4079 lines
110 KiB
JSON
4079 lines
110 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 2.162249864791779,
|
|
"eval_steps": 500,
|
|
"global_step": 2000,
|
|
"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
|
|
}
|
|
],
|
|
"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": 6984519736320000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|