|
"num_tokens": 2110015.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.273185062408447,
|
|
"epoch": 1.0216333153055706,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.1459,
|
|
"mean_token_accuracy": 0.1432407170534134,
|
|
"num_tokens": 2121333.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.2446582317352295,
|
|
"epoch": 1.0270416441319632,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.0343,
|
|
"mean_token_accuracy": 0.15241918861865997,
|
|
"num_tokens": 2131385.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.232630681991577,
|
|
"epoch": 1.0324499729583558,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.0804,
|
|
"mean_token_accuracy": 0.151257885992527,
|
|
"num_tokens": 2143274.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.13126950263977,
|
|
"epoch": 1.0378583017847485,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 5.9638,
|
|
"mean_token_accuracy": 0.1478826306760311,
|
|
"num_tokens": 2154529.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.211151552200318,
|
|
"epoch": 1.043266630611141,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.1174,
|
|
"mean_token_accuracy": 0.14830191433429718,
|
|
"num_tokens": 2165949.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.175490522384644,
|
|
"epoch": 1.0486749594375337,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.099,
|
|
"mean_token_accuracy": 0.14568565934896469,
|
|
"num_tokens": 2176710.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.198600339889526,
|
|
"epoch": 1.0540832882639264,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.0723,
|
|
"mean_token_accuracy": 0.1552335165441036,
|
|
"num_tokens": 2189470.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.1337415218353275,
|
|
"epoch": 1.059491617090319,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.0499,
|
|
"mean_token_accuracy": 0.154169762134552,
|
|
"num_tokens": 2200266.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.177909231185913,
|
|
"epoch": 1.0648999459167117,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.0519,
|
|
"mean_token_accuracy": 0.1501970261335373,
|
|
"num_tokens": 2210736.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.258826351165771,
|
|
"epoch": 1.0703082747431043,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.1764,
|
|
"mean_token_accuracy": 0.14401286840438843,
|
|
"num_tokens": 2221057.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.165603256225586,
|
|
"epoch": 1.075716603569497,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.1478,
|
|
"mean_token_accuracy": 0.1444354996085167,
|
|
"num_tokens": 2232352.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.274627923965454,
|
|
"epoch": 1.0811249323958896,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.0463,
|
|
"mean_token_accuracy": 0.15036138147115707,
|
|
"num_tokens": 2244655.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 1.0811249323958896,
|
|
"eval_entropy": 5.910577596317638,
|
|
"eval_loss": 6.386075496673584,
|
|
"eval_mean_token_accuracy": 0.1436767266216603,
|
|
"eval_num_tokens": 2244655.0,
|
|
"eval_runtime": 2.7446,
|
|
"eval_samples_per_second": 1280.708,
|
|
"eval_steps_per_second": 160.316,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.0580830574035645,
|
|
"epoch": 1.0865332612222822,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999997383518051,
|
|
"loss": 6.0272,
|
|
"mean_token_accuracy": 0.15371852964162827,
|
|
"num_tokens": 2255699.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.108634090423584,
|
|
"epoch": 1.0919415900486749,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999986754070562,
|
|
"loss": 6.0019,
|
|
"mean_token_accuracy": 0.15509194284677505,
|
|
"num_tokens": 2266627.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.244011926651001,
|
|
"epoch": 1.0973499188750675,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004999967948166009,
|
|
"loss": 6.1936,
|
|
"mean_token_accuracy": 0.1431817442178726,
|
|
"num_tokens": 2277814.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.277483749389648,
|
|
"epoch": 1.1027582477014604,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999940965872734,
|
|
"loss": 6.1407,
|
|
"mean_token_accuracy": 0.14654194787144662,
|
|
"num_tokens": 2287344.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 5.961275291442871,
|
|
"epoch": 1.1081665765278528,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999905807288788,
|
|
"loss": 5.9114,
|
|
"mean_token_accuracy": 0.15516552031040193,
|
|
"num_tokens": 2298332.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.2057905197143555,
|
|
"epoch": 1.1135749053542456,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000499986247254194,
|
|
"loss": 6.0525,
|
|
"mean_token_accuracy": 0.14903903752565384,
|
|
"num_tokens": 2309161.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.074467325210572,
|
|
"epoch": 1.1189832341806383,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004999810961789667,
|
|
"loss": 6.0574,
|
|
"mean_token_accuracy": 0.15322347730398178,
|
|
"num_tokens": 2319516.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.072076511383057,
|
|
"epoch": 1.124391563007031,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000499975127521916,
|
|
"loss": 5.9434,
|
|
"mean_token_accuracy": 0.154616841673851,
|
|
"num_tokens": 2330441.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.061796236038208,
|
|
"epoch": 1.1297998918334236,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999683413047319,
|
|
"loss": 5.9652,
|
|
"mean_token_accuracy": 0.1588409036397934,
|
|
"num_tokens": 2340524.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.082087898254395,
|
|
"epoch": 1.1352082206598162,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999607375520759,
|
|
"loss": 5.9699,
|
|
"mean_token_accuracy": 0.14885500371456145,
|
|
"num_tokens": 2353379.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.118402004241943,
|
|
"epoch": 1.1406165494862088,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999523162915794,
|
|
"loss": 6.1425,
|
|
"mean_token_accuracy": 0.14582799524068832,
|
|
"num_tokens": 2365708.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.1052204132080075,
|
|
"epoch": 1.1460248783126015,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999430775538459,
|
|
"loss": 6.0446,
|
|
"mean_token_accuracy": 0.15026993080973625,
|
|
"num_tokens": 2375630.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.074679136276245,
|
|
"epoch": 1.151433207138994,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999330213724487,
|
|
"loss": 6.0351,
|
|
"mean_token_accuracy": 0.14886414706707002,
|
|
"num_tokens": 2386885.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.052271366119385,
|
|
"epoch": 1.1568415359653867,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999221477839319,
|
|
"loss": 6.0917,
|
|
"mean_token_accuracy": 0.14651266634464263,
|
|
"num_tokens": 2397676.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.211881923675537,
|
|
"epoch": 1.1622498647917794,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999104568278104,
|
|
"loss": 6.0803,
|
|
"mean_token_accuracy": 0.1440642461180687,
|
|
"num_tokens": 2409039.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.117143774032593,
|
|
"epoch": 1.167658193618172,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000499897948546569,
|
|
"loss": 6.1282,
|
|
"mean_token_accuracy": 0.14596121460199357,
|
|
"num_tokens": 2420103.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.053849458694458,
|
|
"epoch": 1.1730665224445647,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004998846229856629,
|
|
"loss": 6.0394,
|
|
"mean_token_accuracy": 0.14909014999866485,
|
|
"num_tokens": 2430792.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.164281988143921,
|
|
"epoch": 1.1784748512709573,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004998704801935171,
|
|
"loss": 6.0418,
|
|
"mean_token_accuracy": 0.14994784593582153,
|
|
"num_tokens": 2442165.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.0798987865448,
|
|
"epoch": 1.18388318009735,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004998555202215267,
|
|
"loss": 5.9681,
|
|
"mean_token_accuracy": 0.15097014009952545,
|
|
"num_tokens": 2454231.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.067699670791626,
|
|
"epoch": 1.1892915089237426,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004998397431240561,
|
|
"loss": 5.9975,
|
|
"mean_token_accuracy": 0.15744193494319916,
|
|
"num_tokens": 2464828.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.031586503982544,
|
|
"epoch": 1.1946998377501352,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998231489584395,
|
|
"loss": 6.0548,
|
|
"mean_token_accuracy": 0.14410801753401756,
|
|
"num_tokens": 2476144.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.275937557220459,
|
|
"epoch": 1.2001081665765279,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00049980573778498,
|
|
"loss": 6.2416,
|
|
"mean_token_accuracy": 0.1378739967942238,
|
|
"num_tokens": 2489116.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.189827299118042,
|
|
"epoch": 1.2055164954029205,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004997875096669498,
|
|
"loss": 6.0962,
|
|
"mean_token_accuracy": 0.14992275089025497,
|
|
"num_tokens": 2501609.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 5.979094934463501,
|
|
"epoch": 1.2109248242293131,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004997684646705901,
|
|
"loss": 5.9993,
|
|
"mean_token_accuracy": 0.15765715390443802,
|
|
"num_tokens": 2513338.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.162704706192017,
|
|
"epoch": 1.2163331530557058,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004997486028651103,
|
|
"loss": 6.0171,
|
|
"mean_token_accuracy": 0.15734308660030366,
|
|
"num_tokens": 2523199.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.09754319190979,
|
|
"epoch": 1.2217414818820984,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004997279243226884,
|
|
"loss": 6.0287,
|
|
"mean_token_accuracy": 0.15169059485197067,
|
|
"num_tokens": 2535136.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 5.998078441619873,
|
|
"epoch": 1.227149810708491,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004997064291184701,
|
|
"loss": 5.9507,
|
|
"mean_token_accuracy": 0.14467080906033516,
|
|
"num_tokens": 2546428.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.215820741653443,
|
|
"epoch": 1.2325581395348837,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004996841173305691,
|
|
"loss": 6.1441,
|
|
"mean_token_accuracy": 0.1520974189043045,
|
|
"num_tokens": 2558304.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 5.9105936050415036,
|
|
"epoch": 1.2379664683612763,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004996609890400665,
|
|
"loss": 5.8295,
|
|
"mean_token_accuracy": 0.1567443788051605,
|
|
"num_tokens": 2568717.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.065080881118774,
|
|
"epoch": 1.243374797187669,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004996370443310107,
|
|
"loss": 6.0386,
|
|
"mean_token_accuracy": 0.14542743489146231,
|
|
"num_tokens": 2578748.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.152681446075439,
|
|
"epoch": 1.2487831260140616,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004996122832904165,
|
|
"loss": 5.9983,
|
|
"mean_token_accuracy": 0.14863890111446382,
|
|
"num_tokens": 2590095.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.049049043655396,
|
|
"epoch": 1.2541914548404542,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004995867060082658,
|
|
"loss": 6.0271,
|
|
"mean_token_accuracy": 0.15179531648755074,
|
|
"num_tokens": 2600929.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.060418128967285,
|
|
"epoch": 1.2595997836668469,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004995603125775065,
|
|
"loss": 6.0363,
|
|
"mean_token_accuracy": 0.15505950003862382,
|
|
"num_tokens": 2611830.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.014823246002197,
|
|
"epoch": 1.2650081124932395,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004995331030940523,
|
|
"loss": 5.9914,
|
|
"mean_token_accuracy": 0.14843671172857284,
|
|
"num_tokens": 2623177.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 5.988464593887329,
|
|
"epoch": 1.2704164413196322,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004995050776567826,
|
|
"loss": 5.992,
|
|
"mean_token_accuracy": 0.14856899976730348,
|
|
"num_tokens": 2635563.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.096542739868164,
|
|
"epoch": 1.2758247701460248,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004994762363675421,
|
|
"loss": 5.9693,
|
|
"mean_token_accuracy": 0.1583924800157547,
|
|
"num_tokens": 2647157.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.007678365707397,
|
|
"epoch": 1.2812330989724177,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004994465793311396,
|
|
"loss": 5.9588,
|
|
"mean_token_accuracy": 0.15628022700548172,
|
|
"num_tokens": 2658175.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.013985347747803,
|
|
"epoch": 1.28664142779881,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004994161066553493,
|
|
"loss": 5.9076,
|
|
"mean_token_accuracy": 0.15659919679164885,
|
|
"num_tokens": 2669458.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.058844709396363,
|
|
"epoch": 1.292049756625203,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004993848184509086,
|
|
"loss": 6.117,
|
|
"mean_token_accuracy": 0.15126057639718055,
|
|
"num_tokens": 2679727.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.027663373947144,
|
|
"epoch": 1.2974580854515954,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004993527148315193,
|
|
"loss": 6.0759,
|
|
"mean_token_accuracy": 0.1514742687344551,
|
|
"num_tokens": 2691031.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.038566780090332,
|
|
"epoch": 1.3028664142779882,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004993197959138458,
|
|
"loss": 5.8696,
|
|
"mean_token_accuracy": 0.1627727434039116,
|
|
"num_tokens": 2701699.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 5.943235111236572,
|
|
"epoch": 1.3082747431043806,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004992860618175156,
|
|
"loss": 5.9176,
|
|
"mean_token_accuracy": 0.15399425029754638,
|
|
"num_tokens": 2713038.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 5.947955894470215,
|
|
"epoch": 1.3136830719307735,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004992515126651184,
|
|
"loss": 5.913,
|
|
"mean_token_accuracy": 0.15507251471281053,
|
|
"num_tokens": 2723685.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.03350977897644,
|
|
"epoch": 1.319091400757166,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004992161485822059,
|
|
"loss": 5.9839,
|
|
"mean_token_accuracy": 0.15442734211683273,
|
|
"num_tokens": 2735341.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.001535844802857,
|
|
"epoch": 1.3244997295835588,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004991799696972913,
|
|
"loss": 5.9231,
|
|
"mean_token_accuracy": 0.16096268445253373,
|
|
"num_tokens": 2746052.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 5.954626846313476,
|
|
"epoch": 1.3299080584099512,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004991429761418488,
|
|
"loss": 5.9102,
|
|
"mean_token_accuracy": 0.15518528670072557,
|
|
"num_tokens": 2757155.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.043118524551391,
|
|
"epoch": 1.335316387236344,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000499105168050313,
|
|
"loss": 5.9878,
|
|
"mean_token_accuracy": 0.148323954641819,
|
|
"num_tokens": 2769631.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.098846435546875,
|
|
"epoch": 1.3407247160627367,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004990665455600787,
|
|
"loss": 6.0821,
|
|
"mean_token_accuracy": 0.15610918402671814,
|
|
"num_tokens": 2780797.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 5.944508600234985,
|
|
"epoch": 1.3461330448891293,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004990271088114999,
|
|
"loss": 5.918,
|
|
"mean_token_accuracy": 0.14913789480924605,
|
|
"num_tokens": 2792721.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.023497819900513,
|
|
"epoch": 1.351541373715522,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004989868579478901,
|
|
"loss": 5.9848,
|
|
"mean_token_accuracy": 0.16433796435594558,
|
|
"num_tokens": 2804531.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.046576023101807,
|
|
"epoch": 1.3569497025419146,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004989457931155209,
|
|
"loss": 5.9807,
|
|
"mean_token_accuracy": 0.15308960080146788,
|
|
"num_tokens": 2816666.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.040149736404419,
|
|
"epoch": 1.3623580313683072,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004989039144636219,
|
|
"loss": 6.0016,
|
|
"mean_token_accuracy": 0.15865511894226075,
|
|
"num_tokens": 2827977.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 5.967988777160644,
|
|
"epoch": 1.3677663601946999,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004988612221443804,
|
|
"loss": 5.9876,
|
|
"mean_token_accuracy": 0.1503506377339363,
|
|
"num_tokens": 2839069.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.107842016220093,
|
|
"epoch": 1.3731746890210925,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004988177163129403,
|
|
"loss": 6.0689,
|
|
"mean_token_accuracy": 0.14679001718759538,
|
|
"num_tokens": 2850632.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 5.923655796051025,
|
|
"epoch": 1.3785830178474852,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004987733971274018,
|
|
"loss": 5.9388,
|
|
"mean_token_accuracy": 0.15129240453243256,
|
|
"num_tokens": 2862204.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 5.967949485778808,
|
|
"epoch": 1.3839913466738778,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004987282647488211,
|
|
"loss": 5.8995,
|
|
"mean_token_accuracy": 0.16236116886138915,
|
|
"num_tokens": 2874010.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 5.993822622299194,
|
|
"epoch": 1.3893996755002704,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004986823193412093,
|
|
"loss": 5.9668,
|
|
"mean_token_accuracy": 0.15785306841135024,
|
|
"num_tokens": 2884315.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.162058401107788,
|
|
"epoch": 1.394808004326663,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000498635561071532,
|
|
"loss": 6.1324,
|
|
"mean_token_accuracy": 0.15160465240478516,
|
|
"num_tokens": 2896327.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 5.904622602462768,
|
|
"epoch": 1.4002163331530557,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000498587990109709,
|
|
"loss": 5.889,
|
|
"mean_token_accuracy": 0.15589644089341165,
|
|
"num_tokens": 2907007.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 5.991954374313354,
|
|
"epoch": 1.4056246619794484,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004985396066286133,
|
|
"loss": 5.948,
|
|
"mean_token_accuracy": 0.15928597152233123,
|
|
"num_tokens": 2918852.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 5.967680931091309,
|
|
"epoch": 1.411032990805841,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004984904108040706,
|
|
"loss": 5.8541,
|
|
"mean_token_accuracy": 0.15678130388259887,
|
|
"num_tokens": 2929948.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 5.868888568878174,
|
|
"epoch": 1.4164413196322336,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004984404028148585,
|
|
"loss": 5.9154,
|
|
"mean_token_accuracy": 0.1580822542309761,
|
|
"num_tokens": 2941299.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 5.91631932258606,
|
|
"epoch": 1.4218496484586263,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004983895828427062,
|
|
"loss": 5.8466,
|
|
"mean_token_accuracy": 0.15591808408498764,
|
|
"num_tokens": 2952283.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.187982988357544,
|
|
"epoch": 1.427257977285019,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004983379510722935,
|
|
"loss": 6.214,
|
|
"mean_token_accuracy": 0.14518574103713036,
|
|
"num_tokens": 2962241.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 5.965568780899048,
|
|
"epoch": 1.4326663061114115,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004982855076912506,
|
|
"loss": 5.9814,
|
|
"mean_token_accuracy": 0.15393277555704116,
|
|
"num_tokens": 2973408.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.023658561706543,
|
|
"epoch": 1.4380746349378042,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004982322528901568,
|
|
"loss": 5.9944,
|
|
"mean_token_accuracy": 0.15809524282813073,
|
|
"num_tokens": 2985877.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.089307641983032,
|
|
"epoch": 1.4434829637641968,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00049817818686254,
|
|
"loss": 5.9956,
|
|
"mean_token_accuracy": 0.15151516497135162,
|
|
"num_tokens": 2996935.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 5.8956176280975345,
|
|
"epoch": 1.4488912925905895,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004981233098048765,
|
|
"loss": 5.8806,
|
|
"mean_token_accuracy": 0.16032088696956634,
|
|
"num_tokens": 3007932.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.067206907272339,
|
|
"epoch": 1.454299621416982,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004980676219165894,
|
|
"loss": 5.9844,
|
|
"mean_token_accuracy": 0.15382035970687866,
|
|
"num_tokens": 3019368.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.075215244293213,
|
|
"epoch": 1.4597079502433747,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000498011123400049,
|
|
"loss": 6.0347,
|
|
"mean_token_accuracy": 0.15633052736520767,
|
|
"num_tokens": 3029787.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 5.867668294906617,
|
|
"epoch": 1.4651162790697674,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004979538144605707,
|
|
"loss": 5.9921,
|
|
"mean_token_accuracy": 0.15314557105302812,
|
|
"num_tokens": 3042100.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.0186302185058596,
|
|
"epoch": 1.47052460789616,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004978956953064154,
|
|
"loss": 6.0183,
|
|
"mean_token_accuracy": 0.15329605489969253,
|
|
"num_tokens": 3054160.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 5.989927768707275,
|
|
"epoch": 1.4759329367225527,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004978367661487883,
|
|
"loss": 5.9612,
|
|
"mean_token_accuracy": 0.15613504201173783,
|
|
"num_tokens": 3065401.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 5.977120494842529,
|
|
"epoch": 1.4813412655489453,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004977770272018379,
|
|
"loss": 5.9019,
|
|
"mean_token_accuracy": 0.15995847731828688,
|
|
"num_tokens": 3076137.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.052564191818237,
|
|
"epoch": 1.486749594375338,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000497716478682656,
|
|
"loss": 6.0296,
|
|
"mean_token_accuracy": 0.1589309737086296,
|
|
"num_tokens": 3087274.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 5.983016633987427,
|
|
"epoch": 1.4921579232017308,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004976551208112759,
|
|
"loss": 5.9068,
|
|
"mean_token_accuracy": 0.15948846489191054,
|
|
"num_tokens": 3098476.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 5.9621570110321045,
|
|
"epoch": 1.4975662520281232,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004975929538106721,
|
|
"loss": 5.9269,
|
|
"mean_token_accuracy": 0.15607307106256485,
|
|
"num_tokens": 3110060.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 5.920859289169312,
|
|
"epoch": 1.502974580854516,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004975299779067598,
|
|
"loss": 5.8526,
|
|
"mean_token_accuracy": 0.15449344217777253,
|
|
"num_tokens": 3121032.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 5.9631072044372555,
|
|
"epoch": 1.5083829096809085,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004974661933283936,
|
|
"loss": 5.874,
|
|
"mean_token_accuracy": 0.16314226984977723,
|
|
"num_tokens": 3131999.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 5.821473121643066,
|
|
"epoch": 1.5137912385073014,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004974016003073667,
|
|
"loss": 5.7969,
|
|
"mean_token_accuracy": 0.16583172529935836,
|
|
"num_tokens": 3142693.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 5.825783157348633,
|
|
"epoch": 1.5191995673336938,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004973361990784104,
|
|
"loss": 5.7657,
|
|
"mean_token_accuracy": 0.15736452788114547,
|
|
"num_tokens": 3154179.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 5.94860258102417,
|
|
"epoch": 1.5246078961600866,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004972699898791927,
|
|
"loss": 5.8441,
|
|
"mean_token_accuracy": 0.15827409774065018,
|
|
"num_tokens": 3165314.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 5.866420555114746,
|
|
"epoch": 1.530016224986479,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004972029729503182,
|
|
"loss": 5.963,
|
|
"mean_token_accuracy": 0.16153641790151596,
|
|
"num_tokens": 3177646.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.031915521621704,
|
|
"epoch": 1.535424553812872,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004971351485353265,
|
|
"loss": 5.9518,
|
|
"mean_token_accuracy": 0.16115951240062715,
|
|
"num_tokens": 3188401.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 5.833160400390625,
|
|
"epoch": 1.5408328826392643,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004970665168806917,
|
|
"loss": 5.8347,
|
|
"mean_token_accuracy": 0.15818036198616028,
|
|
"num_tokens": 3200206.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 5.944838333129883,
|
|
"epoch": 1.5462412114656572,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004969970782358214,
|
|
"loss": 5.9487,
|
|
"mean_token_accuracy": 0.15923830419778823,
|
|
"num_tokens": 3211981.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 5.910025930404663,
|
|
"epoch": 1.5516495402920496,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004969268328530557,
|
|
"loss": 5.9799,
|
|
"mean_token_accuracy": 0.15393504351377488,
|
|
"num_tokens": 3221997.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.051741027832032,
|
|
"epoch": 1.5570578691184425,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004968557809876666,
|
|
"loss": 5.9068,
|
|
"mean_token_accuracy": 0.1563706785440445,
|
|
"num_tokens": 3234333.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 5.887247848510742,
|
|
"epoch": 1.562466197944835,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004967839228978568,
|
|
"loss": 5.9412,
|
|
"mean_token_accuracy": 0.15492452681064606,
|
|
"num_tokens": 3245083.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 5.912670040130616,
|
|
"epoch": 1.5678745267712277,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004967112588447585,
|
|
"loss": 5.7802,
|
|
"mean_token_accuracy": 0.17157655358314514,
|
|
"num_tokens": 3255787.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 5.881341552734375,
|
|
"epoch": 1.5732828555976204,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004966377890924332,
|
|
"loss": 5.9659,
|
|
"mean_token_accuracy": 0.15780261754989625,
|
|
"num_tokens": 3267793.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 5.889124965667724,
|
|
"epoch": 1.578691184424013,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004965635139078704,
|
|
"loss": 5.8448,
|
|
"mean_token_accuracy": 0.16613336354494096,
|
|
"num_tokens": 3279100.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 5.949946784973145,
|
|
"epoch": 1.5840995132504057,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004964884335609857,
|
|
"loss": 5.994,
|
|
"mean_token_accuracy": 0.16043369174003602,
|
|
"num_tokens": 3290251.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 5.837384462356567,
|
|
"epoch": 1.5895078420767983,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004964125483246218,
|
|
"loss": 5.8619,
|
|
"mean_token_accuracy": 0.15883424133062363,
|
|
"num_tokens": 3301817.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.027623033523559,
|
|
"epoch": 1.594916170903191,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004963358584745457,
|
|
"loss": 5.937,
|
|
"mean_token_accuracy": 0.15740436017513276,
|
|
"num_tokens": 3313435.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 5.8814623832702635,
|
|
"epoch": 1.6003244997295836,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004962583642894482,
|
|
"loss": 6.02,
|
|
"mean_token_accuracy": 0.15628089159727096,
|
|
"num_tokens": 3325489.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.047384738922119,
|
|
"epoch": 1.6057328285559762,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004961800660509437,
|
|
"loss": 5.8924,
|
|
"mean_token_accuracy": 0.16333561390638351,
|
|
"num_tokens": 3337030.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 5.801163244247436,
|
|
"epoch": 1.6111411573823688,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000496100964043568,
|
|
"loss": 5.8598,
|
|
"mean_token_accuracy": 0.16104877591133118,
|
|
"num_tokens": 3347766.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 5.861564445495605,
|
|
"epoch": 1.6165494862087615,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000496021058554778,
|
|
"loss": 5.7784,
|
|
"mean_token_accuracy": 0.161942058801651,
|
|
"num_tokens": 3360379.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 5.87091178894043,
|
|
"epoch": 1.6219578150351541,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004959403498749505,
|
|
"loss": 5.9122,
|
|
"mean_token_accuracy": 0.162289035320282,
|
|
"num_tokens": 3370962.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.6219578150351541,
|
|
"eval_entropy": 5.879288995265961,
|
|
"eval_loss": 6.163970470428467,
|
|
"eval_mean_token_accuracy": 0.15741102465174414,
|
|
"eval_num_tokens": 3370962.0,
|
|
"eval_runtime": 2.5767,
|
|
"eval_samples_per_second": 1364.125,
|
|
"eval_steps_per_second": 170.758,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 5.936930131912232,
|
|
"epoch": 1.6273661438615468,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004958588382973809,
|
|
"loss": 5.8371,
|
|
"mean_token_accuracy": 0.1634036645293236,
|
|
"num_tokens": 3382648.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 5.8694202423095705,
|
|
"epoch": 1.6327744726879394,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004957765241182826,
|
|
"loss": 5.8901,
|
|
"mean_token_accuracy": 0.15993100851774217,
|
|
"num_tokens": 3393871.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 5.853174686431885,
|
|
"epoch": 1.638182801514332,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004956934076367854,
|
|
"loss": 5.7746,
|
|
"mean_token_accuracy": 0.16616848409175872,
|
|
"num_tokens": 3405148.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 5.868558740615844,
|
|
"epoch": 1.6435911303407247,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000495609489154935,
|
|
"loss": 5.8185,
|
|
"mean_token_accuracy": 0.1695890635251999,
|
|
"num_tokens": 3416647.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 5.882618379592896,
|
|
"epoch": 1.6489994591671173,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004955247689776911,
|
|
"loss": 5.9201,
|
|
"mean_token_accuracy": 0.15620067715644836,
|
|
"num_tokens": 3428331.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 5.811600971221924,
|
|
"epoch": 1.65440778799351,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000495439247412927,
|
|
"loss": 5.778,
|
|
"mean_token_accuracy": 0.16257178112864495,
|
|
"num_tokens": 3440683.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 5.832757234573364,
|
|
"epoch": 1.6598161168199026,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004953529247714285,
|
|
"loss": 5.7696,
|
|
"mean_token_accuracy": 0.16734129041433335,
|
|
"num_tokens": 3452375.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 5.876086807250976,
|
|
"epoch": 1.6652244456462952,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004952658013668922,
|
|
"loss": 5.9255,
|
|
"mean_token_accuracy": 0.15416743606328964,
|
|
"num_tokens": 3463255.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 5.9034727096557615,
|
|
"epoch": 1.670632774472688,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004951778775159246,
|
|
"loss": 5.8613,
|
|
"mean_token_accuracy": 0.15903377681970596,
|
|
"num_tokens": 3473642.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 5.837939214706421,
|
|
"epoch": 1.6760411032990805,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004950891535380414,
|
|
"loss": 5.8262,
|
|
"mean_token_accuracy": 0.15641504824161528,
|
|
"num_tokens": 3484877.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 5.822560214996338,
|
|
"epoch": 1.6814494321254734,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004949996297556656,
|
|
"loss": 5.7607,
|
|
"mean_token_accuracy": 0.164022558927536,
|
|
"num_tokens": 3495500.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 5.884134674072266,
|
|
"epoch": 1.6868577609518658,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004949093064941268,
|
|
"loss": 5.8745,
|
|
"mean_token_accuracy": 0.162730610370636,
|
|
"num_tokens": 3508057.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 5.915961790084839,
|
|
"epoch": 1.6922660897782587,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004948181840816601,
|
|
"loss": 6.0482,
|
|
"mean_token_accuracy": 0.156703981757164,
|
|
"num_tokens": 3521110.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 5.936773014068604,
|
|
"epoch": 1.697674418604651,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004947262628494045,
|
|
"loss": 5.955,
|
|
"mean_token_accuracy": 0.1600995510816574,
|
|
"num_tokens": 3530692.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 5.907471227645874,
|
|
"epoch": 1.703082747431044,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004946335431314018,
|
|
"loss": 5.9069,
|
|
"mean_token_accuracy": 0.15324967950582505,
