0709e1c753
Model: fpadovani/eng-latn-10mb-after-ppt-sort-curriculum_seed3407 Source: Original Platform
7112 lines
193 KiB
JSON
7112 lines
193 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 3.7842076798269337,
|
|
"eval_steps": 500,
|
|
"global_step": 3500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 7.820888805389404,
|
|
"epoch": 0.005408328826392645,
|
|
"grad_norm": 7.65625,
|
|
"learning_rate": 2e-06,
|
|
"loss": 13.4867,
|
|
"mean_token_accuracy": 0.011032939096912742,
|
|
"num_tokens": 10942.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 7.835813283920288,
|
|
"epoch": 0.01081665765278529,
|
|
"grad_norm": 7.8125,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 13.4729,
|
|
"mean_token_accuracy": 0.010750519391149283,
|
|
"num_tokens": 21887.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 7.866043424606323,
|
|
"epoch": 0.016224986479177934,
|
|
"grad_norm": 8.375,
|
|
"learning_rate": 7e-06,
|
|
"loss": 13.4343,
|
|
"mean_token_accuracy": 0.010895622195675969,
|
|
"num_tokens": 34345.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 7.880505704879761,
|
|
"epoch": 0.02163331530557058,
|
|
"grad_norm": 8.75,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 13.3238,
|
|
"mean_token_accuracy": 0.01247827559709549,
|
|
"num_tokens": 45445.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 7.962141990661621,
|
|
"epoch": 0.02704164413196322,
|
|
"grad_norm": 8.75,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 13.2274,
|
|
"mean_token_accuracy": 0.011504577286541463,
|
|
"num_tokens": 58237.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 8.106424951553345,
|
|
"epoch": 0.03244997295835587,
|
|
"grad_norm": 9.625,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 12.88,
|
|
"mean_token_accuracy": 0.01291654328815639,
|
|
"num_tokens": 68098.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 8.39170503616333,
|
|
"epoch": 0.03785830178474851,
|
|
"grad_norm": 9.4375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 12.4981,
|
|
"mean_token_accuracy": 0.01724564200267196,
|
|
"num_tokens": 79867.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 8.898138904571534,
|
|
"epoch": 0.04326663061114116,
|
|
"grad_norm": 6.75,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 11.9446,
|
|
"mean_token_accuracy": 0.019018208887428046,
|
|
"num_tokens": 91257.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 9.481475162506104,
|
|
"epoch": 0.048674959437533805,
|
|
"grad_norm": 4.21875,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 11.4652,
|
|
"mean_token_accuracy": 0.025027596950531007,
|
|
"num_tokens": 103474.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 9.992423725128173,
|
|
"epoch": 0.05408328826392644,
|
|
"grad_norm": 2.8125,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 11.0587,
|
|
"mean_token_accuracy": 0.024848397076129913,
|
|
"num_tokens": 114830.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.25519962310791,
|
|
"epoch": 0.05949161709031909,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 10.8236,
|
|
"mean_token_accuracy": 0.024381351098418234,
|
|
"num_tokens": 126017.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.374946212768554,
|
|
"epoch": 0.06489994591671173,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 10.7281,
|
|
"mean_token_accuracy": 0.028067474253475666,
|
|
"num_tokens": 138509.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.438200187683105,
|
|
"epoch": 0.07030827474310439,
|
|
"grad_norm": 1.9453125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 10.5835,
|
|
"mean_token_accuracy": 0.031993019953370096,
|
|
"num_tokens": 148353.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.488709259033204,
|
|
"epoch": 0.07571660356949703,
|
|
"grad_norm": 1.90625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 10.5393,
|
|
"mean_token_accuracy": 0.036837692372500896,
|
|
"num_tokens": 160269.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.523839092254638,
|
|
"epoch": 0.08112493239588967,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 10.4312,
|
|
"mean_token_accuracy": 0.03589091971516609,
|
|
"num_tokens": 171568.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.546321773529053,
|
|
"epoch": 0.08653326122228232,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 10.3481,
|
|
"mean_token_accuracy": 0.0360051903873682,
|
|
"num_tokens": 184243.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.563430118560792,
|
|
"epoch": 0.09194159004867496,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 10.2337,
|
|
"mean_token_accuracy": 0.03699877411127091,
|
|
"num_tokens": 194887.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.580722332000732,
|
|
"epoch": 0.09734991887506761,
|
|
"grad_norm": 1.953125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 10.1385,
|
|
"mean_token_accuracy": 0.03550931885838508,
|
|
"num_tokens": 206476.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.59913902282715,
|
|
"epoch": 0.10275824770146025,
|
|
"grad_norm": 1.9296875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.9926,
|
|
"mean_token_accuracy": 0.039441929385066035,
|
|
"num_tokens": 216578.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.6052095413208,
|
|
"epoch": 0.10816657652785289,
|
|
"grad_norm": 1.9296875,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.8908,
|
|
"mean_token_accuracy": 0.03636195547878742,
|
|
"num_tokens": 226941.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.607266902923584,
|
|
"epoch": 0.11357490535424554,
|
|
"grad_norm": 1.890625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.7695,
|
|
"mean_token_accuracy": 0.04152456596493721,
|
|
"num_tokens": 238060.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.605393600463866,
|
|
"epoch": 0.11898323418063818,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.6438,
|
|
"mean_token_accuracy": 0.0380807850509882,
|
|
"num_tokens": 249137.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.596975135803223,
|
|
"epoch": 0.12439156300703083,
|
|
"grad_norm": 1.96875,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 9.5365,
|
|
"mean_token_accuracy": 0.039401752315461634,
|
|
"num_tokens": 260712.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.588339138031007,
|
|
"epoch": 0.12979989183342347,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 9.3761,
|
|
"mean_token_accuracy": 0.03739211615175009,
|
|
"num_tokens": 271039.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.577256965637208,
|
|
"epoch": 0.1352082206598161,
|
|
"grad_norm": 2.125,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 9.2529,
|
|
"mean_token_accuracy": 0.04276929795742035,
|
|
"num_tokens": 281353.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.553887462615966,
|
|
"epoch": 0.14061654948620878,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 9.1552,
|
|
"mean_token_accuracy": 0.039082702808082105,
|
|
"num_tokens": 292221.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.525830459594726,
|
|
"epoch": 0.1460248783126014,
|
|
"grad_norm": 1.875,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 9.0356,
|
|
"mean_token_accuracy": 0.04222068060189486,
|
|
"num_tokens": 303353.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.492454433441162,
|
|
"epoch": 0.15143320713899405,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.9125,
|
|
"mean_token_accuracy": 0.04995499067008495,
|
|
"num_tokens": 314165.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.450113677978516,
|
|
"epoch": 0.1568415359653867,
|
|
"grad_norm": 1.90625,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.7544,
|
|
"mean_token_accuracy": 0.04123007263988256,
|
|
"num_tokens": 326387.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 10.41439380645752,
|
|
"epoch": 0.16224986479177933,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.6823,
|
|
"mean_token_accuracy": 0.04394615534693003,
|
|
"num_tokens": 339155.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 10.365491104125976,
|
|
"epoch": 0.167658193618172,
|
|
"grad_norm": 1.9296875,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.5153,
|
|
"mean_token_accuracy": 0.03965858686715364,
|
|
"num_tokens": 350643.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 10.289417934417724,
|
|
"epoch": 0.17306652244456464,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.3842,
|
|
"mean_token_accuracy": 0.043310442194342616,
|
|
"num_tokens": 361977.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 10.205756950378419,
|
|
"epoch": 0.17847485127095727,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.3159,
|
|
"mean_token_accuracy": 0.04538454748690128,
|
|
"num_tokens": 373604.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 10.099522399902344,
|
|
"epoch": 0.1838831800973499,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 8.1886,
|
|
"mean_token_accuracy": 0.04867874421179295,
|
|
"num_tokens": 384306.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 9.964338874816894,
|
|
"epoch": 0.18929150892374255,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 8.0329,
|
|
"mean_token_accuracy": 0.049391471594572064,
|
|
"num_tokens": 396479.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 9.800485134124756,
|
|
"epoch": 0.19469983775013522,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.9052,
|
|
"mean_token_accuracy": 0.049093520268797874,
|
|
"num_tokens": 408489.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 9.617319297790527,
|
|
"epoch": 0.20010816657652786,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.8279,
|
|
"mean_token_accuracy": 0.04878813587129116,
|
|
"num_tokens": 419339.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 9.397418594360351,
|
|
"epoch": 0.2055164954029205,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.7213,
|
|
"mean_token_accuracy": 0.048774123191833496,
|
|
"num_tokens": 430486.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 9.130149364471436,
|
|
"epoch": 0.21092482422931313,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.6542,
|
|
"mean_token_accuracy": 0.05405276417732239,
|
|
"num_tokens": 440414.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.82462682723999,
|
|
"epoch": 0.21633315305570577,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.5581,
|
|
"mean_token_accuracy": 0.05878009013831616,
|
|
"num_tokens": 451475.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.563184356689453,
|
|
"epoch": 0.22174148188209844,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.5674,
|
|
"mean_token_accuracy": 0.06661942265927792,
|
|
"num_tokens": 462053.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.337231540679932,
|
|
"epoch": 0.22714981070849108,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.5121,
|
|
"mean_token_accuracy": 0.06138431690633297,
|
|
"num_tokens": 473207.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.17379379272461,
|
|
"epoch": 0.23255813953488372,
|
|
"grad_norm": 0.73046875,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.4845,
|
|
"mean_token_accuracy": 0.06530819572508335,
|
|
"num_tokens": 484999.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.07261848449707,
|
|
"epoch": 0.23796646836127636,
|
|
"grad_norm": 0.734375,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.4822,
|
|
"mean_token_accuracy": 0.0644452515989542,
|
|
"num_tokens": 496772.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.936752462387085,
|
|
"epoch": 0.24337479718766902,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.4461,
|
|
"mean_token_accuracy": 0.06566701531410217,
|
|
"num_tokens": 507782.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.9283324718475345,
|
|
"epoch": 0.24878312601406166,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.366,
|
|
"mean_token_accuracy": 0.06866429261863231,
|
|
"num_tokens": 518950.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.8786633014678955,
|
|
"epoch": 0.2541914548404543,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.4323,
|
|
"mean_token_accuracy": 0.06760875023901462,
|
|
"num_tokens": 530795.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.858989715576172,
|
|
"epoch": 0.25959978366684694,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.4452,
|
|
"mean_token_accuracy": 0.06729580424726009,
|
|
"num_tokens": 542878.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.774307203292847,
|
|
"epoch": 0.2650081124932396,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.3401,
|
|
"mean_token_accuracy": 0.07460085228085518,
|
|
"num_tokens": 553682.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.707247972488403,
|
|
"epoch": 0.2704164413196322,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.3454,
|
|
"mean_token_accuracy": 0.07869893908500672,
|
|
"num_tokens": 564433.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.7594318866729735,
|
|
"epoch": 0.27582477014602486,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.3708,
|
|
"mean_token_accuracy": 0.07336842231452465,
|
|
"num_tokens": 576001.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.717825031280517,
|
|
"epoch": 0.28123309897241755,
|
|
"grad_norm": 0.66015625,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.3119,
|
|
"mean_token_accuracy": 0.07627563402056695,
|
|
"num_tokens": 587138.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.747607707977295,
|
|
"epoch": 0.2866414277988102,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.2671,
|
|
"mean_token_accuracy": 0.08127573654055595,
|
|
"num_tokens": 597572.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.623431205749512,
|
|
"epoch": 0.2920497566252028,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.2359,
|
|
"mean_token_accuracy": 0.08385064378380776,
|
|
"num_tokens": 608879.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.603728199005127,
|
|
"epoch": 0.29745808545159547,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.2864,
|
|
"mean_token_accuracy": 0.0743965484201908,
|
|
"num_tokens": 620439.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.6806840896606445,
|
|
"epoch": 0.3028664142779881,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.3331,
|
|
"mean_token_accuracy": 0.0795133963227272,
|
|
"num_tokens": 632543.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.567217206954956,
|
|
"epoch": 0.30827474310438074,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.3707,
|
|
"mean_token_accuracy": 0.0822471871972084,
|
|
"num_tokens": 643369.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.696122837066651,
|
|
"epoch": 0.3136830719307734,
|
|
"grad_norm": 0.69140625,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.2433,
|
|
"mean_token_accuracy": 0.07976862639188767,
|
|
"num_tokens": 655328.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.673062705993653,
|
|
"epoch": 0.319091400757166,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.3085,
|
|
"mean_token_accuracy": 0.08202499970793724,
|
|
"num_tokens": 665780.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.630768585205078,
|
|
"epoch": 0.32449972958355866,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.2183,
|
|
"mean_token_accuracy": 0.08841075897216796,
|
|
"num_tokens": 676566.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.61181788444519,
|
|
"epoch": 0.3299080584099513,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.2092,
|
|
"mean_token_accuracy": 0.08559464290738106,
|
|
"num_tokens": 687407.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.709430265426636,
|
|
"epoch": 0.335316387236344,
|
|
"grad_norm": 0.73828125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.1314,
|
|
"mean_token_accuracy": 0.08758908063173294,
|
|
"num_tokens": 698162.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.624366712570191,
|
|
"epoch": 0.34072471606273663,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.1345,
|
|
"mean_token_accuracy": 0.08050442710518838,
|
|
"num_tokens": 708743.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.5649957180023195,
|
|
"epoch": 0.34613304488912927,
|
|
"grad_norm": 0.74609375,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.3075,
|
|
"mean_token_accuracy": 0.08924563452601433,
|
|
"num_tokens": 719828.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.668677186965942,
|
|
"epoch": 0.3515413737155219,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.275,
|
|
"mean_token_accuracy": 0.0808012131601572,
|
|
"num_tokens": 732514.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.571430635452271,
|
|
"epoch": 0.35694970254191455,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.1181,
|
|
"mean_token_accuracy": 0.08540427088737487,
|
|
"num_tokens": 742391.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.561294841766357,
|
|
"epoch": 0.3623580313683072,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.1192,
|
|
"mean_token_accuracy": 0.07976256236433983,
|
|
"num_tokens": 753638.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.643744564056396,
|
|
"epoch": 0.3677663601946998,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.3106,
|
|
"mean_token_accuracy": 0.08193527534604073,
|
|
"num_tokens": 764850.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.528887176513672,
|
|
"epoch": 0.37317468902109246,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.0978,
|
|
"mean_token_accuracy": 0.0877677209675312,
|
|
"num_tokens": 776192.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.488804197311401,
|
|
"epoch": 0.3785830178474851,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.0452,
|
|
"mean_token_accuracy": 0.09009440615773201,
|
|
"num_tokens": 787051.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.453413486480713,
|
|
"epoch": 0.3839913466738778,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.0818,
|
|
"mean_token_accuracy": 0.08464901968836784,
|
|
"num_tokens": 799281.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.47227840423584,
|
|
"epoch": 0.38939967550027044,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.0947,
|
|
"mean_token_accuracy": 0.08406926393508911,
|
|
"num_tokens": 810003.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.510661792755127,
|
|
"epoch": 0.3948080043266631,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.0577,
|
|
"mean_token_accuracy": 0.08685395568609237,
|
|
"num_tokens": 819749.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.452503824234009,
|
|
"epoch": 0.4002163331530557,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.0694,
|
|
"mean_token_accuracy": 0.08692100048065185,
|
|
"num_tokens": 830552.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.486446285247803,
|
|
"epoch": 0.40562466197944835,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.1663,
|
|
"mean_token_accuracy": 0.09074072763323784,
|
|
"num_tokens": 840495.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.408175134658814,
|
|
"epoch": 0.411032990805841,
|
|
"grad_norm": 0.73828125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.0813,
|
|
"mean_token_accuracy": 0.0877058044075966,
|
|
"num_tokens": 852915.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.418009948730469,
|
|
"epoch": 0.41644131963223363,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.0365,
|
|
"mean_token_accuracy": 0.0944710485637188,
|
|
"num_tokens": 862762.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.437512016296386,
|
|
"epoch": 0.42184964845862627,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.9974,
|
|
"mean_token_accuracy": 0.09396275207400322,
|
|
"num_tokens": 873023.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.443737745285034,
|
|
"epoch": 0.4272579772850189,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.0967,
|
|
"mean_token_accuracy": 0.08894659876823426,
|
|
"num_tokens": 885281.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.369609642028808,
|
|
"epoch": 0.43266630611141155,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.0939,
|
|
"mean_token_accuracy": 0.09273864775896072,
|
|
"num_tokens": 896210.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.422218275070191,
|
|
"epoch": 0.43807463493780424,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.0119,
|
|
"mean_token_accuracy": 0.09136949926614761,
|
|
"num_tokens": 909028.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.3562903881073,
|
|
"epoch": 0.4434829637641969,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.9403,
|
|
"mean_token_accuracy": 0.09579638168215751,
|
|
"num_tokens": 919710.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.465255784988403,
|
|
"epoch": 0.4488912925905895,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.0033,
|
|
"mean_token_accuracy": 0.09164737239480018,
|
|
"num_tokens": 930019.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.330646514892578,
|
|
"epoch": 0.45429962141698216,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.9277,
|
|
"mean_token_accuracy": 0.09849832803010941,
|
|
"num_tokens": 940535.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.225619316101074,
|
|
"epoch": 0.4597079502433748,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.8877,
|
|
"mean_token_accuracy": 0.10022106692194939,
|
|
"num_tokens": 951390.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.338269996643066,
|
|
"epoch": 0.46511627906976744,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.9181,
|
|
"mean_token_accuracy": 0.10022875815629959,
|
|
"num_tokens": 963010.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.250108957290649,
|
|
"epoch": 0.4705246078961601,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.8709,
|
|
"mean_token_accuracy": 0.10117963030934334,
|
|
"num_tokens": 974032.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.42537899017334,
|
|
"epoch": 0.4759329367225527,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.9201,
|
|
"mean_token_accuracy": 0.10263018682599068,
|
|
"num_tokens": 983954.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.279693603515625,
|
|
"epoch": 0.48134126554894535,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.9332,
|
|
"mean_token_accuracy": 0.10275022611021996,
|
|
"num_tokens": 994135.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.299053049087524,
|
|
"epoch": 0.48674959437533805,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.8981,
|
|
"mean_token_accuracy": 0.10349252223968505,
|
|
"num_tokens": 1006275.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.163119125366211,
|
|
"epoch": 0.4921579232017307,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.7321,
|
|
"mean_token_accuracy": 0.10733722224831581,
|
|
"num_tokens": 1017767.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.236297607421875,
|
|
"epoch": 0.4975662520281233,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.8507,
|
|
"mean_token_accuracy": 0.1048464372754097,
|
|
"num_tokens": 1028286.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.231361341476441,
|
|
"epoch": 0.502974580854516,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.8625,
|
|
"mean_token_accuracy": 0.1016843743622303,
|
|
"num_tokens": 1040555.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.289519309997559,
|
|
"epoch": 0.5083829096809086,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.9481,
|
|
"mean_token_accuracy": 0.09957458972930908,
|
|
"num_tokens": 1053668.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.211695528030395,
|
|
"epoch": 0.5137912385073012,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.891,
|
|
"mean_token_accuracy": 0.10144442990422249,
|
|
"num_tokens": 1066042.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.23263463973999,
|
|
"epoch": 0.5191995673336939,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.8999,
|
|
"mean_token_accuracy": 0.10248073190450668,
|
|
"num_tokens": 1077843.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.1748435497283936,
|
|
"epoch": 0.5246078961600865,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.794,
|
|
"mean_token_accuracy": 0.10380105674266815,
|
|
"num_tokens": 1089448.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.180902671813965,
|
|
"epoch": 0.5300162249864792,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.8816,
|
|
"mean_token_accuracy": 0.1034023180603981,
|
|
"num_tokens": 1100989.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.378252983093262,
|
|
"epoch": 0.5354245538128718,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.9212,
|
|
"mean_token_accuracy": 0.10497874468564987,
|
|
"num_tokens": 1113012.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.092680644989014,
|
|
"epoch": 0.5408328826392644,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.8695,
|
|
"mean_token_accuracy": 0.10232841521501541,
|
|
"num_tokens": 1123551.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.5408328826392644,
|
|
"eval_entropy": 7.271889178319411,
|
|
"eval_loss": 7.0051140785217285,
|
|
"eval_mean_token_accuracy": 0.10177409166300838,
|
|
"eval_num_tokens": 1123551.0,
|
|
"eval_runtime": 4.518,
|
|
"eval_samples_per_second": 778.002,
|
|
"eval_steps_per_second": 97.389,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.2357734680175785,
|
|
"epoch": 0.5462412114656571,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.7856,
|
|
"mean_token_accuracy": 0.10090178027749061,
|
|
"num_tokens": 1133838.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.1414398670196535,
|
|
"epoch": 0.5516495402920497,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.8278,
|
|
"mean_token_accuracy": 0.10166001990437508,
|
|
"num_tokens": 1146005.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.219015407562256,
|
|
"epoch": 0.5570578691184424,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.8977,
|
|
"mean_token_accuracy": 0.1012024573981762,
|
|
"num_tokens": 1157328.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.0664112091064455,
|
|
"epoch": 0.5624661979448351,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.7056,
|
|
"mean_token_accuracy": 0.111240404099226,
|
|
"num_tokens": 1168592.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.217110013961792,
|
|
"epoch": 0.5678745267712277,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.8235,
|
|
"mean_token_accuracy": 0.10857188254594803,
|
|
"num_tokens": 1178680.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.096605062484741,
|
|
"epoch": 0.5732828555976204,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.7677,
|
|
"mean_token_accuracy": 0.1095486730337143,
|
|
"num_tokens": 1188825.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.168183660507202,
|
|
"epoch": 0.578691184424013,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.8024,
|
|
"mean_token_accuracy": 0.10566585659980773,
|
|
"num_tokens": 1201375.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.183142471313476,
|
|
"epoch": 0.5840995132504057,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.9103,
|
|
"mean_token_accuracy": 0.10791068449616432,
|
|
"num_tokens": 1211915.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.140567922592163,
|
|
"epoch": 0.5895078420767983,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.8295,
|
|
"mean_token_accuracy": 0.10977402180433274,
|
|
"num_tokens": 1223151.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.178637409210205,
|
|
"epoch": 0.5949161709031909,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.7615,
|
|
"mean_token_accuracy": 0.10909788757562637,
|
|
"num_tokens": 1234229.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.1336465835571286,
|
|
"epoch": 0.6003244997295836,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.7672,
|
|
"mean_token_accuracy": 0.10949002876877785,
|
|
"num_tokens": 1245298.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.988934183120728,
|
|
"epoch": 0.6057328285559762,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.7605,
|
|
"mean_token_accuracy": 0.10834354907274246,
|
|
"num_tokens": 1256154.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.085337352752686,
|
|
"epoch": 0.6111411573823688,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.6712,
|
|
"mean_token_accuracy": 0.11431163474917412,
|
|
"num_tokens": 1267404.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.080687570571899,
|
|
"epoch": 0.6165494862087615,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.7296,
|
|
"mean_token_accuracy": 0.11148776188492775,
|
|
"num_tokens": 1278894.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.016615533828736,
|
|
"epoch": 0.6219578150351541,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.7067,
|
|
"mean_token_accuracy": 0.11370179131627083,
|
|
"num_tokens": 1290408.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.077807664871216,
|
|
"epoch": 0.6273661438615468,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.7692,
|
|
"mean_token_accuracy": 0.10994554236531258,
|
|
"num_tokens": 1301508.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.073387670516968,
|
|
"epoch": 0.6327744726879394,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.7459,
|
|
"mean_token_accuracy": 0.1134707696735859,
|
|
"num_tokens": 1312551.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.117682027816772,
|
|
"epoch": 0.638182801514332,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.8539,
|
|
"mean_token_accuracy": 0.10747635588049889,
|
|
"num_tokens": 1324876.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.067667627334595,
|
|
"epoch": 0.6435911303407247,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.7301,
|
|
"mean_token_accuracy": 0.11152606382966042,
|
|
"num_tokens": 1335401.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.943005657196045,
|
|
"epoch": 0.6489994591671173,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.6641,
|
|
"mean_token_accuracy": 0.11002366989850998,
|
|
"num_tokens": 1348265.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.00308403968811,
|
|
"epoch": 0.65440778799351,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.6334,
|
|
"mean_token_accuracy": 0.11659377664327622,
|
|
"num_tokens": 1359078.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.848029279708863,
|
|
"epoch": 0.6598161168199026,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.5246,
|
|
"mean_token_accuracy": 0.1145120233297348,
|
|
"num_tokens": 1369499.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.01754412651062,
|
|
"epoch": 0.6652244456462953,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.6243,
|
|
"mean_token_accuracy": 0.1155681550502777,
|
|
"num_tokens": 1380329.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.9016993045806885,
|
|
"epoch": 0.670632774472688,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.6966,
|
|
"mean_token_accuracy": 0.10963296741247178,
|
|
"num_tokens": 1393047.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.983740949630738,
|
|
"epoch": 0.6760411032990806,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.6993,
|
|
"mean_token_accuracy": 0.11043834760785103,
|
|
"num_tokens": 1405600.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.941162824630737,
|
|
"epoch": 0.6814494321254733,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.6279,
|
|
"mean_token_accuracy": 0.1165043719112873,
|
|
"num_tokens": 1416670.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.917750406265259,
|
|
"epoch": 0.6868577609518659,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.6536,
|
|
"mean_token_accuracy": 0.11544405594468117,
|
|
"num_tokens": 1428929.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.928350114822388,
|
|
"epoch": 0.6922660897782585,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.5784,
|
|
"mean_token_accuracy": 0.11553277149796486,
|
|
"num_tokens": 1439298.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.942523860931397,
|
|
"epoch": 0.6976744186046512,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.6653,
|
|
"mean_token_accuracy": 0.1168369047343731,
|
|
"num_tokens": 1449853.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.9924798011779785,
|
|
"epoch": 0.7030827474310438,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.7393,
|
|
"mean_token_accuracy": 0.11061546951532364,
|
|
"num_tokens": 1460816.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.836633491516113,
|
|
"epoch": 0.7084910762574365,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.5868,
|
|
"mean_token_accuracy": 0.12072381898760795,
|
|
"num_tokens": 1473003.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.978844785690308,
|
|
"epoch": 0.7138994050838291,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.7076,
|
|
"mean_token_accuracy": 0.1100004144012928,
|
|
"num_tokens": 1484745.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.95818305015564,
|
|
"epoch": 0.7193077339102217,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.5842,
|
|
"mean_token_accuracy": 0.11697863414883614,
|
|
"num_tokens": 1495803.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.916927337646484,
|
|
"epoch": 0.7247160627366144,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.6057,
|
|
"mean_token_accuracy": 0.12135127037763596,
|
|
"num_tokens": 1507504.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.949080896377564,
|
|
"epoch": 0.730124391563007,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.6915,
|
|
"mean_token_accuracy": 0.11211434379220009,
|
|
"num_tokens": 1518765.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.811977338790894,
|
|
"epoch": 0.7355327203893997,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.5654,
|
|
"mean_token_accuracy": 0.11754105314612388,
|
|
"num_tokens": 1529481.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.936152696609497,
|
|
"epoch": 0.7409410492157923,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.586,
|
|
"mean_token_accuracy": 0.1217528983950615,
|
|
"num_tokens": 1539612.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.936670732498169,
|
|
"epoch": 0.7463493780421849,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.7353,
|
|
"mean_token_accuracy": 0.11465424224734307,
|
|
"num_tokens": 1551370.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.890514802932739,
|
|
"epoch": 0.7517577068685776,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.6289,
|
|
"mean_token_accuracy": 0.12030648663640023,
|
|
"num_tokens": 1561662.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.9119446754455565,
|
|
"epoch": 0.7571660356949702,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.6066,
|
|
"mean_token_accuracy": 0.11544964835047722,
|
|
"num_tokens": 1573479.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.820972013473511,
|
|
"epoch": 0.7625743645213628,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.5541,
|
|
"mean_token_accuracy": 0.12045397162437439,
|
|
"num_tokens": 1583684.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.84340467453003,
|
|
"epoch": 0.7679826933477556,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.6004,
|
|
"mean_token_accuracy": 0.11728495731949806,
|
|
"num_tokens": 1595694.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.853058099746704,
|
|
"epoch": 0.7733910221741482,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.5926,
|
|
"mean_token_accuracy": 0.11383338570594788,
|
|
"num_tokens": 1606104.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.866402721405029,
|
|
"epoch": 0.7787993510005409,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.6694,
|
|
"mean_token_accuracy": 0.11633143052458764,
|
|
"num_tokens": 1617368.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.938875913619995,
|
|
"epoch": 0.7842076798269335,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.6866,
|
|
"mean_token_accuracy": 0.11025089025497437,
|
|
"num_tokens": 1630485.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.887586164474487,
|
|
"epoch": 0.7896160086533262,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.511,
|
|
"mean_token_accuracy": 0.12038141191005707,
|
|
"num_tokens": 1641411.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.772418260574341,
|
|
"epoch": 0.7950243374797188,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.5445,
|
|
"mean_token_accuracy": 0.1245256394147873,
|
|
"num_tokens": 1653737.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.911994504928589,
|
|
"epoch": 0.8004326663061114,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.711,
|
|
"mean_token_accuracy": 0.11259080916643142,
|
|
"num_tokens": 1665910.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.867116498947143,
|
|
"epoch": 0.8058409951325041,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.4955,
|
|
"mean_token_accuracy": 0.1199782818555832,
|
|
"num_tokens": 1676927.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.749074554443359,
|
|
"epoch": 0.8112493239588967,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.5329,
|
|
"mean_token_accuracy": 0.12157572358846665,
|
|
"num_tokens": 1688167.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.782977724075318,
|
|
"epoch": 0.8166576527852893,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.6009,
|
|
"mean_token_accuracy": 0.12366797998547555,
|
|
"num_tokens": 1699899.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.773900890350342,
|
|
"epoch": 0.822065981611682,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.5265,
|
|
"mean_token_accuracy": 0.12148023024201393,
|
|
"num_tokens": 1709862.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.766888999938965,
|
|
"epoch": 0.8274743104380746,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.5152,
|
|
"mean_token_accuracy": 0.12373604327440262,
|
|
"num_tokens": 1719902.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.849668073654175,
|
|
"epoch": 0.8328826392644673,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.5274,
|
|
"mean_token_accuracy": 0.12188500910997391,
|
|
"num_tokens": 1730351.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.67292013168335,
|
|
"epoch": 0.8382909680908599,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.4828,
|
|
"mean_token_accuracy": 0.11859010457992554,
|
|
"num_tokens": 1741783.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.7774739265441895,
|
|
"epoch": 0.8436992969172525,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.5193,
|
|
"mean_token_accuracy": 0.12553910985589029,
|
|
"num_tokens": 1752952.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.771312952041626,
|
|
"epoch": 0.8491076257436452,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.5523,
|
|
"mean_token_accuracy": 0.1199328914284706,
|
|
"num_tokens": 1764014.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.799132776260376,
|
|
"epoch": 0.8545159545700378,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.5186,
|
|
"mean_token_accuracy": 0.12284265831112862,
|
|
"num_tokens": 1776154.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.771439504623413,
|
|
"epoch": 0.8599242833964305,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.5663,
|
|
"mean_token_accuracy": 0.12119035795331001,
|
|
"num_tokens": 1787292.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.7058521747589115,
|
|
"epoch": 0.8653326122228231,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.5406,
|
|
"mean_token_accuracy": 0.12341648563742638,
|
|
"num_tokens": 1799469.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.716012334823608,
|
|
"epoch": 0.8707409410492158,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.4142,
|
|
"mean_token_accuracy": 0.13010902479290962,
|
|
"num_tokens": 1810091.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.792158555984497,
|
|
"epoch": 0.8761492698756085,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.5778,
|
|
"mean_token_accuracy": 0.12326778918504715,
|
|
"num_tokens": 1820835.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.7224524974822994,
|
|
"epoch": 0.8815575987020011,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.5295,
|
|
"mean_token_accuracy": 0.12325069829821586,
|
|
"num_tokens": 1831868.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.67799129486084,
|
|
"epoch": 0.8869659275283938,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.5106,
|
|
"mean_token_accuracy": 0.12507711797952653,
|
|
"num_tokens": 1842040.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.872373056411743,
|
|
"epoch": 0.8923742563547864,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.5865,
|
|
"mean_token_accuracy": 0.1232219822704792,
|
|
"num_tokens": 1853765.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.725686120986938,
|
|
"epoch": 0.897782585181179,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.5776,
|
|
"mean_token_accuracy": 0.12357532158493996,
|
|
"num_tokens": 1866081.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.687033033370971,
|
|
"epoch": 0.9031909140075717,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.4792,
|
|
"mean_token_accuracy": 0.12714275047183038,
|
|
"num_tokens": 1876754.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.851390647888183,
|
|
"epoch": 0.9085992428339643,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.5838,
|
|
"mean_token_accuracy": 0.12291418090462684,
|
|
"num_tokens": 1888299.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.573156261444092,
|
|
"epoch": 0.914007571660357,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.3628,
|
|
"mean_token_accuracy": 0.13524338006973266,
|
|
"num_tokens": 1899027.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.724084043502808,
|
|
"epoch": 0.9194159004867496,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.5022,
|
|
"mean_token_accuracy": 0.12369738146662712,
|
|
"num_tokens": 1911043.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.700301551818848,
|
|
"epoch": 0.9248242293131422,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.4313,
|
|
"mean_token_accuracy": 0.12529089376330377,
|
|
"num_tokens": 1921295.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.730074739456176,