|
|
"num_tokens": 3542195.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 5.945133829116822,
|
|
"epoch": 1.7084910762574363,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004945400252645959,
|
|
"loss": 5.782,
|
|
"mean_token_accuracy": 0.16795231103897096,
|
|
"num_tokens": 3552676.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 5.7572108745574955,
|
|
"epoch": 1.7138994050838292,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004944457095888308,
|
|
"loss": 5.7593,
|
|
"mean_token_accuracy": 0.1685643956065178,
|
|
"num_tokens": 3563308.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 5.870556497573853,
|
|
"epoch": 1.7193077339102216,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00049435059644685,
|
|
"loss": 5.7839,
|
|
"mean_token_accuracy": 0.16573660224676132,
|
|
"num_tokens": 3574085.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 5.843953418731689,
|
|
"epoch": 1.7247160627366145,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004942546861842947,
|
|
"loss": 5.8642,
|
|
"mean_token_accuracy": 0.15690634548664092,
|
|
"num_tokens": 3585816.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 5.889085102081299,
|
|
"epoch": 1.730124391563007,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004941579791497032,
|
|
"loss": 5.9163,
|
|
"mean_token_accuracy": 0.16075912564992906,
|
|
"num_tokens": 3597776.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.0246764659881595,
|
|
"epoch": 1.7355327203893998,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004940604756945091,
|
|
"loss": 5.9195,
|
|
"mean_token_accuracy": 0.16509058624505996,
|
|
"num_tokens": 3607991.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 5.8152741432189945,
|
|
"epoch": 1.7409410492157922,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004939621761730399,
|
|
"loss": 5.8073,
|
|
"mean_token_accuracy": 0.16705307513475418,
|
|
"num_tokens": 3618204.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 5.836070251464844,
|
|
"epoch": 1.746349378042185,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004938630809425166,
|
|
"loss": 5.9172,
|
|
"mean_token_accuracy": 0.1609561935067177,
|
|
"num_tokens": 3629558.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 5.916400194168091,
|
|
"epoch": 1.7517577068685775,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004937631903630513,
|
|
"loss": 5.8558,
|
|
"mean_token_accuracy": 0.1601060912013054,
|
|
"num_tokens": 3640773.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 5.839791250228882,
|
|
"epoch": 1.7571660356949703,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004936625047976468,
|
|
"loss": 5.7689,
|
|
"mean_token_accuracy": 0.16652738451957702,
|
|
"num_tokens": 3652186.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 5.8121192932128904,
|
|
"epoch": 1.7625743645213627,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004935610246121944,
|
|
"loss": 5.8172,
|
|
"mean_token_accuracy": 0.1682569146156311,
|
|
"num_tokens": 3663068.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 5.832926082611084,
|
|
"epoch": 1.7679826933477556,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004934587501754736,
|
|
"loss": 5.9692,
|
|
"mean_token_accuracy": 0.15569856017827988,
|
|
"num_tokens": 3673803.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.0261900424957275,
|
|
"epoch": 1.7733910221741482,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004933556818591498,
|
|
"loss": 5.9362,
|
|
"mean_token_accuracy": 0.15648741871118546,
|
|
"num_tokens": 3685482.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 5.761137580871582,
|
|
"epoch": 1.7787993510005409,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004932518200377736,
|
|
"loss": 5.7912,
|
|
"mean_token_accuracy": 0.16834952235221862,
|
|
"num_tokens": 3697641.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 5.872494173049927,
|
|
"epoch": 1.7842076798269335,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004931471650887789,
|
|
"loss": 5.8092,
|
|
"mean_token_accuracy": 0.1631079748272896,
|
|
"num_tokens": 3709560.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 5.961399745941162,
|
|
"epoch": 1.7896160086533262,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004930417173924822,
|
|
"loss": 5.9413,
|
|
"mean_token_accuracy": 0.15678294226527215,
|
|
"num_tokens": 3720422.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 5.745348501205444,
|
|
"epoch": 1.7950243374797188,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004929354773320806,
|
|
"loss": 5.7509,
|
|
"mean_token_accuracy": 0.16357783377170562,
|
|
"num_tokens": 3731104.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 5.849882936477661,
|
|
"epoch": 1.8004326663061114,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004928284452936506,
|
|
"loss": 5.9165,
|
|
"mean_token_accuracy": 0.15696647614240647,
|
|
"num_tokens": 3742766.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 5.8943158149719235,
|
|
"epoch": 1.805840995132504,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004927206216661473,
|
|
"loss": 5.8026,
|
|
"mean_token_accuracy": 0.16676504611968995,
|
|
"num_tokens": 3753064.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 5.848807716369629,
|
|
"epoch": 1.8112493239588967,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004926120068414015,
|
|
"loss": 5.8392,
|
|
"mean_token_accuracy": 0.15926389694213866,
|
|
"num_tokens": 3765183.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 5.87504243850708,
|
|
"epoch": 1.8166576527852893,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004925026012141201,
|
|
"loss": 5.8344,
|
|
"mean_token_accuracy": 0.15749894231557846,
|
|
"num_tokens": 3777902.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 5.782968807220459,
|
|
"epoch": 1.822065981611682,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004923924051818832,
|
|
"loss": 5.7928,
|
|
"mean_token_accuracy": 0.16230279207229614,
|
|
"num_tokens": 3788461.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 5.807979583740234,
|
|
"epoch": 1.8274743104380746,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004922814191451435,
|
|
"loss": 5.7236,
|
|
"mean_token_accuracy": 0.16461351066827773,
|
|
"num_tokens": 3799574.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 5.855834150314331,
|
|
"epoch": 1.8328826392644673,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004921696435072245,
|
|
"loss": 5.8815,
|
|
"mean_token_accuracy": 0.16156751960515975,
|
|
"num_tokens": 3811889.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 5.860201358795166,
|
|
"epoch": 1.83829096809086,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004920570786743191,
|
|
"loss": 5.8447,
|
|
"mean_token_accuracy": 0.16074928641319275,
|
|
"num_tokens": 3822949.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 5.883919429779053,
|
|
"epoch": 1.8436992969172525,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004919437250554882,
|
|
"loss": 5.9364,
|
|
"mean_token_accuracy": 0.15667719617486,
|
|
"num_tokens": 3834734.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 5.915875244140625,
|
|
"epoch": 1.8491076257436452,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004918295830626592,
|
|
"loss": 5.7887,
|
|
"mean_token_accuracy": 0.17234719395637513,
|
|
"num_tokens": 3845508.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 5.774176454544067,
|
|
"epoch": 1.8545159545700378,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004917146531106242,
|
|
"loss": 5.7795,
|
|
"mean_token_accuracy": 0.16408951133489608,
|
|
"num_tokens": 3856500.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 5.971676301956177,
|
|
"epoch": 1.8599242833964305,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004915989356170389,
|
|
"loss": 5.8509,
|
|
"mean_token_accuracy": 0.1615983471274376,
|
|
"num_tokens": 3867877.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 5.8224071025848385,
|
|
"epoch": 1.865332612222823,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004914824310024213,
|
|
"loss": 5.8372,
|
|
"mean_token_accuracy": 0.16671036779880524,
|
|
"num_tokens": 3878613.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 5.765814685821534,
|
|
"epoch": 1.870740941049216,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004913651396901489,
|
|
"loss": 5.756,
|
|
"mean_token_accuracy": 0.16456347852945327,
|
|
"num_tokens": 3888418.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 5.885995960235595,
|
|
"epoch": 1.8761492698756084,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000491247062106459,
|
|
"loss": 5.8455,
|
|
"mean_token_accuracy": 0.16331836730241775,
|
|
"num_tokens": 3899359.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 5.918319225311279,
|
|
"epoch": 1.8815575987020012,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004911281986804455,
|
|
"loss": 5.9422,
|
|
"mean_token_accuracy": 0.15653555542230607,
|
|
"num_tokens": 3911568.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 5.870144891738891,
|
|
"epoch": 1.8869659275283936,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004910085498440587,
|
|
"loss": 5.8776,
|
|
"mean_token_accuracy": 0.15643555968999862,
|
|
"num_tokens": 3921597.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 5.810258293151856,
|
|
"epoch": 1.8923742563547865,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004908881160321025,
|
|
"loss": 5.7665,
|
|
"mean_token_accuracy": 0.15902285426855087,
|
|
"num_tokens": 3933124.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.878746509552002,
|
|
"epoch": 1.897782585181179,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004907668976822338,
|
|
"loss": 5.7836,
|
|
"mean_token_accuracy": 0.16795775294303894,
|
|
"num_tokens": 3944388.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 5.8388762950897215,
|
|
"epoch": 1.9031909140075718,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004906448952349603,
|
|
"loss": 5.7588,
|
|
"mean_token_accuracy": 0.16548295170068741,
|
|
"num_tokens": 3955024.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.66992974281311,
|
|
"epoch": 1.9085992428339642,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004905221091336392,
|
|
"loss": 5.7581,
|
|
"mean_token_accuracy": 0.16820060908794404,
|
|
"num_tokens": 3966029.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.8512763500213625,
|
|
"epoch": 1.914007571660357,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004903985398244753,
|
|
"loss": 5.7223,
|
|
"mean_token_accuracy": 0.16350394934415818,
|
|
"num_tokens": 3976066.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 5.862121200561523,
|
|
"epoch": 1.9194159004867495,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004902741877565201,
|
|
"loss": 5.9088,
|
|
"mean_token_accuracy": 0.161722232401371,
|
|
"num_tokens": 3986467.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.893027591705322,
|
|
"epoch": 1.9248242293131423,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000490149053381669,
|
|
"loss": 5.8808,
|
|
"mean_token_accuracy": 0.16341264247894288,
|
|
"num_tokens": 3998208.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.866563510894776,
|
|
"epoch": 1.9302325581395348,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004900231371546608,
|
|
"loss": 5.8284,
|
|
"mean_token_accuracy": 0.1650262013077736,
|
|
"num_tokens": 4008202.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.761198902130127,
|
|
"epoch": 1.9356408869659276,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004898964395330753,
|
|
"loss": 5.6731,
|
|
"mean_token_accuracy": 0.17755473405122757,
|
|
"num_tokens": 4019256.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 5.716888666152954,
|
|
"epoch": 1.94104921579232,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004897689609773318,
|
|
"loss": 5.7038,
|
|
"mean_token_accuracy": 0.16974467188119888,
|
|
"num_tokens": 4029738.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 5.805795907974243,
|
|
"epoch": 1.946457544618713,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004896407019506879,
|
|
"loss": 5.9485,
|
|
"mean_token_accuracy": 0.15304052084684372,
|
|
"num_tokens": 4042277.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.898893117904663,
|
|
"epoch": 1.9518658734451053,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004895116629192371,
|
|
"loss": 5.786,
|
|
"mean_token_accuracy": 0.17360647916793823,
|
|
"num_tokens": 4054019.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.807737588882446,
|
|
"epoch": 1.9572742022714982,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004893818443519074,
|
|
"loss": 5.8634,
|
|
"mean_token_accuracy": 0.1625102698802948,
|
|
"num_tokens": 4065562.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.916156482696533,
|
|
"epoch": 1.9626825310978906,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00048925124672046,
|
|
"loss": 5.9151,
|
|
"mean_token_accuracy": 0.16293887794017792,
|
|
"num_tokens": 4076617.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.870821380615235,
|
|
"epoch": 1.9680908599242835,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004891198704994868,
|
|
"loss": 5.7231,
|
|
"mean_token_accuracy": 0.17624200582504274,
|
|
"num_tokens": 4087191.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.770419502258301,
|
|
"epoch": 1.973499188750676,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004889877161664096,
|
|
"loss": 5.7546,
|
|
"mean_token_accuracy": 0.16992903500795364,
|
|
"num_tokens": 4099150.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.815229177474976,
|
|
"epoch": 1.9789075175770687,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004888547842014771,
|
|
"loss": 5.7672,
|
|
"mean_token_accuracy": 0.16862173080444337,
|
|
"num_tokens": 4110299.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.779698705673217,
|
|
"epoch": 1.9843158464034614,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004887210750877648,
|
|
"loss": 5.7856,
|
|
"mean_token_accuracy": 0.16518506705760955,
|
|
"num_tokens": 4121682.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.789695692062378,
|
|
"epoch": 1.989724175229854,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004885865893111716,
|
|
"loss": 5.7321,
|
|
"mean_token_accuracy": 0.16423704624176025,
|
|
"num_tokens": 4132914.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.79205288887024,
|
|
"epoch": 1.9951325040562466,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004884513273604195,
|
|
"loss": 5.8342,
|
|
"mean_token_accuracy": 0.16368089467287064,
|
|
"num_tokens": 4144213.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 5.850319544474284,
|
|
"epoch": 2.0,
|
|
"grad_norm": 2.109375,
|
|
"learning_rate": 0.0004883152897270502,
|
|
"loss": 5.8523,
|
|
"mean_token_accuracy": 0.1651873919698927,
|
|
"num_tokens": 4153778.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.728305816650391,
|
|
"epoch": 2.005408328826393,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004881784769054253,
|
|
"loss": 5.43,
|
|
"mean_token_accuracy": 0.17241204977035524,
|
|
"num_tokens": 4165536.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.8680202007293705,
|
|
"epoch": 2.0108166576527853,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004880408893927226,
|
|
"loss": 5.4768,
|
|
"mean_token_accuracy": 0.17083195745944976,
|
|
"num_tokens": 4176182.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.649136543273926,
|
|
"epoch": 2.016224986479178,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004879025276889355,
|
|
"loss": 5.4962,
|
|
"mean_token_accuracy": 0.18160250037908554,
|
|
"num_tokens": 4187592.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.651928091049195,
|
|
"epoch": 2.0216333153055706,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00048776339229687077,
|
|
"loss": 5.2851,
|
|
"mean_token_accuracy": 0.1900184229016304,
|
|
"num_tokens": 4198199.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.681820726394653,
|
|
"epoch": 2.0270416441319634,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048762348372214683,
|
|
"loss": 5.5159,
|
|
"mean_token_accuracy": 0.17747106552124023,
|
|
"num_tokens": 4208116.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.700159978866577,
|
|
"epoch": 2.032449972958356,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00048748280247319165,
|
|
"loss": 5.304,
|
|
"mean_token_accuracy": 0.18070727884769439,
|
|
"num_tokens": 4218569.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.663970518112182,
|
|
"epoch": 2.0378583017847487,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004873413490612412,
|
|
"loss": 5.45,
|
|
"mean_token_accuracy": 0.18299697637557982,
|
|
"num_tokens": 4228785.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.678899717330933,
|
|
"epoch": 2.043266630611141,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048719912400033757,
|
|
"loss": 5.4266,
|
|
"mean_token_accuracy": 0.17531348764896393,
|
|
"num_tokens": 4238814.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.651882839202881,
|
|
"epoch": 2.048674959437534,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000487056127807327,
|
|
"loss": 5.3886,
|
|
"mean_token_accuracy": 0.1779720366001129,
|
|
"num_tokens": 4250544.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.652195453643799,
|
|
"epoch": 2.0540832882639264,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048691236100185795,
|
|
"loss": 5.5298,
|
|
"mean_token_accuracy": 0.17636178135871888,
|
|
"num_tokens": 4261806.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.797122573852539,
|
|
"epoch": 2.0594916170903192,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004867678241063792,
|
|
"loss": 5.4707,
|
|
"mean_token_accuracy": 0.17808170318603517,
|
|
"num_tokens": 4272472.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.63343391418457,
|
|
"epoch": 2.0648999459167117,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048662251764613833,
|
|
"loss": 5.4377,
|
|
"mean_token_accuracy": 0.17689038664102555,
|
|
"num_tokens": 4283758.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.698742532730103,
|
|
"epoch": 2.0703082747431045,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004864764421491793,
|
|
"loss": 5.5595,
|
|
"mean_token_accuracy": 0.176458902657032,
|
|
"num_tokens": 4295930.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.803366994857788,
|
|
"epoch": 2.075716603569497,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004863295981463408,
|
|
"loss": 5.4691,
|
|
"mean_token_accuracy": 0.1813316151499748,
|
|
"num_tokens": 4306711.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.576512861251831,
|
|
"epoch": 2.08112493239589,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004861819861712543,
|
|
"loss": 5.4523,
|
|
"mean_token_accuracy": 0.18199085295200348,
|
|
"num_tokens": 4317694.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.773883581161499,
|
|
"epoch": 2.086533261222282,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048603360676034203,
|
|
"loss": 5.5472,
|
|
"mean_token_accuracy": 0.16931245923042298,
|
|
"num_tokens": 4329552.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.705210638046265,
|
|
"epoch": 2.091941590048675,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048588446045281505,
|
|
"loss": 5.4803,
|
|
"mean_token_accuracy": 0.17831153869628907,
|
|
"num_tokens": 4340479.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.603524208068848,
|
|
"epoch": 2.0973499188750675,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004857345477906714,
|
|
"loss": 5.3548,
|
|
"mean_token_accuracy": 0.18533679097890854,
|
|
"num_tokens": 4351147.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.5270696640014645,
|
|
"epoch": 2.1027582477014604,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048558386931869417,
|
|
"loss": 5.3804,
|
|
"mean_token_accuracy": 0.18393090069293977,
|
|
"num_tokens": 4362962.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.697436428070068,
|
|
"epoch": 2.1081665765278528,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004854324255844492,
|
|
"loss": 5.4616,
|
|
"mean_token_accuracy": 0.18124082833528518,
|
|
"num_tokens": 4373723.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.531086158752442,
|
|
"epoch": 2.1135749053542456,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048528021713828336,
|
|
"loss": 5.4444,
|
|
"mean_token_accuracy": 0.17540200650691987,
|
|
"num_tokens": 4383434.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.66317400932312,
|
|
"epoch": 2.118983234180638,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004851272445333225,
|
|
"loss": 5.3929,
|
|
"mean_token_accuracy": 0.18412991017103195,
|
|
"num_tokens": 4394847.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.6545063018798825,
|
|
"epoch": 2.124391563007031,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048497350832546937,
|
|
"loss": 5.4911,
|
|
"mean_token_accuracy": 0.18090608417987825,
|
|
"num_tokens": 4406367.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.736139726638794,
|
|
"epoch": 2.1297998918334233,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004848190090734018,
|
|
"loss": 5.4846,
|
|
"mean_token_accuracy": 0.1810358703136444,
|
|
"num_tokens": 4417636.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.64870662689209,
|
|
"epoch": 2.135208220659816,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048466374733857035,
|
|
"loss": 5.4904,
|
|
"mean_token_accuracy": 0.18504306226968764,
|
|
"num_tokens": 4428473.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.588465929031372,
|
|
"epoch": 2.1406165494862086,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004845077236851966,
|
|
"loss": 5.4432,
|
|
"mean_token_accuracy": 0.17573228031396865,
|
|
"num_tokens": 4439780.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.629347801208496,
|
|
"epoch": 2.1460248783126015,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048435093868027076,
|
|
"loss": 5.3769,
|
|
"mean_token_accuracy": 0.19002774059772493,
|
|
"num_tokens": 4450929.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.605298137664795,
|
|
"epoch": 2.151433207138994,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048419339289354995,
|
|
"loss": 5.4073,
|
|
"mean_token_accuracy": 0.18362201899290084,
|
|
"num_tokens": 4462967.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.7316478252410885,
|
|
"epoch": 2.1568415359653867,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000484035086897556,
|
|
"loss": 5.5117,
|
|
"mean_token_accuracy": 0.18310858607292174,
|
|
"num_tokens": 4473970.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.649066972732544,
|
|
"epoch": 2.162249864791779,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004838760212675732,
|
|
"loss": 5.5202,
|
|
"mean_token_accuracy": 0.1777018651366234,
|
|
"num_tokens": 4485660.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 2.162249864791779,
|
|
"eval_entropy": 5.640687057104977,
|
|
"eval_loss": 6.078622817993164,
|
|
"eval_mean_token_accuracy": 0.1612045247446407,
|
|
"eval_num_tokens": 4485660.0,
|
|
"eval_runtime": 2.5625,
|
|
"eval_samples_per_second": 1371.717,
|
|
"eval_steps_per_second": 171.708,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.666570043563842,
|
|
"epoch": 2.167658193618172,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004837161965816464,
|
|
"loss": 5.4937,
|
|
"mean_token_accuracy": 0.18039914667606355,
|
|
"num_tokens": 4497261.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.612936019897461,
|
|
"epoch": 2.1730665224445644,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048355561342057886,
|
|
"loss": 5.3801,
|
|
"mean_token_accuracy": 0.1858850210905075,
|
|
"num_tokens": 4508120.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.65741605758667,
|
|
"epoch": 2.1784748512709573,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048339427236793024,
|
|
"loss": 5.4575,
|
|
"mean_token_accuracy": 0.18245986104011536,
|
|
"num_tokens": 4519286.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.696688985824585,
|
|
"epoch": 2.1838831800973497,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004832321740100143,
|
|
"loss": 5.5374,
|
|
"mean_token_accuracy": 0.17700066715478896,
|
|
"num_tokens": 4531912.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.71485276222229,
|
|
"epoch": 2.1892915089237426,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048306931893589666,
|
|
"loss": 5.5337,
|
|
"mean_token_accuracy": 0.17527716606855392,
|
|
"num_tokens": 4544370.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.652210235595703,
|
|
"epoch": 2.194699837750135,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048290570773739327,
|
|
"loss": 5.5006,
|
|
"mean_token_accuracy": 0.17758730202913284,
|
|
"num_tokens": 4556013.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.615873527526856,
|
|
"epoch": 2.200108166576528,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048274134100906747,
|
|
"loss": 5.433,
|
|
"mean_token_accuracy": 0.17940177619457245,
|
|
"num_tokens": 4566200.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.650509977340699,
|
|
"epoch": 2.2055164954029207,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048257621934822835,
|
|
"loss": 5.4799,
|
|
"mean_token_accuracy": 0.17791129648685455,
|
|
"num_tokens": 4576899.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.7569211483001705,
|
|
"epoch": 2.210924824229313,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004824103433549285,
|
|
"loss": 5.505,
|
|
"mean_token_accuracy": 0.17102217972278594,
|
|
"num_tokens": 4588983.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.573875045776367,
|
|
"epoch": 2.2163331530557056,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004822437136319616,
|
|
"loss": 5.4586,
|
|
"mean_token_accuracy": 0.1880033642053604,
|
|
"num_tokens": 4599758.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.647621059417725,
|
|
"epoch": 2.2217414818820984,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004820763307848604,
|
|
"loss": 5.4849,
|
|
"mean_token_accuracy": 0.17981284558773042,
|
|
"num_tokens": 4610927.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.677910852432251,
|
|
"epoch": 2.2271498107084913,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048190819542189456,
|
|
"loss": 5.5214,
|
|
"mean_token_accuracy": 0.17770596891641616,
|
|
"num_tokens": 4623719.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.649923467636109,
|
|
"epoch": 2.2325581395348837,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048173930815406847,
|
|
"loss": 5.572,
|
|
"mean_token_accuracy": 0.1757882609963417,
|
|
"num_tokens": 4635389.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.704444122314453,
|
|
"epoch": 2.2379664683612766,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004815696695951187,
|
|
"loss": 5.5151,
|
|
"mean_token_accuracy": 0.17672796845436095,
|
|
"num_tokens": 4645093.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.614984178543091,
|
|
"epoch": 2.243374797187669,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004813992803615123,
|
|
"loss": 5.5133,
|
|
"mean_token_accuracy": 0.17320940941572188,
|
|
"num_tokens": 4656490.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.69649658203125,
|
|
"epoch": 2.248783126014062,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048122814107244414,
|
|
"loss": 5.5241,
|
|
"mean_token_accuracy": 0.1778429329395294,
|
|
"num_tokens": 4667603.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.606650733947754,
|
|
"epoch": 2.2541914548404542,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048105625234983466,
|
|
"loss": 5.4654,
|
|
"mean_token_accuracy": 0.17279692888259887,
|
|
"num_tokens": 4678669.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.552337551116944,
|
|
"epoch": 2.259599783666847,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00048088361481832807,
|
|
"loss": 5.2638,
|
|
"mean_token_accuracy": 0.19510975629091262,
|
|
"num_tokens": 4689757.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.511549425125122,
|
|
"epoch": 2.2650081124932395,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048071022910528934,
|
|
"loss": 5.4504,
|
|
"mean_token_accuracy": 0.18175310343503953,
|
|
"num_tokens": 4699332.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.651071643829345,
|
|
"epoch": 2.2704164413196324,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048053609584080274,
|
|
"loss": 5.5275,
|
|
"mean_token_accuracy": 0.18241235762834548,
|
|
"num_tokens": 4709485.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.613623571395874,
|
|
"epoch": 2.275824770146025,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048036121565766876,
|
|
"loss": 5.4762,
|
|
"mean_token_accuracy": 0.17775042951107026,
|
|
"num_tokens": 4720935.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.540726518630981,
|
|
"epoch": 2.2812330989724177,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004801855891914026,
|
|
"loss": 5.419,
|
|
"mean_token_accuracy": 0.17690307199954985,
|
|
"num_tokens": 4731219.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.686253070831299,
|
|
"epoch": 2.28664142779881,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048000921708023117,
|
|
"loss": 5.5288,
|
|
"mean_token_accuracy": 0.17510719150304793,
|
|
"num_tokens": 4742637.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.61152458190918,
|
|
"epoch": 2.292049756625203,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004798320999650911,
|
|
"loss": 5.4365,
|
|
"mean_token_accuracy": 0.17937714010477065,
|
|
"num_tokens": 4753341.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.7665266513824465,
|
|
"epoch": 2.2974580854515954,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047965423848962636,
|
|
"loss": 5.5726,
|
|
"mean_token_accuracy": 0.1805647626519203,
|
|
"num_tokens": 4765339.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.670615100860596,
|
|
"epoch": 2.302866414277988,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000479475633300186,
|
|
"loss": 5.4346,
|
|
"mean_token_accuracy": 0.183509062230587,
|
|
"num_tokens": 4776149.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.585747480392456,
|
|
"epoch": 2.3082747431043806,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047929628504582173,
|
|
"loss": 5.4544,
|
|
"mean_token_accuracy": 0.18422979414463042,
|
|
"num_tokens": 4788120.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.665250062942505,
|
|
"epoch": 2.3136830719307735,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004791161943782856,
|
|
"loss": 5.5579,
|
|
"mean_token_accuracy": 0.17357907742261885,
|
|
"num_tokens": 4799922.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.523003721237183,
|
|
"epoch": 2.319091400757166,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047893536195202726,
|
|
"loss": 5.489,
|
|
"mean_token_accuracy": 0.17451781630516053,
|
|
"num_tokens": 4810501.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.662677001953125,
|
|
"epoch": 2.3244997295835588,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004787537884241923,
|
|
"loss": 5.424,
|
|
"mean_token_accuracy": 0.18095659017562865,
|
|
"num_tokens": 4820919.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.682862377166748,
|
|
"epoch": 2.329908058409951,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047857147445461943,
|
|
"loss": 5.5198,
|
|
"mean_token_accuracy": 0.1802654057741165,
|
|
"num_tokens": 4833298.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.531137895584107,
|
|
"epoch": 2.335316387236344,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004783884207058379,
|
|
"loss": 5.3992,
|
|
"mean_token_accuracy": 0.1799473449587822,
|
|
"num_tokens": 4845199.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.633423852920532,
|
|
"epoch": 2.3407247160627365,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047820462784306555,
|
|
"loss": 5.4451,
|
|
"mean_token_accuracy": 0.17799740731716157,
|
|
"num_tokens": 4856549.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.581632995605469,
|
|
"epoch": 2.3461330448891293,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000478020096534206,
|
|
"loss": 5.5066,
|
|
"mean_token_accuracy": 0.1754800170660019,
|
|
"num_tokens": 4868446.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.67888994216919,
|
|
"epoch": 2.3515413737155217,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047783482744984644,
|
|
"loss": 5.5785,
|
|
"mean_token_accuracy": 0.1686321973800659,
|
|
"num_tokens": 4879047.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.658730173110962,
|
|
"epoch": 2.3569497025419146,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00047764882126325527,
|
|
"loss": 5.4529,
|
|
"mean_token_accuracy": 0.17824074476957322,
|
|
"num_tokens": 4891591.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.55944242477417,
|
|
"epoch": 2.362358031368307,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004774620786503793,
|
|
"loss": 5.4635,
|
|
"mean_token_accuracy": 0.18263570070266724,
|
|
"num_tokens": 4902710.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.610701179504394,
|
|
"epoch": 2.3677663601947,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004772746002898417,
|
|
"loss": 5.5345,
|
|
"mean_token_accuracy": 0.17826326787471772,
|
|
"num_tokens": 4914668.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.757508754730225,
|
|
"epoch": 2.3731746890210923,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047708638686293903,
|
|
"loss": 5.6764,
|
|
"mean_token_accuracy": 0.17494274377822877,
|
|
"num_tokens": 4926684.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.709033870697022,
|
|
"epoch": 2.378583017847485,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00047689743905363963,
|
|
"loss": 5.5329,
|
|
"mean_token_accuracy": 0.17751119285821915,
|
|
"num_tokens": 4938577.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.445482635498047,
|
|
"epoch": 2.383991346673878,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004767077575485801,
|
|
"loss": 5.4085,
|
|
"mean_token_accuracy": 0.1875528633594513,
|
|
"num_tokens": 4949070.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.642753219604492,
|
|
"epoch": 2.3893996755002704,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004765173430370635,
|
|
"loss": 5.4804,
|
|
"mean_token_accuracy": 0.1852825790643692,
|
|
"num_tokens": 4959558.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.567414951324463,
|
|
"epoch": 2.394808004326663,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004763261962110567,
|
|
"loss": 5.3456,
|
|
"mean_token_accuracy": 0.18535693734884262,
|
|
"num_tokens": 4970310.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.4698888778686525,
|
|
"epoch": 2.4002163331530557,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004761343177651875,
|
|
"loss": 5.4301,
|
|
"mean_token_accuracy": 0.18121608346700668,
|
|
"num_tokens": 4982300.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.637871313095093,
|
|
"epoch": 2.4056246619794486,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004759417083967428,
|
|
"loss": 5.4671,
|
|
"mean_token_accuracy": 0.17587541937828063,
|
|
"num_tokens": 4994425.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.545809364318847,
|
|
"epoch": 2.411032990805841,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047574836880566544,
|
|
"loss": 5.476,
|
|
"mean_token_accuracy": 0.18699895441532136,
|
|
"num_tokens": 5005872.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.563200044631958,
|
|
"epoch": 2.4164413196322334,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004755542996945519,
|
|
"loss": 5.4976,
|
|
"mean_token_accuracy": 0.17911621928215027,
|
|
"num_tokens": 5016903.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.682060432434082,
|
|
"epoch": 2.4218496484586263,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00047535950176864976,
|
|
"loss": 5.5455,
|
|
"mean_token_accuracy": 0.1817258134484291,
|
|
"num_tokens": 5028407.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.673858308792115,
|
|
"epoch": 2.427257977285019,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047516397573585525,
|
|
"loss": 5.5423,
|
|
"mean_token_accuracy": 0.1782567858695984,
|
|
"num_tokens": 5039693.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.67990083694458,
|
|
"epoch": 2.4326663061114115,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047496772230671034,
|
|
"loss": 5.4857,
|
|
"mean_token_accuracy": 0.1797408029437065,
|
|
"num_tokens": 5051279.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.648202419281006,
|
|
"epoch": 2.4380746349378044,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004747707421944004,
|
|
"loss": 5.4378,
|
|
"mean_token_accuracy": 0.18068693876266478,
|
|
"num_tokens": 5062698.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.579425525665283,
|
|
"epoch": 2.443482963764197,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004745730361147517,
|
|
"loss": 5.4244,
|
|
"mean_token_accuracy": 0.18114178627729416,
|
|
"num_tokens": 5074927.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.557804346084595,
|
|
"epoch": 2.4488912925905897,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004743746047862286,
|
|
"loss": 5.5479,
|
|
"mean_token_accuracy": 0.1731310248374939,
|
|
"num_tokens": 5086298.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.70584659576416,
|
|
"epoch": 2.454299621416982,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004741754489299309,
|
|
"loss": 5.4897,
|
|
"mean_token_accuracy": 0.17877588272094727,
|
|
"num_tokens": 5098161.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.502418470382691,
|
|
"epoch": 2.459707950243375,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047397556926959155,
|
|
"loss": 5.4872,
|
|
"mean_token_accuracy": 0.18534856885671616,
|
|
"num_tokens": 5110743.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.712545537948609,
|
|
"epoch": 2.4651162790697674,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047377496653157374,
|
|
"loss": 5.5263,
|
|
"mean_token_accuracy": 0.18395540565252305,
|
|
"num_tokens": 5121838.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.558406400680542,
|
|
"epoch": 2.4705246078961602,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047357364144486806,
|
|
"loss": 5.4617,
|
|
"mean_token_accuracy": 0.178267040848732,
|
|
"num_tokens": 5135047.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.542427253723145,
|
|
"epoch": 2.4759329367225527,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047337159474109057,
|
|
"loss": 5.4542,
|
|
"mean_token_accuracy": 0.17631031572818756,
|
|
"num_tokens": 5146329.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.709754371643067,
|
|
"epoch": 2.4813412655489455,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047316882715447934,
|
|
"loss": 5.5014,
|
|
"mean_token_accuracy": 0.1808954656124115,
|
|
"num_tokens": 5157511.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.573445272445679,
|
|
"epoch": 2.486749594375338,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004729653394218922,
|
|
"loss": 5.4884,
|
|
"mean_token_accuracy": 0.17926492244005204,
|
|
"num_tokens": 5168456.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.701962614059449,
|
|
"epoch": 2.492157923201731,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047276113228280427,
|
|
"loss": 5.5377,
|
|
"mean_token_accuracy": 0.17932706624269484,
|
|
"num_tokens": 5179294.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.5871984481811525,
|
|
"epoch": 2.497566252028123,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004725562064793045,
|
|
"loss": 5.484,
|
|
"mean_token_accuracy": 0.1833512768149376,
|
|
"num_tokens": 5190195.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.450733280181884,
|
|
"epoch": 2.502974580854516,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047235056275609397,
|
|
"loss": 5.3705,
|
|
"mean_token_accuracy": 0.18728333711624146,
|
|
"num_tokens": 5201289.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.59820523262024,
|
|
"epoch": 2.5083829096809085,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004721442018604822,
|
|
"loss": 5.3803,
|
|
"mean_token_accuracy": 0.19116080850362777,
|
|
"num_tokens": 5212295.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.497561740875244,
|
|
"epoch": 2.5137912385073014,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004719371245423853,
|
|
"loss": 5.3806,
|
|
"mean_token_accuracy": 0.18749208599328995,
|
|
"num_tokens": 5223144.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.481751728057861,
|
|
"epoch": 2.5191995673336938,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047172933155432273,
|
|
"loss": 5.401,
|
|
"mean_token_accuracy": 0.1855457529425621,
|
|
"num_tokens": 5232840.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.592643737792969,
|
|
"epoch": 2.5246078961600866,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047152082365141466,
|
|
"loss": 5.4932,
|
|
"mean_token_accuracy": 0.18096424490213395,
|
|
"num_tokens": 5243796.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.636725568771363,
|
|
"epoch": 2.530016224986479,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004713116015913794,
|
|
"loss": 5.5878,
|
|
"mean_token_accuracy": 0.17433768808841704,
|
|
"num_tokens": 5255990.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.666304683685302,
|
|
"epoch": 2.535424553812872,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00047110166613453016,
|
|
"loss": 5.541,
|
|
"mean_token_accuracy": 0.18006350249052047,
|
|
"num_tokens": 5268134.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.448769283294678,
|
|
"epoch": 2.5408328826392643,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047089101804377305,
|
|
"loss": 5.3855,
|
|
"mean_token_accuracy": 0.18617575913667678,
|
|
"num_tokens": 5278662.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.610137462615967,
|
|
"epoch": 2.546241211465657,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004706796580846037,
|
|
"loss": 5.4601,
|
|
"mean_token_accuracy": 0.17706628441810607,
|
|
"num_tokens": 5288382.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.606607484817505,
|
|
"epoch": 2.5516495402920496,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047046758702510465,
|
|
"loss": 5.5141,
|
|
"mean_token_accuracy": 0.17458620965480803,
|
|
"num_tokens": 5299769.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.68546838760376,
|
|
"epoch": 2.5570578691184425,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047025480563594264,
|
|
"loss": 5.4364,
|
|
"mean_token_accuracy": 0.1851160392165184,
|
|
"num_tokens": 5310662.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.5270103931427,
|
|
"epoch": 2.5624661979448353,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047004131469036576,
|
|
"loss": 5.4227,
|
|
"mean_token_accuracy": 0.1756480246782303,
|
|
"num_tokens": 5321785.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.5632617473602295,
|
|
"epoch": 2.5678745267712277,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046982711496420057,
|
|
"loss": 5.4672,
|
|
"mean_token_accuracy": 0.17944474518299103,
|
|
"num_tokens": 5333557.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.64686279296875,
|
|
"epoch": 2.57328285559762,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046961220723584925,
|
|
"loss": 5.5387,
|
|
"mean_token_accuracy": 0.18026718944311143,
|
|
"num_tokens": 5344767.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.537802362442017,
|
|
"epoch": 2.578691184424013,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00046939659228628725,
|
|
"loss": 5.4575,
|
|
"mean_token_accuracy": 0.17610580921173097,
|
|
"num_tokens": 5356673.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.605266571044922,
|
|
"epoch": 2.584099513250406,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00046918027089905956,
|
|
"loss": 5.5482,
|
|
"mean_token_accuracy": 0.17411509305238723,
|
|
"num_tokens": 5367296.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.551157426834107,
|
|
"epoch": 2.5895078420767983,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046896324386027873,
|
|
"loss": 5.4333,
|
|
"mean_token_accuracy": 0.18090079873800277,
|
|
"num_tokens": 5378081.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.695206642150879,
|
|
"epoch": 2.5949161709031907,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046874551195862145,
|
|
"loss": 5.5232,
|
|
"mean_token_accuracy": 0.17625013291835784,
|
|
"num_tokens": 5390244.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.536914730072022,
|
|
"epoch": 2.6003244997295836,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000468527075985326,
|
|
"loss": 5.3809,
|
|
"mean_token_accuracy": 0.1849593326449394,
|
|
"num_tokens": 5400741.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.6184252262115475,
|
|
"epoch": 2.6057328285559764,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046830793673418924,
|
|
"loss": 5.5667,
|
|
"mean_token_accuracy": 0.18291546106338502,
|
|
"num_tokens": 5411440.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.61326961517334,
|
|
"epoch": 2.611141157382369,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046808809500156355,
|
|
"loss": 5.4309,
|
|
"mean_token_accuracy": 0.177603617310524,
|
|
"num_tokens": 5422472.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.485137605667115,
|
|
"epoch": 2.6165494862087613,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004678675515863545,
|
|
"loss": 5.5234,
|
|
"mean_token_accuracy": 0.1755719318985939,
|
|
"num_tokens": 5433958.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.595595932006836,
|
|
"epoch": 2.621957815035154,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004676463072900173,
|
|
"loss": 5.4076,
|
|
"mean_token_accuracy": 0.19072714000940322,
|
|
"num_tokens": 5444343.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.506372499465942,
|
|
"epoch": 2.627366143861547,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00046742436291655415,
|
|
"loss": 5.4181,
|
|
"mean_token_accuracy": 0.17785460203886033,
|
|
"num_tokens": 5456260.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.639331245422364,
|
|
"epoch": 2.6327744726879394,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00046720171927251155,
|
|
"loss": 5.5358,
|
|
"mean_token_accuracy": 0.1824102893471718,
|
|
"num_tokens": 5468281.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.435043001174927,
|
|
"epoch": 2.638182801514332,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004669783771669769,
|
|
"loss": 5.3722,
|
|
"mean_token_accuracy": 0.19723588675260545,
|
|
"num_tokens": 5480068.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.502878093719483,
|
|
"epoch": 2.6435911303407247,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000466754337411576,
|
|
"loss": 5.3648,
|
|
"mean_token_accuracy": 0.18432383686304094,
|
|
"num_tokens": 5490932.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.498383569717407,
|
|
"epoch": 2.6489994591671175,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046652960082046985,
|
|
"loss": 5.4573,
|
|
"mean_token_accuracy": 0.17757296860218047,
|
|
"num_tokens": 5501787.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.602294445037842,
|
|
"epoch": 2.65440778799351,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046630416821035164,
|
|
"loss": 5.4492,
|
|
"mean_token_accuracy": 0.1782757371664047,
|
|
"num_tokens": 5513297.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.5355902194976805,
|
|
"epoch": 2.6598161168199024,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004660780404004441,
|
|
"loss": 5.4396,
|
|
"mean_token_accuracy": 0.18619952499866485,
|
|
"num_tokens": 5523797.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.503725576400757,
|
|
"epoch": 2.6652244456462952,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00046585121821249623,
|
|
"loss": 5.4621,
|
|
"mean_token_accuracy": 0.18713192790746688,
|
|
"num_tokens": 5535808.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.672381591796875,
|
|
"epoch": 2.670632774472688,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00046562370247078025,
|
|
"loss": 5.5562,
|
|
"mean_token_accuracy": 0.1725524201989174,
|
|
"num_tokens": 5548179.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.602239942550659,
|
|
"epoch": 2.6760411032990805,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000465395494002089,
|
|
"loss": 5.5397,
|
|
"mean_token_accuracy": 0.17902407497167588,
|
|
"num_tokens": 5559144.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.63732681274414,
|
|
"epoch": 2.6814494321254734,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004651665936357325,
|
|
"loss": 5.4488,
|
|
"mean_token_accuracy": 0.181016144156456,
|
|
"num_tokens": 5571103.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.510068941116333,
|
|
"epoch": 2.686857760951866,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004649370022035354,
|
|
"loss": 5.3871,
|
|
"mean_token_accuracy": 0.18064596205949784,
|
|
"num_tokens": 5583239.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.466235876083374,
|
|
"epoch": 2.6922660897782587,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004647067205398332,
|
|
"loss": 5.3623,
|
|
"mean_token_accuracy": 0.18435029834508895,
|
|
"num_tokens": 5593602.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.624344062805176,
|
|
"epoch": 2.697674418604651,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046447574948147027,
|
|
"loss": 5.519,
|
|
"mean_token_accuracy": 0.1765885442495346,
|
|
"num_tokens": 5604696.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.555069398880005,
|
|
"epoch": 2.703082747431044,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00046424408986779587,
|
|
"loss": 5.4976,
|
|
"mean_token_accuracy": 0.17446508556604384,
|
|
"num_tokens": 5615668.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.703082747431044,
|
|
"eval_entropy": 5.526033994284544,
|
|
"eval_loss": 5.984447002410889,
|
|
"eval_mean_token_accuracy": 0.1685239829377017,
|
|
"eval_num_tokens": 5615668.0,
|
|
"eval_runtime": 2.5553,
|
|
"eval_samples_per_second": 1375.594,
|
|
"eval_steps_per_second": 172.194,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.563303470611572,
|
|
"epoch": 2.7084910762574363,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004640117425406615,
|
|
"loss": 5.4295,
|
|
"mean_token_accuracy": 0.17890387326478957,
|
|
"num_tokens": 5627344.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.521118402481079,
|
|
"epoch": 2.713899405083829,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000463778708344418,
|
|
"loss": 5.4587,
|
|
"mean_token_accuracy": 0.1767270475625992,
|
|
"num_tokens": 5637472.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.5294698715209964,
|
|
"epoch": 2.7193077339102216,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046354498812591207,
|
|
"loss": 5.46,
|
|
"mean_token_accuracy": 0.1800878256559372,
|
|
"num_tokens": 5647766.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.500148296356201,
|
|
"epoch": 2.7247160627366145,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046331058273448364,
|
|
"loss": 5.3809,
|
|
"mean_token_accuracy": 0.18451676219701768,
|
|
"num_tokens": 5658343.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.51725058555603,
|
|
"epoch": 2.730124391563007,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004630754930219624,
|
|
"loss": 5.3858,
|
|
"mean_token_accuracy": 0.1867602452635765,
|
|
"num_tokens": 5669640.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.384898900985718,
|
|
"epoch": 2.7355327203893998,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004628397198426648,
|
|
"loss": 5.3655,
|
|
"mean_token_accuracy": 0.1731739729642868,
|
|
"num_tokens": 5682421.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.539755153656006,
|
|
"epoch": 2.740941049215792,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046260326405339126,
|
|
"loss": 5.3337,
|
|
"mean_token_accuracy": 0.1887157917022705,
|
|
"num_tokens": 5692797.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.570442628860474,
|
|
"epoch": 2.746349378042185,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004623661265134226,
|
|
"loss": 5.5288,
|
|
"mean_token_accuracy": 0.18444036394357682,
|
|
"num_tokens": 5702779.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.5902628898620605,
|
|
"epoch": 2.7517577068685775,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004621283080845172,
|
|
"loss": 5.4998,
|
|
"mean_token_accuracy": 0.1801348939538002,
|
|
"num_tokens": 5714147.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.488237380981445,
|
|
"epoch": 2.7571660356949703,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00046188980963090783,
|
|
"loss": 5.3275,
|
|
"mean_token_accuracy": 0.18899329304695128,
|
|
"num_tokens": 5725718.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.4276031970977785,
|
|
"epoch": 2.7625743645213627,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046165063201929825,
|
|
"loss": 5.4044,
|
|
"mean_token_accuracy": 0.1894577130675316,
|
|
"num_tokens": 5737441.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.588737821578979,
|
|
"epoch": 2.7679826933477556,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046141077611886074,
|
|
"loss": 5.5026,
|
|
"mean_token_accuracy": 0.18703711032867432,
|
|
"num_tokens": 5748735.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.645930528640747,
|
|
"epoch": 2.773391022174148,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004611702428012319,
|
|
"loss": 5.5659,
|
|
"mean_token_accuracy": 0.17293781340122222,
|
|
"num_tokens": 5760262.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.534254741668701,
|
|
"epoch": 2.778799351000541,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046092903294051057,
|
|
"loss": 5.483,
|
|
"mean_token_accuracy": 0.17904222458601,
|
|
"num_tokens": 5771909.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.440164232254029,
|
|
"epoch": 2.7842076798269337,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046068714741325385,
|
|
"loss": 5.3764,
|
|
"mean_token_accuracy": 0.1812530755996704,
|
|
"num_tokens": 5782028.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.667120456695557,
|
|
"epoch": 2.789616008653326,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004604445870984743,
|
|
"loss": 5.5148,
|
|
"mean_token_accuracy": 0.17830025851726533,
|
|
"num_tokens": 5793641.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.578033447265625,
|
|
"epoch": 2.7950243374797186,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004602013528776367,
|
|
"loss": 5.4543,
|
|
"mean_token_accuracy": 0.1786302700638771,
|
|
"num_tokens": 5805392.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.549567031860351,
|
|
"epoch": 2.8004326663061114,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004599574456346548,
|
|
"loss": 5.447,
|
|
"mean_token_accuracy": 0.1827794209122658,
|
|
"num_tokens": 5816362.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.516026639938355,
|
|
"epoch": 2.8058409951325043,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045971286625588796,
|
|
"loss": 5.3836,
|
|
"mean_token_accuracy": 0.18452833443880082,
|
|
"num_tokens": 5827518.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.594208288192749,
|
|
"epoch": 2.8112493239588967,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004594676156301383,
|
|
"loss": 5.5086,
|
|
"mean_token_accuracy": 0.18318616598844528,
|
|
"num_tokens": 5838455.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.514456796646118,
|
|
"epoch": 2.816657652785289,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045922169464864725,
|
|
"loss": 5.3887,
|
|
"mean_token_accuracy": 0.18833770006895065,
|
|
"num_tokens": 5850065.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.538111162185669,
|
|
"epoch": 2.822065981611682,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000458975104205092,
|
|
"loss": 5.5072,
|
|
"mean_token_accuracy": 0.1778026893734932,
|
|
"num_tokens": 5860444.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.6210051536560055,
|
|
"epoch": 2.827474310438075,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045872784519558304,
|
|
"loss": 5.4841,
|
|
"mean_token_accuracy": 0.17808218896389008,
|
|
"num_tokens": 5873984.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.569654846191407,
|
|
"epoch": 2.8328826392644673,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004584799185186599,
|
|
"loss": 5.5142,
|
|
"mean_token_accuracy": 0.1829293265938759,
|
|
"num_tokens": 5885807.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.507402849197388,
|
|
"epoch": 2.8382909680908597,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045823132507528903,
|
|
"loss": 5.3859,
|
|
"mean_token_accuracy": 0.1851057380437851,
|
|
"num_tokens": 5897122.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.497740459442139,
|
|
"epoch": 2.8436992969172525,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004579820657688593,
|
|
"loss": 5.3956,
|
|
"mean_token_accuracy": 0.18557078689336776,
|
|
"num_tokens": 5909189.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.564215087890625,
|
|
"epoch": 2.8491076257436454,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004577321415051798,
|
|
"loss": 5.5025,
|
|
"mean_token_accuracy": 0.17316450774669648,
|
|
"num_tokens": 5920135.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.596897220611572,
|
|
"epoch": 2.854515954570038,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045748155319247564,
|
|
"loss": 5.4729,
|
|
"mean_token_accuracy": 0.18322434574365615,
|
|
"num_tokens": 5932279.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.46498384475708,
|
|
"epoch": 2.8599242833964302,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004572303017413856,
|
|
"loss": 5.4091,
|
|
"mean_token_accuracy": 0.1800827756524086,
|
|
"num_tokens": 5942082.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.581994104385376,
|
|
"epoch": 2.865332612222823,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004569783880649578,
|
|
"loss": 5.4414,
|
|
"mean_token_accuracy": 0.18471686393022538,
|
|
"num_tokens": 5952856.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.411898756027222,
|
|
"epoch": 2.870740941049216,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004567258130786474,
|
|
"loss": 5.3529,
|
|
"mean_token_accuracy": 0.18910276293754577,
|
|
"num_tokens": 5963937.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.530285358428955,
|
|
"epoch": 2.8761492698756084,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045647257770031215,
|
|
"loss": 5.4204,
|
|
"mean_token_accuracy": 0.18503426760435104,
|
|
"num_tokens": 5974852.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.53133192062378,
|
|
"epoch": 2.8815575987020012,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045621868285021023,
|
|
"loss": 5.439,
|
|
"mean_token_accuracy": 0.18998821079730988,
|
|
"num_tokens": 5985616.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.506141233444214,
|
|
"epoch": 2.8869659275283936,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00045596412945099615,
|
|
"loss": 5.3829,
|
|
"mean_token_accuracy": 0.1894236460328102,
|
|
"num_tokens": 5997199.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.579552412033081,
|
|
"epoch": 2.8923742563547865,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045570891842771727,
|
|
"loss": 5.4504,
|
|
"mean_token_accuracy": 0.18665963858366014,
|
|
"num_tokens": 6008654.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.541787099838257,
|
|
"epoch": 2.897782585181179,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045545305070781135,
|
|
"loss": 5.502,
|
|
"mean_token_accuracy": 0.17659719586372374,
|
|
"num_tokens": 6020662.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.568135452270508,
|
|
"epoch": 2.903190914007572,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045519652722110206,
|
|
"loss": 5.3739,
|
|
"mean_token_accuracy": 0.20091857761144638,
|
|
"num_tokens": 6031793.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.433980703353882,
|
|
"epoch": 2.908599242833964,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045493934889979655,
|
|
"loss": 5.3681,
|
|
"mean_token_accuracy": 0.1835208550095558,
|
|
"num_tokens": 6042181.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.419794702529908,
|
|
"epoch": 2.914007571660357,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045468151667848116,
|
|
"loss": 5.353,
|
|
"mean_token_accuracy": 0.18946526497602462,
|
|
"num_tokens": 6052774.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.553079557418823,
|
|
"epoch": 2.9194159004867495,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000454423031494119,
|
|
"loss": 5.3302,
|
|
"mean_token_accuracy": 0.19462115317583084,
|
|
"num_tokens": 6063273.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.412002182006836,
|
|
"epoch": 2.9248242293131423,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004541638942860458,
|
|
"loss": 5.3746,
|
|
"mean_token_accuracy": 0.18324050307273865,
|
|
"num_tokens": 6075121.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.473644542694092,
|
|
"epoch": 2.9302325581395348,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004539041059959667,
|
|
"loss": 5.3371,
|
|
"mean_token_accuracy": 0.18794260621070863,
|
|
"num_tokens": 6085666.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.504933309555054,
|
|
"epoch": 2.9356408869659276,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045364366756795295,
|
|
"loss": 5.3503,
|
|
"mean_token_accuracy": 0.19261666089296342,
|
|
"num_tokens": 6096619.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.5451802730560305,
|
|
"epoch": 2.94104921579232,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045338257994843845,
|
|
"loss": 5.5144,
|
|
"mean_token_accuracy": 0.177358578145504,
|
|
"num_tokens": 6108800.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.51861047744751,
|
|
"epoch": 2.946457544618713,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004531208440862162,
|
|
"loss": 5.435,
|
|
"mean_token_accuracy": 0.17969423681497573,
|
|
"num_tokens": 6119789.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.540252923965454,
|
|
"epoch": 2.9518658734451053,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045285846093243485,
|
|
"loss": 5.4167,
|
|
"mean_token_accuracy": 0.1801318868994713,
|
|
"num_tokens": 6131786.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.604067373275757,
|
|
"epoch": 2.957274202271498,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045259543144059557,
|
|
"loss": 5.5011,
|
|
"mean_token_accuracy": 0.18367851078510283,
|
|
"num_tokens": 6143012.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.4651960849761965,
|
|
"epoch": 2.9626825310978906,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004523317565665479,
|
|
"loss": 5.3884,
|
|
"mean_token_accuracy": 0.18817291408777237,
|
|
"num_tokens": 6153972.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.580589056015015,
|
|
"epoch": 2.9680908599242835,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045206743726848716,
|
|
"loss": 5.543,
|
|
"mean_token_accuracy": 0.17746709138154984,
|
|
"num_tokens": 6166202.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.424905300140381,
|
|
"epoch": 2.973499188750676,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004518024745069502,
|
|
"loss": 5.2846,
|
|
"mean_token_accuracy": 0.19504003524780272,
|
|
"num_tokens": 6177253.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.553038740158081,
|
|
"epoch": 2.9789075175770687,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004515368692448123,
|
|
"loss": 5.5595,
|
|
"mean_token_accuracy": 0.17424825578927994,
|
|
"num_tokens": 6188857.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.609841537475586,
|
|
"epoch": 2.9843158464034616,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045127062244728353,
|
|
"loss": 5.5072,
|
|
"mean_token_accuracy": 0.17310477495193483,
|
|
"num_tokens": 6199945.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.505488681793213,
|
|
"epoch": 2.989724175229854,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004510037350819055,
|
|
"loss": 5.4901,
|
|
"mean_token_accuracy": 0.1852865383028984,
|
|
"num_tokens": 6209866.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.556007671356201,
|
|
"epoch": 2.9951325040562464,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00045073620811854744,
|
|
"loss": 5.4565,
|
|
"mean_token_accuracy": 0.1760351225733757,
|
|
"num_tokens": 6221636.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.51098993089464,
|
|
"epoch": 3.0,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 0.00045046804252940294,
|
|
"loss": 5.3473,
|
|
"mean_token_accuracy": 0.1814789358112547,
|
|
"num_tokens": 6230667.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.506106376647949,
|
|
"epoch": 3.005408328826393,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045019923928898644,
|
|
"loss": 5.0946,
|
|
"mean_token_accuracy": 0.20731269121170043,
|
|
"num_tokens": 6242242.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.546751260757446,
|
|
"epoch": 3.0108166576527853,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004499297993741294,
|
|
"loss": 5.1412,
|
|
"mean_token_accuracy": 0.19812405854463577,
|
|
"num_tokens": 6253337.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.417823028564453,
|
|
"epoch": 3.016224986479178,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004496597237639773,
|
|
"loss": 5.0159,
|
|
"mean_token_accuracy": 0.20064211934804915,
|
|
"num_tokens": 6264582.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.45553674697876,
|
|
"epoch": 3.0216333153055706,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00044938901343998536,
|
|
"loss": 5.0643,
|
|
"mean_token_accuracy": 0.20520399361848832,
|
|
"num_tokens": 6276172.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.421815633773804,
|
|
"epoch": 3.0270416441319634,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004491176693859156,
|
|
"loss": 4.9369,
|
|
"mean_token_accuracy": 0.2152072533965111,
|
|
"num_tokens": 6286038.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.399924230575562,
|
|
"epoch": 3.032449972958356,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004488456925878329,
|
|
"loss": 5.0239,
|
|
"mean_token_accuracy": 0.20057855397462845,
|
|
"num_tokens": 6296546.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.419568252563477,
|
|
"epoch": 3.0378583017847487,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044857308403410163,
|
|
"loss": 5.1143,
|
|
"mean_token_accuracy": 0.209441077709198,
|
|
"num_tokens": 6308001.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.440928983688354,
|
|
"epoch": 3.043266630611141,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00044829984471538183,
|
|
"loss": 5.1133,
|
|
"mean_token_accuracy": 0.2006756529211998,
|
|
"num_tokens": 6320896.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.482006931304932,
|
|
"epoch": 3.048674959437534,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004480259756246258,
|
|
"loss": 5.1682,
|
|
"mean_token_accuracy": 0.1964373841881752,
|
|
"num_tokens": 6332710.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.423978805541992,
|
|
"epoch": 3.0540832882639264,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00044775147775707465,
|
|
"loss": 5.0418,
|
|
"mean_token_accuracy": 0.20230309367179872,
|
|
"num_tokens": 6343563.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.4204973697662355,
|
|
"epoch": 3.0594916170903192,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000447476352110254,
|
|
"loss": 5.1584,
|
|
"mean_token_accuracy": 0.19732941836118698,
|
|
"num_tokens": 6354064.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.394972229003907,
|
|
"epoch": 3.0648999459167117,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004472005996839712,
|
|
"loss": 5.0709,
|
|
"mean_token_accuracy": 0.19945697784423827,
|
|
"num_tokens": 6366551.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.371446132659912,
|
|
"epoch": 3.0703082747431045,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004469242214803112,
|
|
"loss": 5.0048,
|
|
"mean_token_accuracy": 0.20002005994319916,
|
|
"num_tokens": 6377086.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.435087013244629,
|
|
"epoch": 3.075716603569497,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004466472185036331,
|
|
"loss": 5.1211,
|
|