|
|
"epoch": 0.9302325581395349,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.5241,
|
|
"mean_token_accuracy": 0.12679863050580026,
|
|
"num_tokens": 1932944.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.617708730697632,
|
|
"epoch": 0.9356408869659275,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.4668,
|
|
"mean_token_accuracy": 0.1202548824250698,
|
|
"num_tokens": 1944072.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.705368375778198,
|
|
"epoch": 0.9410492157923201,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.462,
|
|
"mean_token_accuracy": 0.1302238143980503,
|
|
"num_tokens": 1954798.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.665252828598023,
|
|
"epoch": 0.9464575446187128,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.4217,
|
|
"mean_token_accuracy": 0.13559063598513604,
|
|
"num_tokens": 1965304.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.698717737197876,
|
|
"epoch": 0.9518658734451054,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.4574,
|
|
"mean_token_accuracy": 0.1233592003583908,
|
|
"num_tokens": 1976797.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.673433637619018,
|
|
"epoch": 0.9572742022714981,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.4792,
|
|
"mean_token_accuracy": 0.1290605790913105,
|
|
"num_tokens": 1987448.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.445412921905517,
|
|
"epoch": 0.9626825310978907,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.3189,
|
|
"mean_token_accuracy": 0.12513260841369628,
|
|
"num_tokens": 2000948.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.692562246322632,
|
|
"epoch": 0.9680908599242833,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.4284,
|
|
"mean_token_accuracy": 0.1312178909778595,
|
|
"num_tokens": 2011807.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.67714900970459,
|
|
"epoch": 0.9734991887506761,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.4909,
|
|
"mean_token_accuracy": 0.12214648351073265,
|
|
"num_tokens": 2022835.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.528797817230225,
|
|
"epoch": 0.9789075175770687,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.4073,
|
|
"mean_token_accuracy": 0.1290948636829853,
|
|
"num_tokens": 2034331.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.6936122417449955,
|
|
"epoch": 0.9843158464034614,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.4058,
|
|
"mean_token_accuracy": 0.12213764488697051,
|
|
"num_tokens": 2047002.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.5391932964324955,
|
|
"epoch": 0.989724175229854,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.3452,
|
|
"mean_token_accuracy": 0.1322292260825634,
|
|
"num_tokens": 2056299.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.66591796875,
|
|
"epoch": 0.9951325040562466,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.4762,
|
|
"mean_token_accuracy": 0.126188213378191,
|
|
"num_tokens": 2068031.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.616532802581787,
|
|
"epoch": 1.0,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.403,
|
|
"mean_token_accuracy": 0.1335077327158716,
|
|
"num_tokens": 2076889.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.620981454849243,
|
|
"epoch": 1.0054083288263926,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.3081,
|
|
"mean_token_accuracy": 0.13363704532384874,
|
|
"num_tokens": 2087316.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.650969076156616,
|
|
"epoch": 1.0108166576527853,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.3096,
|
|
"mean_token_accuracy": 0.13058339804410934,
|
|
"num_tokens": 2098522.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.422043418884277,
|
|
"epoch": 1.016224986479178,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.2819,
|
|
"mean_token_accuracy": 0.13444022387266158,
|
|
"num_tokens": 2110015.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.650893974304199,
|
|
"epoch": 1.0216333153055706,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.3811,
|
|
"mean_token_accuracy": 0.12852860391139984,
|
|
"num_tokens": 2121333.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.6168934345245365,
|
|
"epoch": 1.0270416441319632,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.277,
|
|
"mean_token_accuracy": 0.1385727897286415,
|
|
"num_tokens": 2131385.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.5812818050384525,
|
|
"epoch": 1.0324499729583558,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.3407,
|
|
"mean_token_accuracy": 0.13214196115732194,
|
|
"num_tokens": 2143274.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.391562795639038,
|
|
"epoch": 1.0378583017847485,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.199,
|
|
"mean_token_accuracy": 0.13773884251713753,
|
|
"num_tokens": 2154529.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.626508951187134,
|
|
"epoch": 1.043266630611141,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.3531,
|
|
"mean_token_accuracy": 0.1296512059867382,
|
|
"num_tokens": 2165949.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.510818290710449,
|
|
"epoch": 1.0486749594375337,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.321,
|
|
"mean_token_accuracy": 0.13480681478977202,
|
|
"num_tokens": 2176710.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.509940958023071,
|
|
"epoch": 1.0540832882639264,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.3071,
|
|
"mean_token_accuracy": 0.13666068464517594,
|
|
"num_tokens": 2189470.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.516313314437866,
|
|
"epoch": 1.059491617090319,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.259,
|
|
"mean_token_accuracy": 0.13776395097374916,
|
|
"num_tokens": 2200266.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.556395244598389,
|
|
"epoch": 1.0648999459167117,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.2656,
|
|
"mean_token_accuracy": 0.13476681634783744,
|
|
"num_tokens": 2210736.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.5328137397766115,
|
|
"epoch": 1.0703082747431043,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.3809,
|
|
"mean_token_accuracy": 0.1317829020321369,
|
|
"num_tokens": 2221057.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.504324674606323,
|
|
"epoch": 1.075716603569497,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.3542,
|
|
"mean_token_accuracy": 0.13191068023443223,
|
|
"num_tokens": 2232352.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.625280714035034,
|
|
"epoch": 1.0811249323958896,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.2763,
|
|
"mean_token_accuracy": 0.1364446148276329,
|
|
"num_tokens": 2244655.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 1.0811249323958896,
|
|
"eval_entropy": 6.191254384951158,
|
|
"eval_loss": 6.552407741546631,
|
|
"eval_mean_token_accuracy": 0.13381302994252606,
|
|
"eval_num_tokens": 2244655.0,
|
|
"eval_runtime": 4.4463,
|
|
"eval_samples_per_second": 790.539,
|
|
"eval_steps_per_second": 98.958,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.450396490097046,
|
|
"epoch": 1.0865332612222822,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999997383518051,
|
|
"loss": 6.2292,
|
|
"mean_token_accuracy": 0.14541730880737305,
|
|
"num_tokens": 2255699.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.452439308166504,
|
|
"epoch": 1.0919415900486749,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999986754070562,
|
|
"loss": 6.2214,
|
|
"mean_token_accuracy": 0.1412962906062603,
|
|
"num_tokens": 2266627.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.620412015914917,
|
|
"epoch": 1.0973499188750675,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999967948166009,
|
|
"loss": 6.3971,
|
|
"mean_token_accuracy": 0.12971344217658043,
|
|
"num_tokens": 2277814.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.5968804359436035,
|
|
"epoch": 1.1027582477014604,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004999940965872734,
|
|
"loss": 6.346,
|
|
"mean_token_accuracy": 0.13398940712213517,
|
|
"num_tokens": 2287344.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.305682420730591,
|
|
"epoch": 1.1081665765278528,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004999905807288788,
|
|
"loss": 6.1519,
|
|
"mean_token_accuracy": 0.1420295163989067,
|
|
"num_tokens": 2298332.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.5274440288543705,
|
|
"epoch": 1.1135749053542456,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000499986247254194,
|
|
"loss": 6.2551,
|
|
"mean_token_accuracy": 0.1386675715446472,
|
|
"num_tokens": 2309161.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.500128126144409,
|
|
"epoch": 1.1189832341806383,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999810961789667,
|
|
"loss": 6.2543,
|
|
"mean_token_accuracy": 0.13907705023884773,
|
|
"num_tokens": 2319516.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.414528656005859,
|
|
"epoch": 1.124391563007031,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000499975127521916,
|
|
"loss": 6.1628,
|
|
"mean_token_accuracy": 0.13905372619628906,
|
|
"num_tokens": 2330441.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.43261022567749,
|
|
"epoch": 1.1297998918334236,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999683413047319,
|
|
"loss": 6.1652,
|
|
"mean_token_accuracy": 0.14364721179008483,
|
|
"num_tokens": 2340524.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.3701033115386965,
|
|
"epoch": 1.1352082206598162,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004999607375520759,
|
|
"loss": 6.1644,
|
|
"mean_token_accuracy": 0.1395645372569561,
|
|
"num_tokens": 2353379.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.495369243621826,
|
|
"epoch": 1.1406165494862088,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004999523162915794,
|
|
"loss": 6.3563,
|
|
"mean_token_accuracy": 0.1315511830151081,
|
|
"num_tokens": 2365708.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.4393187999725345,
|
|
"epoch": 1.1460248783126015,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999430775538459,
|
|
"loss": 6.244,
|
|
"mean_token_accuracy": 0.13950775936245918,
|
|
"num_tokens": 2375630.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.420579957962036,
|
|
"epoch": 1.151433207138994,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999330213724487,
|
|
"loss": 6.2415,
|
|
"mean_token_accuracy": 0.13685646206140517,
|
|
"num_tokens": 2386885.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.404643249511719,
|
|
"epoch": 1.1568415359653867,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999221477839319,
|
|
"loss": 6.2821,
|
|
"mean_token_accuracy": 0.1356322333216667,
|
|
"num_tokens": 2397676.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.48656849861145,
|
|
"epoch": 1.1622498647917794,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004999104568278104,
|
|
"loss": 6.2798,
|
|
"mean_token_accuracy": 0.13633241057395934,
|
|
"num_tokens": 2409039.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.491619968414307,
|
|
"epoch": 1.167658193618172,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000499897948546569,
|
|
"loss": 6.3141,
|
|
"mean_token_accuracy": 0.13397775068879128,
|
|
"num_tokens": 2420103.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.458958578109741,
|
|
"epoch": 1.1730665224445647,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004998846229856629,
|
|
"loss": 6.2385,
|
|
"mean_token_accuracy": 0.14006805866956712,
|
|
"num_tokens": 2430792.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.476471424102783,
|
|
"epoch": 1.1784748512709573,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004998704801935171,
|
|
"loss": 6.2481,
|
|
"mean_token_accuracy": 0.13575137481093408,
|
|
"num_tokens": 2442165.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.351526546478271,
|
|
"epoch": 1.18388318009735,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004998555202215267,
|
|
"loss": 6.1523,
|
|
"mean_token_accuracy": 0.14058220088481904,
|
|
"num_tokens": 2454231.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.49529128074646,
|
|
"epoch": 1.1892915089237426,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004998397431240561,
|
|
"loss": 6.2026,
|
|
"mean_token_accuracy": 0.1469121038913727,
|
|
"num_tokens": 2464828.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.318851900100708,
|
|
"epoch": 1.1946998377501352,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004998231489584395,
|
|
"loss": 6.2649,
|
|
"mean_token_accuracy": 0.1295018821954727,
|
|
"num_tokens": 2476144.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.604259729385376,
|
|
"epoch": 1.2001081665765279,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00049980573778498,
|
|
"loss": 6.432,
|
|
"mean_token_accuracy": 0.1328103795647621,
|
|
"num_tokens": 2489116.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.4929886817932125,
|
|
"epoch": 1.2055164954029205,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004997875096669498,
|
|
"loss": 6.281,
|
|
"mean_token_accuracy": 0.13723942190408706,
|
|
"num_tokens": 2501609.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.315258407592774,
|
|
"epoch": 1.2109248242293131,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004997684646705901,
|
|
"loss": 6.1823,
|
|
"mean_token_accuracy": 0.14818599075078964,
|
|
"num_tokens": 2513338.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.52914867401123,
|
|
"epoch": 1.2163331530557058,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004997486028651103,
|
|
"loss": 6.2172,
|
|
"mean_token_accuracy": 0.14651284366846085,
|
|
"num_tokens": 2523199.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.42001051902771,
|
|
"epoch": 1.2217414818820984,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004997279243226884,
|
|
"loss": 6.2313,
|
|
"mean_token_accuracy": 0.1383729934692383,
|
|
"num_tokens": 2535136.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.3943792343139645,
|
|
"epoch": 1.227149810708491,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004997064291184701,
|
|
"loss": 6.1331,
|
|
"mean_token_accuracy": 0.13703298419713975,
|
|
"num_tokens": 2546428.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.485766887664795,
|
|
"epoch": 1.2325581395348837,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004996841173305691,
|
|
"loss": 6.3232,
|
|
"mean_token_accuracy": 0.13487154394388198,
|
|
"num_tokens": 2558304.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.263403987884521,
|
|
"epoch": 1.2379664683612763,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004996609890400665,
|
|
"loss": 6.0197,
|
|
"mean_token_accuracy": 0.1503302812576294,
|
|
"num_tokens": 2568717.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.406390810012818,
|
|
"epoch": 1.243374797187669,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004996370443310107,
|
|
"loss": 6.232,
|
|
"mean_token_accuracy": 0.1359953410923481,
|
|
"num_tokens": 2578748.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.418755197525025,
|
|
"epoch": 1.2487831260140616,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004996122832904165,
|
|
"loss": 6.2031,
|
|
"mean_token_accuracy": 0.13639421239495278,
|
|
"num_tokens": 2590095.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.432490587234497,
|
|
"epoch": 1.2541914548404542,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004995867060082658,
|
|
"loss": 6.2509,
|
|
"mean_token_accuracy": 0.13807161152362823,
|
|
"num_tokens": 2600929.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.422495651245117,
|
|
"epoch": 1.2595997836668469,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004995603125775065,
|
|
"loss": 6.2334,
|
|
"mean_token_accuracy": 0.1417832314968109,
|
|
"num_tokens": 2611830.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.377342653274536,
|
|
"epoch": 1.2650081124932395,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004995331030940523,
|
|
"loss": 6.1746,
|
|
"mean_token_accuracy": 0.14085349589586257,
|
|
"num_tokens": 2623177.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.2920270442962645,
|
|
"epoch": 1.2704164413196322,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004995050776567826,
|
|
"loss": 6.1754,
|
|
"mean_token_accuracy": 0.1398814782500267,
|
|
"num_tokens": 2635563.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.4678558826446535,
|
|
"epoch": 1.2758247701460248,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004994762363675421,
|
|
"loss": 6.1697,
|
|
"mean_token_accuracy": 0.14567856714129448,
|
|
"num_tokens": 2647157.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.339539527893066,
|
|
"epoch": 1.2812330989724177,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004994465793311396,
|
|
"loss": 6.1378,
|
|
"mean_token_accuracy": 0.14672928750514985,
|
|
"num_tokens": 2658175.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.3544481754302975,
|
|
"epoch": 1.28664142779881,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004994161066553493,
|
|
"loss": 6.1089,
|
|
"mean_token_accuracy": 0.14247029796242713,
|
|
"num_tokens": 2669458.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.496858882904053,
|
|
"epoch": 1.292049756625203,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004993848184509086,
|
|
"loss": 6.2973,
|
|
"mean_token_accuracy": 0.14422353953123093,
|
|
"num_tokens": 2679727.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.325573873519898,
|
|
"epoch": 1.2974580854515954,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004993527148315193,
|
|
"loss": 6.2567,
|
|
"mean_token_accuracy": 0.13874245584011077,
|
|
"num_tokens": 2691031.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.37370719909668,
|
|
"epoch": 1.3028664142779882,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004993197959138458,
|
|
"loss": 6.0528,
|
|
"mean_token_accuracy": 0.15171269029378892,
|
|
"num_tokens": 2701699.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.30531415939331,
|
|
"epoch": 1.3082747431043806,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004992860618175156,
|
|
"loss": 6.1096,
|
|
"mean_token_accuracy": 0.14267323017120362,
|
|
"num_tokens": 2713038.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.320184707641602,
|
|
"epoch": 1.3136830719307735,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004992515126651184,
|
|
"loss": 6.1026,
|
|
"mean_token_accuracy": 0.14341416358947753,
|
|
"num_tokens": 2723685.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.3012495040893555,
|
|
"epoch": 1.319091400757166,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004992161485822059,
|
|
"loss": 6.1734,
|
|
"mean_token_accuracy": 0.14591606706380844,
|
|
"num_tokens": 2735341.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.374839544296265,
|
|
"epoch": 1.3244997295835588,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004991799696972913,
|
|
"loss": 6.1203,
|
|
"mean_token_accuracy": 0.14721479415893554,
|
|
"num_tokens": 2746052.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.2311437129974365,
|
|
"epoch": 1.3299080584099512,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004991429761418488,
|
|
"loss": 6.1062,
|
|
"mean_token_accuracy": 0.14226374998688698,
|
|
"num_tokens": 2757155.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.372267866134644,
|
|
"epoch": 1.335316387236344,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000499105168050313,
|
|
"loss": 6.1563,
|
|
"mean_token_accuracy": 0.1399580717086792,
|
|
"num_tokens": 2769631.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.4830890655517575,
|
|
"epoch": 1.3407247160627367,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004990665455600787,
|
|
"loss": 6.2694,
|
|
"mean_token_accuracy": 0.14426804780960084,
|
|
"num_tokens": 2780797.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.3079752922058105,
|
|
"epoch": 1.3461330448891293,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004990271088114999,
|
|
"loss": 6.1137,
|
|
"mean_token_accuracy": 0.1408174991607666,
|
|
"num_tokens": 2792721.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.320454740524292,
|
|
"epoch": 1.351541373715522,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004989868579478901,
|
|
"loss": 6.1844,
|
|
"mean_token_accuracy": 0.15411919802427293,
|
|
"num_tokens": 2804531.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.418113660812378,
|
|
"epoch": 1.3569497025419146,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004989457931155209,
|
|
"loss": 6.1699,
|
|
"mean_token_accuracy": 0.13872058242559432,
|
|
"num_tokens": 2816666.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.334762191772461,
|
|
"epoch": 1.3623580313683072,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004989039144636219,