"mean_token_accuracy": 0.2017862692475319,
|
|
"num_tokens": 6388795.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.459153413772583,
|
|
"epoch": 3.08112493239589,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004463695917605663,
|
|
"loss": 5.1073,
|
|
"mean_token_accuracy": 0.1999192863702774,
|
|
"num_tokens": 6400174.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.364331817626953,
|
|
"epoch": 3.086533261222282,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000446091342260007,
|
|
"loss": 5.0374,
|
|
"mean_token_accuracy": 0.20255780816078187,
|
|
"num_tokens": 6411297.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.463997936248779,
|
|
"epoch": 3.091941590048675,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004458124710131145,
|
|
"loss": 5.0717,
|
|
"mean_token_accuracy": 0.20193164944648742,
|
|
"num_tokens": 6423487.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.49720196723938,
|
|
"epoch": 3.0973499188750675,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00044553297903330766,
|
|
"loss": 5.2802,
|
|
"mean_token_accuracy": 0.188761006295681,
|
|
"num_tokens": 6435460.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.497384643554687,
|
|
"epoch": 3.1027582477014604,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00044525286733626085,
|
|
"loss": 5.1985,
|
|
"mean_token_accuracy": 0.20143782645463942,
|
|
"num_tokens": 6445664.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.453384685516357,
|
|
"epoch": 3.1081665765278528,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004449721369399008,
|
|
"loss": 5.0726,
|
|
"mean_token_accuracy": 0.19608601778745652,
|
|
"num_tokens": 6457182.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.319879484176636,
|
|
"epoch": 3.1135749053542456,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00044469078886440227,
|
|
"loss": 4.9311,
|
|
"mean_token_accuracy": 0.21415926665067672,
|
|
"num_tokens": 6468378.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.343447351455689,
|
|
"epoch": 3.118983234180638,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00044440882413218484,
|
|
"loss": 5.1541,
|
|
"mean_token_accuracy": 0.19760852009058,
|
|
"num_tokens": 6479625.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.356214094161987,
|
|
"epoch": 3.124391563007031,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044412624376790905,
|
|
"loss": 5.0665,
|
|
"mean_token_accuracy": 0.20471785217523575,
|
|
"num_tokens": 6490713.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.356153106689453,
|
|
"epoch": 3.1297998918334233,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004438430487984726,
|
|
"loss": 5.1249,
|
|
"mean_token_accuracy": 0.18504695147275924,
|
|
"num_tokens": 6502161.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.420809507369995,
|
|
"epoch": 3.135208220659816,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004435592402530066,
|
|
"loss": 5.0217,
|
|
"mean_token_accuracy": 0.20395056903362274,
|
|
"num_tokens": 6513639.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.342472314834595,
|
|
"epoch": 3.1406165494862086,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000443274819162872,
|
|
"loss": 5.0657,
|
|
"mean_token_accuracy": 0.20275074392557144,
|
|
"num_tokens": 6524893.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.333340501785278,
|
|
"epoch": 3.1460248783126015,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004429897865616557,
|
|
"loss": 5.0517,
|
|
"mean_token_accuracy": 0.20374273657798767,
|
|
"num_tokens": 6535797.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.416862916946411,
|
|
"epoch": 3.151433207138994,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004427041434851668,
|
|
"loss": 5.1532,
|
|
"mean_token_accuracy": 0.19501399844884873,
|
|
"num_tokens": 6547117.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.421797132492065,
|
|
"epoch": 3.1568415359653867,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00044241789097143293,
|
|
"loss": 5.1805,
|
|
"mean_token_accuracy": 0.19888945370912553,
|
|
"num_tokens": 6558771.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.40197229385376,
|
|
"epoch": 3.162249864791779,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044213103006069635,
|
|
"loss": 5.1443,
|
|
"mean_token_accuracy": 0.19872353821992875,
|
|
"num_tokens": 6569922.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.459195470809936,
|
|
"epoch": 3.167658193618172,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00044184356179541035,
|
|
"loss": 5.1423,
|
|
"mean_token_accuracy": 0.19692348688840866,
|
|
"num_tokens": 6580561.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.343637275695801,
|
|
"epoch": 3.1730665224445644,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004415554872202352,
|
|
"loss": 4.9686,
|
|
"mean_token_accuracy": 0.20476706624031066,
|
|
"num_tokens": 6591636.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.405619382858276,
|
|
"epoch": 3.1784748512709573,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00044126680738203466,
|
|
"loss": 5.2043,
|
|
"mean_token_accuracy": 0.19140468090772628,
|
|
"num_tokens": 6603736.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.478578901290893,
|
|
"epoch": 3.1838831800973497,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004409775233298718,
|
|
"loss": 5.1182,
|
|
"mean_token_accuracy": 0.19861625432968139,
|
|
"num_tokens": 6615415.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.388946771621704,
|
|
"epoch": 3.1892915089237426,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004406876361150056,
|
|
"loss": 5.1703,
|
|
"mean_token_accuracy": 0.19311941415071487,
|
|
"num_tokens": 6625970.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.434563112258911,
|
|
"epoch": 3.194699837750135,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004403971467908869,
|
|
"loss": 5.1058,
|
|
"mean_token_accuracy": 0.19681683033704758,
|
|
"num_tokens": 6636326.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.380080366134644,
|
|
"epoch": 3.200108166576528,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004401060564131545,
|
|
"loss": 5.1694,
|
|
"mean_token_accuracy": 0.19399933516979218,
|
|
"num_tokens": 6646825.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.427671909332275,
|
|
"epoch": 3.2055164954029207,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004398143660396315,
|
|
"loss": 5.1435,
|
|
"mean_token_accuracy": 0.2005398988723755,
|
|
"num_tokens": 6657748.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.312004518508911,
|
|
"epoch": 3.210924824229313,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004395220767303215,
|
|
"loss": 5.0449,
|
|
"mean_token_accuracy": 0.20731872618198394,
|
|
"num_tokens": 6668422.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.440751314163208,
|
|
"epoch": 3.2163331530557056,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004392291895474045,
|
|
"loss": 5.1414,
|
|
"mean_token_accuracy": 0.19433858394622802,
|
|
"num_tokens": 6678898.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.420091581344605,
|
|
"epoch": 3.2217414818820984,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004389357055552331,
|
|
"loss": 5.1619,
|
|
"mean_token_accuracy": 0.20023332685232162,
|
|
"num_tokens": 6688691.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.3957414627075195,
|
|
"epoch": 3.2271498107084913,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00043864162582032897,
|
|
"loss": 5.0428,
|
|
"mean_token_accuracy": 0.20716743916273117,
|
|
"num_tokens": 6700279.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.362940406799316,
|
|
"epoch": 3.2325581395348837,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004383469514113784,
|
|
"loss": 5.0068,
|
|
"mean_token_accuracy": 0.21174349635839462,
|
|
"num_tokens": 6710440.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.326319885253906,
|
|
"epoch": 3.2379664683612766,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004380516833992289,
|
|
"loss": 5.1627,
|
|
"mean_token_accuracy": 0.19876041412353515,
|
|
"num_tokens": 6722347.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.419641590118408,
|
|
"epoch": 3.243374797187669,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000437755822856885,
|
|
"loss": 5.1903,
|
|
"mean_token_accuracy": 0.1981905683875084,
|
|
"num_tokens": 6732936.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 3.243374797187669,
|
|
"eval_entropy": 5.393885960362175,
|
|
"eval_loss": 5.970073699951172,
|
|
"eval_mean_token_accuracy": 0.17068068043074824,
|
|
"eval_num_tokens": 6732936.0,
|
|
"eval_runtime": 2.5585,
|
|
"eval_samples_per_second": 1373.842,
|
|
"eval_steps_per_second": 171.975,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.459097576141358,
|
|
"epoch": 3.248783126014062,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004374593708595049,
|
|
"loss": 5.2064,
|
|
"mean_token_accuracy": 0.19572709053754805,
|
|
"num_tokens": 6744552.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.411734962463379,
|
|
"epoch": 3.2541914548404542,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00043716232848439555,
|
|
"loss": 5.1361,
|
|
"mean_token_accuracy": 0.20123752653598787,
|
|
"num_tokens": 6755749.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.3295823097229,
|
|
"epoch": 3.259599783666847,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00043686469681100985,
|
|
"loss": 5.059,
|
|
"mean_token_accuracy": 0.20840994417667388,
|
|
"num_tokens": 6766119.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.333610153198242,
|
|
"epoch": 3.2650081124932395,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00043656647692094186,
|
|
"loss": 5.1312,
|
|
"mean_token_accuracy": 0.19330950379371642,
|
|
"num_tokens": 6776491.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.5096197605133055,
|
|
"epoch": 3.2704164413196324,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004362676698979233,
|
|
"loss": 5.2314,
|
|
"mean_token_accuracy": 0.19391171634197235,
|
|
"num_tokens": 6788795.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.43352255821228,
|
|
"epoch": 3.275824770146025,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00043596827682781974,
|
|
"loss": 5.2119,
|
|
"mean_token_accuracy": 0.20098159462213516,
|
|
"num_tokens": 6799706.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.399108934402466,
|
|
"epoch": 3.2812330989724177,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004356682987986262,
|
|
"loss": 5.1505,
|
|
"mean_token_accuracy": 0.19465389251708984,
|
|
"num_tokens": 6812454.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.439399337768554,
|
|
"epoch": 3.28664142779881,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004353677369004635,
|
|
"loss": 5.0936,
|
|
"mean_token_accuracy": 0.19978901445865632,
|
|
"num_tokens": 6823950.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.429735565185547,
|
|
"epoch": 3.292049756625203,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00043506659222557426,
|
|
"loss": 5.2819,
|
|
"mean_token_accuracy": 0.18565253615379335,
|
|
"num_tokens": 6835028.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.41392822265625,
|
|
"epoch": 3.2974580854515954,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000434764865868319,
|
|
"loss": 5.1731,
|
|
"mean_token_accuracy": 0.1960723280906677,
|
|
"num_tokens": 6846316.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.354257297515869,
|
|
"epoch": 3.302866414277988,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000434462558925172,
|
|
"loss": 5.1006,
|
|
"mean_token_accuracy": 0.20529791861772537,
|
|
"num_tokens": 6856463.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.409739351272583,
|
|
"epoch": 3.3082747431043806,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00043415967249471734,
|
|
"loss": 5.0998,
|
|
"mean_token_accuracy": 0.20486413985490798,
|
|
"num_tokens": 6867869.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.390873813629151,
|
|
"epoch": 3.3136830719307735,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000433856207677645,
|
|
"loss": 5.1002,
|
|
"mean_token_accuracy": 0.1998838797211647,
|
|
"num_tokens": 6878545.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.319905614852905,
|
|
"epoch": 3.319091400757166,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000433552165576747,
|
|
"loss": 5.1641,
|
|
"mean_token_accuracy": 0.1956935003399849,
|
|
"num_tokens": 6889399.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.46302604675293,
|
|
"epoch": 3.3244997295835588,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00043324754729691275,
|
|
"loss": 5.1322,
|
|
"mean_token_accuracy": 0.20623356848955154,
|
|
"num_tokens": 6900428.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.484996366500854,
|
|
"epoch": 3.329908058409951,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00043294235394512594,
|
|
"loss": 5.2547,
|
|
"mean_token_accuracy": 0.19137114733457566,
|
|
"num_tokens": 6911112.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.352552700042724,
|
|
"epoch": 3.335316387236344,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004326365866304599,
|
|
"loss": 5.0247,
|
|
"mean_token_accuracy": 0.20902705043554307,
|
|
"num_tokens": 6922383.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.348286581039429,
|
|
"epoch": 3.3407247160627365,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004323302464640738,
|
|
"loss": 5.2128,
|
|
"mean_token_accuracy": 0.19261810034513474,
|
|
"num_tokens": 6935153.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.536054515838623,
|
|
"epoch": 3.3461330448891293,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00043202333455920843,
|
|
"loss": 5.2517,
|
|
"mean_token_accuracy": 0.19047711789608002,
|
|
"num_tokens": 6946814.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.337420845031739,
|
|
"epoch": 3.3515413737155217,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004317158520311824,
|
|
"loss": 5.0838,
|
|
"mean_token_accuracy": 0.20277185887098312,
|
|
"num_tokens": 6957230.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.272858572006226,
|
|
"epoch": 3.3569497025419146,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004314077999973879,
|
|
"loss": 5.1203,
|
|
"mean_token_accuracy": 0.19302735328674317,
|
|
"num_tokens": 6968060.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.422040939331055,
|
|
"epoch": 3.362358031368307,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004310991795772869,
|
|
"loss": 5.0164,
|
|
"mean_token_accuracy": 0.2077233999967575,
|
|
"num_tokens": 6977782.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.367076301574707,
|
|
"epoch": 3.3677663601947,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004307899918924065,
|
|
"loss": 5.106,
|
|
"mean_token_accuracy": 0.1994634449481964,
|
|
"num_tokens": 6988252.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.418995189666748,
|
|
"epoch": 3.3731746890210923,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004304802380663356,
|
|
"loss": 5.1286,
|
|
"mean_token_accuracy": 0.19716998785734177,
|
|
"num_tokens": 6998523.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.395699501037598,
|
|
"epoch": 3.378583017847485,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00043016991922472015,
|
|
"loss": 5.1599,
|
|
"mean_token_accuracy": 0.19909669011831282,
|
|
"num_tokens": 7010058.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.397343540191651,
|
|
"epoch": 3.383991346673878,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00042985903649525977,
|
|
"loss": 5.2652,
|
|
"mean_token_accuracy": 0.18819963783025742,
|
|
"num_tokens": 7021289.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.5298584461212155,
|
|
"epoch": 3.3893996755002704,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00042954759100770275,
|
|
"loss": 5.278,
|
|
"mean_token_accuracy": 0.18610745072364807,
|
|
"num_tokens": 7032649.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.395885562896728,
|
|
"epoch": 3.394808004326663,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00042923558389384303,
|
|
"loss": 5.1412,
|
|
"mean_token_accuracy": 0.20260152965784073,
|
|
"num_tokens": 7044120.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.3889978408813475,
|
|
"epoch": 3.4002163331530557,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004289230162875149,
|
|
"loss": 5.162,
|
|
"mean_token_accuracy": 0.19930318593978882,
|
|
"num_tokens": 7054526.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.254993057250976,
|
|
"epoch": 3.4056246619794486,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00042860988932458987,
|
|
"loss": 5.0222,
|
|
"mean_token_accuracy": 0.2123810201883316,
|
|
"num_tokens": 7064744.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.4115931510925295,
|
|
"epoch": 3.411032990805841,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004282962041429721,
|
|
"loss": 5.1335,
|
|
"mean_token_accuracy": 0.19928616732358934,
|
|
"num_tokens": 7076360.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.389709711074829,
|
|
"epoch": 3.4164413196322334,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00042798196188259427,
|
|
"loss": 5.1625,
|
|
"mean_token_accuracy": 0.2029835984110832,
|
|
"num_tokens": 7088114.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.413705825805664,
|
|
"epoch": 3.4218496484586263,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004276671636854133,
|
|
"loss": 5.1856,
|
|
"mean_token_accuracy": 0.1875985786318779,
|
|
"num_tokens": 7099125.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.387046480178833,
|
|
"epoch": 3.427257977285019,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00042735181069540677,
|
|
"loss": 5.122,
|
|
"mean_token_accuracy": 0.2014174282550812,
|
|
"num_tokens": 7109875.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.418265676498413,
|
|
"epoch": 3.4326663061114115,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004270359040585681,
|
|
"loss": 5.1804,
|
|
"mean_token_accuracy": 0.19509298503398895,
|
|
"num_tokens": 7121388.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.405701160430908,
|
|
"epoch": 3.4380746349378044,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00042671944492290277,
|
|
"loss": 5.2067,
|
|
"mean_token_accuracy": 0.1919805273413658,
|
|
"num_tokens": 7134161.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.378969001770019,
|
|
"epoch": 3.443482963764197,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004264024344384239,
|
|
"loss": 5.2012,
|
|
"mean_token_accuracy": 0.19525598734617233,
|
|
"num_tokens": 7146580.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.419828271865844,
|
|
"epoch": 3.4488912925905897,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00042608487375714847,
|
|
"loss": 5.143,
|
|
"mean_token_accuracy": 0.19293057769536973,
|
|
"num_tokens": 7156969.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.326986169815063,
|
|
"epoch": 3.454299621416982,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00042576676403309265,
|
|
"loss": 5.0841,
|
|
"mean_token_accuracy": 0.207579605281353,
|
|
"num_tokens": 7168687.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.4504303455352785,
|
|
"epoch": 3.459707950243375,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004254481064222679,
|
|
"loss": 5.1642,
|
|
"mean_token_accuracy": 0.2007901191711426,
|
|
"num_tokens": 7179996.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.407182025909424,
|
|
"epoch": 3.4651162790697674,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00042512890208267677,
|
|
"loss": 5.2014,
|
|
"mean_token_accuracy": 0.19162100702524185,
|
|
"num_tokens": 7191214.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.455952405929565,
|
|
"epoch": 3.4705246078961602,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00042480915217430853,
|
|
"loss": 5.2122,
|
|
"mean_token_accuracy": 0.1950717955827713,
|
|
"num_tokens": 7203491.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.438957786560058,
|
|
"epoch": 3.4759329367225527,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00042448885785913507,
|
|
"loss": 5.1465,
|
|
"mean_token_accuracy": 0.19635559171438216,
|
|
"num_tokens": 7215018.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.4492463111877445,
|
|
"epoch": 3.4813412655489455,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00042416802030110677,
|
|
"loss": 5.0973,
|
|
"mean_token_accuracy": 0.2053507462143898,
|
|
"num_tokens": 7226209.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.361258220672608,
|
|
"epoch": 3.486749594375338,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00042384664066614787,
|
|
"loss": 5.1146,
|
|
"mean_token_accuracy": 0.199769027531147,
|
|
"num_tokens": 7236505.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.307104682922363,
|
|
"epoch": 3.492157923201731,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004235247201221528,
|
|
"loss": 5.0188,
|
|
"mean_token_accuracy": 0.19949624687433243,
|
|
"num_tokens": 7247414.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.399760103225708,
|
|
"epoch": 3.497566252028123,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004232022598389816,
|
|
"loss": 5.225,
|
|
"mean_token_accuracy": 0.195807746052742,
|
|
"num_tokens": 7258666.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.414032936096191,
|
|
"epoch": 3.502974580854516,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004228792609884557,
|
|
"loss": 5.1309,
|
|
"mean_token_accuracy": 0.20133870989084243,
|
|
"num_tokens": 7270201.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.411387491226196,
|
|
"epoch": 3.5083829096809085,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00042255572474435364,
|
|
"loss": 5.2206,
|
|
"mean_token_accuracy": 0.19567420333623886,
|
|
"num_tokens": 7282254.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.441634464263916,
|
|
"epoch": 3.5137912385073014,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004222316522824069,
|
|
"loss": 5.2237,
|
|
"mean_token_accuracy": 0.19139660000801087,
|
|
"num_tokens": 7294579.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.376083612442017,
|
|
"epoch": 3.5191995673336938,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004219070447802955,
|
|
"loss": 5.1961,
|
|
"mean_token_accuracy": 0.19730623066425323,
|
|
"num_tokens": 7305822.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.454420709609986,
|
|
"epoch": 3.5246078961600866,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00042158190341764384,
|
|
"loss": 5.2024,
|
|
"mean_token_accuracy": 0.19293610453605653,
|
|
"num_tokens": 7317543.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.370056962966919,
|
|
"epoch": 3.530016224986479,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00042125622937601645,
|
|
"loss": 5.1405,
|
|
"mean_token_accuracy": 0.19678603410720824,
|
|
"num_tokens": 7327804.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.4196186542510985,
|
|
"epoch": 3.535424553812872,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004209300238389135,
|
|
"loss": 5.1081,
|
|
"mean_token_accuracy": 0.20407064706087114,
|
|
"num_tokens": 7337045.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.3320306777954105,
|
|
"epoch": 3.5408328826392643,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004206032879917667,
|
|
"loss": 5.1991,
|
|
"mean_token_accuracy": 0.20033071488142012,
|
|
"num_tokens": 7348184.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.401216077804565,
|
|
"epoch": 3.546241211465657,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004202760230219348,
|
|
"loss": 5.0868,
|
|
"mean_token_accuracy": 0.2016289561986923,
|
|
"num_tokens": 7359571.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.3284759521484375,
|
|
"epoch": 3.5516495402920496,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00041994823011869955,
|
|
"loss": 5.1288,
|
|
"mean_token_accuracy": 0.19803138077259064,
|
|
"num_tokens": 7370429.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.361565113067627,
|
|
"epoch": 3.5570578691184425,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00041961991047326116,
|
|
"loss": 5.065,
|
|
"mean_token_accuracy": 0.1963614195585251,
|
|
"num_tokens": 7381569.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.345052814483642,
|
|
"epoch": 3.5624661979448353,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004192910652787338,
|
|
"loss": 5.0864,
|
|
"mean_token_accuracy": 0.20247837901115417,
|
|
"num_tokens": 7392746.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.3618645668029785,
|
|
"epoch": 3.5678745267712277,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004189616957301419,
|
|
"loss": 5.1937,
|
|
"mean_token_accuracy": 0.18934139460325242,
|
|
"num_tokens": 7405520.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.332915019989014,
|
|
"epoch": 3.57328285559762,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004186318030244149,
|
|
"loss": 5.1438,
|
|
"mean_token_accuracy": 0.19418755769729615,
|
|
"num_tokens": 7417788.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.413018369674683,
|
|
"epoch": 3.578691184424013,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00041830138836038383,
|
|
"loss": 5.2202,
|
|
"mean_token_accuracy": 0.1925087407231331,
|
|
"num_tokens": 7428912.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.378791809082031,
|
|
"epoch": 3.584099513250406,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004179704529387763,
|
|
"loss": 5.0853,
|
|
"mean_token_accuracy": 0.20182125121355057,
|
|
"num_tokens": 7439797.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.316334438323975,
|
|
"epoch": 3.5895078420767983,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004176389979622125,
|
|
"loss": 5.0871,
|
|
"mean_token_accuracy": 0.2002997487783432,
|
|
"num_tokens": 7451755.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.362658929824829,
|
|
"epoch": 3.5949161709031907,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004173070246352003,
|
|
"loss": 5.1791,
|
|
"mean_token_accuracy": 0.19661410450935363,
|
|
"num_tokens": 7463393.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.31440224647522,
|
|
"epoch": 3.6003244997295836,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00041697453416413184,
|
|
"loss": 5.1128,
|
|
"mean_token_accuracy": 0.19766476452350618,
|
|
"num_tokens": 7474404.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.418523120880127,
|
|
"epoch": 3.6057328285559764,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000416641527757278,
|
|
"loss": 5.1158,
|
|
"mean_token_accuracy": 0.19920457750558854,
|
|
"num_tokens": 7485495.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.292885065078735,
|
|
"epoch": 3.611141157382369,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00041630800662478514,
|
|
"loss": 5.0863,
|
|
"mean_token_accuracy": 0.20488675236701964,
|
|
"num_tokens": 7497305.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.298235464096069,
|
|
"epoch": 3.6165494862087613,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00041597397197866957,
|
|
"loss": 5.0363,
|
|
"mean_token_accuracy": 0.2036196172237396,
|
|
"num_tokens": 7509008.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.274233770370484,
|
|
"epoch": 3.621957815035154,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004156394250328141,
|
|
"loss": 5.0985,
|
|
"mean_token_accuracy": 0.20638637542724608,
|
|
"num_tokens": 7520079.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.359864997863769,
|
|
"epoch": 3.627366143861547,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000415304367002963,
|
|
"loss": 5.1185,
|
|
"mean_token_accuracy": 0.2037244990468025,
|
|
"num_tokens": 7531294.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.365013694763183,
|
|
"epoch": 3.6327744726879394,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00041496879910671796,
|
|
"loss": 5.136,
|
|
"mean_token_accuracy": 0.19820253551006317,
|
|
"num_tokens": 7542508.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.344226741790772,
|
|
"epoch": 3.638182801514332,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004146327225635336,
|
|
"loss": 5.073,
|
|
"mean_token_accuracy": 0.2047281250357628,
|
|
"num_tokens": 7553288.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.363565349578858,
|
|
"epoch": 3.6435911303407247,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004142961385947127,
|
|
"loss": 5.2884,
|
|
"mean_token_accuracy": 0.1972485288977623,
|
|
"num_tokens": 7564532.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.364321899414063,
|
|
"epoch": 3.6489994591671175,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00041395904842340215,
|
|
"loss": 5.1007,
|
|
"mean_token_accuracy": 0.20422061532735825,
|
|
"num_tokens": 7575109.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.286902332305909,
|
|
"epoch": 3.65440778799351,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004136214532745887,
|
|
"loss": 5.0942,
|
|
"mean_token_accuracy": 0.20465875118970872,
|
|
"num_tokens": 7585871.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.39935941696167,
|
|
"epoch": 3.6598161168199024,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004132833543750936,
|
|
"loss": 5.1727,
|
|
"mean_token_accuracy": 0.2083156079053879,
|
|
"num_tokens": 7596821.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.314902496337891,
|
|
"epoch": 3.6652244456462952,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00041294475295356917,
|
|
"loss": 5.1135,
|
|
"mean_token_accuracy": 0.19837460070848464,
|
|
"num_tokens": 7608101.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.341395854949951,
|
|
"epoch": 3.670632774472688,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004126056502404937,
|
|
"loss": 5.1137,
|
|
"mean_token_accuracy": 0.19502026587724686,
|
|
"num_tokens": 7618722.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.316238498687744,
|
|
"epoch": 3.6760411032990805,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004122660474681672,
|
|
"loss": 5.1108,
|
|
"mean_token_accuracy": 0.20419516861438752,
|
|
"num_tokens": 7630061.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.344601106643677,
|
|
"epoch": 3.6814494321254734,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00041192594587070706,
|
|
"loss": 5.1088,
|
|
"mean_token_accuracy": 0.20161224156618118,
|
|
"num_tokens": 7641830.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.339562797546387,
|
|
"epoch": 3.686857760951866,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00041158534668404325,
|
|
"loss": 5.1477,
|
|
"mean_token_accuracy": 0.19715045839548112,
|
|
"num_tokens": 7652750.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.273939752578736,
|
|
"epoch": 3.6922660897782587,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00041124425114591386,
|
|
"loss": 5.048,
|
|
"mean_token_accuracy": 0.2082615986466408,
|
|
"num_tokens": 7663405.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.267064046859741,
|
|
"epoch": 3.697674418604651,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00041090266049586107,
|
|
"loss": 5.0732,
|
|
"mean_token_accuracy": 0.19478354305028917,
|
|
"num_tokens": 7673958.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.307836246490479,
|
|
"epoch": 3.703082747431044,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004105605759752258,
|
|
"loss": 5.0806,
|
|
"mean_token_accuracy": 0.19744422733783723,
|
|
"num_tokens": 7685475.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.268500423431396,
|
|
"epoch": 3.7084910762574363,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00041021799882714436,
|
|
"loss": 5.0679,
|
|
"mean_token_accuracy": 0.20464238822460173,
|
|
"num_tokens": 7697296.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.330860757827759,
|
|
"epoch": 3.713899405083829,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004098749302965426,
|
|
"loss": 5.2124,
|
|
"mean_token_accuracy": 0.19693688601255416,
|
|
"num_tokens": 7709414.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.386068534851074,
|
|
"epoch": 3.7193077339102216,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004095313716301324,
|
|
"loss": 5.1855,
|
|
"mean_token_accuracy": 0.19180702716112136,
|
|
"num_tokens": 7720965.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.364685916900635,
|
|
"epoch": 3.7247160627366145,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004091873240764069,
|
|
"loss": 5.1401,
|
|
"mean_token_accuracy": 0.19018883407115936,
|
|
"num_tokens": 7733044.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.419634675979614,
|
|
"epoch": 3.730124391563007,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00040884278888563566,
|
|
"loss": 5.1377,
|
|
"mean_token_accuracy": 0.19863850623369217,
|
|
"num_tokens": 7743878.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.290865468978882,
|
|
"epoch": 3.7355327203893998,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00040849776730986014,
|
|
"loss": 5.1108,
|
|
"mean_token_accuracy": 0.20561688989400864,
|
|
"num_tokens": 7755445.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.371950769424439,
|
|
"epoch": 3.740941049215792,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004081522606028899,
|
|
"loss": 5.2033,
|
|
"mean_token_accuracy": 0.1917797088623047,
|
|
"num_tokens": 7765990.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.292093944549561,
|
|
"epoch": 3.746349378042185,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00040780627002029674,
|
|
"loss": 4.9907,
|
|
"mean_token_accuracy": 0.21034044474363328,
|
|
"num_tokens": 7777064.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.210793590545654,
|
|
"epoch": 3.7517577068685775,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004074597968194116,
|
|
"loss": 5.0367,
|
|
"mean_token_accuracy": 0.20004954785108567,
|
|
"num_tokens": 7788618.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.301958799362183,
|
|
"epoch": 3.7571660356949703,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004071128422593186,
|
|
"loss": 5.1675,
|
|
"mean_token_accuracy": 0.19086567163467408,
|
|
"num_tokens": 7798484.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.372506332397461,
|
|
"epoch": 3.7625743645213627,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004067654076008515,