|
|
"loss": 6.1896,
|
|
"mean_token_accuracy": 0.14479968696832657,
|
|
"num_tokens": 2827977.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.282640838623047,
|
|
"epoch": 1.3677663601946999,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004988612221443804,
|
|
"loss": 6.1628,
|
|
"mean_token_accuracy": 0.14071345701813698,
|
|
"num_tokens": 2839069.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.4330846786499025,
|
|
"epoch": 1.3731746890210925,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004988177163129403,
|
|
"loss": 6.2356,
|
|
"mean_token_accuracy": 0.1398556999862194,
|
|
"num_tokens": 2850632.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.29965443611145,
|
|
"epoch": 1.3785830178474852,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004987733971274018,
|
|
"loss": 6.1213,
|
|
"mean_token_accuracy": 0.14067689031362535,
|
|
"num_tokens": 2862204.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.256815624237061,
|
|
"epoch": 1.3839913466738778,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004987282647488211,
|
|
"loss": 6.0849,
|
|
"mean_token_accuracy": 0.15381099730730058,
|
|
"num_tokens": 2874010.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.362789487838745,
|
|
"epoch": 1.3893996755002704,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004986823193412093,
|
|
"loss": 6.1446,
|
|
"mean_token_accuracy": 0.15051770359277725,
|
|
"num_tokens": 2884315.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.394857549667359,
|
|
"epoch": 1.394808004326663,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000498635561071532,
|
|
"loss": 6.2989,
|
|
"mean_token_accuracy": 0.13909848481416703,
|
|
"num_tokens": 2896327.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.338388347625733,
|
|
"epoch": 1.4002163331530557,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000498587990109709,
|
|
"loss": 6.0656,
|
|
"mean_token_accuracy": 0.14448106661438942,
|
|
"num_tokens": 2907007.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.3123034000396725,
|
|
"epoch": 1.4056246619794484,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004985396066286133,
|
|
"loss": 6.1136,
|
|
"mean_token_accuracy": 0.15292371809482574,
|
|
"num_tokens": 2918852.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.231387519836426,
|
|
"epoch": 1.411032990805841,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004984904108040706,
|
|
"loss": 6.0242,
|
|
"mean_token_accuracy": 0.14866252094507218,
|
|
"num_tokens": 2929948.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.256463861465454,
|
|
"epoch": 1.4164413196322336,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004984404028148585,
|
|
"loss": 6.1033,
|
|
"mean_token_accuracy": 0.14884472787380218,
|
|
"num_tokens": 2941299.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.2421142101287845,
|
|
"epoch": 1.4218496484586263,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004983895828427062,
|
|
"loss": 6.0316,
|
|
"mean_token_accuracy": 0.14848915934562684,
|
|
"num_tokens": 2952283.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.555593776702881,
|
|
"epoch": 1.427257977285019,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004983379510722935,
|
|
"loss": 6.3871,
|
|
"mean_token_accuracy": 0.14026519209146499,
|
|
"num_tokens": 2962241.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.315966176986694,
|
|
"epoch": 1.4326663061114115,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004982855076912506,
|
|
"loss": 6.1417,
|
|
"mean_token_accuracy": 0.14470569640398026,
|
|
"num_tokens": 2973408.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.26273889541626,
|
|
"epoch": 1.4380746349378042,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004982322528901568,
|
|
"loss": 6.1779,
|
|
"mean_token_accuracy": 0.14446252807974816,
|
|
"num_tokens": 2985877.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.439066934585571,
|
|
"epoch": 1.4434829637641968,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.00049817818686254,
|
|
"loss": 6.1773,
|
|
"mean_token_accuracy": 0.14509968981146812,
|
|
"num_tokens": 2996935.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.2080278396606445,
|
|
"epoch": 1.4488912925905895,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004981233098048765,
|
|
"loss": 6.063,
|
|
"mean_token_accuracy": 0.15094534754753114,
|
|
"num_tokens": 3007932.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.359493446350098,
|
|
"epoch": 1.454299621416982,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004980676219165894,
|
|
"loss": 6.154,
|
|
"mean_token_accuracy": 0.1459366887807846,
|
|
"num_tokens": 3019368.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.34571681022644,
|
|
"epoch": 1.4597079502433747,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.000498011123400049,
|
|
"loss": 6.2202,
|
|
"mean_token_accuracy": 0.140365319699049,
|
|
"num_tokens": 3029787.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.237887668609619,
|
|
"epoch": 1.4651162790697674,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004979538144605707,
|
|
"loss": 6.1487,
|
|
"mean_token_accuracy": 0.139919201284647,
|
|
"num_tokens": 3042100.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.3007818222045895,
|
|
"epoch": 1.47052460789616,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004978956953064154,
|
|
"loss": 6.1784,
|
|
"mean_token_accuracy": 0.14781156331300735,
|
|
"num_tokens": 3054160.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.314223575592041,
|
|
"epoch": 1.4759329367225527,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004978367661487883,
|
|
"loss": 6.1417,
|
|
"mean_token_accuracy": 0.14655674770474433,
|
|
"num_tokens": 3065401.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.263867568969727,
|
|
"epoch": 1.4813412655489453,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004977770272018379,
|
|
"loss": 6.0533,
|
|
"mean_token_accuracy": 0.15251464024186134,
|
|
"num_tokens": 3076137.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.355552768707275,
|
|
"epoch": 1.486749594375338,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000497716478682656,
|
|
"loss": 6.1912,
|
|
"mean_token_accuracy": 0.14727144837379455,
|
|
"num_tokens": 3087274.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.287002611160278,
|
|
"epoch": 1.4921579232017308,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004976551208112759,
|
|
"loss": 6.0899,
|
|
"mean_token_accuracy": 0.15184621065855025,
|
|
"num_tokens": 3098476.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.294669342041016,
|
|
"epoch": 1.4975662520281232,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004975929538106721,
|
|
"loss": 6.1194,
|
|
"mean_token_accuracy": 0.14649061784148215,
|
|
"num_tokens": 3110060.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.156888437271118,
|
|
"epoch": 1.502974580854516,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004975299779067598,
|
|
"loss": 6.0017,
|
|
"mean_token_accuracy": 0.15018683224916457,
|
|
"num_tokens": 3121032.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.350030136108399,
|
|
"epoch": 1.5083829096809085,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004974661933283936,
|
|
"loss": 6.0342,
|
|
"mean_token_accuracy": 0.15411721616983415,
|
|
"num_tokens": 3131999.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.105301761627198,
|
|
"epoch": 1.5137912385073014,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004974016003073667,
|
|
"loss": 5.9535,
|
|
"mean_token_accuracy": 0.15385461300611497,
|
|
"num_tokens": 3142693.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.189932203292846,
|
|
"epoch": 1.5191995673336938,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004973361990784104,
|
|
"loss": 5.9247,
|
|
"mean_token_accuracy": 0.14769245386123658,
|
|
"num_tokens": 3154179.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.279983186721802,
|
|
"epoch": 1.5246078961600866,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004972699898791927,
|
|
"loss": 6.0183,
|
|
"mean_token_accuracy": 0.15297574549913406,
|
|
"num_tokens": 3165314.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.2220347881317135,
|
|
"epoch": 1.530016224986479,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004972029729503182,
|
|
"loss": 6.1264,
|
|
"mean_token_accuracy": 0.15253533720970153,
|
|
"num_tokens": 3177646.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.3068013191223145,
|
|
"epoch": 1.535424553812872,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004971351485353265,
|
|
"loss": 6.1241,
|
|
"mean_token_accuracy": 0.15275899022817613,
|
|
"num_tokens": 3188401.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.181515836715699,
|
|
"epoch": 1.5408328826392643,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004970665168806917,
|
|
"loss": 5.9982,
|
|
"mean_token_accuracy": 0.15052082240581513,
|
|
"num_tokens": 3200206.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.251823472976684,
|
|
"epoch": 1.5462412114656572,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004969970782358214,
|
|
"loss": 6.1061,
|
|
"mean_token_accuracy": 0.14980608224868774,
|
|
"num_tokens": 3211981.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.2617998123168945,
|
|
"epoch": 1.5516495402920496,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004969268328530557,
|
|
"loss": 6.1326,
|
|
"mean_token_accuracy": 0.14618899673223495,
|
|
"num_tokens": 3221997.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.312466096878052,
|
|
"epoch": 1.5570578691184425,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004968557809876666,
|
|
"loss": 6.0647,
|
|
"mean_token_accuracy": 0.1465691164135933,
|
|
"num_tokens": 3234333.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.2024046897888185,
|
|
"epoch": 1.562466197944835,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004967839228978568,
|
|
"loss": 6.1101,
|
|
"mean_token_accuracy": 0.1412971571087837,
|
|
"num_tokens": 3245083.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.268845701217652,
|
|
"epoch": 1.5678745267712277,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004967112588447585,
|
|
"loss": 5.9556,
|
|
"mean_token_accuracy": 0.15390751361846924,
|
|
"num_tokens": 3255787.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.106264925003051,
|
|
"epoch": 1.5732828555976204,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004966377890924332,
|
|
"loss": 6.1241,
|
|
"mean_token_accuracy": 0.14512548372149467,
|
|
"num_tokens": 3267793.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.223533487319946,
|
|
"epoch": 1.578691184424013,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004965635139078704,
|
|
"loss": 5.9987,
|
|
"mean_token_accuracy": 0.1574643447995186,
|
|
"num_tokens": 3279100.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.284828519821167,
|
|
"epoch": 1.5840995132504057,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004964884335609857,
|
|
"loss": 6.1626,
|
|
"mean_token_accuracy": 0.14985712692141534,
|
|
"num_tokens": 3290251.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.214255952835083,
|
|
"epoch": 1.5895078420767983,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004964125483246218,
|
|
"loss": 6.0203,
|
|
"mean_token_accuracy": 0.14933666437864304,
|
|
"num_tokens": 3301817.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.303307151794433,
|
|
"epoch": 1.594916170903191,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004963358584745457,
|
|
"loss": 6.0644,
|
|
"mean_token_accuracy": 0.1487608626484871,
|
|
"num_tokens": 3313435.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.239393138885498,
|
|
"epoch": 1.6003244997295836,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004962583642894482,
|
|
"loss": 6.1937,
|
|
"mean_token_accuracy": 0.14459390044212342,
|
|
"num_tokens": 3325489.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.331709337234497,
|
|
"epoch": 1.6057328285559762,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004961800660509437,
|
|
"loss": 6.0633,
|
|
"mean_token_accuracy": 0.15172976553440093,
|
|
"num_tokens": 3337030.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.137213468551636,
|
|
"epoch": 1.6111411573823688,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000496100964043568,
|
|
"loss": 6.0347,
|
|
"mean_token_accuracy": 0.1523049309849739,
|
|
"num_tokens": 3347766.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.132486057281494,
|
|
"epoch": 1.6165494862087615,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000496021058554778,
|
|
"loss": 5.9263,
|
|
"mean_token_accuracy": 0.15191623717546462,
|
|
"num_tokens": 3360379.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.191946172714234,
|
|
"epoch": 1.6219578150351541,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004959403498749505,
|
|
"loss": 6.0763,
|
|
"mean_token_accuracy": 0.15376296490430832,
|
|
"num_tokens": 3370962.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.6219578150351541,
|
|
"eval_entropy": 6.064782928336751,
|
|
"eval_loss": 6.276134490966797,
|
|
"eval_mean_token_accuracy": 0.14995028799907728,
|
|
"eval_num_tokens": 3370962.0,
|
|
"eval_runtime": 4.4865,
|
|
"eval_samples_per_second": 783.455,
|
|
"eval_steps_per_second": 98.071,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.232034826278687,
|
|
"epoch": 1.6273661438615468,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004958588382973809,
|
|
"loss": 5.9937,
|
|
"mean_token_accuracy": 0.14936962574720383,
|
|
"num_tokens": 3382648.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.198745536804199,
|
|
"epoch": 1.6327744726879394,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004957765241182826,
|
|
"loss": 6.0457,
|
|
"mean_token_accuracy": 0.15271211713552474,
|
|
"num_tokens": 3393871.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.140760469436645,
|
|
"epoch": 1.638182801514332,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004956934076367854,
|
|
"loss": 5.9333,
|
|
"mean_token_accuracy": 0.15750416219234467,
|
|
"num_tokens": 3405148.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.1533373355865475,
|
|
"epoch": 1.6435911303407247,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000495609489154935,
|
|
"loss": 5.9815,
|
|
"mean_token_accuracy": 0.16143201887607575,
|
|
"num_tokens": 3416647.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.237495803833008,
|
|
"epoch": 1.6489994591671173,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004955247689776911,
|
|
"loss": 6.0662,
|
|
"mean_token_accuracy": 0.14711438715457917,
|
|
"num_tokens": 3428331.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.069284343719483,
|
|
"epoch": 1.65440778799351,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000495439247412927,
|
|
"loss": 5.923,
|
|
"mean_token_accuracy": 0.1526483453810215,
|
|
"num_tokens": 3440683.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.1490607261657715,
|
|
"epoch": 1.6598161168199026,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004953529247714285,
|
|
"loss": 5.9202,
|
|
"mean_token_accuracy": 0.15594317615032197,
|
|
"num_tokens": 3452375.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.1787848472595215,
|
|
"epoch": 1.6652244456462952,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004952658013668922,
|
|
"loss": 6.076,
|
|
"mean_token_accuracy": 0.14493782445788383,
|
|
"num_tokens": 3463255.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.237282943725586,
|
|
"epoch": 1.670632774472688,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004951778775159246,
|
|
"loss": 6.0122,
|
|
"mean_token_accuracy": 0.14876080676913261,
|
|
"num_tokens": 3473642.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.156322288513183,
|
|
"epoch": 1.6760411032990805,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004950891535380414,
|
|
"loss": 5.9817,
|
|
"mean_token_accuracy": 0.15054196119308472,
|
|
"num_tokens": 3484877.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.161394357681274,
|
|
"epoch": 1.6814494321254734,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004949996297556656,
|
|
"loss": 5.9028,
|
|
"mean_token_accuracy": 0.15873813927173613,
|
|
"num_tokens": 3495500.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.198855972290039,
|
|
"epoch": 1.6868577609518658,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004949093064941268,
|
|
"loss": 6.0283,
|
|
"mean_token_accuracy": 0.15485089272260666,
|
|
"num_tokens": 3508057.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.192743921279908,
|
|
"epoch": 1.6922660897782587,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004948181840816601,
|
|
"loss": 6.2052,
|
|
"mean_token_accuracy": 0.14864519387483596,
|
|
"num_tokens": 3521110.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.252610301971435,
|
|
"epoch": 1.697674418604651,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004947262628494045,
|
|
"loss": 6.1151,
|
|
"mean_token_accuracy": 0.15217255800962448,
|
|
"num_tokens": 3530692.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.248164892196655,
|
|
"epoch": 1.703082747431044,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004946335431314018,
|
|
"loss": 6.046,
|
|
"mean_token_accuracy": 0.14736961275339128,
|
|
"num_tokens": 3542195.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.203562545776367,
|
|
"epoch": 1.7084910762574363,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004945400252645959,
|
|
"loss": 5.949,
|
|
"mean_token_accuracy": 0.1598360300064087,
|
|
"num_tokens": 3552676.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.076257991790771,
|
|
"epoch": 1.7138994050838292,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004944457095888308,
|
|
"loss": 5.9107,
|
|
"mean_token_accuracy": 0.15307197719812393,
|
|
"num_tokens": 3563308.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.141274070739746,
|
|
"epoch": 1.7193077339102216,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00049435059644685,
|
|
"loss": 5.9323,
|
|
"mean_token_accuracy": 0.15914579331874848,
|
|
"num_tokens": 3574085.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.195391607284546,
|
|
"epoch": 1.7247160627366145,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004942546861842947,
|
|
"loss": 6.0122,
|
|
"mean_token_accuracy": 0.1477242238819599,
|
|
"num_tokens": 3585816.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.18343596458435,
|
|
"epoch": 1.730124391563007,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004941579791497032,
|
|
"loss": 6.0595,
|
|
"mean_token_accuracy": 0.15503935813903807,
|
|
"num_tokens": 3597776.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.289033079147339,
|
|
"epoch": 1.7355327203893998,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004940604756945091,
|
|
"loss": 6.0674,
|
|
"mean_token_accuracy": 0.15241216123104095,
|
|
"num_tokens": 3607991.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.068208932876587,
|
|
"epoch": 1.7409410492157922,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004939621761730399,
|
|
"loss": 5.9614,
|
|
"mean_token_accuracy": 0.16285738497972488,
|
|
"num_tokens": 3618204.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.176714992523193,
|
|
"epoch": 1.746349378042185,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004938630809425166,
|
|
"loss": 6.0594,
|
|
"mean_token_accuracy": 0.14926779568195342,
|
|
"num_tokens": 3629558.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.20286602973938,
|
|
"epoch": 1.7517577068685775,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004937631903630513,
|
|
"loss": 6.001,
|
|
"mean_token_accuracy": 0.1529402181506157,
|
|
"num_tokens": 3640773.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.170701408386231,
|
|
"epoch": 1.7571660356949703,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004936625047976468,
|
|
"loss": 5.9144,
|
|
"mean_token_accuracy": 0.15648894011974335,
|
|
"num_tokens": 3652186.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.091852855682373,
|
|
"epoch": 1.7625743645213627,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004935610246121944,
|
|
"loss": 5.9775,
|
|
"mean_token_accuracy": 0.15743271261453629,
|
|
"num_tokens": 3663068.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.1932361125946045,
|
|
"epoch": 1.7679826933477556,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004934587501754736,
|
|
"loss": 6.1074,
|
|
"mean_token_accuracy": 0.14869039058685302,
|
|
"num_tokens": 3673803.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.2811558723449705,
|
|
"epoch": 1.7733910221741482,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004933556818591498,
|
|
"loss": 6.0727,
|
|
"mean_token_accuracy": 0.14556089416146278,
|
|
"num_tokens": 3685482.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.1613668441772464,
|
|
"epoch": 1.7787993510005409,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004932518200377736,
|
|
"loss": 5.94,
|
|
"mean_token_accuracy": 0.16360433846712114,
|
|
"num_tokens": 3697641.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.114334917068481,
|
|
"epoch": 1.7842076798269335,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004931471650887789,
|
|
"loss": 5.9546,
|
|
"mean_token_accuracy": 0.15964469611644744,
|
|
"num_tokens": 3709560.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.221502876281738,
|
|
"epoch": 1.7896160086533262,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004930417173924822,
|
|
"loss": 6.0517,
|
|
"mean_token_accuracy": 0.15098627731204034,
|
|
"num_tokens": 3720422.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.0206968784332275,
|
|
"epoch": 1.7950243374797188,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004929354773320806,
|
|
"loss": 5.8816,
|
|
"mean_token_accuracy": 0.15427074134349822,
|
|
"num_tokens": 3731104.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.189523124694825,
|
|
"epoch": 1.8004326663061114,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004928284452936506,
|
|
"loss": 6.063,
|
|
"mean_token_accuracy": 0.1498106002807617,
|
|
"num_tokens": 3742766.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.172471809387207,
|
|
"epoch": 1.805840995132504,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004927206216661473,
|
|
"loss": 5.9374,
|
|
"mean_token_accuracy": 0.15604119896888732,
|
|
"num_tokens": 3753064.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.169418382644653,
|
|
"epoch": 1.8112493239588967,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004926120068414015,
|
|
"loss": 5.9773,
|
|
"mean_token_accuracy": 0.15225915387272834,
|
|
"num_tokens": 3765183.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.134690809249878,
|
|
"epoch": 1.8166576527852893,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004925026012141201,
|
|
"loss": 5.9651,
|
|
"mean_token_accuracy": 0.14956604242324828,
|
|
"num_tokens": 3777902.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.1190526485443115,
|
|
"epoch": 1.822065981611682,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004923924051818832,
|
|
"loss": 5.9411,
|
|
"mean_token_accuracy": 0.15607824176549911,
|
|
"num_tokens": 3788461.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.095177125930786,
|
|
"epoch": 1.8274743104380746,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004922814191451435,
|
|
"loss": 5.8659,
|
|
"mean_token_accuracy": 0.16016081720590591,
|
|
"num_tokens": 3799574.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.149827575683593,
|
|
"epoch": 1.8328826392644673,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004921696435072245,
|
|
"loss": 6.0158,
|
|
"mean_token_accuracy": 0.1539915531873703,
|
|
"num_tokens": 3811889.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.15166015625,
|
|
"epoch": 1.83829096809086,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004920570786743191,
|
|
"loss": 5.9791,
|
|
"mean_token_accuracy": 0.15387122333049774,
|
|
"num_tokens": 3822949.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.1611213207244875,
|
|
"epoch": 1.8436992969172525,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004919437250554882,
|
|
"loss": 6.0731,
|
|
"mean_token_accuracy": 0.15032268762588502,
|
|
"num_tokens": 3834734.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.173774862289429,
|
|
"epoch": 1.8491076257436452,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004918295830626592,
|
|
"loss": 5.9376,
|
|
"mean_token_accuracy": 0.16012347787618636,
|
|
"num_tokens": 3845508.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.0878442287445065,
|
|
"epoch": 1.8545159545700378,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004917146531106242,
|
|
"loss": 5.9151,
|
|
"mean_token_accuracy": 0.15681960731744765,
|
|
"num_tokens": 3856500.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.238417530059815,
|
|
"epoch": 1.8599242833964305,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004915989356170389,
|
|
"loss": 6.0013,
|
|
"mean_token_accuracy": 0.15328494608402252,
|
|
"num_tokens": 3867877.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.131330776214599,
|
|
"epoch": 1.865332612222823,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004914824310024213,
|
|
"loss": 5.993,
|
|
"mean_token_accuracy": 0.15654735416173934,
|
|
"num_tokens": 3878613.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.126480054855347,
|
|
"epoch": 1.870740941049216,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004913651396901489,
|
|
"loss": 5.9127,
|
|
"mean_token_accuracy": 0.1572210296988487,
|
|
"num_tokens": 3888418.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.141847562789917,
|
|
"epoch": 1.8761492698756084,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000491247062106459,
|
|
"loss": 5.9766,
|
|
"mean_token_accuracy": 0.15648199170827864,
|
|
"num_tokens": 3899359.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.1703675270080565,
|
|
"epoch": 1.8815575987020012,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004911281986804455,
|
|
"loss": 6.0667,
|
|
"mean_token_accuracy": 0.1507619395852089,
|
|
"num_tokens": 3911568.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.130995368957519,
|
|
"epoch": 1.8869659275283936,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004910085498440587,
|
|
"loss": 6.0184,
|
|
"mean_token_accuracy": 0.1529741957783699,
|
|
"num_tokens": 3921597.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.116496229171753,
|
|
"epoch": 1.8923742563547865,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004908881160321025,
|
|
"loss": 5.9068,
|
|
"mean_token_accuracy": 0.15113782584667207,
|
|
"num_tokens": 3933124.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.1459725856781,
|
|
"epoch": 1.897782585181179,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004907668976822338,
|
|
"loss": 5.9208,
|
|
"mean_token_accuracy": 0.16208045780658722,
|
|
"num_tokens": 3944388.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.111624908447266,
|
|
"epoch": 1.9031909140075718,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004906448952349603,
|
|
"loss": 5.8962,
|
|
"mean_token_accuracy": 0.1609338641166687,
|
|
"num_tokens": 3955024.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.990694236755371,
|
|
"epoch": 1.9085992428339642,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004905221091336392,
|
|
"loss": 5.8861,
|
|
"mean_token_accuracy": 0.16136262863874434,
|
|
"num_tokens": 3966029.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.073043870925903,
|
|
"epoch": 1.914007571660357,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004903985398244753,
|
|
"loss": 5.8478,
|
|
"mean_token_accuracy": 0.15888635516166688,
|
|
"num_tokens": 3976066.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.208917856216431,
|
|
"epoch": 1.9194159004867495,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004902741877565201,
|
|
"loss": 6.0554,
|
|
"mean_token_accuracy": 0.15132782012224197,
|
|
"num_tokens": 3986467.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.163874626159668,
|
|
"epoch": 1.9248242293131423,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000490149053381669,
|
|
"loss": 6.0127,
|
|
"mean_token_accuracy": 0.15224549025297165,
|
|
"num_tokens": 3998208.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.168772029876709,
|
|
"epoch": 1.9302325581395348,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004900231371546608,
|
|
"loss": 5.9849,
|
|
"mean_token_accuracy": 0.15751034170389175,
|
|
"num_tokens": 4008202.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.98819580078125,
|
|
"epoch": 1.9356408869659276,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004898964395330753,
|
|
"loss": 5.7913,
|
|
"mean_token_accuracy": 0.17117448449134826,
|
|
"num_tokens": 4019256.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.018447399139404,
|
|
"epoch": 1.94104921579232,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004897689609773318,
|
|
"loss": 5.8344,
|
|
"mean_token_accuracy": 0.1589585065841675,
|
|
"num_tokens": 4029738.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.147535371780395,
|
|
"epoch": 1.946457544618713,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004896407019506879,
|
|
"loss": 6.0712,
|
|
"mean_token_accuracy": 0.14708023965358735,
|
|
"num_tokens": 4042277.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.114105272293091,
|
|
"epoch": 1.9518658734451053,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004895116629192371,
|
|
"loss": 5.935,
|
|
"mean_token_accuracy": 0.16128160059452057,
|
|
"num_tokens": 4054019.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.060033559799194,
|
|
"epoch": 1.9572742022714982,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004893818443519074,
|
|
"loss": 6.0045,
|
|
"mean_token_accuracy": 0.1560356706380844,
|
|
"num_tokens": 4065562.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.209696435928345,
|
|
"epoch": 1.9626825310978906,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00048925124672046,
|
|
"loss": 6.037,
|
|
"mean_token_accuracy": 0.15175911486148835,
|
|
"num_tokens": 4076617.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.065681266784668,
|
|
"epoch": 1.9680908599242835,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004891198704994868,
|
|
"loss": 5.8614,
|
|
"mean_token_accuracy": 0.1639748439192772,
|
|
"num_tokens": 4087191.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.115711641311646,
|
|
"epoch": 1.973499188750676,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004889877161664096,
|
|
"loss": 5.8814,
|
|
"mean_token_accuracy": 0.1596323437988758,
|
|
"num_tokens": 4099150.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.085807132720947,
|
|
"epoch": 1.9789075175770687,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004888547842014771,
|
|
"loss": 5.8895,
|
|
"mean_token_accuracy": 0.1610187590122223,
|
|
"num_tokens": 4110299.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.024379587173462,
|
|
"epoch": 1.9843158464034614,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004887210750877648,
|
|
"loss": 5.9215,
|
|
"mean_token_accuracy": 0.1599341869354248,
|
|
"num_tokens": 4121682.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.091310596466064,
|
|
"epoch": 1.989724175229854,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004885865893111716,
|
|
"loss": 5.8479,
|
|
"mean_token_accuracy": 0.15902262330055236,
|
|
"num_tokens": 4132914.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.070318126678467,
|
|
"epoch": 1.9951325040562466,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004884513273604195,
|
|
"loss": 5.964,
|
|
"mean_token_accuracy": 0.15529086738824843,
|
|
"num_tokens": 4144213.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.109223577711317,
|
|
"epoch": 2.0,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 0.0004883152897270502,
|
|
"loss": 5.9687,
|
|
"mean_token_accuracy": 0.1574669513437483,
|
|
"num_tokens": 4153778.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.02831130027771,
|
|
"epoch": 2.005408328826393,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004881784769054253,
|
|
"loss": 5.633,
|
|
"mean_token_accuracy": 0.16491630375385286,
|
|
"num_tokens": 4165536.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.117471361160279,
|
|
"epoch": 2.0108166576527853,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004880408893927226,
|
|
"loss": 5.7059,
|
|
"mean_token_accuracy": 0.15841729640960694,
|
|
"num_tokens": 4176182.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.918351554870606,
|
|
"epoch": 2.016224986479178,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004879025276889355,
|
|
"loss": 5.6896,
|
|
"mean_token_accuracy": 0.17206783294677735,
|
|
"num_tokens": 4187592.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.023003816604614,
|
|
"epoch": 2.0216333153055706,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00048776339229687077,
|
|
"loss": 5.5093,
|
|
"mean_token_accuracy": 0.17836080193519593,
|
|
"num_tokens": 4198199.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.950564765930176,
|
|
"epoch": 2.0270416441319634,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00048762348372214683,
|
|
"loss": 5.7382,
|
|
"mean_token_accuracy": 0.1662365898489952,
|
|
"num_tokens": 4208116.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.001375579833985,
|
|
"epoch": 2.032449972958356,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00048748280247319165,
|
|
"loss": 5.5155,
|
|
"mean_token_accuracy": 0.1725028410553932,
|
|
"num_tokens": 4218569.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.9467418670654295,
|
|
"epoch": 2.0378583017847487,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004873413490612412,
|
|
"loss": 5.673,
|
|
"mean_token_accuracy": 0.17331435084342955,
|
|
"num_tokens": 4228785.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.982183408737183,
|
|
"epoch": 2.043266630611141,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048719912400033757,
|
|
"loss": 5.6194,
|
|
"mean_token_accuracy": 0.17110280096530914,
|
|
"num_tokens": 4238814.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.916099309921265,
|
|
"epoch": 2.048674959437534,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000487056127807327,
|
|
"loss": 5.6005,
|
|
"mean_token_accuracy": 0.16790374964475632,
|
|
"num_tokens": 4250544.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.986256456375122,
|
|
"epoch": 2.0540832882639264,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048691236100185795,
|
|
"loss": 5.7198,
|
|
"mean_token_accuracy": 0.1698155000805855,
|
|
"num_tokens": 4261806.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.111843156814575,
|
|
"epoch": 2.0594916170903192,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004867678241063792,
|
|
"loss": 5.6801,
|
|
"mean_token_accuracy": 0.17007888108491898,
|
|
"num_tokens": 4272472.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.9632566452026365,
|
|
"epoch": 2.0648999459167117,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00048662251764613833,
|
|
"loss": 5.6451,
|
|
"mean_token_accuracy": 0.16867664754390715,
|
|
"num_tokens": 4283758.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.963249921798706,
|
|
"epoch": 2.0703082747431045,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004864764421491793,
|
|
"loss": 5.7658,
|
|
"mean_token_accuracy": 0.1674502156674862,
|
|
"num_tokens": 4295930.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.111072015762329,
|
|
"epoch": 2.075716603569497,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004863295981463408,
|
|
"loss": 5.6949,
|
|
"mean_token_accuracy": 0.1755090221762657,
|
|
"num_tokens": 4306711.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.9135904788970945,
|
|
"epoch": 2.08112493239589,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004861819861712543,
|
|
"loss": 5.6592,
|
|
"mean_token_accuracy": 0.1705186188220978,
|
|
"num_tokens": 4317694.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.058824586868286,
|
|
"epoch": 2.086533261222282,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048603360676034203,
|
|
"loss": 5.7448,
|
|
"mean_token_accuracy": 0.16260779947042464,
|
|
"num_tokens": 4329552.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.980276346206665,
|
|
"epoch": 2.091941590048675,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00048588446045281505,
|
|
"loss": 5.6812,
|
|
"mean_token_accuracy": 0.16590352952480317,
|
|
"num_tokens": 4340479.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.880865144729614,
|
|
"epoch": 2.0973499188750675,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004857345477906714,
|
|
"loss": 5.5449,
|
|
"mean_token_accuracy": 0.17517983317375183,
|
|
"num_tokens": 4351147.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.872055387496948,
|
|
"epoch": 2.1027582477014604,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00048558386931869417,
|
|
"loss": 5.5464,
|
|
"mean_token_accuracy": 0.1755799114704132,
|
|
"num_tokens": 4362962.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.981302118301391,
|
|
"epoch": 2.1081665765278528,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004854324255844492,
|
|
"loss": 5.6735,
|
|
"mean_token_accuracy": 0.17112574130296707,
|
|
"num_tokens": 4373723.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.885309553146362,
|
|
"epoch": 2.1135749053542456,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048528021713828336,
|
|
"loss": 5.6396,
|
|
"mean_token_accuracy": 0.17210001200437547,
|
|
"num_tokens": 4383434.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.941852664947509,
|
|
"epoch": 2.118983234180638,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004851272445333225,
|
|
"loss": 5.592,
|
|
"mean_token_accuracy": 0.17496410757303238,
|
|
"num_tokens": 4394847.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.967831993103028,
|
|
"epoch": 2.124391563007031,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00048497350832546937,
|
|
"loss": 5.7211,
|
|
"mean_token_accuracy": 0.16780611425638198,
|
|
"num_tokens": 4406367.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.0558301448822025,
|
|
"epoch": 2.1297998918334233,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004848190090734018,
|
|
"loss": 5.6956,
|
|
"mean_token_accuracy": 0.17166045010089875,
|
|
"num_tokens": 4417636.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.9655835151672365,
|
|
"epoch": 2.135208220659816,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00048466374733857035,
|
|
"loss": 5.7189,
|
|
"mean_token_accuracy": 0.1743161901831627,
|
|
"num_tokens": 4428473.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.929123258590698,
|
|
"epoch": 2.1406165494862086,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004845077236851966,
|
|
"loss": 5.6357,
|
|
"mean_token_accuracy": 0.1691516235470772,
|
|
"num_tokens": 4439780.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.883416795730591,
|
|
"epoch": 2.1460248783126015,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00048435093868027076,
|
|
"loss": 5.5858,
|
|
"mean_token_accuracy": 0.18319243639707566,
|
|
"num_tokens": 4450929.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.9608971118927006,
|
|
"epoch": 2.151433207138994,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048419339289354995,
|
|
"loss": 5.6229,
|
|
"mean_token_accuracy": 0.17188669890165328,
|
|
"num_tokens": 4462967.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.974660348892212,
|
|
"epoch": 2.1568415359653867,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000484035086897556,
|
|
"loss": 5.7109,
|
|
"mean_token_accuracy": 0.17126106917858125,
|
|
"num_tokens": 4473970.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.982066917419433,
|
|
"epoch": 2.162249864791779,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004838760212675732,
|
|
"loss": 5.7023,
|
|
"mean_token_accuracy": 0.1717233881354332,
|
|
"num_tokens": 4485660.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 2.162249864791779,
|
|
"eval_entropy": 5.949298878149553,
|
|
"eval_loss": 6.125514507293701,
|
|
"eval_mean_token_accuracy": 0.1598722877150232,
|
|
"eval_num_tokens": 4485660.0,
|
|
"eval_runtime": 4.4626,
|
|
"eval_samples_per_second": 787.66,
|
|
"eval_steps_per_second": 98.598,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.982108306884766,
|
|
"epoch": 2.167658193618172,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004837161965816464,
|
|
"loss": 5.6775,
|
|
"mean_token_accuracy": 0.1760273039340973,
|
|
"num_tokens": 4497261.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.929099082946777,
|
|
"epoch": 2.1730665224445644,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00048355561342057886,
|
|
"loss": 5.5625,
|
|
"mean_token_accuracy": 0.1747855231165886,
|
|
"num_tokens": 4508120.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.995627117156983,
|
|
"epoch": 2.1784748512709573,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048339427236793024,
|
|
"loss": 5.6588,
|
|
"mean_token_accuracy": 0.17412116080522538,
|
|
"num_tokens": 4519286.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.992680311203003,
|
|
"epoch": 2.1838831800973497,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004832321740100143,
|
|
"loss": 5.7212,
|
|
"mean_token_accuracy": 0.16483382433652877,
|
|
"num_tokens": 4531912.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.025650262832642,
|
|
"epoch": 2.1892915089237426,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048306931893589666,