|
|
"loss": 5.126,
|
|
"mean_token_accuracy": 0.19856498688459395,
|
|
"num_tokens": 7810270.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.328622817993164,
|
|
"epoch": 3.7679826933477556,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00040641749410658877,
|
|
"loss": 5.2249,
|
|
"mean_token_accuracy": 0.19223035424947738,
|
|
"num_tokens": 7822616.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.363778209686279,
|
|
"epoch": 3.773391022174148,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004060691030408487,
|
|
"loss": 5.1275,
|
|
"mean_token_accuracy": 0.20397243201732634,
|
|
"num_tokens": 7834933.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.373229455947876,
|
|
"epoch": 3.778799351000541,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00040572023566968536,
|
|
"loss": 5.1433,
|
|
"mean_token_accuracy": 0.20085810124874115,
|
|
"num_tokens": 7847686.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.274054431915284,
|
|
"epoch": 3.7842076798269337,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00040537089326088345,
|
|
"loss": 5.1251,
|
|
"mean_token_accuracy": 0.19822696596384048,
|
|
"num_tokens": 7859087.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 3.7842076798269337,
|
|
"eval_entropy": 5.343244110454212,
|
|
"eval_loss": 5.908792018890381,
|
|
"eval_mean_token_accuracy": 0.17334189984778112,
|
|
"eval_num_tokens": 7859087.0,
|
|
"eval_runtime": 2.543,
|
|
"eval_samples_per_second": 1382.245,
|
|
"eval_steps_per_second": 173.026,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.419901657104492,
|
|
"epoch": 3.789616008653326,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004050210770839541,
|
|
"loss": 5.2018,
|
|
"mean_token_accuracy": 0.19709425717592238,
|
|
"num_tokens": 7870684.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.391724538803101,
|
|
"epoch": 3.7950243374797186,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004046707884101303,
|
|
"loss": 5.1274,
|
|
"mean_token_accuracy": 0.20245471149682998,
|
|
"num_tokens": 7880017.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.239365196228027,
|
|
"epoch": 3.8004326663061114,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00040432002851236164,
|
|
"loss": 5.058,
|
|
"mean_token_accuracy": 0.20143700689077376,
|
|
"num_tokens": 7891751.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.371563720703125,
|
|
"epoch": 3.8058409951325043,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004039687986653106,
|
|
"loss": 5.1819,
|
|
"mean_token_accuracy": 0.19514155238866807,
|
|
"num_tokens": 7902602.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.308206176757812,
|
|
"epoch": 3.8112493239588967,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00040361710014534714,
|
|
"loss": 5.0894,
|
|
"mean_token_accuracy": 0.1993850812315941,
|
|
"num_tokens": 7913548.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.422957563400269,
|
|
"epoch": 3.816657652785289,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004032649342305445,
|
|
"loss": 5.1082,
|
|
"mean_token_accuracy": 0.19650279581546784,
|
|
"num_tokens": 7925042.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.4093037128448485,
|
|
"epoch": 3.822065981611682,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004029123022006745,
|
|
"loss": 5.1712,
|
|
"mean_token_accuracy": 0.19517712146043778,
|
|
"num_tokens": 7937113.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.306524038314819,
|
|
"epoch": 3.827474310438075,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004025592053372027,
|
|
"loss": 5.089,
|
|
"mean_token_accuracy": 0.20707519501447677,
|
|
"num_tokens": 7947796.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.336840391159058,
|
|
"epoch": 3.8328826392644673,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004022056449232839,
|
|
"loss": 5.1356,
|
|
"mean_token_accuracy": 0.19625027924776078,
|
|
"num_tokens": 7959642.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.396138381958008,
|
|
"epoch": 3.8382909680908597,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004018516222437575,
|
|
"loss": 5.0876,
|
|
"mean_token_accuracy": 0.1988441213965416,
|
|
"num_tokens": 7969996.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.2749635696411135,
|
|
"epoch": 3.8436992969172525,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004014971385851428,
|
|
"loss": 5.0694,
|
|
"mean_token_accuracy": 0.20175705403089522,
|
|
"num_tokens": 7980079.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.3151074886322025,
|
|
"epoch": 3.8491076257436454,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000401142195235634,
|
|
"loss": 5.1467,
|
|
"mean_token_accuracy": 0.20246297121047974,
|
|
"num_tokens": 7990453.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.427642774581909,
|
|
"epoch": 3.854515954570038,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004007867934850963,
|
|
"loss": 5.2152,
|
|
"mean_token_accuracy": 0.19654237180948259,
|
|
"num_tokens": 8001810.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.40740818977356,
|
|
"epoch": 3.8599242833964302,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004004309346250603,
|
|
"loss": 5.1106,
|
|
"mean_token_accuracy": 0.21052826792001725,
|
|
"num_tokens": 8011821.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.25329704284668,
|
|
"epoch": 3.865332612222823,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004000746199487181,
|
|
"loss": 5.076,
|
|
"mean_token_accuracy": 0.20677282512187958,
|
|
"num_tokens": 8024106.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.264664125442505,
|
|
"epoch": 3.870740941049216,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000399717850750918,
|
|
"loss": 5.1023,
|
|
"mean_token_accuracy": 0.20512110441923143,
|
|
"num_tokens": 8035630.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.388676977157592,
|
|
"epoch": 3.8761492698756084,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00039936062832816,
|
|
"loss": 5.1078,
|
|
"mean_token_accuracy": 0.2056085005402565,
|
|
"num_tokens": 8047130.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.28256049156189,
|
|
"epoch": 3.8815575987020012,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003990029539785913,
|
|
"loss": 5.0955,
|
|
"mean_token_accuracy": 0.19555404931306838,
|
|
"num_tokens": 8059484.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.378173017501831,
|
|
"epoch": 3.8869659275283936,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00039864482900200135,
|
|
"loss": 5.185,
|
|
"mean_token_accuracy": 0.19763617664575578,
|
|
"num_tokens": 8069467.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.434432077407837,
|
|
"epoch": 3.8923742563547865,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003982862546998169,
|
|
"loss": 5.1857,
|
|
"mean_token_accuracy": 0.19747186899185182,
|
|
"num_tokens": 8081390.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.255051565170288,
|
|
"epoch": 3.897782585181179,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003979272323750981,
|
|
"loss": 5.0096,
|
|
"mean_token_accuracy": 0.21605978459119796,
|
|
"num_tokens": 8091202.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.257553005218506,
|
|
"epoch": 3.903190914007572,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003975677633325327,
|
|
"loss": 5.1108,
|
|
"mean_token_accuracy": 0.1953676998615265,
|
|
"num_tokens": 8101900.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.387345552444458,
|
|
"epoch": 3.908599242833964,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003972078488784321,
|
|
"loss": 5.1337,
|
|
"mean_token_accuracy": 0.19343214482069016,
|
|
"num_tokens": 8114500.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.277220296859741,
|
|
"epoch": 3.914007571660357,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003968474903207263,
|
|
"loss": 5.0054,
|
|
"mean_token_accuracy": 0.21512256413698197,
|
|
"num_tokens": 8125319.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.3228062152862545,
|
|
"epoch": 3.9194159004867495,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003964866889689592,
|
|
"loss": 5.2014,
|
|
"mean_token_accuracy": 0.19903041273355485,
|
|
"num_tokens": 8137077.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.278341817855835,
|
|
"epoch": 3.9248242293131423,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003961254461342834,
|
|
"loss": 5.075,
|
|
"mean_token_accuracy": 0.2023685947060585,
|
|
"num_tokens": 8148430.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.305149698257447,
|
|
"epoch": 3.9302325581395348,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003957637631294566,
|
|
"loss": 5.0903,
|
|
"mean_token_accuracy": 0.2067402720451355,
|
|
"num_tokens": 8159250.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.361065006256103,
|
|
"epoch": 3.9356408869659276,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00039540164126883554,
|
|
"loss": 5.1842,
|
|
"mean_token_accuracy": 0.19624228477478028,
|
|
"num_tokens": 8170177.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.356918525695801,
|
|
"epoch": 3.94104921579232,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003950390818683718,
|
|
"loss": 5.1321,
|
|
"mean_token_accuracy": 0.20411310344934464,
|
|
"num_tokens": 8181351.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.30003867149353,
|
|
"epoch": 3.946457544618713,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003946760862456071,
|
|
"loss": 5.1425,
|
|
"mean_token_accuracy": 0.19613187164068221,
|
|
"num_tokens": 8193472.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.359955453872681,
|
|
"epoch": 3.9518658734451053,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003943126557196685,
|
|
"loss": 5.2337,
|
|
"mean_token_accuracy": 0.20510112047195433,
|
|
"num_tokens": 8205449.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.2841283798217775,
|
|
"epoch": 3.957274202271498,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00039394879161126333,
|
|
"loss": 4.9911,
|
|
"mean_token_accuracy": 0.20808043628931044,
|
|
"num_tokens": 8216520.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.328678226470947,
|
|
"epoch": 3.9626825310978906,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003935844952426745,
|
|
"loss": 5.2106,
|
|
"mean_token_accuracy": 0.19062691181898117,
|
|
"num_tokens": 8228727.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.355085849761963,
|
|
"epoch": 3.9680908599242835,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00039321976793775586,
|
|
"loss": 5.1314,
|
|
"mean_token_accuracy": 0.19883581697940828,
|
|
"num_tokens": 8240423.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.223177194595337,
|
|
"epoch": 3.973499188750676,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003928546110219275,
|
|
"loss": 5.0504,
|
|
"mean_token_accuracy": 0.2023528516292572,
|
|
"num_tokens": 8252757.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.206526517868042,
|
|
"epoch": 3.9789075175770687,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00039248902582217024,
|
|
"loss": 5.016,
|
|
"mean_token_accuracy": 0.200811967253685,
|
|
"num_tokens": 8263462.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.303615093231201,
|
|
"epoch": 3.9843158464034616,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00039212301366702153,
|
|
"loss": 5.0858,
|
|
"mean_token_accuracy": 0.199895241856575,
|
|
"num_tokens": 8274364.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.291612958908081,
|
|
"epoch": 3.989724175229854,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00039175657588657063,
|
|
"loss": 5.1062,
|
|
"mean_token_accuracy": 0.20076947510242463,
|
|
"num_tokens": 8286026.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.405158662796021,
|
|
"epoch": 3.9951325040562464,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003913897138124531,
|
|
"loss": 5.1586,
|
|
"mean_token_accuracy": 0.2002616137266159,
|
|
"num_tokens": 8297990.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.390300750732422,
|
|
"epoch": 4.0,
|
|
"grad_norm": 2.375,
|
|
"learning_rate": 0.0003910224287778466,
|
|
"loss": 5.1908,
|
|
"mean_token_accuracy": 0.19133670462502372,
|
|
"num_tokens": 8307556.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.353345108032227,
|
|
"epoch": 4.005408328826393,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00039065472211746586,
|
|
"loss": 4.8891,
|
|
"mean_token_accuracy": 0.21125887781381608,
|
|
"num_tokens": 8320805.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.304692602157592,
|
|
"epoch": 4.010816657652786,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003902865951675576,
|
|
"loss": 4.8605,
|
|
"mean_token_accuracy": 0.21419780552387238,
|
|
"num_tokens": 8332390.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.272004556655884,
|
|
"epoch": 4.016224986479178,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00038991804926589616,
|
|
"loss": 4.7686,
|
|
"mean_token_accuracy": 0.2157680422067642,
|
|
"num_tokens": 8342325.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.285643720626831,
|
|
"epoch": 4.0216333153055706,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000389549085751778,
|
|
"loss": 4.7775,
|
|
"mean_token_accuracy": 0.21727175861597062,
|
|
"num_tokens": 8352115.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.3047772407531735,
|
|
"epoch": 4.027041644131963,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00038917970596601765,
|
|
"loss": 4.8532,
|
|
"mean_token_accuracy": 0.2236403450369835,
|
|
"num_tokens": 8364265.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.420230531692505,
|
|
"epoch": 4.032449972958356,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003888099112509419,
|
|
"loss": 4.8083,
|
|
"mean_token_accuracy": 0.22125934362411498,
|
|
"num_tokens": 8375979.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.296147060394287,
|
|
"epoch": 4.037858301784748,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00038843970295038564,
|
|
"loss": 4.808,
|
|
"mean_token_accuracy": 0.21896344274282456,
|
|
"num_tokens": 8387794.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.23355712890625,
|
|
"epoch": 4.043266630611141,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00038806908240968666,
|
|
"loss": 4.7482,
|
|
"mean_token_accuracy": 0.2159058153629303,
|
|
"num_tokens": 8398796.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.257886791229248,
|
|
"epoch": 4.048674959437534,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000387698050975681,
|
|
"loss": 4.8399,
|
|
"mean_token_accuracy": 0.22372966557741164,
|
|
"num_tokens": 8408788.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.304345893859863,
|
|
"epoch": 4.054083288263927,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003873266099966976,
|
|
"loss": 4.8645,
|
|
"mean_token_accuracy": 0.20975825637578965,
|
|
"num_tokens": 8420373.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.209354591369629,
|
|
"epoch": 4.059491617090319,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00038695476082255395,
|
|
"loss": 4.8049,
|
|
"mean_token_accuracy": 0.22424950301647187,
|
|
"num_tokens": 8430399.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.340624904632568,
|
|
"epoch": 4.064899945916712,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003865825048045507,
|
|
"loss": 4.9233,
|
|
"mean_token_accuracy": 0.20802856385707855,
|
|
"num_tokens": 8442649.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.405769968032837,
|
|
"epoch": 4.0703082747431045,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003862098432954672,
|
|
"loss": 4.9342,
|
|
"mean_token_accuracy": 0.21130404770374298,
|
|
"num_tokens": 8453985.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.35463581085205,
|
|
"epoch": 4.075716603569497,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003858367776495561,
|
|
"loss": 4.9347,
|
|
"mean_token_accuracy": 0.2114950880408287,
|
|
"num_tokens": 8465566.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.211642408370972,
|
|
"epoch": 4.081124932395889,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00038546330922253886,
|
|
"loss": 4.7324,
|
|
"mean_token_accuracy": 0.22888267636299134,
|
|
"num_tokens": 8477426.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.199952125549316,
|
|
"epoch": 4.086533261222282,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003850894393716007,
|
|
"loss": 4.8478,
|
|
"mean_token_accuracy": 0.22065728157758713,
|
|
"num_tokens": 8489654.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.233488702774048,
|
|
"epoch": 4.091941590048675,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003847151694553855,
|
|
"loss": 4.821,
|
|
"mean_token_accuracy": 0.2131299704313278,
|
|
"num_tokens": 8500398.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.188064861297607,
|
|
"epoch": 4.097349918875068,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003843405008339908,
|
|
"loss": 4.7735,
|
|
"mean_token_accuracy": 0.21535448729991913,
|
|
"num_tokens": 8511974.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.261436843872071,
|
|
"epoch": 4.10275824770146,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00038396543486896354,
|
|
"loss": 4.8317,
|
|
"mean_token_accuracy": 0.21560191065073014,
|
|
"num_tokens": 8523140.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.292982959747315,
|
|
"epoch": 4.108166576527853,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00038358997292329406,
|
|
"loss": 4.8515,
|
|
"mean_token_accuracy": 0.21860649138689042,
|
|
"num_tokens": 8534261.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.316209983825684,
|
|
"epoch": 4.113574905354246,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00038321411636141214,
|
|
"loss": 4.8962,
|
|
"mean_token_accuracy": 0.21073623597621918,
|
|
"num_tokens": 8544620.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.222609424591065,
|
|
"epoch": 4.1189832341806385,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003828378665491811,
|
|
"loss": 4.8535,
|
|
"mean_token_accuracy": 0.21865524500608444,
|
|
"num_tokens": 8555427.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.280526876449585,
|
|
"epoch": 4.1243915630070305,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000382461224853894,
|
|
"loss": 4.797,
|
|
"mean_token_accuracy": 0.21900686621665955,
|
|
"num_tokens": 8566956.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.311002922058106,
|
|
"epoch": 4.129799891833423,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003820841926442674,
|
|
"loss": 4.9727,
|
|
"mean_token_accuracy": 0.20906080007553102,
|
|
"num_tokens": 8579373.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.253505563735962,
|
|
"epoch": 4.135208220659816,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003817067712904372,
|
|
"loss": 4.7903,
|
|
"mean_token_accuracy": 0.21533195078372955,
|
|
"num_tokens": 8590979.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.2259361743927,
|
|
"epoch": 4.140616549486209,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00038132896216395355,
|
|
"loss": 4.8325,
|
|
"mean_token_accuracy": 0.2139109864830971,
|
|
"num_tokens": 8602514.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.176524639129639,
|
|
"epoch": 4.146024878312601,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003809507666377758,
|
|
"loss": 4.7926,
|
|
"mean_token_accuracy": 0.22516833394765853,
|
|
"num_tokens": 8613766.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.3135552406311035,
|
|
"epoch": 4.151433207138994,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00038057218608626714,
|
|
"loss": 4.8992,
|
|
"mean_token_accuracy": 0.20659841001033782,
|
|
"num_tokens": 8625245.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.279325151443482,
|
|
"epoch": 4.156841535965387,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003801932218851903,
|
|
"loss": 4.8327,
|
|
"mean_token_accuracy": 0.22115242183208467,
|
|
"num_tokens": 8635653.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.199054002761841,
|
|
"epoch": 4.16224986479178,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000379813875411702,
|
|
"loss": 4.7864,
|
|
"mean_token_accuracy": 0.21551311761140823,
|
|
"num_tokens": 8646671.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.167051315307617,
|
|
"epoch": 4.167658193618172,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00037943414804434826,
|
|
"loss": 4.74,
|
|
"mean_token_accuracy": 0.22161268144845964,
|
|
"num_tokens": 8656441.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.234937572479248,
|
|
"epoch": 4.173066522444564,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003790540411630592,
|
|
"loss": 4.829,
|
|
"mean_token_accuracy": 0.22128485888242722,
|
|
"num_tokens": 8666632.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.193305778503418,
|
|
"epoch": 4.178474851270957,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00037867355614914404,
|
|
"loss": 4.8615,
|
|
"mean_token_accuracy": 0.21823760718107224,
|
|
"num_tokens": 8678269.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.260924386978149,
|
|
"epoch": 4.18388318009735,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003782926943852864,
|
|
"loss": 4.8871,
|
|
"mean_token_accuracy": 0.2139583259820938,
|
|
"num_tokens": 8690004.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.341515779495239,
|
|
"epoch": 4.189291508923742,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00037791145725553853,
|
|
"loss": 4.9087,
|
|
"mean_token_accuracy": 0.20802224576473236,
|
|
"num_tokens": 8702306.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.245114517211914,
|
|
"epoch": 4.194699837750135,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003775298461453172,
|
|
"loss": 4.9339,
|
|
"mean_token_accuracy": 0.20960736721754075,
|
|
"num_tokens": 8714070.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.264302349090576,
|
|
"epoch": 4.200108166576528,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000377147862441398,
|
|
"loss": 4.8631,
|
|
"mean_token_accuracy": 0.2178913712501526,
|
|
"num_tokens": 8724433.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.234062433242798,
|
|
"epoch": 4.205516495402921,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00037676550753191075,
|
|
"loss": 4.8044,
|
|
"mean_token_accuracy": 0.22861745208501816,
|
|
"num_tokens": 8735485.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.244630575180054,
|
|
"epoch": 4.210924824229313,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003763827828063339,
|
|
"loss": 4.8388,
|
|
"mean_token_accuracy": 0.20746962130069732,
|
|
"num_tokens": 8746356.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.25743613243103,
|
|
"epoch": 4.2163331530557056,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00037599968965549,
|
|
"loss": 4.8557,
|
|
"mean_token_accuracy": 0.21454470604658127,
|
|
"num_tokens": 8757667.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.188058233261108,
|
|
"epoch": 4.221741481882098,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003756162294715405,
|
|
"loss": 4.8204,
|
|
"mean_token_accuracy": 0.21627243012189865,
|
|
"num_tokens": 8768565.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.185556936264038,
|
|
"epoch": 4.227149810708491,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00037523240364798063,
|
|
"loss": 4.8506,
|
|
"mean_token_accuracy": 0.21275447607040404,
|
|
"num_tokens": 8779608.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.261024904251099,
|
|
"epoch": 4.232558139534884,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00037484821357963423,
|
|
"loss": 4.8696,
|
|
"mean_token_accuracy": 0.2193668633699417,
|
|
"num_tokens": 8791492.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.22265362739563,
|
|
"epoch": 4.237966468361276,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000374463660662649,
|
|
"loss": 4.887,
|
|
"mean_token_accuracy": 0.2145410358905792,
|
|
"num_tokens": 8803280.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.258722496032715,
|
|
"epoch": 4.243374797187669,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000374078746294491,
|
|
"loss": 4.9179,
|
|
"mean_token_accuracy": 0.21171808242797852,
|
|
"num_tokens": 8814626.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.215199756622314,
|
|
"epoch": 4.248783126014062,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00037369347187393996,
|
|
"loss": 4.7358,
|
|
"mean_token_accuracy": 0.22830508053302764,
|
|
"num_tokens": 8825520.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.1555475234985355,
|
|
"epoch": 4.254191454840455,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003733078388010842,
|
|
"loss": 4.7736,
|
|
"mean_token_accuracy": 0.2244316890835762,
|
|
"num_tokens": 8835980.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.287052154541016,
|
|
"epoch": 4.259599783666847,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003729218484773149,
|
|
"loss": 4.9619,
|
|
"mean_token_accuracy": 0.20417773723602295,
|
|
"num_tokens": 8846714.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.296690177917481,
|
|
"epoch": 4.2650081124932395,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003725355023053221,
|
|
"loss": 4.877,
|
|
"mean_token_accuracy": 0.21835277676582338,
|
|
"num_tokens": 8858215.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.16245322227478,
|
|
"epoch": 4.270416441319632,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003721488016890884,
|
|
"loss": 4.7798,
|
|
"mean_token_accuracy": 0.223819500207901,
|
|
"num_tokens": 8870587.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.134206867218017,
|
|
"epoch": 4.275824770146025,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00037176174803388496,
|
|
"loss": 4.844,
|
|
"mean_token_accuracy": 0.2206039696931839,
|
|
"num_tokens": 8881335.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.160008955001831,
|
|
"epoch": 4.281233098972417,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00037137434274626545,
|
|
"loss": 4.8293,
|
|
"mean_token_accuracy": 0.21559333205223083,
|
|
"num_tokens": 8893294.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.29144549369812,
|
|
"epoch": 4.28664142779881,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003709865872340618,
|
|
"loss": 4.8942,
|
|
"mean_token_accuracy": 0.2130601614713669,
|
|
"num_tokens": 8903914.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.2798316955566404,
|
|
"epoch": 4.292049756625203,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003705984829063782,
|
|
"loss": 4.8703,
|
|
"mean_token_accuracy": 0.21196929216384888,
|
|
"num_tokens": 8915591.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.226657199859619,
|
|
"epoch": 4.297458085451596,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003702100311735869,
|
|
"loss": 4.8569,
|
|
"mean_token_accuracy": 0.21135589927434922,
|
|
"num_tokens": 8927031.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.157093191146851,
|
|
"epoch": 4.302866414277988,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00036982123344732223,
|
|
"loss": 4.7787,
|
|
"mean_token_accuracy": 0.22054984271526337,
|
|
"num_tokens": 8937220.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.147545337677002,
|
|
"epoch": 4.308274743104381,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00036943209114047613,
|
|
"loss": 4.8068,
|
|
"mean_token_accuracy": 0.21458472907543183,
|
|
"num_tokens": 8948032.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.190038347244263,
|
|
"epoch": 4.3136830719307735,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003690426056671926,
|
|
"loss": 4.9116,
|
|
"mean_token_accuracy": 0.21072200536727906,
|
|
"num_tokens": 8960735.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.205200624465943,
|
|
"epoch": 4.319091400757166,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003686527784428627,
|
|
"loss": 4.7713,
|
|
"mean_token_accuracy": 0.22340385913848876,
|
|
"num_tokens": 8971801.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.260702228546142,
|
|
"epoch": 4.324499729583558,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00036826261088411955,
|
|
"loss": 4.9015,
|
|
"mean_token_accuracy": 0.20845765769481658,
|
|
"num_tokens": 8983041.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 4.324499729583558,
|
|
"eval_entropy": 5.13772134997628,
|
|
"eval_loss": 5.912678241729736,
|
|
"eval_mean_token_accuracy": 0.17654112712903455,
|
|
"eval_num_tokens": 8983041.0,
|
|
"eval_runtime": 2.7661,
|
|
"eval_samples_per_second": 1270.735,
|
|
"eval_steps_per_second": 159.068,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 5.1894816875457765,
|
|
"epoch": 4.329908058409951,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003678721044088329,
|
|
"loss": 4.8544,
|
|
"mean_token_accuracy": 0.21754262447357178,
|
|
"num_tokens": 8993935.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 5.2399632930755615,
|
|
"epoch": 4.335316387236344,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003674812604361043,
|
|
"loss": 4.8314,
|
|
"mean_token_accuracy": 0.2137932300567627,
|
|
"num_tokens": 9003896.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 5.231870031356811,
|
|
"epoch": 4.340724716062737,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003670900803862615,
|
|
"loss": 4.8706,
|
|
"mean_token_accuracy": 0.21803625971078872,
|
|
"num_tokens": 9014323.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 5.2589808940887455,
|
|
"epoch": 4.346133044889129,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003666985656808537,
|
|
"loss": 4.9243,
|
|
"mean_token_accuracy": 0.2163531720638275,
|
|
"num_tokens": 9024252.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 5.3169597625732425,
|
|
"epoch": 4.351541373715522,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00036630671774264635,
|
|
"loss": 4.9475,
|
|
"mean_token_accuracy": 0.20517051219940186,
|
|
"num_tokens": 9035178.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 5.262640523910522,
|
|
"epoch": 4.356949702541915,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003659145379956158,
|
|
"loss": 4.8655,
|
|
"mean_token_accuracy": 0.21626094579696656,
|
|
"num_tokens": 9045807.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 5.315378189086914,
|
|
"epoch": 4.3623580313683075,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00036552202786494393,
|
|
"loss": 4.8846,
|
|
"mean_token_accuracy": 0.21611111462116242,
|
|
"num_tokens": 9056927.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 5.209165906906128,
|
|
"epoch": 4.367766360194699,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00036512918877701367,
|
|
"loss": 4.8787,
|
|
"mean_token_accuracy": 0.2174328163266182,
|
|
"num_tokens": 9070141.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 5.149424695968628,
|
|
"epoch": 4.373174689021092,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00036473602215940316,
|
|
"loss": 4.7977,
|
|
"mean_token_accuracy": 0.22122627943754197,
|
|
"num_tokens": 9081595.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 5.253453874588013,
|
|
"epoch": 4.378583017847485,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003643425294408807,
|
|
"loss": 4.8771,
|
|
"mean_token_accuracy": 0.2091942772269249,
|
|
"num_tokens": 9092158.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 5.227699851989746,
|
|
"epoch": 4.383991346673878,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00036394871205139984,
|
|
"loss": 4.8691,
|
|
"mean_token_accuracy": 0.2166379228234291,
|
|
"num_tokens": 9102958.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 5.226753807067871,
|
|
"epoch": 4.38939967550027,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003635545714220938,
|
|
"loss": 4.8243,
|
|
"mean_token_accuracy": 0.21546567231416702,
|
|
"num_tokens": 9113793.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 5.253853797912598,
|
|
"epoch": 4.394808004326663,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003631601089852706,
|
|
"loss": 4.9181,
|
|
"mean_token_accuracy": 0.21917014718055725,
|
|
"num_tokens": 9125149.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 5.145966720581055,
|
|
"epoch": 4.400216333153056,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003627653261744075,
|
|
"loss": 4.7359,
|
|
"mean_token_accuracy": 0.22841183394193648,
|
|
"num_tokens": 9136222.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 5.245536136627197,
|
|
"epoch": 4.405624661979449,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00036237022442414636,
|
|
"loss": 4.9184,
|
|
"mean_token_accuracy": 0.21662650108337403,
|
|
"num_tokens": 9147933.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 5.206748533248901,
|
|
"epoch": 4.411032990805841,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003619748051702878,
|
|
"loss": 4.9308,
|
|
"mean_token_accuracy": 0.21570136696100234,
|
|
"num_tokens": 9159421.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 5.244329023361206,
|
|
"epoch": 4.416441319632233,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003615790698497862,
|
|
"loss": 4.8337,
|
|
"mean_token_accuracy": 0.2218611642718315,
|
|
"num_tokens": 9170676.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 5.250214147567749,
|
|
"epoch": 4.421849648458626,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00036118301990074463,
|
|
"loss": 4.8274,
|
|
"mean_token_accuracy": 0.22138022184371947,
|
|
"num_tokens": 9181267.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 5.154590749740601,
|
|
"epoch": 4.427257977285019,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003607866567624097,
|
|
"loss": 4.8855,
|
|
"mean_token_accuracy": 0.2154363512992859,
|
|
"num_tokens": 9192999.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 5.213081884384155,
|
|
"epoch": 4.432666306111411,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00036038998187516583,
|
|
"loss": 4.8127,
|
|
"mean_token_accuracy": 0.2144768178462982,
|
|
"num_tokens": 9203778.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 5.19510235786438,
|
|
"epoch": 4.438074634937804,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00035999299668053043,
|
|
"loss": 4.8411,
|
|
"mean_token_accuracy": 0.2213861122727394,
|
|
"num_tokens": 9214331.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.201658439636231,
|
|
"epoch": 4.443482963764197,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003595957026211487,
|
|
"loss": 4.8791,
|
|
"mean_token_accuracy": 0.2191414311528206,
|
|
"num_tokens": 9224809.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 5.285597467422486,
|
|
"epoch": 4.44889129259059,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00035919810114078815,
|
|
"loss": 4.9908,
|
|
"mean_token_accuracy": 0.20575572401285172,
|
|
"num_tokens": 9235852.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 5.242265129089356,
|
|
"epoch": 4.4542996214169825,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00035880019368433354,
|
|
"loss": 4.8393,
|
|
"mean_token_accuracy": 0.22336905300617219,
|
|
"num_tokens": 9246296.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 5.248835849761963,
|
|
"epoch": 4.4597079502433745,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003584019816977813,
|
|
"loss": 4.8376,
|
|
"mean_token_accuracy": 0.22735059559345244,
|
|
"num_tokens": 9257761.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 5.1857036590576175,
|
|
"epoch": 4.465116279069767,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00035800346662823493,
|
|
"loss": 4.9239,
|
|
"mean_token_accuracy": 0.2154724955558777,
|
|
"num_tokens": 9268796.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 5.331050014495849,
|
|
"epoch": 4.47052460789616,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000357604649923899,
|
|
"loss": 4.9667,
|
|
"mean_token_accuracy": 0.20768869519233704,
|
|
"num_tokens": 9280625.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 5.184424209594726,
|
|
"epoch": 4.475932936722553,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003572055330340744,
|
|
"loss": 4.7872,
|
|
"mean_token_accuracy": 0.22010961621999742,
|
|
"num_tokens": 9291096.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 5.173257827758789,
|
|
"epoch": 4.481341265548945,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00035680611740915264,
|
|
"loss": 4.8006,
|
|
"mean_token_accuracy": 0.22125547826290132,
|
|
"num_tokens": 9301664.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 5.162072610855103,
|
|
"epoch": 4.486749594375338,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003564064045006113,
|
|
"loss": 4.8355,
|
|
"mean_token_accuracy": 0.21493352502584456,
|
|
"num_tokens": 9312907.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 5.210605764389038,
|
|
"epoch": 4.492157923201731,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00035600639576100786,
|
|
"loss": 4.9271,
|
|
"mean_token_accuracy": 0.2129287376999855,
|
|
"num_tokens": 9323806.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 5.2831458568573,
|
|
"epoch": 4.497566252028124,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00035560609264397497,
|
|
"loss": 4.9441,
|
|
"mean_token_accuracy": 0.20415034145116806,
|
|
"num_tokens": 9336394.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 5.205191612243652,
|
|
"epoch": 4.502974580854516,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00035520549660421515,
|
|
"loss": 4.8653,
|
|
"mean_token_accuracy": 0.2189026415348053,
|
|
"num_tokens": 9349177.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 5.281127834320069,
|
|
"epoch": 4.5083829096809085,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003548046090974954,
|
|
"loss": 4.9184,
|
|
"mean_token_accuracy": 0.21316232532262802,
|
|
"num_tokens": 9360033.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 5.172052001953125,
|
|
"epoch": 4.513791238507301,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00035440343158064176,
|
|
"loss": 4.8778,
|
|
"mean_token_accuracy": 0.2110334888100624,
|
|
"num_tokens": 9371300.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 5.1869689464569095,
|
|
"epoch": 4.519199567333694,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003540019655115341,
|
|
"loss": 4.8946,
|
|
"mean_token_accuracy": 0.21620258837938308,
|
|
"num_tokens": 9383112.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 5.171368598937988,
|
|
"epoch": 4.524607896160086,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00035360021234910133,
|
|
"loss": 4.7929,
|
|
"mean_token_accuracy": 0.22111257016658784,
|
|
"num_tokens": 9394062.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 5.29243540763855,
|
|
"epoch": 4.530016224986479,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00035319817355331537,
|
|
"loss": 4.9137,
|
|
"mean_token_accuracy": 0.2167750209569931,
|
|
"num_tokens": 9406016.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 5.267550659179688,
|
|
"epoch": 4.535424553812872,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003527958505851858,
|
|
"loss": 4.9463,
|
|
"mean_token_accuracy": 0.20864920318126678,
|
|
"num_tokens": 9417286.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 5.146933126449585,
|
|
"epoch": 4.540832882639265,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00035239324490675546,
|
|
"loss": 4.7951,
|
|
"mean_token_accuracy": 0.21861904561519624,
|
|
"num_tokens": 9428794.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 5.22603440284729,
|
|
"epoch": 4.546241211465657,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003519903579810943,
|
|
"loss": 4.919,
|
|
"mean_token_accuracy": 0.2128203496336937,
|
|
"num_tokens": 9440524.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 5.191568756103516,
|
|
"epoch": 4.55164954029205,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003515871912722944,
|
|
"loss": 4.7956,
|
|
"mean_token_accuracy": 0.226128551363945,
|
|
"num_tokens": 9450796.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 5.222783041000366,
|
|
"epoch": 4.5570578691184425,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003511837462454643,
|
|
"loss": 4.9411,
|
|
"mean_token_accuracy": 0.21448575258255004,
|
|
"num_tokens": 9462498.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 5.284732627868652,
|
|
"epoch": 4.562466197944835,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00035078002436672443,
|
|
"loss": 4.8541,
|
|
"mean_token_accuracy": 0.22078913003206252,
|
|
"num_tokens": 9473395.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 5.220347213745117,
|
|
"epoch": 4.567874526771227,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003503760271032007,
|
|
"loss": 4.9212,
|
|
"mean_token_accuracy": 0.21168201565742492,
|
|
"num_tokens": 9485099.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 5.303788900375366,
|
|
"epoch": 4.57328285559762,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003499717559230205,
|
|
"loss": 4.9071,
|
|
"mean_token_accuracy": 0.2082112804055214,
|
|
"num_tokens": 9495129.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 5.218679761886596,
|
|
"epoch": 4.578691184424013,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00034956721229530586,
|
|
"loss": 4.9494,
|
|
"mean_token_accuracy": 0.2112857386469841,
|
|
"num_tokens": 9506933.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 5.215309476852417,
|
|
"epoch": 4.584099513250406,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003491623976901695,
|
|
"loss": 4.8136,
|
|
"mean_token_accuracy": 0.22183329313993455,
|
|
"num_tokens": 9518262.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 5.236282777786255,
|
|
"epoch": 4.589507842076799,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003487573135787085,
|
|
"loss": 4.8562,
|
|
"mean_token_accuracy": 0.21417921483516694,
|
|
"num_tokens": 9530829.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 5.143803453445434,
|
|
"epoch": 4.594916170903191,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00034835196143299956,
|
|
"loss": 4.7514,
|
|
"mean_token_accuracy": 0.22633299380540847,
|
|
"num_tokens": 9541762.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 5.158538007736206,
|
|
"epoch": 4.600324499729584,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00034794634272609313,
|
|
"loss": 4.8453,
|
|
"mean_token_accuracy": 0.21947379559278488,
|
|
"num_tokens": 9552344.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 5.0739751815795895,
|
|
"epoch": 4.605732828555976,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003475404589320089,
|
|
"loss": 4.7869,
|
|
"mean_token_accuracy": 0.2136710524559021,
|
|
"num_tokens": 9563806.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 5.270521783828736,
|
|
"epoch": 4.611141157382368,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003471343115257291,
|
|
"loss": 4.9146,
|
|
"mean_token_accuracy": 0.2133626863360405,
|
|
"num_tokens": 9575305.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 5.300551319122315,
|
|
"epoch": 4.616549486208761,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00034672790198319453,
|
|
"loss": 4.9506,
|
|
"mean_token_accuracy": 0.21007315814495087,
|
|
"num_tokens": 9586155.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 5.217075443267822,
|
|
"epoch": 4.621957815035154,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003463212317812985,
|
|
"loss": 4.9328,
|
|
"mean_token_accuracy": 0.21347327381372452,
|
|
"num_tokens": 9597731.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 5.363858890533447,
|
|
"epoch": 4.627366143861547,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003459143023978813,
|
|
"loss": 4.9183,
|
|
"mean_token_accuracy": 0.21418648362159728,
|
|
"num_tokens": 9609438.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 5.242604494094849,
|
|
"epoch": 4.63277447268794,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003455071153117252,
|
|
"loss": 4.8787,
|
|
"mean_token_accuracy": 0.2110474407672882,
|
|
"num_tokens": 9620665.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 5.217668056488037,
|
|
"epoch": 4.638182801514332,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003450996720025493,
|
|
"loss": 4.9157,
|
|
"mean_token_accuracy": 0.2141656830906868,
|
|
"num_tokens": 9631868.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 5.167800188064575,
|
|
"epoch": 4.643591130340725,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00034469197395100323,
|
|
"loss": 4.7913,
|
|
"mean_token_accuracy": 0.22088928818702697,
|
|
"num_tokens": 9643879.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 5.196749448776245,
|
|
"epoch": 4.6489994591671175,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00034428402263866285,
|
|
"loss": 4.8458,
|
|
"mean_token_accuracy": 0.2248424082994461,
|
|
"num_tokens": 9656277.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 5.197022819519043,
|
|
"epoch": 4.6544077879935095,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00034387581954802406,
|
|
"loss": 4.9277,
|
|
"mean_token_accuracy": 0.21301621198654175,
|
|
"num_tokens": 9666590.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 5.2290057182312015,
|
|
"epoch": 4.659816116819902,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00034346736616249786,
|
|
"loss": 4.8769,
|
|
"mean_token_accuracy": 0.2151239976286888,
|
|
"num_tokens": 9677248.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 5.2219993591308596,
|
|
"epoch": 4.665224445646295,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00034305866396640473,
|
|
"loss": 4.8427,
|
|
"mean_token_accuracy": 0.21257670521736144,
|
|
"num_tokens": 9689617.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 5.221749114990234,
|
|
"epoch": 4.670632774472688,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00034264971444496954,
|
|
"loss": 4.9439,
|
|
"mean_token_accuracy": 0.2121889054775238,
|
|
"num_tokens": 9702663.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 5.216382932662964,
|
|
"epoch": 4.676041103299081,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00034224051908431574,
|
|
"loss": 4.9568,
|
|
"mean_token_accuracy": 0.2102925792336464,
|
|
"num_tokens": 9714005.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 5.2628706932067875,
|
|
"epoch": 4.681449432125473,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003418310793714602,
|
|
"loss": 4.8318,
|
|
"mean_token_accuracy": 0.2253083735704422,
|
|
"num_tokens": 9724931.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 5.180553960800171,
|
|
"epoch": 4.686857760951866,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003414213967943077,
|
|
"loss": 4.8407,
|
|
"mean_token_accuracy": 0.21361288875341417,
|
|
"num_tokens": 9734948.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 5.158719873428344,
|
|
"epoch": 4.692266089778259,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00034101147284164583,
|
|
"loss": 4.8095,
|
|
"mean_token_accuracy": 0.21615717858076094,
|
|
"num_tokens": 9746424.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 5.351026248931885,
|
|
"epoch": 4.6976744186046515,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00034060130900313905,
|
|
"loss": 4.9483,
|
|
"mean_token_accuracy": 0.21695078611373902,
|
|
"num_tokens": 9758434.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 5.130232906341552,
|
|
"epoch": 4.7030827474310435,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00034019090676932383,
|
|
"loss": 4.8446,
|
|
"mean_token_accuracy": 0.21692127138376235,
|
|
"num_tokens": 9768869.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 5.255046606063843,
|
|
"epoch": 4.708491076257436,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003397802676316027,
|
|
"loss": 5.0076,
|
|
"mean_token_accuracy": 0.20929261893033982,
|
|
"num_tokens": 9780001.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 5.282883787155152,
|
|
"epoch": 4.713899405083829,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00033936939308223935,
|
|
"loss": 4.9279,
|
|
"mean_token_accuracy": 0.215233413875103,
|
|
"num_tokens": 9790609.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 5.215489053726197,
|
|
"epoch": 4.719307733910222,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00033895828461435275,
|
|
"loss": 4.8646,
|
|
"mean_token_accuracy": 0.21731891632080078,
|
|
"num_tokens": 9802064.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 5.180708503723144,
|
|
"epoch": 4.724716062736614,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003385469437219121,
|
|
"loss": 4.9136,
|
|
"mean_token_accuracy": 0.21607225239276887,
|
|
"num_tokens": 9812124.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 5.190913200378418,
|
|
"epoch": 4.730124391563007,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00033813537189973105,
|
|
"loss": 4.777,
|
|
"mean_token_accuracy": 0.2290421262383461,
|
|
"num_tokens": 9824371.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 5.17042088508606,
|
|
"epoch": 4.7355327203894,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00033772357064346266,
|
|
"loss": 4.8219,
|
|
"mean_token_accuracy": 0.21956049650907516,
|
|
"num_tokens": 9834919.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 5.199196147918701,
|
|
"epoch": 4.740941049215793,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003373115414495935,
|
|
"loss": 4.8788,
|
|
"mean_token_accuracy": 0.21201282888650894,
|
|
"num_tokens": 9845435.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 5.230615472793579,
|
|
"epoch": 4.746349378042185,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003368992858154389,
|
|
"loss": 4.8403,
|
|
"mean_token_accuracy": 0.2236371800303459,
|
|
"num_tokens": 9856091.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 5.129108047485351,
|
|
"epoch": 4.7517577068685775,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00033648680523913643,
|
|
"loss": 4.7188,
|
|
"mean_token_accuracy": 0.22776069343090058,
|
|
"num_tokens": 9866861.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 5.141558408737183,
|
|
"epoch": 4.75716603569497,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003360741012196417,
|
|
"loss": 4.8231,
|
|
"mean_token_accuracy": 0.2069469854235649,
|
|
"num_tokens": 9877852.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 5.17921290397644,
|
|
"epoch": 4.762574364521363,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000335661175256722,
|
|
"loss": 4.7644,
|
|
"mean_token_accuracy": 0.23025994151830673,
|
|
"num_tokens": 9888938.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 5.236938047409057,
|
|
"epoch": 4.767982693347756,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003352480288509513,
|
|
"loss": 4.845,
|
|
"mean_token_accuracy": 0.22896136343479156,
|
|
"num_tokens": 9901782.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 5.149921846389771,
|
|
"epoch": 4.773391022174148,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00033483466350370445,
|
|
"loss": 4.9714,
|
|
"mean_token_accuracy": 0.20430240780115128,
|
|
"num_tokens": 9913254.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 5.2382423877716064,
|
|
"epoch": 4.778799351000541,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00033442108071715215,
|
|
"loss": 4.8683,
|
|
"mean_token_accuracy": 0.21967888474464417,
|
|
"num_tokens": 9923170.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 5.2672521591186525,
|
|
"epoch": 4.784207679826934,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00033400728199425525,
|
|
"loss": 4.8988,
|
|
"mean_token_accuracy": 0.21837569922208785,
|
|
"num_tokens": 9934087.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 5.13267822265625,
|
|
"epoch": 4.789616008653326,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00033359326883875917,
|
|
"loss": 4.9792,
|
|
"mean_token_accuracy": 0.20426354706287383,
|
|
"num_tokens": 9945463.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 5.183151578903198,
|
|
"epoch": 4.795024337479719,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00033317904275518873,
|
|
"loss": 4.792,
|
|
"mean_token_accuracy": 0.21514214426279069,
|
|
"num_tokens": 9955602.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 5.215031290054322,
|
|
"epoch": 4.800432666306111,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003327646052488425,
|
|
"loss": 4.8153,
|
|
"mean_token_accuracy": 0.21860183477401735,
|
|
"num_tokens": 9965899.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 5.233831262588501,
|
|
"epoch": 4.805840995132504,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003323499578257873,
|
|
"loss": 4.9827,
|
|
"mean_token_accuracy": 0.2063148081302643,
|
|
"num_tokens": 9978047.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 5.344552993774414,
|
|
"epoch": 4.811249323958897,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00033193510199285286,
|
|
"loss": 5.0546,
|
|
"mean_token_accuracy": 0.20674531757831574,
|
|
"num_tokens": 9991760.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 5.252249431610108,
|
|
"epoch": 4.816657652785289,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00033152003925762615,
|
|
"loss": 4.8767,
|
|
"mean_token_accuracy": 0.21397899389266967,
|
|
"num_tokens": 10002540.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 5.217188358306885,
|
|
"epoch": 4.822065981611682,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003311047711284462,
|
|
"loss": 4.8942,
|
|
"mean_token_accuracy": 0.21213582009077073,
|
|
"num_tokens": 10013955.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 5.2355588436126705,
|
|
"epoch": 4.827474310438075,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003306892991143983,
|
|
"loss": 4.8786,
|
|
"mean_token_accuracy": 0.21068003475666047,
|
|
"num_tokens": 10024845.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 5.235380363464356,
|
|
"epoch": 4.832882639264467,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003302736247253087,
|
|
"loss": 4.9161,
|
|
"mean_token_accuracy": 0.21594039648771285,
|
|
"num_tokens": 10036440.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 5.260376977920532,
|
|
"epoch": 4.83829096809086,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003298577494717391,
|
|
"loss": 4.9223,
|
|
"mean_token_accuracy": 0.21162653267383574,
|
|
"num_tokens": 10049891.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 5.153290319442749,
|
|
"epoch": 4.8436992969172525,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003294416748649812,
|
|
"loss": 4.7871,
|
|
"mean_token_accuracy": 0.22590140700340272,
|
|
"num_tokens": 10061105.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 5.17445125579834,
|
|
"epoch": 4.849107625743645,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00032902540241705106,
|
|
"loss": 4.8498,
|
|
"mean_token_accuracy": 0.21788168400526048,
|
|
"num_tokens": 10072721.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 5.203429317474365,
|
|
"epoch": 4.854515954570038,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003286089336406837,
|
|
"loss": 4.8434,
|
|
"mean_token_accuracy": 0.2164887249469757,
|
|
"num_tokens": 10083288.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 5.258232116699219,
|
|
"epoch": 4.85992428339643,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00032819227004932745,
|
|
"loss": 4.8967,
|
|
"mean_token_accuracy": 0.21811196058988572,
|
|
"num_tokens": 10093879.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 5.176331281661987,
|
|
"epoch": 4.865332612222823,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000327775413157139,
|
|
"loss": 4.8243,
|
|
"mean_token_accuracy": 0.21997172981500626,
|
|
"num_tokens": 10104861.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"epoch": 4.865332612222823,
|
|
"eval_entropy": 5.101167671246962,
|
|
"eval_loss": 5.887608528137207,
|
|
"eval_mean_token_accuracy": 0.18027714209800416,
|
|
"eval_num_tokens": 10104861.0,
|
|
"eval_runtime": 2.5522,
|
|
"eval_samples_per_second": 1377.233,
|
|
"eval_steps_per_second": 172.399,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 5.176019096374512,
|
|
"epoch": 4.870740941049216,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003273583644789772,
|
|
"loss": 4.8279,
|
|
"mean_token_accuracy": 0.21740572154521942,
|
|
"num_tokens": 10116420.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 5.1739624500274655,
|
|
"epoch": 4.876149269875609,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00032694112553039795,
|
|
"loss": 4.8221,
|
|
"mean_token_accuracy": 0.21809208244085312,
|
|
"num_tokens": 10127289.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 5.267616081237793,
|
|
"epoch": 4.881557598702001,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003265236978276484,
|
|
"loss": 4.9799,
|
|
"mean_token_accuracy": 0.20410214364528656,
|
|
"num_tokens": 10139216.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 5.264618635177612,
|
|
"epoch": 4.886965927528394,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00032610608288766185,
|
|
"loss": 4.8405,
|
|
"mean_token_accuracy": 0.22492994666099547,
|
|
"num_tokens": 10150010.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 5.191774082183838,
|
|
"epoch": 4.8923742563547865,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00032568828222805216,
|
|
"loss": 4.842,
|
|
"mean_token_accuracy": 0.22070289999246598,
|
|
"num_tokens": 10161317.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 5.188334321975708,
|
|
"epoch": 4.897782585181179,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003252702973671075,
|
|
"loss": 4.8601,
|
|
"mean_token_accuracy": 0.22831744104623794,
|
|
"num_tokens": 10172681.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 5.197067356109619,
|
|
"epoch": 4.903190914007571,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00032485212982378615,
|
|
"loss": 4.8687,
|
|
"mean_token_accuracy": 0.2179756835103035,
|
|
"num_tokens": 10184755.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 5.138756656646729,
|
|
"epoch": 4.908599242833964,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003244337811177098,
|
|
"loss": 4.8367,
|
|
"mean_token_accuracy": 0.2261403977870941,
|
|
"num_tokens": 10195252.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 5.289881849288941,
|
|
"epoch": 4.914007571660357,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003240152527691586,
|
|
"loss": 4.9512,
|
|
"mean_token_accuracy": 0.2024526536464691,
|
|
"num_tokens": 10207660.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 5.214212417602539,
|
|
"epoch": 4.91941590048675,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00032359654629906555,
|
|
"loss": 4.8756,
|
|
"mean_token_accuracy": 0.21816073954105378,
|
|
"num_tokens": 10218154.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 5.156635046005249,
|
|
"epoch": 4.924824229313142,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00032317766322901106,
|
|
"loss": 4.819,
|
|
"mean_token_accuracy": 0.21561871021986007,
|
|
"num_tokens": 10229624.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 5.189560031890869,
|
|
"epoch": 4.930232558139535,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000322758605081217,
|
|
"loss": 4.7998,
|
|
"mean_token_accuracy": 0.22236124873161317,
|
|
"num_tokens": 10240338.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 5.189955425262451,
|
|
"epoch": 4.935640886965928,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00032233937337854167,
|
|
"loss": 4.8445,
|
|
"mean_token_accuracy": 0.22020443379878998,
|
|
"num_tokens": 10250721.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 5.203208780288696,
|
|
"epoch": 4.9410492157923205,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003219199696444741,
|
|
"loss": 4.9041,
|
|
"mean_token_accuracy": 0.20997923016548156,
|
|
"num_tokens": 10262596.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 5.293448352813721,
|
|
"epoch": 4.9464575446187125,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00032150039540312843,
|
|
"loss": 5.0121,
|
|
"mean_token_accuracy": 0.20630424320697785,
|
|
"num_tokens": 10275043.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 5.2475017547607425,
|
|
"epoch": 4.951865873445105,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00032108065217923836,
|
|
"loss": 4.947,
|
|
"mean_token_accuracy": 0.20821117758750915,
|
|
"num_tokens": 10286593.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 5.344606494903564,
|
|
"epoch": 4.957274202271498,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00032066074149815184,
|
|
"loss": 4.985,
|
|
"mean_token_accuracy": 0.20951248258352279,
|
|
"num_tokens": 10297459.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 5.108428478240967,
|
|
"epoch": 4.962682531097891,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00032024066488582517,
|
|
"loss": 4.693,
|
|
"mean_token_accuracy": 0.22870682924985886,
|
|
"num_tokens": 10308397.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 5.2634690284729,
|
|
"epoch": 4.968090859924283,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003198204238688178,
|
|
"loss": 4.987,
|
|
"mean_token_accuracy": 0.20976692736148833,
|
|
"num_tokens": 10319648.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 5.297033309936523,
|
|
"epoch": 4.973499188750676,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00031940001997428645,
|
|
"loss": 4.9528,
|
|
"mean_token_accuracy": 0.2128663420677185,
|
|
"num_tokens": 10330922.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 5.2219380855560305,
|
|
"epoch": 4.978907517577069,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00031897945472997996,
|
|
"loss": 4.8379,
|
|
"mean_token_accuracy": 0.21470965147018434,
|
|
"num_tokens": 10341959.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 5.2132651805877686,
|
|
"epoch": 4.984315846403462,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003185587296642333,
|
|
"loss": 4.9158,
|
|
"mean_token_accuracy": 0.22163569331169128,
|
|
"num_tokens": 10353125.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 5.271954202651978,
|
|
"epoch": 4.9897241752298545,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003181378463059625,
|
|
"loss": 4.9513,
|
|
"mean_token_accuracy": 0.20910066813230516,
|
|
"num_tokens": 10364528.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 5.207727146148682,
|
|
"epoch": 4.995132504056246,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003177168061846584,
|
|
"loss": 4.8094,
|
|
"mean_token_accuracy": 0.2225380539894104,
|
|
"num_tokens": 10375252.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 5.154112180074056,
|
|
"epoch": 5.0,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 0.0003172956108303819,
|
|
"loss": 4.7966,
|
|
"mean_token_accuracy": 0.22260207268926832,
|
|
"num_tokens": 10384445.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 5.183626556396485,
|
|
"epoch": 5.005408328826393,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003168742617737579,
|
|
"loss": 4.6906,
|
|
"mean_token_accuracy": 0.2302063673734665,
|
|
"num_tokens": 10396122.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 5.264249467849732,
|
|
"epoch": 5.010816657652786,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00031645276054597006,
|
|
"loss": 4.632,
|
|
"mean_token_accuracy": 0.23431840538978577,
|
|
"num_tokens": 10409135.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 5.221957778930664,
|
|
"epoch": 5.016224986479178,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00031603110867875435,
|
|
"loss": 4.6118,
|
|
"mean_token_accuracy": 0.24040763229131698,
|
|
"num_tokens": 10421068.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 5.144167900085449,
|
|
"epoch": 5.0216333153055706,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000315609307704395,
|
|
"loss": 4.572,
|
|
"mean_token_accuracy": 0.23828354775905608,
|
|
"num_tokens": 10432117.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 5.25443320274353,
|
|
"epoch": 5.027041644131963,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003151873591557176,
|
|
"loss": 4.6278,
|
|
"mean_token_accuracy": 0.2296961948275566,
|
|
"num_tokens": 10443547.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 5.174649572372436,
|
|
"epoch": 5.032449972958356,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00031476526456608415,
|
|
"loss": 4.6088,
|
|
"mean_token_accuracy": 0.23213777393102647,
|
|
"num_tokens": 10455038.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 5.189857769012451,
|
|
"epoch": 5.037858301784748,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003143430254693873,
|
|
"loss": 4.681,
|
|
"mean_token_accuracy": 0.22490702122449874,
|
|
"num_tokens": 10466682.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 5.183123588562012,
|
|
"epoch": 5.043266630611141,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003139206434000451,
|
|
"loss": 4.636,
|
|
"mean_token_accuracy": 0.2351364314556122,
|
|
"num_tokens": 10478575.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 5.218813276290893,
|
|
"epoch": 5.048674959437534,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003134981198929948,
|
|
"loss": 4.6257,
|
|
"mean_token_accuracy": 0.2343133881688118,
|
|
"num_tokens": 10490088.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 5.146064233779907,
|
|
"epoch": 5.054083288263927,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00031307545648368777,
|
|
"loss": 4.4957,
|
|
"mean_token_accuracy": 0.24293478727340698,
|
|
"num_tokens": 10499877.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 5.0958555221557615,
|
|
"epoch": 5.059491617090319,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00031265265470808393,
|
|
"loss": 4.5224,
|
|
"mean_token_accuracy": 0.24178962260484696,
|
|
"num_tokens": 10509830.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 5.0940409183502195,
|
|
"epoch": 5.064899945916712,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00031222971610264595,
|
|
"loss": 4.5417,
|
|
"mean_token_accuracy": 0.24098313897848128,
|
|
"num_tokens": 10520072.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 5.180801677703857,
|
|
"epoch": 5.0703082747431045,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00031180664220433347,
|
|
"loss": 4.72,
|
|
"mean_token_accuracy": 0.22770989090204238,
|
|
"num_tokens": 10531260.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 5.131834936141968,
|
|
"epoch": 5.075716603569497,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00031138343455059823,
|
|
"loss": 4.5772,
|
|
"mean_token_accuracy": 0.23766616135835647,
|
|
"num_tokens": 10542971.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 5.12978835105896,
|
|
"epoch": 5.081124932395889,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003109600946793778,
|
|
"loss": 4.5699,
|
|
"mean_token_accuracy": 0.2390519469976425,
|
|
"num_tokens": 10554101.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 5.289145421981812,
|
|
"epoch": 5.086533261222282,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00031053662412909024,
|
|
"loss": 4.7099,
|
|
"mean_token_accuracy": 0.22744055986404418,
|
|
"num_tokens": 10565866.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 5.086775588989258,
|
|
"epoch": 5.091941590048675,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003101130244386283,
|
|
"loss": 4.5373,
|
|
"mean_token_accuracy": 0.23784517347812653,
|
|
"num_tokens": 10577276.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 5.214812898635865,
|
|
"epoch": 5.097349918875068,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00030968929714735456,
|
|
"loss": 4.7203,
|
|
"mean_token_accuracy": 0.2304264187812805,
|
|
"num_tokens": 10588625.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 5.108019495010376,
|
|
"epoch": 5.10275824770146,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003092654437950947,
|
|
"loss": 4.6251,
|
|
"mean_token_accuracy": 0.23543978184461595,
|
|
"num_tokens": 10599833.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 5.139721298217774,
|
|
"epoch": 5.108166576527853,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003088414659221331,
|
|
"loss": 4.6144,
|
|
"mean_token_accuracy": 0.23015127032995225,
|
|
"num_tokens": 10610156.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 5.15448899269104,
|
|
"epoch": 5.113574905354246,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000308417365069206,
|
|
"loss": 4.6165,
|
|
"mean_token_accuracy": 0.23851192742586136,
|
|
"num_tokens": 10619839.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 5.13133807182312,
|
|
"epoch": 5.1189832341806385,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00030799314277749697,
|
|
"loss": 4.6024,
|
|
"mean_token_accuracy": 0.2315606042742729,