|
|
"loss": 5.7478,
|
|
"mean_token_accuracy": 0.17140780687332152,
|
|
"num_tokens": 4544370.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.91739673614502,
|
|
"epoch": 2.194699837750135,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00048290570773739327,
|
|
"loss": 5.6842,
|
|
"mean_token_accuracy": 0.16955092698335647,
|
|
"num_tokens": 4556013.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.887455940246582,
|
|
"epoch": 2.200108166576528,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048274134100906747,
|
|
"loss": 5.6377,
|
|
"mean_token_accuracy": 0.17268437147140503,
|
|
"num_tokens": 4566200.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.9667966842651365,
|
|
"epoch": 2.2055164954029207,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00048257621934822835,
|
|
"loss": 5.6777,
|
|
"mean_token_accuracy": 0.17119603008031845,
|
|
"num_tokens": 4576899.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.065783500671387,
|
|
"epoch": 2.210924824229313,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004824103433549285,
|
|
"loss": 5.6817,
|
|
"mean_token_accuracy": 0.16497973054647447,
|
|
"num_tokens": 4588983.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.92646484375,
|
|
"epoch": 2.2163331530557056,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004822437136319616,
|
|
"loss": 5.6565,
|
|
"mean_token_accuracy": 0.1765311762690544,
|
|
"num_tokens": 4599758.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.91777982711792,
|
|
"epoch": 2.2217414818820984,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004820763307848604,
|
|
"loss": 5.6875,
|
|
"mean_token_accuracy": 0.1702496513724327,
|
|
"num_tokens": 4610927.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.936834144592285,
|
|
"epoch": 2.2271498107084913,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048190819542189456,
|
|
"loss": 5.7031,
|
|
"mean_token_accuracy": 0.16890842765569686,
|
|
"num_tokens": 4623719.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.967677927017212,
|
|
"epoch": 2.2325581395348837,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048173930815406847,
|
|
"loss": 5.7741,
|
|
"mean_token_accuracy": 0.16684289872646332,
|
|
"num_tokens": 4635389.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.9489861011505125,
|
|
"epoch": 2.2379664683612766,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004815696695951187,
|
|
"loss": 5.7082,
|
|
"mean_token_accuracy": 0.17083445340394973,
|
|
"num_tokens": 4645093.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.914329242706299,
|
|
"epoch": 2.243374797187669,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004813992803615123,
|
|
"loss": 5.6878,
|
|
"mean_token_accuracy": 0.16323266327381133,
|
|
"num_tokens": 4656490.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.025431823730469,
|
|
"epoch": 2.248783126014062,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048122814107244414,
|
|
"loss": 5.7137,
|
|
"mean_token_accuracy": 0.1705820456147194,
|
|
"num_tokens": 4667603.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.911839437484741,
|
|
"epoch": 2.2541914548404542,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00048105625234983466,
|
|
"loss": 5.6472,
|
|
"mean_token_accuracy": 0.16425488889217377,
|
|
"num_tokens": 4678669.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.847280263900757,
|
|
"epoch": 2.259599783666847,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048088361481832807,
|
|
"loss": 5.4507,
|
|
"mean_token_accuracy": 0.18409539759159088,
|
|
"num_tokens": 4689757.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.854237985610962,
|
|
"epoch": 2.2650081124932395,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048071022910528934,
|
|
"loss": 5.6433,
|
|
"mean_token_accuracy": 0.1749528542160988,
|
|
"num_tokens": 4699332.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.975234699249268,
|
|
"epoch": 2.2704164413196324,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048053609584080274,
|
|
"loss": 5.717,
|
|
"mean_token_accuracy": 0.17401882410049438,
|
|
"num_tokens": 4709485.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.873265790939331,
|
|
"epoch": 2.275824770146025,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00048036121565766876,
|
|
"loss": 5.6389,
|
|
"mean_token_accuracy": 0.17150153815746308,
|
|
"num_tokens": 4720935.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.8582350730896,
|
|
"epoch": 2.2812330989724177,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004801855891914026,
|
|
"loss": 5.5948,
|
|
"mean_token_accuracy": 0.16869606524705888,
|
|
"num_tokens": 4731219.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.979439306259155,
|
|
"epoch": 2.28664142779881,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048000921708023117,
|
|
"loss": 5.7208,
|
|
"mean_token_accuracy": 0.1678316444158554,
|
|
"num_tokens": 4742637.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.920532274246216,
|
|
"epoch": 2.292049756625203,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004798320999650911,
|
|
"loss": 5.6241,
|
|
"mean_token_accuracy": 0.1700661599636078,
|
|
"num_tokens": 4753341.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.088420104980469,
|
|
"epoch": 2.2974580854515954,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00047965423848962636,
|
|
"loss": 5.7706,
|
|
"mean_token_accuracy": 0.17284098863601685,
|
|
"num_tokens": 4765339.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.93036732673645,
|
|
"epoch": 2.302866414277988,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000479475633300186,
|
|
"loss": 5.6271,
|
|
"mean_token_accuracy": 0.17804240435361862,
|
|
"num_tokens": 4776149.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.847447538375855,
|
|
"epoch": 2.3082747431043806,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00047929628504582173,
|
|
"loss": 5.6529,
|
|
"mean_token_accuracy": 0.1765597864985466,
|
|
"num_tokens": 4788120.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.995581007003784,
|
|
"epoch": 2.3136830719307735,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004791161943782856,
|
|
"loss": 5.7351,
|
|
"mean_token_accuracy": 0.1670808896422386,
|
|
"num_tokens": 4799922.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.819302654266357,
|
|
"epoch": 2.319091400757166,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00047893536195202726,
|
|
"loss": 5.6632,
|
|
"mean_token_accuracy": 0.16916958391666412,
|
|
"num_tokens": 4810501.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.933983278274536,
|
|
"epoch": 2.3244997295835588,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004787537884241923,
|
|
"loss": 5.6107,
|
|
"mean_token_accuracy": 0.17808097451925278,
|
|
"num_tokens": 4820919.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.9785120487213135,
|
|
"epoch": 2.329908058409951,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00047857147445461943,
|
|
"loss": 5.711,
|
|
"mean_token_accuracy": 0.1713182508945465,
|
|
"num_tokens": 4833298.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.829991483688355,
|
|
"epoch": 2.335316387236344,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004783884207058379,
|
|
"loss": 5.5811,
|
|
"mean_token_accuracy": 0.17137516885995865,
|
|
"num_tokens": 4845199.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.845594596862793,
|
|
"epoch": 2.3407247160627365,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047820462784306555,
|
|
"loss": 5.6374,
|
|
"mean_token_accuracy": 0.172796031832695,
|
|
"num_tokens": 4856549.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.947533798217774,
|
|
"epoch": 2.3461330448891293,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000478020096534206,
|
|
"loss": 5.6901,
|
|
"mean_token_accuracy": 0.16817506551742553,
|
|
"num_tokens": 4868446.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.981800317764282,
|
|
"epoch": 2.3515413737155217,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00047783482744984644,
|
|
"loss": 5.7254,
|
|
"mean_token_accuracy": 0.16550086438655853,
|
|
"num_tokens": 4879047.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.911879682540894,
|
|
"epoch": 2.3569497025419146,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00047764882126325527,
|
|
"loss": 5.6327,
|
|
"mean_token_accuracy": 0.16818329840898513,
|
|
"num_tokens": 4891591.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.880240488052368,
|
|
"epoch": 2.362358031368307,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004774620786503793,
|
|
"loss": 5.6431,
|
|
"mean_token_accuracy": 0.17590303868055343,
|
|
"num_tokens": 4902710.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.984390211105347,
|
|
"epoch": 2.3677663601947,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004772746002898417,
|
|
"loss": 5.7205,
|
|
"mean_token_accuracy": 0.1710437521338463,
|
|
"num_tokens": 4914668.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.01737380027771,
|
|
"epoch": 2.3731746890210923,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047708638686293903,
|
|
"loss": 5.8558,
|
|
"mean_token_accuracy": 0.16607069075107575,
|
|
"num_tokens": 4926684.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.030994319915772,
|
|
"epoch": 2.378583017847485,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00047689743905363963,
|
|
"loss": 5.7237,
|
|
"mean_token_accuracy": 0.16687667220830918,
|
|
"num_tokens": 4938577.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.798695087432861,
|
|
"epoch": 2.383991346673878,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004767077575485801,
|
|
"loss": 5.5823,
|
|
"mean_token_accuracy": 0.176284322142601,
|
|
"num_tokens": 4949070.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.948533344268799,
|
|
"epoch": 2.3893996755002704,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004765173430370635,
|
|
"loss": 5.6508,
|
|
"mean_token_accuracy": 0.1710529163479805,
|
|
"num_tokens": 4959558.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.829991817474365,
|
|
"epoch": 2.394808004326663,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004763261962110567,
|
|
"loss": 5.522,
|
|
"mean_token_accuracy": 0.17642372101545334,
|
|
"num_tokens": 4970310.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.794989013671875,
|
|
"epoch": 2.4002163331530557,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004761343177651875,
|
|
"loss": 5.6002,
|
|
"mean_token_accuracy": 0.17396020144224167,
|
|
"num_tokens": 4982300.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.873043870925903,
|
|
"epoch": 2.4056246619794486,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004759417083967428,
|
|
"loss": 5.628,
|
|
"mean_token_accuracy": 0.17426924854516984,
|
|
"num_tokens": 4994425.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.856123733520508,
|
|
"epoch": 2.411032990805841,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00047574836880566544,
|
|
"loss": 5.6198,
|
|
"mean_token_accuracy": 0.17820293754339217,
|
|
"num_tokens": 5005872.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.90379900932312,
|
|
"epoch": 2.4164413196322334,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004755542996945519,
|
|
"loss": 5.67,
|
|
"mean_token_accuracy": 0.17322854548692704,
|
|
"num_tokens": 5016903.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.953499126434326,
|
|
"epoch": 2.4218496484586263,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00047535950176864976,
|
|
"loss": 5.7316,
|
|
"mean_token_accuracy": 0.17420125156641006,
|
|
"num_tokens": 5028407.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.972252464294433,
|
|
"epoch": 2.427257977285019,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00047516397573585525,
|
|
"loss": 5.7123,
|
|
"mean_token_accuracy": 0.16774825155735015,
|
|
"num_tokens": 5039693.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.947587585449218,
|
|
"epoch": 2.4326663061114115,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047496772230671034,
|
|
"loss": 5.6617,
|
|
"mean_token_accuracy": 0.17488781958818436,
|
|
"num_tokens": 5051279.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.891977834701538,
|
|
"epoch": 2.4380746349378044,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004747707421944004,
|
|
"loss": 5.6058,
|
|
"mean_token_accuracy": 0.16987852603197098,
|
|
"num_tokens": 5062698.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.910605096817017,
|
|
"epoch": 2.443482963764197,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004745730361147517,
|
|
"loss": 5.607,
|
|
"mean_token_accuracy": 0.17366028428077698,
|
|
"num_tokens": 5074927.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.872031497955322,
|
|
"epoch": 2.4488912925905897,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004743746047862286,
|
|
"loss": 5.7347,
|
|
"mean_token_accuracy": 0.16416713446378708,
|
|
"num_tokens": 5086298.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.988913822174072,
|
|
"epoch": 2.454299621416982,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004741754489299309,
|
|
"loss": 5.6607,
|
|
"mean_token_accuracy": 0.1701670527458191,
|
|
"num_tokens": 5098161.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.839615058898926,
|
|
"epoch": 2.459707950243375,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00047397556926959155,
|
|
"loss": 5.663,
|
|
"mean_token_accuracy": 0.17833108305931092,
|
|
"num_tokens": 5110743.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 6.012667226791382,
|
|
"epoch": 2.4651162790697674,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00047377496653157374,
|
|
"loss": 5.7117,
|
|
"mean_token_accuracy": 0.17241982966661454,
|
|
"num_tokens": 5121838.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.828307723999023,
|
|
"epoch": 2.4705246078961602,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047357364144486806,
|
|
"loss": 5.6357,
|
|
"mean_token_accuracy": 0.17124500423669814,
|
|
"num_tokens": 5135047.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.8528220653533936,
|
|
"epoch": 2.4759329367225527,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00047337159474109057,
|
|
"loss": 5.6099,
|
|
"mean_token_accuracy": 0.16912028789520264,
|
|
"num_tokens": 5146329.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.88716950416565,
|
|
"epoch": 2.4813412655489455,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047316882715447934,
|
|
"loss": 5.6853,
|
|
"mean_token_accuracy": 0.17450718134641646,
|
|
"num_tokens": 5157511.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.836351585388184,
|
|
"epoch": 2.486749594375338,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004729653394218922,
|
|
"loss": 5.6611,
|
|
"mean_token_accuracy": 0.17262056916952134,
|
|
"num_tokens": 5168456.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.032444190979004,
|
|
"epoch": 2.492157923201731,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00047276113228280427,
|
|
"loss": 5.7015,
|
|
"mean_token_accuracy": 0.173800228536129,
|
|
"num_tokens": 5179294.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.8588666915893555,
|
|
"epoch": 2.497566252028123,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004725562064793045,
|
|
"loss": 5.6664,
|
|
"mean_token_accuracy": 0.17490550577640535,
|
|
"num_tokens": 5190195.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.79622712135315,
|
|
"epoch": 2.502974580854516,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047235056275609397,
|
|
"loss": 5.5371,
|
|
"mean_token_accuracy": 0.17794119864702224,
|
|
"num_tokens": 5201289.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.890370321273804,
|
|
"epoch": 2.5083829096809085,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004721442018604822,
|
|
"loss": 5.5715,
|
|
"mean_token_accuracy": 0.18150455504655838,
|
|
"num_tokens": 5212295.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.755840110778808,
|
|
"epoch": 2.5137912385073014,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004719371245423853,
|
|
"loss": 5.5382,
|
|
"mean_token_accuracy": 0.17813810110092163,
|
|
"num_tokens": 5223144.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.81374192237854,
|
|
"epoch": 2.5191995673336938,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047172933155432273,
|
|
"loss": 5.5595,
|
|
"mean_token_accuracy": 0.17341333329677583,
|
|
"num_tokens": 5232840.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.873983573913574,
|
|
"epoch": 2.5246078961600866,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047152082365141466,
|
|
"loss": 5.6725,
|
|
"mean_token_accuracy": 0.17399543821811675,
|
|
"num_tokens": 5243796.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.965564870834351,
|
|
"epoch": 2.530016224986479,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004713116015913794,
|
|
"loss": 5.7633,
|
|
"mean_token_accuracy": 0.17094511389732361,
|
|
"num_tokens": 5255990.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.943576765060425,
|
|
"epoch": 2.535424553812872,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00047110166613453016,
|
|
"loss": 5.7045,
|
|
"mean_token_accuracy": 0.1707229271531105,
|
|
"num_tokens": 5268134.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.747303247451782,
|
|
"epoch": 2.5408328826392643,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00047089101804377305,
|
|
"loss": 5.5371,
|
|
"mean_token_accuracy": 0.1794818714261055,
|
|
"num_tokens": 5278662.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.9200972557067875,
|
|
"epoch": 2.546241211465657,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004706796580846037,
|
|
"loss": 5.6197,
|
|
"mean_token_accuracy": 0.1728336125612259,
|
|
"num_tokens": 5288382.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.868596982955933,
|
|
"epoch": 2.5516495402920496,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00047046758702510465,
|
|
"loss": 5.6541,
|
|
"mean_token_accuracy": 0.17161357700824736,
|
|
"num_tokens": 5299769.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.949838733673095,
|
|
"epoch": 2.5570578691184425,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00047025480563594264,
|
|
"loss": 5.6247,
|
|
"mean_token_accuracy": 0.1787155970931053,
|
|
"num_tokens": 5310662.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.8381787776947025,
|
|
"epoch": 2.5624661979448353,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00047004131469036576,
|
|
"loss": 5.5911,
|
|
"mean_token_accuracy": 0.17437346577644347,
|
|
"num_tokens": 5321785.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.884653806686401,
|
|
"epoch": 2.5678745267712277,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00046982711496420057,
|
|
"loss": 5.6297,
|
|
"mean_token_accuracy": 0.16985892057418822,
|
|
"num_tokens": 5333557.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.9346517562866214,
|
|
"epoch": 2.57328285559762,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00046961220723584925,
|
|
"loss": 5.68,
|
|
"mean_token_accuracy": 0.1674743965268135,
|
|
"num_tokens": 5344767.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.871582412719727,
|
|
"epoch": 2.578691184424013,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00046939659228628725,
|
|
"loss": 5.6006,
|
|
"mean_token_accuracy": 0.17152523398399352,
|
|
"num_tokens": 5356673.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.839425945281983,
|
|
"epoch": 2.584099513250406,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046918027089905956,
|
|
"loss": 5.715,
|
|
"mean_token_accuracy": 0.16710921823978425,
|
|
"num_tokens": 5367296.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.829170274734497,
|
|
"epoch": 2.5895078420767983,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00046896324386027873,
|
|
"loss": 5.5849,
|
|
"mean_token_accuracy": 0.17569882422685623,
|
|
"num_tokens": 5378081.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.930244874954224,
|
|
"epoch": 2.5949161709031907,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00046874551195862145,
|
|
"loss": 5.6729,
|
|
"mean_token_accuracy": 0.16581428945064544,
|
|
"num_tokens": 5390244.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.78630428314209,
|
|
"epoch": 2.6003244997295836,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000468527075985326,
|
|
"loss": 5.5476,
|
|
"mean_token_accuracy": 0.17874917536973953,
|
|
"num_tokens": 5400741.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.890814304351807,
|
|
"epoch": 2.6057328285559764,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00046830793673418924,
|
|
"loss": 5.7354,
|
|
"mean_token_accuracy": 0.17858280539512633,
|
|
"num_tokens": 5411440.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.874212265014648,
|
|