|
|
"num_tokens": 10631671.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 5.133159446716308,
|
|
"epoch": 5.1243915630070305,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00030756880058863085,
|
|
"loss": 4.6163,
|
|
"mean_token_accuracy": 0.2361859440803528,
|
|
"num_tokens": 10642597.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 5.223555994033814,
|
|
"epoch": 5.129799891833423,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00030714434004466823,
|
|
"loss": 4.6866,
|
|
"mean_token_accuracy": 0.22340937405824662,
|
|
"num_tokens": 10653323.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 5.236954307556152,
|
|
"epoch": 5.135208220659816,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00030671976268809947,
|
|
"loss": 4.6985,
|
|
"mean_token_accuracy": 0.22893307358026505,
|
|
"num_tokens": 10665572.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 5.107706022262573,
|
|
"epoch": 5.140616549486209,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00030629507006183995,
|
|
"loss": 4.5056,
|
|
"mean_token_accuracy": 0.2415497213602066,
|
|
"num_tokens": 10676686.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 5.134891414642334,
|
|
"epoch": 5.146024878312601,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003058702637092234,
|
|
"loss": 4.7168,
|
|
"mean_token_accuracy": 0.22589138597249986,
|
|
"num_tokens": 10687188.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 5.1993269443511965,
|
|
"epoch": 5.151433207138994,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00030544534517399724,
|
|
"loss": 4.615,
|
|
"mean_token_accuracy": 0.2293543353676796,
|
|
"num_tokens": 10698343.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 5.1077189445495605,
|
|
"epoch": 5.156841535965387,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003050203160003163,
|
|
"loss": 4.6085,
|
|
"mean_token_accuracy": 0.23088880628347397,
|
|
"num_tokens": 10710732.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 5.20466194152832,
|
|
"epoch": 5.16224986479178,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003045951777327377,
|
|
"loss": 4.6813,
|
|
"mean_token_accuracy": 0.23435860872268677,
|
|
"num_tokens": 10722397.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 5.103364753723144,
|
|
"epoch": 5.167658193618172,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003041699319162148,
|
|
"loss": 4.5788,
|
|
"mean_token_accuracy": 0.2383410796523094,
|
|
"num_tokens": 10733714.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 5.1291268825531,
|
|
"epoch": 5.173066522444564,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00030374458009609193,
|
|
"loss": 4.6379,
|
|
"mean_token_accuracy": 0.23427872210741044,
|
|
"num_tokens": 10745748.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 5.229328536987305,
|
|
"epoch": 5.178474851270957,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003033191238180985,
|
|
"loss": 4.6549,
|
|
"mean_token_accuracy": 0.23590377867221832,
|
|
"num_tokens": 10756094.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 5.173886346817016,
|
|
"epoch": 5.18388318009735,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003028935646283438,
|
|
"loss": 4.736,
|
|
"mean_token_accuracy": 0.22904880791902543,
|
|
"num_tokens": 10768068.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 5.11404128074646,
|
|
"epoch": 5.189291508923742,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00030246790407331063,
|
|
"loss": 4.5894,
|
|
"mean_token_accuracy": 0.2352805554866791,
|
|
"num_tokens": 10778917.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 5.126118135452271,
|
|
"epoch": 5.194699837750135,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00030204214369985066,
|
|
"loss": 4.6177,
|
|
"mean_token_accuracy": 0.23723979741334916,
|
|
"num_tokens": 10790552.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 5.179587888717651,
|
|
"epoch": 5.200108166576528,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00030161628505517797,
|
|
"loss": 4.6579,
|
|
"mean_token_accuracy": 0.23024794459342957,
|
|
"num_tokens": 10803199.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 5.08224835395813,
|
|
"epoch": 5.205516495402921,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00030119032968686397,
|
|
"loss": 4.5776,
|
|
"mean_token_accuracy": 0.2430170789361,
|
|
"num_tokens": 10814339.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 5.168409299850464,
|
|
"epoch": 5.210924824229313,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00030076427914283127,
|
|
"loss": 4.665,
|
|
"mean_token_accuracy": 0.22666335999965667,
|
|
"num_tokens": 10825776.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 5.178348445892334,
|
|
"epoch": 5.2163331530557056,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00030033813497134864,
|
|
"loss": 4.5855,
|
|
"mean_token_accuracy": 0.2354062333703041,
|
|
"num_tokens": 10837179.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 5.087699127197266,
|
|
"epoch": 5.221741481882098,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000299911898721025,
|
|
"loss": 4.6416,
|
|
"mean_token_accuracy": 0.23371794670820237,
|
|
"num_tokens": 10847205.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 5.158977317810058,
|
|
"epoch": 5.227149810708491,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00029948557194080385,
|
|
"loss": 4.6228,
|
|
"mean_token_accuracy": 0.2304420843720436,
|
|
"num_tokens": 10859895.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 5.065768718719482,
|
|
"epoch": 5.232558139534884,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002990591561799576,
|
|
"loss": 4.6443,
|
|
"mean_token_accuracy": 0.22677141129970552,
|
|
"num_tokens": 10871286.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 5.093819189071655,
|
|
"epoch": 5.237966468361276,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002986326529880823,
|
|
"loss": 4.498,
|
|
"mean_token_accuracy": 0.24408798068761825,
|
|
"num_tokens": 10883479.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 5.195363569259643,
|
|
"epoch": 5.243374797187669,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002982060639150913,
|
|
"loss": 4.6615,
|
|
"mean_token_accuracy": 0.23751705437898635,
|
|
"num_tokens": 10895432.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 5.118285417556763,
|
|
"epoch": 5.248783126014062,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002977793905112106,
|
|
"loss": 4.7015,
|
|
"mean_token_accuracy": 0.2321558937430382,
|
|
"num_tokens": 10906891.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 5.12140645980835,
|
|
"epoch": 5.254191454840455,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00029735263432697183,
|
|
"loss": 4.6988,
|
|
"mean_token_accuracy": 0.22902847975492477,
|
|
"num_tokens": 10917706.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 5.136018991470337,
|
|
"epoch": 5.259599783666847,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002969257969132085,
|
|
"loss": 4.6658,
|
|
"mean_token_accuracy": 0.22978196293115616,
|
|
"num_tokens": 10929325.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 5.215710067749024,
|
|
"epoch": 5.2650081124932395,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002964988798210485,
|
|
"loss": 4.6789,
|
|
"mean_token_accuracy": 0.21959993690252305,
|
|
"num_tokens": 10939899.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 5.146029806137085,
|
|
"epoch": 5.270416441319632,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002960718846019095,
|
|
"loss": 4.6198,
|
|
"mean_token_accuracy": 0.2347557798027992,
|
|
"num_tokens": 10949801.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 4.984179592132568,
|
|
"epoch": 5.275824770146025,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002956448128074931,
|
|
"loss": 4.501,
|
|
"mean_token_accuracy": 0.24598432034254075,
|
|
"num_tokens": 10960191.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 5.2069847106933596,
|
|
"epoch": 5.281233098972417,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002952176659897794,
|
|
"loss": 4.737,
|
|
"mean_token_accuracy": 0.23088638186454774,
|
|
"num_tokens": 10972818.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 5.195149612426758,
|
|
"epoch": 5.28664142779881,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002947904457010207,
|
|
"loss": 4.6787,
|
|
"mean_token_accuracy": 0.22410915046930313,
|
|
"num_tokens": 10984320.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 5.108417129516601,
|
|
"epoch": 5.292049756625203,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002943631534937366,
|
|
"loss": 4.6012,
|
|
"mean_token_accuracy": 0.23733832836151122,
|
|
"num_tokens": 10995262.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 5.069682502746582,
|
|
"epoch": 5.297458085451596,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002939357909207079,
|
|
"loss": 4.6342,
|
|
"mean_token_accuracy": 0.23058290928602218,
|
|
"num_tokens": 11005788.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 5.149464893341064,
|
|
"epoch": 5.302866414277988,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002935083595349713,
|
|
"loss": 4.6274,
|
|
"mean_token_accuracy": 0.23659219294786454,
|
|
"num_tokens": 11016772.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 5.177496242523193,
|
|
"epoch": 5.308274743104381,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00029308086088981335,
|
|
"loss": 4.6964,
|
|
"mean_token_accuracy": 0.22869670540094375,
|
|
"num_tokens": 11028917.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 5.174309396743775,
|
|
"epoch": 5.3136830719307735,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00029265329653876515,
|
|
"loss": 4.6785,
|
|
"mean_token_accuracy": 0.2267504572868347,
|
|
"num_tokens": 11040500.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 5.121609926223755,
|
|
"epoch": 5.319091400757166,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002922256680355965,
|
|
"loss": 4.6176,
|
|
"mean_token_accuracy": 0.2351036936044693,
|
|
"num_tokens": 11052173.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 5.173959636688233,
|
|
"epoch": 5.324499729583558,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00029179797693431046,
|
|
"loss": 4.6229,
|
|
"mean_token_accuracy": 0.23528208434581757,
|
|
"num_tokens": 11061917.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 5.161088657379151,
|
|
"epoch": 5.329908058409951,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002913702247891374,
|
|
"loss": 4.633,
|
|
"mean_token_accuracy": 0.24165104031562806,
|
|
"num_tokens": 11072820.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 5.192854261398315,
|
|
"epoch": 5.335316387236344,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002909424131545296,
|
|
"loss": 4.7038,
|
|
"mean_token_accuracy": 0.22601987272500992,
|
|
"num_tokens": 11083140.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 5.192864942550659,
|
|
"epoch": 5.340724716062737,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00029051454358515553,
|
|
"loss": 4.6773,
|
|
"mean_token_accuracy": 0.2267790526151657,
|
|
"num_tokens": 11094314.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 5.162869548797607,
|
|
"epoch": 5.346133044889129,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00029008661763589436,
|
|
"loss": 4.7076,
|
|
"mean_token_accuracy": 0.22938373535871506,
|
|
"num_tokens": 11106748.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 5.157003259658813,
|
|
"epoch": 5.351541373715522,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002896586368618297,
|
|
"loss": 4.6919,
|
|
"mean_token_accuracy": 0.23329251408576965,
|
|
"num_tokens": 11118702.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 5.14311032295227,
|
|
"epoch": 5.356949702541915,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00028923060281824483,
|
|
"loss": 4.6732,
|
|
"mean_token_accuracy": 0.2248365581035614,
|
|
"num_tokens": 11130782.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 5.1537618160247805,
|
|
"epoch": 5.3623580313683075,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002888025170606165,
|
|
"loss": 4.6368,
|
|
"mean_token_accuracy": 0.23271929770708083,
|
|
"num_tokens": 11142536.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 5.19008150100708,
|
|
"epoch": 5.367766360194699,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00028837438114460924,
|
|
"loss": 4.674,
|
|
"mean_token_accuracy": 0.22938187271356583,
|
|
"num_tokens": 11154027.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 5.112209272384644,
|
|
"epoch": 5.373174689021092,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002879461966260699,
|
|
"loss": 4.6687,
|
|
"mean_token_accuracy": 0.22808918952941895,
|
|
"num_tokens": 11164759.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 5.190345382690429,
|
|
"epoch": 5.378583017847485,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00028751796506102224,
|
|
"loss": 4.6763,
|
|
"mean_token_accuracy": 0.23074355721473694,
|
|
"num_tokens": 11176341.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 5.161106395721435,
|
|
"epoch": 5.383991346673878,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00028708968800566056,
|
|
"loss": 4.6306,
|
|
"mean_token_accuracy": 0.23221389353275299,
|
|
"num_tokens": 11188958.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 5.158840894699097,
|
|
"epoch": 5.38939967550027,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00028666136701634474,
|
|
"loss": 4.721,
|
|
"mean_token_accuracy": 0.22434271425008773,
|
|
"num_tokens": 11200895.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 5.208285093307495,
|
|
"epoch": 5.394808004326663,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002862330036495942,
|
|
"loss": 4.6618,
|
|
"mean_token_accuracy": 0.23629681318998336,
|
|
"num_tokens": 11211563.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 5.236965847015381,
|
|
"epoch": 5.400216333153056,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002858045994620825,
|
|
"loss": 4.7459,
|
|
"mean_token_accuracy": 0.2252361312508583,
|
|
"num_tokens": 11223323.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 5.155903387069702,
|
|
"epoch": 5.405624661979449,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00028537615601063133,
|
|
"loss": 4.6376,
|
|
"mean_token_accuracy": 0.22755652368068696,
|
|
"num_tokens": 11235991.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 5.405624661979449,
|
|
"eval_entropy": 5.087796183065935,
|
|
"eval_loss": 5.888650894165039,
|
|
"eval_mean_token_accuracy": 0.17962701530619102,
|
|
"eval_num_tokens": 11235991.0,
|
|
"eval_runtime": 2.5675,
|
|
"eval_samples_per_second": 1369.034,
|
|
"eval_steps_per_second": 171.373,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"entropy": 5.115123462677002,
|
|
"epoch": 5.411032990805841,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002849476748522052,
|
|
"loss": 4.637,
|
|
"mean_token_accuracy": 0.22837891280651093,
|
|
"num_tokens": 11247406.0,
|
|
"step": 5005
|
|
},
|
|
{
|
|
"entropy": 5.056633710861206,
|
|
"epoch": 5.416441319632233,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002845191575439057,
|
|
"loss": 4.6158,
|
|
"mean_token_accuracy": 0.22495330274105071,
|
|
"num_tokens": 11259031.0,
|
|
"step": 5010
|
|
},
|
|
{
|
|
"entropy": 5.132718181610107,
|
|
"epoch": 5.421849648458626,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002840906056429658,
|
|
"loss": 4.6856,
|
|
"mean_token_accuracy": 0.2258858785033226,
|
|
"num_tokens": 11270321.0,
|
|
"step": 5015
|
|
},
|
|
{
|
|
"entropy": 5.180118083953857,
|
|
"epoch": 5.427257977285019,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002836620207067439,
|
|
"loss": 4.7893,
|
|
"mean_token_accuracy": 0.21195855140686035,
|
|
"num_tokens": 11282209.0,
|
|
"step": 5020
|
|
},
|
|
{
|
|
"entropy": 5.177636528015137,
|
|
"epoch": 5.432666306111411,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00028323340429271875,
|
|
"loss": 4.6598,
|
|
"mean_token_accuracy": 0.23256608098745346,
|
|
"num_tokens": 11292987.0,
|
|
"step": 5025
|
|
},
|
|
{
|
|
"entropy": 5.197863292694092,
|
|
"epoch": 5.438074634937804,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002828047579584833,
|
|
"loss": 4.6559,
|
|
"mean_token_accuracy": 0.2316030815243721,
|
|
"num_tokens": 11303197.0,
|
|
"step": 5030
|
|
},
|
|
{
|
|
"entropy": 5.056801748275757,
|
|
"epoch": 5.443482963764197,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00028237608326173944,
|
|
"loss": 4.5681,
|
|
"mean_token_accuracy": 0.24357122629880906,
|
|
"num_tokens": 11315229.0,
|
|
"step": 5035
|
|
},
|
|
{
|
|
"entropy": 5.129215335845947,
|
|
"epoch": 5.44889129259059,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00028194738176029206,
|
|
"loss": 4.6745,
|
|
"mean_token_accuracy": 0.2339717075228691,
|
|
"num_tokens": 11326033.0,
|
|
"step": 5040
|
|
},
|
|
{
|
|
"entropy": 5.128978967666626,
|
|
"epoch": 5.4542996214169825,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00028151865501204316,
|
|
"loss": 4.6681,
|
|
"mean_token_accuracy": 0.22977797985076903,
|
|
"num_tokens": 11335670.0,
|
|
"step": 5045
|
|
},
|
|
{
|
|
"entropy": 5.134216785430908,
|
|
"epoch": 5.4597079502433745,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00028108990457498696,
|
|
"loss": 4.6033,
|
|
"mean_token_accuracy": 0.23949703127145766,
|
|
"num_tokens": 11346882.0,
|
|
"step": 5050
|
|
},
|
|
{
|
|
"entropy": 5.111203193664551,
|
|
"epoch": 5.465116279069767,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00028066113200720343,
|
|
"loss": 4.6063,
|
|
"mean_token_accuracy": 0.23930297940969467,
|
|
"num_tokens": 11358141.0,
|
|
"step": 5055
|
|
},
|
|
{
|
|
"entropy": 5.0704785823822025,
|
|
"epoch": 5.47052460789616,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00028023233886685314,
|
|
"loss": 4.6401,
|
|
"mean_token_accuracy": 0.22616044878959657,
|
|
"num_tokens": 11368981.0,
|
|
"step": 5060
|
|
},
|
|
{
|
|
"entropy": 5.196547794342041,
|
|
"epoch": 5.475932936722553,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002798035267121713,
|
|
"loss": 4.7502,
|
|
"mean_token_accuracy": 0.22203879058361053,
|
|
"num_tokens": 11381613.0,
|
|
"step": 5065
|
|
},
|
|
{
|
|
"entropy": 5.221626424789429,
|
|
"epoch": 5.481341265548945,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002793746971014623,
|
|
"loss": 4.669,
|
|
"mean_token_accuracy": 0.23037578761577607,
|
|
"num_tokens": 11392321.0,
|
|
"step": 5070
|
|
},
|
|
{
|
|
"entropy": 5.115049552917481,
|
|
"epoch": 5.486749594375338,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002789458515930939,
|
|
"loss": 4.6863,
|
|
"mean_token_accuracy": 0.22719697654247284,
|
|
"num_tokens": 11403546.0,
|
|
"step": 5075
|
|
},
|
|
{
|
|
"entropy": 5.097189426422119,
|
|
"epoch": 5.492157923201731,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00027851699174549164,
|
|
"loss": 4.6121,
|
|
"mean_token_accuracy": 0.2367207318544388,
|
|
"num_tokens": 11415167.0,
|
|
"step": 5080
|
|
},
|
|
{
|
|
"entropy": 5.148455619812012,
|
|
"epoch": 5.497566252028124,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002780881191171333,
|
|
"loss": 4.6189,
|
|
"mean_token_accuracy": 0.239136403799057,
|
|
"num_tokens": 11426715.0,
|
|
"step": 5085
|
|
},
|
|
{
|
|
"entropy": 5.117072439193725,
|
|
"epoch": 5.502974580854516,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00027765923526654304,
|
|
"loss": 4.6021,
|
|
"mean_token_accuracy": 0.23602851629257202,
|
|
"num_tokens": 11437760.0,
|
|
"step": 5090
|
|
},
|
|
{
|
|
"entropy": 5.156884670257568,
|
|
"epoch": 5.5083829096809085,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00027723034175228554,
|
|
"loss": 4.6741,
|
|
"mean_token_accuracy": 0.23744746744632722,
|
|
"num_tokens": 11448329.0,
|
|
"step": 5095
|
|
},
|
|
{
|
|
"entropy": 5.049209403991699,
|
|
"epoch": 5.513791238507301,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000276801440132961,
|
|
"loss": 4.5005,
|
|
"mean_token_accuracy": 0.24241209030151367,
|
|
"num_tokens": 11459173.0,
|
|
"step": 5100
|
|
},
|
|
{
|
|
"entropy": 5.099529695510864,
|
|
"epoch": 5.519199567333694,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002763725319671989,
|
|
"loss": 4.6106,
|
|
"mean_token_accuracy": 0.22774458080530166,
|
|
"num_tokens": 11470232.0,
|
|
"step": 5105
|
|
},
|
|
{
|
|
"entropy": 5.068125247955322,
|
|
"epoch": 5.524607896160086,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002759436188136525,
|
|
"loss": 4.6171,
|
|
"mean_token_accuracy": 0.2360287368297577,
|
|
"num_tokens": 11481181.0,
|
|
"step": 5110
|
|
},
|
|
{
|
|
"entropy": 5.134531545639038,
|
|
"epoch": 5.530016224986479,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00027551470223099306,
|
|
"loss": 4.654,
|
|
"mean_token_accuracy": 0.22781693190336227,
|
|
"num_tokens": 11491772.0,
|
|
"step": 5115
|
|
},
|
|
{
|
|
"entropy": 5.1614173412322994,
|
|
"epoch": 5.535424553812872,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002750857837779045,
|
|
"loss": 4.6179,
|
|
"mean_token_accuracy": 0.23431987762451173,
|
|
"num_tokens": 11503895.0,
|
|
"step": 5120
|
|
},
|
|
{
|
|
"entropy": 5.131058216094971,
|
|
"epoch": 5.540832882639265,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002746568650130775,
|
|
"loss": 4.6189,
|
|
"mean_token_accuracy": 0.23728804290294647,
|
|
"num_tokens": 11514189.0,
|
|
"step": 5125
|
|
},
|
|
{
|
|
"entropy": 5.040120363235474,
|
|
"epoch": 5.546241211465657,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002742279474952039,
|
|
"loss": 4.645,
|
|
"mean_token_accuracy": 0.23493632823228836,
|
|
"num_tokens": 11526142.0,
|
|
"step": 5130
|
|
},
|
|
{
|
|
"entropy": 5.180205678939819,
|
|
"epoch": 5.55164954029205,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00027379903278297086,
|
|
"loss": 4.6787,
|
|
"mean_token_accuracy": 0.22211943715810775,
|
|
"num_tokens": 11537061.0,
|
|
"step": 5135
|
|
},
|
|
{
|
|
"entropy": 5.234910774230957,
|
|
"epoch": 5.5570578691184425,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00027337012243505536,
|
|
"loss": 4.7705,
|
|
"mean_token_accuracy": 0.21657010316848754,
|
|
"num_tokens": 11547319.0,
|
|
"step": 5140
|
|
},
|
|
{
|
|
"entropy": 5.170358943939209,
|
|
"epoch": 5.562466197944835,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00027294121801011887,
|
|
"loss": 4.6677,
|
|
"mean_token_accuracy": 0.23267145305871964,
|
|
"num_tokens": 11558766.0,
|
|
"step": 5145
|
|
},
|
|
{
|
|
"entropy": 5.097517633438111,
|
|
"epoch": 5.567874526771227,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00027251232106680084,
|
|
"loss": 4.6077,
|
|
"mean_token_accuracy": 0.23632706254720687,
|
|
"num_tokens": 11570039.0,
|
|
"step": 5150
|
|
},
|
|
{
|
|
"entropy": 5.04287428855896,
|
|
"epoch": 5.57328285559762,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00027208343316371383,
|
|
"loss": 4.6447,
|
|
"mean_token_accuracy": 0.23549768179655076,
|
|
"num_tokens": 11581664.0,
|
|
"step": 5155
|
|
},
|
|
{
|
|
"entropy": 5.121302652359009,
|
|
"epoch": 5.578691184424013,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002716545558594373,
|
|
"loss": 4.6765,
|
|
"mean_token_accuracy": 0.22490311712026595,
|
|
"num_tokens": 11592689.0,
|
|
"step": 5160
|
|
},
|
|
{
|
|
"entropy": 5.134082555770874,
|
|
"epoch": 5.584099513250406,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002712256907125126,
|
|
"loss": 4.5725,
|
|
"mean_token_accuracy": 0.24146390706300735,
|
|
"num_tokens": 11604315.0,
|
|
"step": 5165
|
|
},
|
|
{
|
|
"entropy": 5.171964120864868,
|
|
"epoch": 5.589507842076799,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002707968392814366,
|
|
"loss": 4.68,
|
|
"mean_token_accuracy": 0.23471864312887192,
|
|
"num_tokens": 11614510.0,
|
|
"step": 5170
|
|
},
|
|
{
|
|
"entropy": 5.124934530258178,
|
|
"epoch": 5.594916170903191,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00027036800312465635,
|
|
"loss": 4.6327,
|
|
"mean_token_accuracy": 0.23509945422410966,
|
|
"num_tokens": 11625501.0,
|
|
"step": 5175
|
|
},
|
|
{
|
|
"entropy": 5.177364158630371,
|
|
"epoch": 5.600324499729584,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002699391838005634,
|
|
"loss": 4.6857,
|
|
"mean_token_accuracy": 0.23527527302503587,
|
|
"num_tokens": 11636841.0,
|
|
"step": 5180
|
|
},
|
|
{
|
|
"entropy": 5.07881727218628,
|
|
"epoch": 5.605732828555976,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00026951038286748824,
|
|
"loss": 4.5943,
|
|
"mean_token_accuracy": 0.2398626759648323,
|
|
"num_tokens": 11647475.0,
|
|
"step": 5185
|
|
},
|
|
{
|
|
"entropy": 5.177493572235107,
|
|
"epoch": 5.611141157382368,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002690816018836945,
|
|
"loss": 4.6982,
|
|
"mean_token_accuracy": 0.22178054302930833,
|
|
"num_tokens": 11658887.0,
|
|
"step": 5190
|
|
},
|
|
{
|
|
"entropy": 5.159755039215088,
|
|
"epoch": 5.616549486208761,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002686528424073731,
|
|
"loss": 4.7617,
|
|
"mean_token_accuracy": 0.22504634112119676,
|
|
"num_tokens": 11669658.0,
|
|
"step": 5195
|
|
},
|
|
{
|
|
"entropy": 5.172432088851929,
|
|
"epoch": 5.621957815035154,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00026822410599663704,
|
|
"loss": 4.7123,
|
|
"mean_token_accuracy": 0.2242208406329155,
|
|
"num_tokens": 11681792.0,
|
|
"step": 5200
|
|
},
|
|
{
|
|
"entropy": 5.1599345207214355,
|
|
"epoch": 5.627366143861547,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00026779539420951554,
|
|
"loss": 4.6634,
|
|
"mean_token_accuracy": 0.23229407966136933,
|
|
"num_tokens": 11691903.0,
|
|
"step": 5205
|
|
},
|
|
{
|
|
"entropy": 5.1540144920349125,
|
|
"epoch": 5.63277447268794,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002673667086039481,
|
|
"loss": 4.6961,
|
|
"mean_token_accuracy": 0.23588108718395234,
|
|
"num_tokens": 11703383.0,
|
|
"step": 5210
|
|
},
|
|
{
|
|
"entropy": 5.111746406555175,
|
|
"epoch": 5.638182801514332,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002669380507377792,
|
|
"loss": 4.6596,
|
|
"mean_token_accuracy": 0.23208150118589402,
|
|
"num_tokens": 11714721.0,
|
|
"step": 5215
|
|
},
|
|
{
|
|
"entropy": 5.153729391098023,
|
|
"epoch": 5.643591130340725,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002665094221687525,
|
|
"loss": 4.6984,
|
|
"mean_token_accuracy": 0.22700366228818894,
|
|
"num_tokens": 11726016.0,
|
|
"step": 5220
|
|
},
|
|
{
|
|
"entropy": 5.095989418029785,
|
|
"epoch": 5.6489994591671175,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002660808244545054,
|
|
"loss": 4.5116,
|
|
"mean_token_accuracy": 0.24052972942590714,
|
|
"num_tokens": 11737589.0,
|
|
"step": 5225
|
|
},
|
|
{
|
|
"entropy": 5.097131443023682,
|
|
"epoch": 5.6544077879935095,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00026565225915256276,
|
|
"loss": 4.6083,
|
|
"mean_token_accuracy": 0.24009061306715013,
|
|
"num_tokens": 11747969.0,
|
|
"step": 5230
|
|
},
|
|
{
|
|
"entropy": 5.071250867843628,
|
|
"epoch": 5.659816116819902,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000265223727820332,
|
|
"loss": 4.5905,
|
|
"mean_token_accuracy": 0.2303268149495125,
|
|
"num_tokens": 11759535.0,
|
|
"step": 5235
|
|
},
|
|
{
|
|
"entropy": 5.12540111541748,
|
|
"epoch": 5.665224445646295,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002647952320150969,
|
|
"loss": 4.7305,
|
|
"mean_token_accuracy": 0.2214445501565933,
|
|
"num_tokens": 11770863.0,
|
|
"step": 5240
|
|
},
|
|
{
|
|
"entropy": 5.165092515945434,
|
|
"epoch": 5.670632774472688,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002643667732940122,
|
|
"loss": 4.6362,
|
|
"mean_token_accuracy": 0.2316431522369385,
|
|
"num_tokens": 11782643.0,
|
|
"step": 5245
|
|
},
|
|
{
|
|
"entropy": 5.159668207168579,
|
|
"epoch": 5.676041103299081,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00026393835321409796,
|
|
"loss": 4.6721,
|
|
"mean_token_accuracy": 0.23681631833314895,
|
|
"num_tokens": 11794468.0,
|
|
"step": 5250
|
|
},
|
|
{
|
|
"entropy": 5.129774379730224,
|
|
"epoch": 5.681449432125473,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00026350997333223375,
|
|
"loss": 4.6291,
|
|
"mean_token_accuracy": 0.2382364645600319,
|
|
"num_tokens": 11807051.0,
|
|
"step": 5255
|
|
},
|
|
{
|
|
"entropy": 5.075513172149658,
|
|
"epoch": 5.686857760951866,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00026308163520515293,
|
|
"loss": 4.6556,
|
|
"mean_token_accuracy": 0.22641628980636597,
|
|
"num_tokens": 11817067.0,
|
|
"step": 5260
|
|
},
|
|
{
|
|
"entropy": 5.087302207946777,
|
|
"epoch": 5.692266089778259,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00026265334038943735,
|
|
"loss": 4.6652,
|
|
"mean_token_accuracy": 0.23625146299600602,
|
|
"num_tokens": 11828111.0,
|
|
"step": 5265
|
|
},
|
|
{
|
|
"entropy": 5.175454711914062,
|
|
"epoch": 5.6976744186046515,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002622250904415113,
|
|
"loss": 4.6421,
|
|
"mean_token_accuracy": 0.23426122069358826,
|
|
"num_tokens": 11838536.0,
|
|
"step": 5270
|
|
},
|
|
{
|
|
"entropy": 5.186965131759644,
|
|
"epoch": 5.7030827474310435,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002617968869176362,
|
|
"loss": 4.6788,
|
|
"mean_token_accuracy": 0.23464035391807556,
|
|
"num_tokens": 11848899.0,
|
|
"step": 5275
|
|
},
|
|
{
|
|
"entropy": 5.104984474182129,
|
|
"epoch": 5.708491076257436,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002613687313739044,
|
|
"loss": 4.7192,
|
|
"mean_token_accuracy": 0.23104217797517776,
|
|
"num_tokens": 11861023.0,
|
|
"step": 5280
|
|
},
|
|
{
|
|
"entropy": 5.105616474151612,
|
|
"epoch": 5.713899405083829,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002609406253662344,
|
|
"loss": 4.6765,
|
|
"mean_token_accuracy": 0.22527302354574202,
|
|
"num_tokens": 11871953.0,
|
|
"step": 5285
|
|
},
|
|
{
|
|
"entropy": 5.189648151397705,
|
|
"epoch": 5.719307733910222,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002605125704503643,
|
|
"loss": 4.6951,
|
|
"mean_token_accuracy": 0.2301825389266014,
|
|
"num_tokens": 11882874.0,
|
|
"step": 5290
|
|
},
|
|
{
|
|
"entropy": 5.215121936798096,
|
|
"epoch": 5.724716062736614,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002600845681818467,
|
|
"loss": 4.6066,
|
|
"mean_token_accuracy": 0.23581512719392778,
|
|
"num_tokens": 11893431.0,
|
|
"step": 5295
|
|
},
|
|
{
|
|
"entropy": 5.056259632110596,
|
|
"epoch": 5.730124391563007,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002596566201160427,
|
|
"loss": 4.6309,
|
|
"mean_token_accuracy": 0.23407334238290786,
|
|
"num_tokens": 11905311.0,
|
|
"step": 5300
|
|
},
|
|
{
|
|
"entropy": 5.070898914337159,
|
|
"epoch": 5.7355327203894,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00025922872780811664,
|
|
"loss": 4.7192,
|
|
"mean_token_accuracy": 0.2241229459643364,
|
|
"num_tokens": 11916342.0,
|
|
"step": 5305
|
|
},
|
|
{
|
|
"entropy": 5.158933305740357,
|
|
"epoch": 5.740941049215793,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00025880089281303027,
|
|
"loss": 4.6688,
|
|
"mean_token_accuracy": 0.22884765565395354,
|
|
"num_tokens": 11927485.0,
|
|
"step": 5310
|
|
},
|
|
{
|
|
"entropy": 5.075602436065674,
|
|
"epoch": 5.746349378042185,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00025837311668553683,
|
|
"loss": 4.5637,
|
|
"mean_token_accuracy": 0.24008084386587142,
|
|
"num_tokens": 11937968.0,
|
|
"step": 5315
|
|
},
|
|
{
|
|
"entropy": 5.084715557098389,
|
|
"epoch": 5.7517577068685775,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002579454009801757,
|
|
"loss": 4.5358,
|
|
"mean_token_accuracy": 0.24598098695278167,
|
|
"num_tokens": 11947510.0,
|
|
"step": 5320
|
|
},
|
|
{
|
|
"entropy": 5.125357389450073,
|
|
"epoch": 5.75716603569497,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002575177472512669,
|
|
"loss": 4.6972,
|
|
"mean_token_accuracy": 0.2251044288277626,
|
|
"num_tokens": 11960002.0,
|
|
"step": 5325
|
|
},
|
|
{
|
|
"entropy": 5.07122163772583,
|
|
"epoch": 5.762574364521363,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000257090157052905,
|
|
"loss": 4.5977,
|
|
"mean_token_accuracy": 0.22701413482427596,
|
|
"num_tokens": 11970895.0,
|
|
"step": 5330
|
|
},
|
|
{
|
|
"entropy": 5.127821445465088,
|
|
"epoch": 5.767982693347756,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00025666263193895375,
|
|
"loss": 4.6326,
|
|
"mean_token_accuracy": 0.22893535792827607,
|
|
"num_tokens": 11982910.0,
|
|
"step": 5335
|
|
},
|
|
{
|
|
"entropy": 5.170942306518555,
|
|
"epoch": 5.773391022174148,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00025623517346304027,
|
|
"loss": 4.6926,
|
|
"mean_token_accuracy": 0.2247198238968849,
|
|
"num_tokens": 11994066.0,
|
|
"step": 5340
|
|
},
|
|
{
|
|
"entropy": 5.1133551597595215,
|
|
"epoch": 5.778799351000541,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00025580778317855,
|
|
"loss": 4.6162,
|
|
"mean_token_accuracy": 0.23568190038204193,
|
|
"num_tokens": 12004821.0,
|
|
"step": 5345
|
|
},
|
|
{
|
|
"entropy": 5.169846153259277,
|
|
"epoch": 5.784207679826934,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002553804626386199,
|
|
"loss": 4.6674,
|
|
"mean_token_accuracy": 0.2330174595117569,
|
|
"num_tokens": 12014973.0,
|
|
"step": 5350
|
|
},
|
|
{
|
|
"entropy": 5.201066970825195,
|
|
"epoch": 5.789616008653326,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002549532133961339,
|
|
"loss": 4.7757,
|
|
"mean_token_accuracy": 0.2223259463906288,
|
|
"num_tokens": 12026528.0,
|
|
"step": 5355
|
|
},
|
|
{
|
|
"entropy": 5.227397871017456,
|
|
"epoch": 5.795024337479719,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00025452603700371675,
|
|
"loss": 4.7952,
|
|
"mean_token_accuracy": 0.22488296926021575,
|
|
"num_tokens": 12038733.0,
|
|
"step": 5360
|
|
},
|
|
{
|
|
"entropy": 5.201702880859375,
|
|
"epoch": 5.800432666306111,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002540989350137285,
|
|
"loss": 4.7192,
|
|
"mean_token_accuracy": 0.22585128098726273,
|
|
"num_tokens": 12049821.0,
|
|
"step": 5365
|
|
},
|
|
{
|
|
"entropy": 5.238588094711304,
|
|
"epoch": 5.805840995132504,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00025367190897825873,
|
|
"loss": 4.7145,
|
|
"mean_token_accuracy": 0.22480454444885253,
|
|
"num_tokens": 12060982.0,