"epoch": 2.611141157382369,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046808809500156355,
|
|
"loss": 5.5845,
|
|
"mean_token_accuracy": 0.16877447217702865,
|
|
"num_tokens": 5422472.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.8111409664154055,
|
|
"epoch": 2.6165494862087613,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004678675515863545,
|
|
"loss": 5.6691,
|
|
"mean_token_accuracy": 0.17181654423475265,
|
|
"num_tokens": 5433958.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.886745548248291,
|
|
"epoch": 2.621957815035154,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004676463072900173,
|
|
"loss": 5.567,
|
|
"mean_token_accuracy": 0.18356151431798934,
|
|
"num_tokens": 5444343.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.813747978210449,
|
|
"epoch": 2.627366143861547,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00046742436291655415,
|
|
"loss": 5.5834,
|
|
"mean_token_accuracy": 0.17102641314268113,
|
|
"num_tokens": 5456260.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.866550350189209,
|
|
"epoch": 2.6327744726879394,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00046720171927251155,
|
|
"loss": 5.6719,
|
|
"mean_token_accuracy": 0.17833822816610337,
|
|
"num_tokens": 5468281.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.70713701248169,
|
|
"epoch": 2.638182801514332,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004669783771669769,
|
|
"loss": 5.5304,
|
|
"mean_token_accuracy": 0.18463329076766968,
|
|
"num_tokens": 5480068.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.779549694061279,
|
|
"epoch": 2.6435911303407247,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000466754337411576,
|
|
"loss": 5.5239,
|
|
"mean_token_accuracy": 0.17508679926395415,
|
|
"num_tokens": 5490932.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.7630411148071286,
|
|
"epoch": 2.6489994591671175,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00046652960082046985,
|
|
"loss": 5.6117,
|
|
"mean_token_accuracy": 0.17406541109085083,
|
|
"num_tokens": 5501787.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.9030004978179935,
|
|
"epoch": 2.65440778799351,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046630416821035164,
|
|
"loss": 5.6031,
|
|
"mean_token_accuracy": 0.1745482236146927,
|
|
"num_tokens": 5513297.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.8007895946502686,
|
|
"epoch": 2.6598161168199024,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004660780404004441,
|
|
"loss": 5.5828,
|
|
"mean_token_accuracy": 0.1781241253018379,
|
|
"num_tokens": 5523797.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.776331567764283,
|
|
"epoch": 2.6652244456462952,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00046585121821249623,
|
|
"loss": 5.6111,
|
|
"mean_token_accuracy": 0.18064478039741516,
|
|
"num_tokens": 5535808.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.948901510238647,
|
|
"epoch": 2.670632774472688,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.00046562370247078025,
|
|
"loss": 5.6991,
|
|
"mean_token_accuracy": 0.16583070158958435,
|
|
"num_tokens": 5548179.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.937611246109009,
|
|
"epoch": 2.6760411032990805,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000465395494002089,
|
|
"loss": 5.7155,
|
|
"mean_token_accuracy": 0.17039823085069655,
|
|
"num_tokens": 5559144.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.839483022689819,
|
|
"epoch": 2.6814494321254734,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004651665936357325,
|
|
"loss": 5.611,
|
|
"mean_token_accuracy": 0.1743037223815918,
|
|
"num_tokens": 5571103.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.805416870117187,
|
|
"epoch": 2.686857760951866,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004649370022035354,
|
|
"loss": 5.5366,
|
|
"mean_token_accuracy": 0.16802931427955628,
|
|
"num_tokens": 5583239.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.797738885879516,
|
|
"epoch": 2.6922660897782587,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004647067205398332,
|
|
"loss": 5.4966,
|
|
"mean_token_accuracy": 0.17762224674224852,
|
|
"num_tokens": 5593602.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.891375541687012,
|
|
"epoch": 2.697674418604651,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00046447574948147027,
|
|
"loss": 5.6704,
|
|
"mean_token_accuracy": 0.17071135342121124,
|
|
"num_tokens": 5604696.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.800823068618774,
|
|
"epoch": 2.703082747431044,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00046424408986779587,
|
|
"loss": 5.6439,
|
|
"mean_token_accuracy": 0.17097456455230714,
|
|
"num_tokens": 5615668.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.703082747431044,
|
|
"eval_entropy": 5.791831820661371,
|
|
"eval_loss": 6.004801273345947,
|
|
"eval_mean_token_accuracy": 0.1671279453079809,
|
|
"eval_num_tokens": 5615668.0,
|
|
"eval_runtime": 4.462,
|
|
"eval_samples_per_second": 787.757,
|
|
"eval_steps_per_second": 98.61,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.904642820358276,
|
|
"epoch": 2.7084910762574363,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004640117425406615,
|
|
"loss": 5.5984,
|
|
"mean_token_accuracy": 0.17170237898826599,
|
|
"num_tokens": 5627344.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.793451404571533,
|
|
"epoch": 2.713899405083829,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000463778708344418,
|
|
"loss": 5.6224,
|
|
"mean_token_accuracy": 0.17248727530241012,
|
|
"num_tokens": 5637472.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.821416330337525,
|
|
"epoch": 2.7193077339102216,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046354498812591207,
|
|
"loss": 5.5978,
|
|
"mean_token_accuracy": 0.18123650401830674,
|
|
"num_tokens": 5647766.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.773206901550293,
|
|
"epoch": 2.7247160627366145,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00046331058273448364,
|
|
"loss": 5.526,
|
|
"mean_token_accuracy": 0.17756171226501466,
|
|
"num_tokens": 5658343.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.798688793182373,
|
|
"epoch": 2.730124391563007,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004630754930219624,
|
|
"loss": 5.5224,
|
|
"mean_token_accuracy": 0.1786252036690712,
|
|
"num_tokens": 5669640.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.7096422672271725,
|
|
"epoch": 2.7355327203893998,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004628397198426648,
|
|
"loss": 5.515,
|
|
"mean_token_accuracy": 0.16967434287071229,
|
|
"num_tokens": 5682421.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.780844831466675,
|
|
"epoch": 2.740941049215792,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00046260326405339126,
|
|
"loss": 5.4915,
|
|
"mean_token_accuracy": 0.18447836190462114,
|
|
"num_tokens": 5692797.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.855740785598755,
|
|
"epoch": 2.746349378042185,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004623661265134226,
|
|
"loss": 5.6774,
|
|
"mean_token_accuracy": 0.17448816150426866,
|
|
"num_tokens": 5702779.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.876797199249268,
|
|
"epoch": 2.7517577068685775,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004621283080845172,
|
|
"loss": 5.6602,
|
|
"mean_token_accuracy": 0.1762245014309883,
|
|
"num_tokens": 5714147.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.767714166641236,
|
|
"epoch": 2.7571660356949703,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.00046188980963090783,
|
|
"loss": 5.4937,
|
|
"mean_token_accuracy": 0.17743666023015975,
|
|
"num_tokens": 5725718.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.75641975402832,
|
|
"epoch": 2.7625743645213627,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00046165063201929825,
|
|
"loss": 5.5637,
|
|
"mean_token_accuracy": 0.18259367793798448,
|
|
"num_tokens": 5737441.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.842602872848511,
|
|
"epoch": 2.7679826933477556,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046141077611886074,
|
|
"loss": 5.6708,
|
|
"mean_token_accuracy": 0.17596835792064666,
|
|
"num_tokens": 5748735.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.970414352416992,
|
|
"epoch": 2.773391022174148,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004611702428012319,
|
|
"loss": 5.7106,
|
|
"mean_token_accuracy": 0.1694123312830925,
|
|
"num_tokens": 5760262.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.795847988128662,
|
|
"epoch": 2.778799351000541,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00046092903294051057,
|
|
"loss": 5.626,
|
|
"mean_token_accuracy": 0.17107931077480315,
|
|
"num_tokens": 5771909.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.695746946334839,
|
|
"epoch": 2.7842076798269337,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00046068714741325385,
|
|
"loss": 5.5086,
|
|
"mean_token_accuracy": 0.17800214141607285,
|
|
"num_tokens": 5782028.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.941395235061646,
|
|
"epoch": 2.789616008653326,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004604445870984743,
|
|
"loss": 5.6755,
|
|
"mean_token_accuracy": 0.17159214466810227,
|
|
"num_tokens": 5793641.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.905786848068237,
|
|
"epoch": 2.7950243374797186,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004602013528776367,
|
|
"loss": 5.6097,
|
|
"mean_token_accuracy": 0.16968271732330323,
|
|
"num_tokens": 5805392.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.822073650360108,
|
|
"epoch": 2.8004326663061114,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004599574456346548,
|
|
"loss": 5.5884,
|
|
"mean_token_accuracy": 0.1765652135014534,
|
|
"num_tokens": 5816362.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.763013935089111,
|
|
"epoch": 2.8058409951325043,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00045971286625588796,
|
|
"loss": 5.5299,
|
|
"mean_token_accuracy": 0.17811470478773117,
|
|
"num_tokens": 5827518.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.906323337554932,
|
|
"epoch": 2.8112493239588967,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004594676156301383,
|
|
"loss": 5.675,
|
|
"mean_token_accuracy": 0.17632317095994948,
|
|
"num_tokens": 5838455.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.801389694213867,
|
|
"epoch": 2.816657652785289,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00045922169464864725,
|
|
"loss": 5.5351,
|
|
"mean_token_accuracy": 0.18242559134960173,
|
|
"num_tokens": 5850065.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.844778633117675,
|
|
"epoch": 2.822065981611682,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000458975104205092,
|
|
"loss": 5.6667,
|
|
"mean_token_accuracy": 0.16942819356918334,
|
|
"num_tokens": 5860444.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.93465518951416,
|
|
"epoch": 2.827474310438075,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00045872784519558304,
|
|
"loss": 5.6441,
|
|
"mean_token_accuracy": 0.17248206734657287,
|
|
"num_tokens": 5873984.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.794377946853638,
|
|
"epoch": 2.8328826392644673,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004584799185186599,
|
|
"loss": 5.6636,
|
|
"mean_token_accuracy": 0.17575643211603165,
|
|
"num_tokens": 5885807.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.823089170455932,
|
|
"epoch": 2.8382909680908597,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00045823132507528903,
|
|
"loss": 5.5434,
|
|
"mean_token_accuracy": 0.18048701137304307,
|
|
"num_tokens": 5897122.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.793164682388306,
|
|
"epoch": 2.8436992969172525,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004579820657688593,
|
|
"loss": 5.5584,
|
|
"mean_token_accuracy": 0.17940302193164825,
|
|
"num_tokens": 5909189.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.869535636901856,
|
|
"epoch": 2.8491076257436454,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004577321415051798,
|
|
"loss": 5.6477,
|
|
"mean_token_accuracy": 0.16871020793914795,
|
|
"num_tokens": 5920135.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.850260829925537,
|
|
"epoch": 2.854515954570038,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00045748155319247564,
|
|
"loss": 5.6114,
|
|
"mean_token_accuracy": 0.1819598764181137,
|
|
"num_tokens": 5932279.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.758639717102051,
|
|
"epoch": 2.8599242833964302,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004572303017413856,
|
|
"loss": 5.5358,
|
|
"mean_token_accuracy": 0.17633006423711778,
|
|
"num_tokens": 5942082.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.856132173538208,
|
|
"epoch": 2.865332612222823,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004569783880649578,
|
|
"loss": 5.6053,
|
|
"mean_token_accuracy": 0.17732162177562713,
|
|
"num_tokens": 5952856.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.679220962524414,
|
|
"epoch": 2.870740941049216,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004567258130786474,
|
|
"loss": 5.4865,
|
|
"mean_token_accuracy": 0.18453991413116455,
|
|
"num_tokens": 5963937.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.8489007472991945,
|
|
"epoch": 2.8761492698756084,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045647257770031215,
|
|
"loss": 5.578,
|
|
"mean_token_accuracy": 0.17881677001714708,
|
|
"num_tokens": 5974852.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.7810591697692875,
|
|
"epoch": 2.8815575987020012,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00045621868285021023,
|
|
"loss": 5.5805,
|
|
"mean_token_accuracy": 0.18164530098438264,
|
|
"num_tokens": 5985616.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.8331770420074465,
|
|
"epoch": 2.8869659275283936,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045596412945099615,
|
|
"loss": 5.5497,
|
|
"mean_token_accuracy": 0.18524349927902223,
|
|
"num_tokens": 5997199.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.79961633682251,
|
|
"epoch": 2.8923742563547865,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00045570891842771727,
|
|
"loss": 5.6021,
|
|
"mean_token_accuracy": 0.17862610816955565,
|
|
"num_tokens": 6008654.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.8329901695251465,
|
|
"epoch": 2.897782585181179,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00045545305070781135,
|
|
"loss": 5.6495,
|
|
"mean_token_accuracy": 0.1718444138765335,
|
|
"num_tokens": 6020662.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.8480823040008545,
|
|
"epoch": 2.903190914007572,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045519652722110206,
|
|
"loss": 5.5348,
|
|
"mean_token_accuracy": 0.1957799568772316,
|
|
"num_tokens": 6031793.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.739364099502564,
|
|
"epoch": 2.908599242833964,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045493934889979655,
|
|
"loss": 5.5054,
|
|
"mean_token_accuracy": 0.1793569028377533,
|
|
"num_tokens": 6042181.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.6632387161254885,
|
|
"epoch": 2.914007571660357,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00045468151667848116,
|
|
"loss": 5.4686,
|
|
"mean_token_accuracy": 0.1869698643684387,
|
|
"num_tokens": 6052774.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.888607597351074,
|
|
"epoch": 2.9194159004867495,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000454423031494119,
|
|
"loss": 5.5005,
|
|
"mean_token_accuracy": 0.19342892169952391,
|
|
"num_tokens": 6063273.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.729899120330811,
|
|
"epoch": 2.9248242293131423,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004541638942860458,
|
|
"loss": 5.5,
|
|
"mean_token_accuracy": 0.17978276014328004,
|
|
"num_tokens": 6075121.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.65375394821167,
|
|
"epoch": 2.9302325581395348,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004539041059959667,
|
|
"loss": 5.4741,
|
|
"mean_token_accuracy": 0.18108482509851456,
|
|
"num_tokens": 6085666.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.822653675079346,
|
|
"epoch": 2.9356408869659276,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00045364366756795295,
|
|
"loss": 5.492,
|
|
"mean_token_accuracy": 0.18550640642642974,
|
|
"num_tokens": 6096619.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.756463050842285,
|
|
"epoch": 2.94104921579232,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00045338257994843845,
|
|
"loss": 5.6667,
|
|
"mean_token_accuracy": 0.17645267397165298,
|
|
"num_tokens": 6108800.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.880536460876465,
|
|
"epoch": 2.946457544618713,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004531208440862162,
|
|
"loss": 5.5705,
|
|
"mean_token_accuracy": 0.17858628928661346,
|
|
"num_tokens": 6119789.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.700832796096802,
|
|
"epoch": 2.9518658734451053,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00045285846093243485,
|
|
"loss": 5.5604,
|
|
"mean_token_accuracy": 0.1742287591099739,
|
|
"num_tokens": 6131786.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.91920223236084,
|
|
"epoch": 2.957274202271498,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045259543144059557,
|
|
"loss": 5.6545,
|
|
"mean_token_accuracy": 0.17816822826862336,
|
|
"num_tokens": 6143012.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.800297260284424,
|
|
"epoch": 2.9626825310978906,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004523317565665479,
|
|
"loss": 5.52,
|
|
"mean_token_accuracy": 0.17961374372243882,
|
|
"num_tokens": 6153972.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.798951578140259,
|
|
"epoch": 2.9680908599242835,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00045206743726848716,
|
|
"loss": 5.69,
|
|
"mean_token_accuracy": 0.17037637531757355,
|
|
"num_tokens": 6166202.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.770240068435669,
|
|
"epoch": 2.973499188750676,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004518024745069502,
|
|
"loss": 5.4091,
|
|
"mean_token_accuracy": 0.19165886789560319,
|
|
"num_tokens": 6177253.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.771560144424439,
|
|
"epoch": 2.9789075175770687,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004515368692448123,
|
|
"loss": 5.6833,
|
|
"mean_token_accuracy": 0.16807670146226883,
|
|
"num_tokens": 6188857.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.848395681381225,
|
|
"epoch": 2.9843158464034616,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045127062244728353,
|
|
"loss": 5.6372,
|
|
"mean_token_accuracy": 0.16932384967803954,
|
|
"num_tokens": 6199945.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.822900199890137,
|
|
"epoch": 2.989724175229854,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004510037350819055,
|
|
"loss": 5.6194,
|
|
"mean_token_accuracy": 0.17950630336999893,
|
|
"num_tokens": 6209866.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.8538178443908695,
|
|
"epoch": 2.9951325040562464,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.00045073620811854744,
|
|
"loss": 5.6144,
|
|
"mean_token_accuracy": 0.16657373160123826,
|
|
"num_tokens": 6221636.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.739393658108181,
|
|
"epoch": 3.0,
|
|
"grad_norm": 2.109375,
|
|
"learning_rate": 0.00045046804252940294,
|
|
"loss": 5.4862,
|
|
"mean_token_accuracy": 0.18007244831985897,
|
|
"num_tokens": 6230667.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.780527019500733,
|
|
"epoch": 3.005408328826393,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00045019923928898644,
|
|
"loss": 5.3024,
|
|
"mean_token_accuracy": 0.2013756215572357,
|
|
"num_tokens": 6242242.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.826884126663208,
|
|
"epoch": 3.0108166576527853,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004499297993741294,
|
|
"loss": 5.3392,
|
|
"mean_token_accuracy": 0.1889042779803276,
|
|
"num_tokens": 6253337.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.729806137084961,
|
|
"epoch": 3.016224986479178,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004496597237639773,
|
|
"loss": 5.2188,
|
|
"mean_token_accuracy": 0.19127930104732513,
|
|
"num_tokens": 6264582.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.748434972763062,
|
|
"epoch": 3.0216333153055706,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00044938901343998536,
|
|
"loss": 5.2751,
|
|