|
|
"step": 5370
|
|
},
|
|
{
|
|
"entropy": 5.110591125488281,
|
|
"epoch": 5.811249323958897,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002532449604491209,
|
|
"loss": 4.5983,
|
|
"mean_token_accuracy": 0.2373492956161499,
|
|
"num_tokens": 12070842.0,
|
|
"step": 5375
|
|
},
|
|
{
|
|
"entropy": 5.0790657043457035,
|
|
"epoch": 5.816657652785289,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002528180909778471,
|
|
"loss": 4.5958,
|
|
"mean_token_accuracy": 0.24220499098300935,
|
|
"num_tokens": 12080992.0,
|
|
"step": 5380
|
|
},
|
|
{
|
|
"entropy": 5.1046778678894045,
|
|
"epoch": 5.822065981611682,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00025239130211568206,
|
|
"loss": 4.6653,
|
|
"mean_token_accuracy": 0.23016827553510666,
|
|
"num_tokens": 12091905.0,
|
|
"step": 5385
|
|
},
|
|
{
|
|
"entropy": 5.165070247650147,
|
|
"epoch": 5.827474310438075,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00025196459541357725,
|
|
"loss": 4.5532,
|
|
"mean_token_accuracy": 0.24282920509576797,
|
|
"num_tokens": 12101577.0,
|
|
"step": 5390
|
|
},
|
|
{
|
|
"entropy": 5.213714504241944,
|
|
"epoch": 5.832882639264467,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00025153797242218604,
|
|
"loss": 4.7669,
|
|
"mean_token_accuracy": 0.22002027928829193,
|
|
"num_tokens": 12112593.0,
|
|
"step": 5395
|
|
},
|
|
{
|
|
"entropy": 5.118342876434326,
|
|
"epoch": 5.83829096809086,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00025111143469185715,
|
|
"loss": 4.5869,
|
|
"mean_token_accuracy": 0.23804302811622619,
|
|
"num_tokens": 12122947.0,
|
|
"step": 5400
|
|
},
|
|
{
|
|
"entropy": 5.0096839427947994,
|
|
"epoch": 5.8436992969172525,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002506849837726299,
|
|
"loss": 4.5128,
|
|
"mean_token_accuracy": 0.2520808681845665,
|
|
"num_tokens": 12133557.0,
|
|
"step": 5405
|
|
},
|
|
{
|
|
"entropy": 5.1979992389678955,
|
|
"epoch": 5.849107625743645,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002502586212142278,
|
|
"loss": 4.7172,
|
|
"mean_token_accuracy": 0.22161097675561905,
|
|
"num_tokens": 12144735.0,
|
|
"step": 5410
|
|
},
|
|
{
|
|
"entropy": 5.124048948287964,
|
|
"epoch": 5.854515954570038,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00024983234856605337,
|
|
"loss": 4.663,
|
|
"mean_token_accuracy": 0.23320786654949188,
|
|
"num_tokens": 12155786.0,
|
|
"step": 5415
|
|
},
|
|
{
|
|
"entropy": 5.135622644424439,
|
|
"epoch": 5.85992428339643,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00024940616737718235,
|
|
"loss": 4.7425,
|
|
"mean_token_accuracy": 0.22887424230575562,
|
|
"num_tokens": 12168846.0,
|
|
"step": 5420
|
|
},
|
|
{
|
|
"entropy": 5.107228755950928,
|
|
"epoch": 5.865332612222823,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002489800791963583,
|
|
"loss": 4.5878,
|
|
"mean_token_accuracy": 0.23133180141448975,
|
|
"num_tokens": 12179743.0,
|
|
"step": 5425
|
|
},
|
|
{
|
|
"entropy": 5.172464275360108,
|
|
"epoch": 5.870740941049216,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00024855408557198657,
|
|
"loss": 4.6872,
|
|
"mean_token_accuracy": 0.23701113909482957,
|
|
"num_tokens": 12190754.0,
|
|
"step": 5430
|
|
},
|
|
{
|
|
"entropy": 5.1805438041687015,
|
|
"epoch": 5.876149269875609,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000248128188052129,
|
|
"loss": 4.685,
|
|
"mean_token_accuracy": 0.22996634989976883,
|
|
"num_tokens": 12201293.0,
|
|
"step": 5435
|
|
},
|
|
{
|
|
"entropy": 5.139066457748413,
|
|
"epoch": 5.881557598702001,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00024770238818449804,
|
|
"loss": 4.7503,
|
|
"mean_token_accuracy": 0.23059774190187454,
|
|
"num_tokens": 12214361.0,
|
|
"step": 5440
|
|
},
|
|
{
|
|
"entropy": 5.206919717788696,
|
|
"epoch": 5.886965927528394,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00024727668751645164,
|
|
"loss": 4.8121,
|
|
"mean_token_accuracy": 0.218590846657753,
|
|
"num_tokens": 12226766.0,
|
|
"step": 5445
|
|
},
|
|
{
|
|
"entropy": 5.16403374671936,
|
|
"epoch": 5.8923742563547865,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002468510875949868,
|
|
"loss": 4.704,
|
|
"mean_token_accuracy": 0.23323660343885422,
|
|
"num_tokens": 12237339.0,
|
|
"step": 5450
|
|
},
|
|
{
|
|
"entropy": 5.084883975982666,
|
|
"epoch": 5.897782585181179,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00024642558996673474,
|
|
"loss": 4.6055,
|
|
"mean_token_accuracy": 0.24564282447099686,
|
|
"num_tokens": 12248567.0,
|
|
"step": 5455
|
|
},
|
|
{
|
|
"entropy": 5.132214689254761,
|
|
"epoch": 5.903190914007571,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 0.00024600019617795465,
|
|
"loss": 4.6229,
|
|
"mean_token_accuracy": 0.23880134522914886,
|
|
"num_tokens": 12258968.0,
|
|
"step": 5460
|
|
},
|
|
{
|
|
"entropy": 5.194768571853638,
|
|
"epoch": 5.908599242833964,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002455749077745288,
|
|
"loss": 4.6723,
|
|
"mean_token_accuracy": 0.2397978574037552,
|
|
"num_tokens": 12271233.0,
|
|
"step": 5465
|
|
},
|
|
{
|
|
"entropy": 5.153114748001099,
|
|
"epoch": 5.914007571660357,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00024514972630195603,
|
|
"loss": 4.6448,
|
|
"mean_token_accuracy": 0.22066182941198348,
|
|
"num_tokens": 12280634.0,
|
|
"step": 5470
|
|
},
|
|
{
|
|
"entropy": 5.0837061405181885,
|
|
"epoch": 5.91941590048675,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002447246533053469,
|
|
"loss": 4.6781,
|
|
"mean_token_accuracy": 0.23044090270996093,
|
|
"num_tokens": 12292282.0,
|
|
"step": 5475
|
|
},
|
|
{
|
|
"entropy": 5.15794153213501,
|
|
"epoch": 5.924824229313142,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00024429969032941756,
|
|
"loss": 4.725,
|
|
"mean_token_accuracy": 0.22953017801046371,
|
|
"num_tokens": 12303940.0,
|
|
"step": 5480
|
|
},
|
|
{
|
|
"entropy": 5.169911956787109,
|
|
"epoch": 5.930232558139535,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00024387483891848456,
|
|
"loss": 4.6837,
|
|
"mean_token_accuracy": 0.233815798163414,
|
|
"num_tokens": 12315400.0,
|
|
"step": 5485
|
|
},
|
|
{
|
|
"entropy": 5.195281648635865,
|
|
"epoch": 5.935640886965928,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00024345010061645889,
|
|
"loss": 4.7207,
|
|
"mean_token_accuracy": 0.22349406629800797,
|
|
"num_tokens": 12327462.0,
|
|
"step": 5490
|
|
},
|
|
{
|
|
"entropy": 5.138352346420288,
|
|
"epoch": 5.9410492157923205,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00024302547696684047,
|
|
"loss": 4.6781,
|
|
"mean_token_accuracy": 0.23432565182447435,
|
|
"num_tokens": 12338966.0,
|
|
"step": 5495
|
|
},
|
|
{
|
|
"entropy": 5.028692150115967,
|
|
"epoch": 5.9464575446187125,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00024260096951271253,
|
|
"loss": 4.5538,
|
|
"mean_token_accuracy": 0.24259533137083053,
|
|
"num_tokens": 12349663.0,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"epoch": 5.9464575446187125,
|
|
"eval_entropy": 5.069843965226954,
|
|
"eval_loss": 5.858776092529297,
|
|
"eval_mean_token_accuracy": 0.18189243970608168,
|
|
"eval_num_tokens": 12349663.0,
|
|
"eval_runtime": 2.5623,
|
|
"eval_samples_per_second": 1371.803,
|
|
"eval_steps_per_second": 171.719,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"entropy": 5.093153285980224,
|
|
"epoch": 5.951865873445105,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00024217657979673634,
|
|
"loss": 4.6216,
|
|
"mean_token_accuracy": 0.23165910243988036,
|
|
"num_tokens": 12361304.0,
|
|
"step": 5505
|
|
},
|
|
{
|
|
"entropy": 5.061578226089478,
|
|
"epoch": 5.957274202271498,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00024175230936114507,
|
|
"loss": 4.5685,
|
|
"mean_token_accuracy": 0.24055711179971695,
|
|
"num_tokens": 12372162.0,
|
|
"step": 5510
|
|
},
|
|
{
|
|
"entropy": 5.121871709823608,
|
|
"epoch": 5.962682531097891,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002413281597477384,
|
|
"loss": 4.6154,
|
|
"mean_token_accuracy": 0.23273260295391082,
|
|
"num_tokens": 12383916.0,
|
|
"step": 5515
|
|
},
|
|
{
|
|
"entropy": 5.154853773117066,
|
|
"epoch": 5.968090859924283,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00024090413249787696,
|
|
"loss": 4.7141,
|
|
"mean_token_accuracy": 0.23405831307172775,
|
|
"num_tokens": 12394787.0,
|
|
"step": 5520
|
|
},
|
|
{
|
|
"entropy": 5.131857395172119,
|
|
"epoch": 5.973499188750676,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002404802291524768,
|
|
"loss": 4.6224,
|
|
"mean_token_accuracy": 0.24263519942760467,
|
|
"num_tokens": 12405905.0,
|
|
"step": 5525
|
|
},
|
|
{
|
|
"entropy": 5.12526388168335,
|
|
"epoch": 5.978907517577069,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00024005645125200375,
|
|
"loss": 4.6797,
|
|
"mean_token_accuracy": 0.22981744408607482,
|
|
"num_tokens": 12416785.0,
|
|
"step": 5530
|
|
},
|
|
{
|
|
"entropy": 5.065377140045166,
|
|
"epoch": 5.984315846403462,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002396328003364675,
|
|
"loss": 4.6005,
|
|
"mean_token_accuracy": 0.23356126844882966,
|
|
"num_tokens": 12427900.0,
|
|
"step": 5535
|
|
},
|
|
{
|
|
"entropy": 5.174858903884887,
|
|
"epoch": 5.9897241752298545,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00023920927794541665,
|
|
"loss": 4.7204,
|
|
"mean_token_accuracy": 0.23408112525939942,
|
|
"num_tokens": 12439823.0,
|
|
"step": 5540
|
|
},
|
|
{
|
|
"entropy": 5.131950569152832,
|
|
"epoch": 5.995132504056246,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002387858856179324,
|
|
"loss": 4.6496,
|
|
"mean_token_accuracy": 0.227919602394104,
|
|
"num_tokens": 12450988.0,
|
|
"step": 5545
|
|
},
|
|
{
|
|
"entropy": 5.247280120849609,
|
|
"epoch": 6.0,
|
|
"grad_norm": 1.875,
|
|
"learning_rate": 0.00023836262489262361,
|
|
"loss": 4.7147,
|
|
"mean_token_accuracy": 0.23430228730042776,
|
|
"num_tokens": 12461334.0,
|
|
"step": 5550
|
|
},
|
|
{
|
|
"entropy": 5.097964429855347,
|
|
"epoch": 6.005408328826393,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00023793949730762049,
|
|
"loss": 4.4037,
|
|
"mean_token_accuracy": 0.25196374356746676,
|
|
"num_tokens": 12473206.0,
|
|
"step": 5555
|
|
},
|
|
{
|
|
"entropy": 5.13435115814209,
|
|
"epoch": 6.010816657652786,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00023751650440056982,
|
|
"loss": 4.5394,
|
|
"mean_token_accuracy": 0.24281908720731735,
|
|
"num_tokens": 12484247.0,
|
|
"step": 5560
|
|
},
|
|
{
|
|
"entropy": 5.081768941879273,
|
|
"epoch": 6.016224986479178,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00023709364770862885,
|
|
"loss": 4.4357,
|
|
"mean_token_accuracy": 0.2508038252592087,
|
|
"num_tokens": 12495234.0,
|
|
"step": 5565
|
|
},
|
|
{
|
|
"entropy": 5.128761291503906,
|
|
"epoch": 6.0216333153055706,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00023667092876845976,
|
|
"loss": 4.5587,
|
|
"mean_token_accuracy": 0.24328985959291458,
|
|
"num_tokens": 12506516.0,
|
|
"step": 5570
|
|
},
|
|
{
|
|
"entropy": 5.14731502532959,
|
|
"epoch": 6.027041644131963,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002362483491162239,
|
|
"loss": 4.4551,
|
|
"mean_token_accuracy": 0.24906942546367644,
|
|
"num_tokens": 12517238.0,
|
|
"step": 5575
|
|
},
|
|
{
|
|
"entropy": 5.188383150100708,
|
|
"epoch": 6.032449972958356,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00023582591028757706,
|
|
"loss": 4.4732,
|
|
"mean_token_accuracy": 0.24781478196382523,
|
|
"num_tokens": 12527212.0,
|
|
"step": 5580
|
|
},
|
|
{
|
|
"entropy": 5.102848958969116,
|
|
"epoch": 6.037858301784748,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00023540361381766274,
|
|
"loss": 4.4407,
|
|
"mean_token_accuracy": 0.25573561042547227,
|
|
"num_tokens": 12537939.0,
|
|
"step": 5585
|
|
},
|
|
{
|
|
"entropy": 5.0774894714355465,
|
|
"epoch": 6.043266630611141,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002349814612411073,
|
|
"loss": 4.4587,
|
|
"mean_token_accuracy": 0.2469854548573494,
|
|
"num_tokens": 12547987.0,
|
|
"step": 5590
|
|
},
|
|
{
|
|
"entropy": 5.098888254165649,
|
|
"epoch": 6.048674959437534,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00023455945409201411,
|
|
"loss": 4.4606,
|
|
"mean_token_accuracy": 0.24783782809972763,
|
|
"num_tokens": 12558505.0,
|
|
"step": 5595
|
|
},
|
|
{
|
|
"entropy": 5.097957992553711,
|
|
"epoch": 6.054083288263927,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00023413759390395828,
|
|
"loss": 4.3966,
|
|
"mean_token_accuracy": 0.2569520816206932,
|
|
"num_tokens": 12569748.0,
|
|
"step": 5600
|
|
},
|
|
{
|
|
"entropy": 5.053547048568726,
|
|
"epoch": 6.059491617090319,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002337158822099807,
|
|
"loss": 4.4104,
|
|
"mean_token_accuracy": 0.24684248715639115,
|
|
"num_tokens": 12581420.0,
|
|
"step": 5605
|
|
},
|
|
{
|
|
"entropy": 5.152947330474854,
|
|
"epoch": 6.064899945916712,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002332943205425825,
|
|
"loss": 4.4779,
|
|
"mean_token_accuracy": 0.2521220833063126,
|
|
"num_tokens": 12593781.0,
|
|
"step": 5610
|
|
},
|
|
{
|
|
"entropy": 5.054061317443848,
|
|
"epoch": 6.0703082747431045,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00023287291043371978,
|
|
"loss": 4.4265,
|
|
"mean_token_accuracy": 0.24702493101358414,
|
|
"num_tokens": 12606264.0,
|
|
"step": 5615
|
|
},
|
|
{
|
|
"entropy": 5.167685890197754,
|
|
"epoch": 6.075716603569497,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00023245165341479796,
|
|
"loss": 4.4587,
|
|
"mean_token_accuracy": 0.2525437191128731,
|
|
"num_tokens": 12617419.0,
|
|
"step": 5620
|
|
},
|
|
{
|
|
"entropy": 5.108767890930176,
|
|
"epoch": 6.081124932395889,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00023203055101666593,
|
|
"loss": 4.4134,
|
|
"mean_token_accuracy": 0.2486819237470627,
|
|
"num_tokens": 12628628.0,
|
|
"step": 5625
|
|
},
|
|
{
|
|
"entropy": 5.0853352546691895,
|
|
"epoch": 6.086533261222282,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002316096047696108,
|
|
"loss": 4.4182,
|
|
"mean_token_accuracy": 0.25006395280361177,
|
|
"num_tokens": 12638375.0,
|
|
"step": 5630
|
|
},
|
|
{
|
|
"entropy": 5.036393404006958,
|
|
"epoch": 6.091941590048675,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002311888162033522,
|
|
"loss": 4.4408,
|
|
"mean_token_accuracy": 0.24180681705474855,
|
|
"num_tokens": 12649174.0,
|
|
"step": 5635
|
|
},
|
|
{
|
|
"entropy": 5.0512079238891605,
|
|
"epoch": 6.097349918875068,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00023076818684703682,
|
|
"loss": 4.4164,
|
|
"mean_token_accuracy": 0.253505203127861,
|
|
"num_tokens": 12659454.0,
|
|
"step": 5640
|
|
},
|
|
{
|
|
"entropy": 5.081248712539673,
|
|
"epoch": 6.10275824770146,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00023034771822923268,
|
|
"loss": 4.4421,
|
|
"mean_token_accuracy": 0.2529549762606621,
|
|
"num_tokens": 12672093.0,
|
|
"step": 5645
|
|
},
|
|
{
|
|
"entropy": 5.135938501358032,
|
|
"epoch": 6.108166576527853,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002299274118779236,
|
|
"loss": 4.458,
|
|
"mean_token_accuracy": 0.25059467405080793,
|
|
"num_tokens": 12683628.0,
|
|
"step": 5650
|
|
},
|
|
{
|
|
"entropy": 5.075101041793824,
|
|
"epoch": 6.113574905354246,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00022950726932050386,
|
|
"loss": 4.4182,
|
|
"mean_token_accuracy": 0.2507233828306198,
|
|
"num_tokens": 12694726.0,
|
|
"step": 5655
|
|
},
|
|
{
|
|
"entropy": 5.121889591217041,
|
|
"epoch": 6.1189832341806385,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00022908729208377267,
|
|
"loss": 4.4814,
|
|
"mean_token_accuracy": 0.24927080124616624,
|
|
"num_tokens": 12705292.0,
|
|
"step": 5660
|
|
},
|
|
{
|
|
"entropy": 5.140897130966186,
|
|
"epoch": 6.1243915630070305,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00022866748169392815,
|
|
"loss": 4.5067,
|
|
"mean_token_accuracy": 0.2453337773680687,
|
|
"num_tokens": 12717721.0,
|
|
"step": 5665
|
|
},
|
|
{
|
|
"entropy": 5.129576635360718,
|
|
"epoch": 6.129799891833423,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00022824783967656208,
|
|
"loss": 4.4741,
|
|
"mean_token_accuracy": 0.24447181671857834,
|
|
"num_tokens": 12728360.0,
|
|
"step": 5670
|
|
},
|
|
{
|
|
"entropy": 5.069584321975708,
|
|
"epoch": 6.135208220659816,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00022782836755665475,
|
|
"loss": 4.4968,
|
|
"mean_token_accuracy": 0.24051816165447235,
|
|
"num_tokens": 12739771.0,
|
|
"step": 5675
|
|
},
|
|
{
|
|
"entropy": 5.111352443695068,
|
|
"epoch": 6.140616549486209,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00022740906685856876,
|
|
"loss": 4.4713,
|
|
"mean_token_accuracy": 0.24546525478363038,
|
|
"num_tokens": 12750024.0,
|
|
"step": 5680
|
|
},
|
|
{
|
|
"entropy": 5.072743701934814,
|
|
"epoch": 6.146024878312601,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00022698993910604383,
|
|
"loss": 4.4304,
|
|
"mean_token_accuracy": 0.25067793726921084,
|
|
"num_tokens": 12761160.0,
|
|
"step": 5685
|
|
},
|
|
{
|
|
"entropy": 5.112010526657104,
|
|
"epoch": 6.151433207138994,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000226570985822191,
|
|
"loss": 4.42,
|
|
"mean_token_accuracy": 0.25379774123430254,
|
|
"num_tokens": 12773043.0,
|
|
"step": 5690
|
|
},
|
|
{
|
|
"entropy": 5.057463359832764,
|
|
"epoch": 6.156841535965387,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00022615220852948765,
|
|
"loss": 4.4357,
|
|
"mean_token_accuracy": 0.246117402613163,
|
|
"num_tokens": 12783479.0,
|
|
"step": 5695
|
|
},
|
|
{
|
|
"entropy": 5.1814055919647215,
|
|
"epoch": 6.16224986479178,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00022573360874977145,
|
|
"loss": 4.5204,
|
|
"mean_token_accuracy": 0.24575739353895187,
|
|
"num_tokens": 12794889.0,
|
|
"step": 5700
|
|
},
|
|
{
|
|
"entropy": 5.1277049541473385,
|
|
"epoch": 6.167658193618172,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00022531518800423484,
|
|
"loss": 4.5113,
|
|
"mean_token_accuracy": 0.23952895551919937,
|
|
"num_tokens": 12806161.0,
|
|
"step": 5705
|
|
},
|
|
{
|
|
"entropy": 5.12687578201294,
|
|
"epoch": 6.173066522444564,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002248969478134199,
|
|
"loss": 4.5132,
|
|
"mean_token_accuracy": 0.24108312427997589,
|
|
"num_tokens": 12817829.0,
|
|
"step": 5710
|
|
},
|
|
{
|
|
"entropy": 5.101347923278809,
|
|
"epoch": 6.178474851270957,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00022447888969721232,
|
|
"loss": 4.5663,
|
|
"mean_token_accuracy": 0.23082333505153657,
|
|
"num_tokens": 12828212.0,
|
|
"step": 5715
|
|
},
|
|
{
|
|
"entropy": 5.153430557250976,
|
|
"epoch": 6.18388318009735,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00022406101517483636,
|
|
"loss": 4.505,
|
|
"mean_token_accuracy": 0.2424133762717247,
|
|
"num_tokens": 12840058.0,
|
|
"step": 5720
|
|
},
|
|
{
|
|
"entropy": 5.1044210433959964,
|
|
"epoch": 6.189291508923742,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00022364332576484897,
|
|
"loss": 4.4974,
|
|
"mean_token_accuracy": 0.24557265192270278,
|
|
"num_tokens": 12852833.0,
|
|
"step": 5725
|
|
},
|
|
{
|
|
"entropy": 5.094424486160278,
|
|
"epoch": 6.194699837750135,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002232258229851345,
|
|
"loss": 4.4736,
|
|
"mean_token_accuracy": 0.24467359334230424,
|
|
"num_tokens": 12863487.0,
|
|
"step": 5730
|
|
},
|
|
{
|
|
"entropy": 5.074555730819702,
|
|
"epoch": 6.200108166576528,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00022280850835289885,
|
|
"loss": 4.409,
|
|
"mean_token_accuracy": 0.25091417133808136,
|
|
"num_tokens": 12873952.0,
|
|
"step": 5735
|
|
},
|
|
{
|
|
"entropy": 5.036533546447754,
|
|
"epoch": 6.205516495402921,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00022239138338466452,
|
|
"loss": 4.4771,
|
|
"mean_token_accuracy": 0.2453603595495224,
|
|
"num_tokens": 12884462.0,
|
|
"step": 5740
|
|
},
|
|
{
|
|
"entropy": 5.081358623504639,
|
|
"epoch": 6.210924824229313,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00022197444959626451,
|
|
"loss": 4.4521,
|
|
"mean_token_accuracy": 0.2548902094364166,
|
|
"num_tokens": 12894996.0,
|
|
"step": 5745
|
|
},
|
|
{
|
|
"entropy": 5.071122121810913,
|
|
"epoch": 6.2163331530557056,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00022155770850283723,
|
|
"loss": 4.4109,
|
|
"mean_token_accuracy": 0.2501024067401886,
|
|
"num_tokens": 12906256.0,
|
|
"step": 5750
|
|
},
|
|
{
|
|
"entropy": 5.0838953971862795,
|
|
"epoch": 6.221741481882098,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00022114116161882065,
|
|
"loss": 4.5355,
|
|
"mean_token_accuracy": 0.24561379700899125,
|
|
"num_tokens": 12918738.0,
|
|
"step": 5755
|
|
},
|
|
{
|
|
"entropy": 5.067653179168701,
|
|
"epoch": 6.227149810708491,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00022072481045794733,
|
|
"loss": 4.446,
|
|
"mean_token_accuracy": 0.24226458072662355,
|
|
"num_tokens": 12929918.0,
|
|
"step": 5760
|
|
},
|
|
{
|
|
"entropy": 5.147340774536133,
|
|
"epoch": 6.232558139534884,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00022030865653323823,
|
|
"loss": 4.5116,
|
|
"mean_token_accuracy": 0.2437214508652687,
|
|
"num_tokens": 12941427.0,
|
|
"step": 5765
|
|
},
|
|
{
|
|
"entropy": 5.097586059570313,
|
|
"epoch": 6.237966468361276,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002198927013569977,
|
|
"loss": 4.4593,
|
|
"mean_token_accuracy": 0.24690070748329163,
|
|
"num_tokens": 12952889.0,
|
|
"step": 5770
|
|
},
|
|
{
|
|
"entropy": 5.09056282043457,
|
|
"epoch": 6.243374797187669,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002194769464408076,
|
|
"loss": 4.4604,
|
|
"mean_token_accuracy": 0.25365334302186965,
|
|
"num_tokens": 12962928.0,
|
|
"step": 5775
|
|
},
|
|
{
|
|
"entropy": 5.0598729133605955,
|
|
"epoch": 6.248783126014062,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002190613932955226,
|
|
"loss": 4.4224,
|
|
"mean_token_accuracy": 0.25400429368019106,
|
|
"num_tokens": 12972901.0,
|
|
"step": 5780
|
|
},
|
|
{
|
|
"entropy": 5.0984124660491945,
|
|
"epoch": 6.254191454840455,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00021864604343126365,
|
|
"loss": 4.51,
|
|
"mean_token_accuracy": 0.2480001524090767,
|
|
"num_tokens": 12985296.0,
|
|
"step": 5785
|
|
},
|
|
{
|
|
"entropy": 5.160121059417724,
|
|
"epoch": 6.259599783666847,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00021823089835741312,
|
|
"loss": 4.5908,
|
|
"mean_token_accuracy": 0.23779818266630173,
|
|
"num_tokens": 12997069.0,
|
|
"step": 5790
|
|
},
|
|
{
|
|
"entropy": 5.117949914932251,
|
|
"epoch": 6.2650081124932395,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000217815959582609,
|
|
"loss": 4.4505,
|
|
"mean_token_accuracy": 0.24992359131574632,
|
|
"num_tokens": 13007147.0,
|
|
"step": 5795
|
|
},
|
|
{
|
|
"entropy": 5.084199333190918,
|
|
"epoch": 6.270416441319632,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00021740122861473992,
|
|
"loss": 4.4817,
|
|
"mean_token_accuracy": 0.2546451151371002,
|
|
"num_tokens": 13019084.0,
|
|
"step": 5800
|
|
},
|
|
{
|
|
"entropy": 5.102336978912353,
|
|
"epoch": 6.275824770146025,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00021698670696093932,
|
|
"loss": 4.4907,
|
|
"mean_token_accuracy": 0.24857060760259628,
|
|
"num_tokens": 13030751.0,
|
|
"step": 5805
|
|
},
|
|
{
|
|
"entropy": 5.08336911201477,
|
|
"epoch": 6.281233098972417,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00021657239612757963,
|
|
"loss": 4.4491,
|
|
"mean_token_accuracy": 0.24939932972192763,
|
|
"num_tokens": 13042731.0,
|
|
"step": 5810
|
|
},
|
|
{
|
|
"entropy": 5.097407388687134,
|
|
"epoch": 6.28664142779881,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00021615829762026734,
|
|
"loss": 4.4332,
|
|
"mean_token_accuracy": 0.2454545333981514,
|
|
"num_tokens": 13053442.0,
|
|
"step": 5815
|
|
},
|
|
{
|
|
"entropy": 5.115796136856079,
|
|
"epoch": 6.292049756625203,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00021574441294383752,
|
|
"loss": 4.4947,
|
|
"mean_token_accuracy": 0.24055950492620468,
|
|
"num_tokens": 13063911.0,
|
|
"step": 5820
|
|
},
|
|
{
|
|
"entropy": 5.127041673660278,
|
|
"epoch": 6.297458085451596,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00021533074360234806,
|
|
"loss": 4.5531,
|
|
"mean_token_accuracy": 0.23869300484657288,
|
|
"num_tokens": 13075887.0,
|
|
"step": 5825
|
|
},
|
|
{
|
|
"entropy": 5.055332040786743,
|
|
"epoch": 6.302866414277988,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00021491729109907414,
|
|
"loss": 4.4232,
|
|
"mean_token_accuracy": 0.2530810981988907,
|
|
"num_tokens": 13087320.0,
|
|
"step": 5830
|
|
},
|
|
{
|
|
"entropy": 5.107473278045655,
|
|
"epoch": 6.308274743104381,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00021450405693650333,
|
|
"loss": 4.4404,
|
|
"mean_token_accuracy": 0.25520368814468386,
|
|
"num_tokens": 13099078.0,
|
|
"step": 5835
|
|
},
|
|
{
|
|
"entropy": 5.1563458919525145,
|
|
"epoch": 6.3136830719307735,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00021409104261632932,
|
|
"loss": 4.5047,
|
|
"mean_token_accuracy": 0.25065077692270277,
|
|
"num_tokens": 13111909.0,
|
|
"step": 5840
|
|
},
|
|
{
|
|
"entropy": 5.012200450897216,
|
|
"epoch": 6.319091400757166,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00021367824963944727,
|
|
"loss": 4.424,
|
|
"mean_token_accuracy": 0.2532737761735916,
|
|
"num_tokens": 13123297.0,
|
|
"step": 5845
|
|
},
|
|
{
|
|
"entropy": 5.023969650268555,
|
|
"epoch": 6.324499729583558,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00021326567950594766,
|
|
"loss": 4.4976,
|
|
"mean_token_accuracy": 0.2412753015756607,
|
|
"num_tokens": 13133959.0,
|
|
"step": 5850
|
|
},
|
|
{
|
|
"entropy": 5.031236743927002,
|
|
"epoch": 6.329908058409951,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002128533337151114,
|
|
"loss": 4.3428,
|
|
"mean_token_accuracy": 0.2588138684630394,
|
|
"num_tokens": 13144858.0,
|
|
"step": 5855
|
|
},
|
|
{
|
|
"entropy": 5.057790660858155,
|
|
"epoch": 6.335316387236344,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002124412137654041,
|
|
"loss": 4.4955,
|
|
"mean_token_accuracy": 0.24278368651866913,
|
|
"num_tokens": 13156437.0,
|
|
"step": 5860
|
|
},
|
|
{
|
|
"entropy": 5.10028395652771,
|
|
"epoch": 6.340724716062737,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00021202932115447056,
|
|
"loss": 4.4246,
|
|
"mean_token_accuracy": 0.2527458414435387,
|
|
"num_tokens": 13166860.0,
|
|
"step": 5865
|
|
},
|
|
{
|
|
"entropy": 5.123906373977661,
|
|
"epoch": 6.346133044889129,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00021161765737912937,
|
|
"loss": 4.5253,
|
|
"mean_token_accuracy": 0.2467150166630745,
|
|
"num_tokens": 13178563.0,
|
|
"step": 5870
|
|
},
|
|
{
|
|
"entropy": 5.078815984725952,
|
|
"epoch": 6.351541373715522,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00021120622393536775,
|
|
"loss": 4.4689,
|
|
"mean_token_accuracy": 0.2442133381962776,
|
|
"num_tokens": 13189915.0,
|
|
"step": 5875
|
|
},
|
|
{
|
|
"entropy": 5.095115661621094,
|
|
"epoch": 6.356949702541915,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00021079502231833585,
|
|
"loss": 4.4643,
|
|
"mean_token_accuracy": 0.24681447446346283,
|
|
"num_tokens": 13201767.0,
|
|
"step": 5880
|
|
},
|
|
{
|
|
"entropy": 5.081144762039185,
|
|
"epoch": 6.3623580313683075,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00021038405402234116,
|
|
"loss": 4.4692,
|
|
"mean_token_accuracy": 0.25071409791707994,
|
|
"num_tokens": 13212265.0,
|
|
"step": 5885
|
|
},
|
|
{
|
|
"entropy": 5.056689882278443,
|
|
"epoch": 6.367766360194699,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002099733205408435,
|
|
"loss": 4.4998,
|
|
"mean_token_accuracy": 0.24432076811790465,
|
|
"num_tokens": 13223503.0,
|
|
"step": 5890
|
|
},
|
|
{
|
|
"entropy": 5.108233118057251,
|
|
"epoch": 6.373174689021092,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002095628233664492,
|
|
"loss": 4.4951,
|
|
"mean_token_accuracy": 0.24592576473951339,
|
|
"num_tokens": 13234797.0,
|
|
"step": 5895
|
|
},
|
|
{
|
|
"entropy": 5.074272632598877,
|
|
"epoch": 6.378583017847485,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002091525639909061,
|
|
"loss": 4.4291,
|
|
"mean_token_accuracy": 0.2502945214509964,
|
|
"num_tokens": 13246421.0,
|
|
"step": 5900
|
|
},
|
|
{
|
|
"entropy": 5.108126592636109,
|
|
"epoch": 6.383991346673878,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00020874254390509757,
|
|
"loss": 4.442,
|
|
"mean_token_accuracy": 0.2535440444946289,
|
|
"num_tokens": 13256747.0,
|
|
"step": 5905
|
|
},
|
|
{
|
|
"entropy": 5.106110525131226,
|
|
"epoch": 6.38939967550027,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00020833276459903766,
|
|
"loss": 4.4525,
|
|
"mean_token_accuracy": 0.24672785103321077,
|
|
"num_tokens": 13267620.0,
|
|
"step": 5910
|
|
},
|
|
{
|
|
"entropy": 5.138248491287231,
|
|
"epoch": 6.394808004326663,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002079232275618651,
|
|
"loss": 4.4097,
|
|
"mean_token_accuracy": 0.24966918230056762,
|
|
"num_tokens": 13277818.0,
|
|
"step": 5915
|
|
},
|
|
{
|
|
"entropy": 4.970045948028565,
|
|
"epoch": 6.400216333153056,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002075139342818388,
|
|
"loss": 4.4021,
|
|
"mean_token_accuracy": 0.2511683166027069,
|
|
"num_tokens": 13289905.0,
|
|
"step": 5920
|
|
},
|
|
{
|
|
"entropy": 4.999069023132324,
|
|
"epoch": 6.405624661979449,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00020710488624633118,
|
|
"loss": 4.3946,
|
|
"mean_token_accuracy": 0.2529490053653717,
|
|
"num_tokens": 13302465.0,
|
|
"step": 5925
|
|
},
|
|
{
|
|
"entropy": 5.109829139709473,
|
|
"epoch": 6.411032990805841,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00020669608494182384,
|
|
"loss": 4.4981,
|
|
"mean_token_accuracy": 0.24396926462650298,
|
|
"num_tokens": 13314366.0,
|
|
"step": 5930
|
|
},
|
|
{
|
|
"entropy": 4.955432796478272,
|
|
"epoch": 6.416441319632233,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00020628753185390164,
|
|
"loss": 4.329,
|
|
"mean_token_accuracy": 0.26145068258047105,
|
|
"num_tokens": 13324981.0,
|
|
"step": 5935
|
|
},
|
|
{
|
|
"entropy": 5.054769229888916,
|
|
"epoch": 6.421849648458626,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002058792284672475,
|
|
"loss": 4.4555,
|
|
"mean_token_accuracy": 0.25324306041002276,
|
|
"num_tokens": 13335705.0,
|
|
"step": 5940
|
|
},
|
|
{
|
|
"entropy": 4.9640850067138675,
|
|
"epoch": 6.427257977285019,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002054711762656369,
|
|
"loss": 4.4065,
|
|
"mean_token_accuracy": 0.2554488331079483,
|
|
"num_tokens": 13346515.0,
|
|
"step": 5945
|
|
},
|
|
{
|
|
"entropy": 5.070720148086548,
|
|
"epoch": 6.432666306111411,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00020506337673193236,
|
|
"loss": 4.4333,
|
|
"mean_token_accuracy": 0.24604305773973464,
|
|
"num_tokens": 13357780.0,
|
|
"step": 5950
|
|
},
|
|
{
|
|
"entropy": 5.090268850326538,
|
|
"epoch": 6.438074634937804,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00020465583134807836,
|
|
"loss": 4.4373,
|
|
"mean_token_accuracy": 0.25184490978717805,
|
|
"num_tokens": 13369549.0,
|
|
"step": 5955
|
|
},
|
|
{
|
|
"entropy": 5.127700996398926,
|
|
"epoch": 6.443482963764197,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00020424854159509587,
|
|
"loss": 4.5098,
|
|
"mean_token_accuracy": 0.24484294205904006,
|
|
"num_tokens": 13380259.0,
|
|
"step": 5960
|
|
},
|
|
{
|
|
"entropy": 5.073073196411133,
|
|
"epoch": 6.44889129259059,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002038415089530767,
|
|
"loss": 4.5338,
|
|
"mean_token_accuracy": 0.2439861625432968,
|
|
"num_tokens": 13392049.0,
|
|
"step": 5965
|
|
},
|
|
{
|
|
"entropy": 5.098212766647339,
|
|
"epoch": 6.4542996214169825,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00020343473490117846,
|
|
"loss": 4.4966,
|
|
"mean_token_accuracy": 0.24678318053483964,
|
|
"num_tokens": 13403874.0,
|
|
"step": 5970
|
|
},
|
|
{
|
|
"entropy": 5.062539529800415,
|
|
"epoch": 6.4597079502433745,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00020302822091761884,
|
|
"loss": 4.4805,
|
|
"mean_token_accuracy": 0.2513225793838501,
|
|
"num_tokens": 13415194.0,
|
|
"step": 5975
|
|
},
|
|
{
|
|
"entropy": 5.093571996688842,
|
|
"epoch": 6.465116279069767,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00020262196847967091,
|
|
"loss": 4.4837,
|
|
"mean_token_accuracy": 0.2453096866607666,
|
|
"num_tokens": 13425159.0,
|
|
"step": 5980
|
|
},
|
|
{
|
|
"entropy": 5.148891162872315,
|
|
"epoch": 6.47052460789616,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00020221597906365667,
|
|
"loss": 4.4852,
|
|
"mean_token_accuracy": 0.2466130316257477,
|
|
"num_tokens": 13437279.0,
|
|
"step": 5985
|
|
},
|
|
{
|
|
"entropy": 5.115873146057129,
|
|
"epoch": 6.475932936722553,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00020181025414494285,
|
|
"loss": 4.4752,
|
|
"mean_token_accuracy": 0.2467889219522476,
|
|
"num_tokens": 13449170.0,
|
|
"step": 5990
|
|
},
|
|
{
|
|
"entropy": 5.103183317184448,
|
|
"epoch": 6.481341265548945,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002014047951979344,
|
|
"loss": 4.5456,
|
|
"mean_token_accuracy": 0.24151896089315414,
|
|
"num_tokens": 13460216.0,
|
|
"step": 5995
|
|
},
|
|
{
|
|
"entropy": 5.068713760375976,
|
|
"epoch": 6.486749594375338,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002009996036960704,
|
|
"loss": 4.5186,
|
|
"mean_token_accuracy": 0.24359928369522094,
|
|
"num_tokens": 13470468.0,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"epoch": 6.486749594375338,
|
|
"eval_entropy": 5.021412039886822,
|
|
"eval_loss": 5.878503322601318,
|
|
"eval_mean_token_accuracy": 0.1820899965580214,
|
|
"eval_num_tokens": 13470468.0,
|
|
"eval_runtime": 2.753,
|
|
"eval_samples_per_second": 1276.782,
|
|
"eval_steps_per_second": 159.825,
|
|
"step": 6000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 9240,
|
|
"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": 2.1042957828096e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|