"mean_token_accuracy": 0.19821389019489288,
|
|
"num_tokens": 6276172.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.720718765258789,
|
|
"epoch": 3.0270416441319634,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004491176693859156,
|
|
"loss": 5.142,
|
|
"mean_token_accuracy": 0.20424381047487258,
|
|
"num_tokens": 6286038.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.648404550552368,
|
|
"epoch": 3.032449972958356,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004488456925878329,
|
|
"loss": 5.2165,
|
|
"mean_token_accuracy": 0.20037273317575455,
|
|
"num_tokens": 6296546.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.728126144409179,
|
|
"epoch": 3.0378583017847487,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00044857308403410163,
|
|
"loss": 5.3161,
|
|
"mean_token_accuracy": 0.20054817497730254,
|
|
"num_tokens": 6308001.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.733685302734375,
|
|
"epoch": 3.043266630611141,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00044829984471538183,
|
|
"loss": 5.3204,
|
|
"mean_token_accuracy": 0.19108995348215102,
|
|
"num_tokens": 6320896.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.802718687057495,
|
|
"epoch": 3.048674959437534,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004480259756246258,
|
|
"loss": 5.3802,
|
|
"mean_token_accuracy": 0.1868284061551094,
|
|
"num_tokens": 6332710.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.756425619125366,
|
|
"epoch": 3.0540832882639264,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00044775147775707465,
|
|
"loss": 5.248,
|
|
"mean_token_accuracy": 0.19477226287126542,
|
|
"num_tokens": 6343563.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.654908561706543,
|
|
"epoch": 3.0594916170903192,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000447476352110254,
|
|
"loss": 5.3618,
|
|
"mean_token_accuracy": 0.1951124608516693,
|
|
"num_tokens": 6354064.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.683624362945556,
|
|
"epoch": 3.0648999459167117,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004472005996839712,
|
|
"loss": 5.2692,
|
|
"mean_token_accuracy": 0.19286440163850785,
|
|
"num_tokens": 6366551.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.704025650024414,
|
|
"epoch": 3.0703082747431045,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004469242214803112,
|
|
"loss": 5.2239,
|
|
"mean_token_accuracy": 0.19485385715961456,
|
|
"num_tokens": 6377086.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.735103273391724,
|
|
"epoch": 3.075716603569497,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004466472185036331,
|
|
"loss": 5.3236,
|
|
"mean_token_accuracy": 0.1923196718096733,
|
|
"num_tokens": 6388795.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.7453389167785645,
|
|
"epoch": 3.08112493239589,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004463695917605663,
|
|
"loss": 5.3077,
|
|
"mean_token_accuracy": 0.19795713871717452,
|
|
"num_tokens": 6400174.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.672986268997192,
|
|
"epoch": 3.086533261222282,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000446091342260007,
|
|
"loss": 5.2473,
|
|
"mean_token_accuracy": 0.1956798553466797,
|
|
"num_tokens": 6411297.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.67154016494751,
|
|
"epoch": 3.091941590048675,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004458124710131145,
|
|
"loss": 5.2835,
|
|
"mean_token_accuracy": 0.19642237722873687,
|
|
"num_tokens": 6423487.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.832068014144897,
|
|
"epoch": 3.0973499188750675,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00044553297903330766,
|
|
"loss": 5.4855,
|
|
"mean_token_accuracy": 0.1803535133600235,
|
|
"num_tokens": 6435460.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.816472434997559,
|
|
"epoch": 3.1027582477014604,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00044525286733626085,
|
|
"loss": 5.4138,
|
|
"mean_token_accuracy": 0.1969684287905693,
|
|
"num_tokens": 6445664.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.656048011779785,
|
|
"epoch": 3.1081665765278528,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004449721369399008,
|
|
"loss": 5.2626,
|
|
"mean_token_accuracy": 0.19232904464006423,
|
|
"num_tokens": 6457182.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.62199821472168,
|
|
"epoch": 3.1135749053542456,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00044469078886440227,
|
|
"loss": 5.1274,
|
|
"mean_token_accuracy": 0.2075447306036949,
|
|
"num_tokens": 6468378.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.656603050231934,
|
|
"epoch": 3.118983234180638,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00044440882413218484,
|
|
"loss": 5.3532,
|
|
"mean_token_accuracy": 0.1889808937907219,
|
|
"num_tokens": 6479625.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.702569961547852,
|
|
"epoch": 3.124391563007031,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00044412624376790905,
|
|
"loss": 5.2654,
|
|
"mean_token_accuracy": 0.19832505136728287,
|
|
"num_tokens": 6490713.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.584449338912964,
|
|
"epoch": 3.1297998918334233,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004438430487984726,
|
|
"loss": 5.2993,
|
|
"mean_token_accuracy": 0.1844169393181801,
|
|
"num_tokens": 6502161.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.7165929794311525,
|
|
"epoch": 3.135208220659816,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004435592402530066,
|
|
"loss": 5.2222,
|
|
"mean_token_accuracy": 0.19488259106874467,
|
|
"num_tokens": 6513639.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.627776002883911,
|
|
"epoch": 3.1406165494862086,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000443274819162872,
|
|
"loss": 5.2387,
|
|
"mean_token_accuracy": 0.19704062044620513,
|
|
"num_tokens": 6524893.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.557360315322876,
|
|
"epoch": 3.1460248783126015,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004429897865616557,
|
|
"loss": 5.2326,
|
|
"mean_token_accuracy": 0.19775411933660508,
|
|
"num_tokens": 6535797.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.697722244262695,
|
|
"epoch": 3.151433207138994,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004427041434851668,
|
|
"loss": 5.3525,
|
|
"mean_token_accuracy": 0.1884851336479187,
|
|
"num_tokens": 6547117.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.732564306259155,
|
|
"epoch": 3.1568415359653867,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00044241789097143293,
|
|
"loss": 5.376,
|
|
"mean_token_accuracy": 0.19378221035003662,
|
|
"num_tokens": 6558771.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.662858390808106,
|
|
"epoch": 3.162249864791779,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00044213103006069635,
|
|
"loss": 5.3345,
|
|
"mean_token_accuracy": 0.18811459839344025,
|
|
"num_tokens": 6569922.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.733915901184082,
|
|
"epoch": 3.167658193618172,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044184356179541035,
|
|
"loss": 5.3512,
|
|
"mean_token_accuracy": 0.19083004891872407,
|
|
"num_tokens": 6580561.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.639116287231445,
|
|
"epoch": 3.1730665224445644,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004415554872202352,
|
|
"loss": 5.1692,
|
|
"mean_token_accuracy": 0.19876095056533813,
|
|
"num_tokens": 6591636.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.676800918579102,
|
|
"epoch": 3.1784748512709573,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00044126680738203466,
|
|
"loss": 5.3959,
|
|
"mean_token_accuracy": 0.19027754068374633,
|
|
"num_tokens": 6603736.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.810044813156128,
|
|
"epoch": 3.1838831800973497,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004409775233298718,
|
|
"loss": 5.3238,
|
|
"mean_token_accuracy": 0.1942296579480171,
|
|
"num_tokens": 6615415.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.681856250762939,
|
|
"epoch": 3.1892915089237426,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004406876361150056,
|
|
"loss": 5.3563,
|
|
"mean_token_accuracy": 0.18749646693468094,
|
|
"num_tokens": 6625970.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.706464433670044,
|
|
"epoch": 3.194699837750135,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004403971467908869,
|
|
"loss": 5.3102,
|
|
"mean_token_accuracy": 0.1898470938205719,
|
|
"num_tokens": 6636326.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.6406721591949465,
|
|
"epoch": 3.200108166576528,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004401060564131545,
|
|
"loss": 5.3296,
|
|
"mean_token_accuracy": 0.19487152397632598,
|
|
"num_tokens": 6646825.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.702856636047363,
|
|
"epoch": 3.2055164954029207,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004398143660396315,
|
|
"loss": 5.3255,
|
|
"mean_token_accuracy": 0.19223950654268265,
|
|
"num_tokens": 6657748.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.632095003128052,
|
|
"epoch": 3.210924824229313,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004395220767303215,
|
|
"loss": 5.2559,
|
|
"mean_token_accuracy": 0.19639974981546401,
|
|
"num_tokens": 6668422.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.711442565917968,
|
|
"epoch": 3.2163331530557056,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004392291895474045,
|
|
"loss": 5.3215,
|
|
"mean_token_accuracy": 0.18598974943161012,
|
|
"num_tokens": 6678898.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.765573167800904,
|
|
"epoch": 3.2217414818820984,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004389357055552331,
|
|
"loss": 5.3694,
|
|
"mean_token_accuracy": 0.19876599311828613,
|
|
"num_tokens": 6688691.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.610404682159424,
|
|
"epoch": 3.2271498107084913,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00043864162582032897,
|
|
"loss": 5.2485,
|
|
"mean_token_accuracy": 0.2039194718003273,
|
|
"num_tokens": 6700279.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.638318824768066,
|
|
"epoch": 3.2325581395348837,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004383469514113784,
|
|
"loss": 5.1766,
|
|
"mean_token_accuracy": 0.19922815561294555,
|
|
"num_tokens": 6710440.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.617186164855957,
|
|
"epoch": 3.2379664683612766,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004380516833992289,
|
|
"loss": 5.3549,
|
|
"mean_token_accuracy": 0.19068751931190492,
|
|
"num_tokens": 6722347.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.689956188201904,
|
|
"epoch": 3.243374797187669,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000437755822856885,
|
|
"loss": 5.3729,
|
|
"mean_token_accuracy": 0.19313010275363923,
|
|
"num_tokens": 6732936.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 3.243374797187669,
|
|
"eval_entropy": 5.465103129907088,
|
|
"eval_loss": 5.975052356719971,
|
|
"eval_mean_token_accuracy": 0.17053426035933875,
|
|
"eval_num_tokens": 6732936.0,
|
|
"eval_runtime": 7.3488,
|
|
"eval_samples_per_second": 478.311,
|
|
"eval_steps_per_second": 59.874,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.639364004135132,
|
|
"epoch": 3.248783126014062,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004374593708595049,
|
|
"loss": 5.3878,
|
|
"mean_token_accuracy": 0.19232976734638213,
|
|
"num_tokens": 6744552.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.775102853775024,
|
|
"epoch": 3.2541914548404542,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00043716232848439555,
|
|
"loss": 5.3372,
|
|
"mean_token_accuracy": 0.19323543310165406,
|
|
"num_tokens": 6755749.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.666355323791504,
|
|
"epoch": 3.259599783666847,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00043686469681100985,
|
|
"loss": 5.2495,
|
|
"mean_token_accuracy": 0.1992473542690277,
|
|
"num_tokens": 6766119.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.578059434890747,
|
|
"epoch": 3.2650081124932395,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00043656647692094186,
|
|
"loss": 5.3065,
|
|
"mean_token_accuracy": 0.19038026034832,
|
|
"num_tokens": 6776491.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.811544036865234,
|
|
"epoch": 3.2704164413196324,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004362676698979233,
|
|
"loss": 5.4162,
|
|
"mean_token_accuracy": 0.186884406208992,
|
|
"num_tokens": 6788795.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.739661073684692,
|
|
"epoch": 3.275824770146025,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00043596827682781974,
|
|
"loss": 5.4179,
|
|
"mean_token_accuracy": 0.18891228437423707,
|
|
"num_tokens": 6799706.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.688111877441406,
|
|
"epoch": 3.2812330989724177,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004356682987986262,
|
|
"loss": 5.3421,
|
|
"mean_token_accuracy": 0.1869387611746788,
|
|
"num_tokens": 6812454.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.710467100143433,
|
|
"epoch": 3.28664142779881,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004353677369004635,
|
|
"loss": 5.2855,
|
|
"mean_token_accuracy": 0.19317322820425034,
|
|
"num_tokens": 6823950.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.755735731124878,
|
|
"epoch": 3.292049756625203,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00043506659222557426,
|
|
"loss": 5.4924,
|
|
"mean_token_accuracy": 0.18100904673337936,
|
|
"num_tokens": 6835028.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.645953369140625,
|
|
"epoch": 3.2974580854515954,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000434764865868319,
|
|
"loss": 5.3436,
|
|
"mean_token_accuracy": 0.18870311379432678,
|
|
"num_tokens": 6846316.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.645599222183227,
|
|
"epoch": 3.302866414277988,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000434462558925172,
|
|
"loss": 5.2843,
|
|
"mean_token_accuracy": 0.1932820901274681,
|
|
"num_tokens": 6856463.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.71835355758667,
|
|
"epoch": 3.3082747431043806,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00043415967249471734,
|
|
"loss": 5.3088,
|
|
"mean_token_accuracy": 0.19949909895658494,
|
|
"num_tokens": 6867869.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.637714099884033,
|
|
"epoch": 3.3136830719307735,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000433856207677645,
|
|
"loss": 5.2919,
|
|
"mean_token_accuracy": 0.19391226172447204,
|
|
"num_tokens": 6878545.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.625209522247315,
|
|
"epoch": 3.319091400757166,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000433552165576747,
|
|
"loss": 5.3477,
|
|
"mean_token_accuracy": 0.19106703698635102,
|
|
"num_tokens": 6889399.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.745428371429443,
|
|
"epoch": 3.3244997295835588,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00043324754729691275,
|
|
"loss": 5.3326,
|
|
"mean_token_accuracy": 0.19991680681705476,
|
|
"num_tokens": 6900428.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.7930199146270756,
|
|
"epoch": 3.329908058409951,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00043294235394512594,
|
|
"loss": 5.4434,
|
|
"mean_token_accuracy": 0.1890667587518692,
|
|
"num_tokens": 6911112.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.595087623596191,
|
|
"epoch": 3.335316387236344,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004326365866304599,
|
|
"loss": 5.1908,
|
|
"mean_token_accuracy": 0.20456220656633378,
|
|
"num_tokens": 6922383.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.634238338470459,
|
|
"epoch": 3.3407247160627365,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004323302464640738,
|
|
"loss": 5.3946,
|
|
"mean_token_accuracy": 0.18521609604358674,
|
|
"num_tokens": 6935153.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.789434766769409,
|
|
"epoch": 3.3461330448891293,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00043202333455920843,
|
|
"loss": 5.4388,
|
|
"mean_token_accuracy": 0.18687722235918044,
|
|
"num_tokens": 6946814.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.66050910949707,
|
|
"epoch": 3.3515413737155217,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004317158520311824,
|
|
"loss": 5.2761,
|
|
"mean_token_accuracy": 0.19619374573230744,
|
|
"num_tokens": 6957230.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.544762706756591,
|
|
"epoch": 3.3569497025419146,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004314077999973879,
|
|
"loss": 5.2906,
|
|
"mean_token_accuracy": 0.1911259561777115,
|
|
"num_tokens": 6968060.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.64546766281128,
|
|
"epoch": 3.362358031368307,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004310991795772869,
|
|
"loss": 5.2187,
|
|
"mean_token_accuracy": 0.1955324590206146,
|
|
"num_tokens": 6977782.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.7000940322875975,
|
|
"epoch": 3.3677663601947,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004307899918924065,
|
|
"loss": 5.2992,
|
|
"mean_token_accuracy": 0.18789736479520797,
|
|
"num_tokens": 6988252.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.693121099472046,
|
|
"epoch": 3.3731746890210923,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004304802380663356,
|
|
"loss": 5.2959,
|
|
"mean_token_accuracy": 0.1919243276119232,
|
|
"num_tokens": 6998523.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.653825855255127,
|
|
"epoch": 3.378583017847485,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00043016991922472015,
|
|
"loss": 5.3455,
|
|
"mean_token_accuracy": 0.189077490568161,
|
|
"num_tokens": 7010058.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.702905321121216,
|
|
"epoch": 3.383991346673878,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00042985903649525977,
|
|
"loss": 5.4412,
|
|
"mean_token_accuracy": 0.18205490112304687,
|
|
"num_tokens": 7021289.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.780404376983642,
|
|
"epoch": 3.3893996755002704,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00042954759100770275,
|
|
"loss": 5.4647,
|
|
"mean_token_accuracy": 0.1802703931927681,
|
|
"num_tokens": 7032649.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.7563704490661625,
|
|
"epoch": 3.394808004326663,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00042923558389384303,
|
|
"loss": 5.3257,
|
|
"mean_token_accuracy": 0.1927846908569336,
|
|
"num_tokens": 7044120.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.632722425460815,
|
|
"epoch": 3.4002163331530557,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004289230162875149,
|
|
"loss": 5.3204,
|
|
"mean_token_accuracy": 0.1936240464448929,
|
|
"num_tokens": 7054526.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.510391044616699,
|
|
"epoch": 3.4056246619794486,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00042860988932458987,
|
|
"loss": 5.1995,
|
|
"mean_token_accuracy": 0.203300541639328,
|
|
"num_tokens": 7064744.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.7097033023834225,
|
|
"epoch": 3.411032990805841,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004282962041429721,
|
|
"loss": 5.3159,
|
|
"mean_token_accuracy": 0.1921289324760437,
|
|
"num_tokens": 7076360.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.70991702079773,
|
|
"epoch": 3.4164413196322334,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00042798196188259427,
|
|
"loss": 5.3372,
|
|
"mean_token_accuracy": 0.19764234572649003,
|
|
"num_tokens": 7088114.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.668651533126831,
|
|
"epoch": 3.4218496484586263,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004276671636854133,
|
|
"loss": 5.3759,
|
|
"mean_token_accuracy": 0.1824158936738968,
|
|
"num_tokens": 7099125.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.68306450843811,
|
|
"epoch": 3.427257977285019,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00042735181069540677,
|
|
"loss": 5.3238,
|
|
"mean_token_accuracy": 0.19332555532455445,
|
|
"num_tokens": 7109875.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.672682762145996,
|
|
"epoch": 3.4326663061114115,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004270359040585681,
|
|
"loss": 5.3448,
|
|
"mean_token_accuracy": 0.18851479440927504,
|
|
"num_tokens": 7121388.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.65582766532898,
|
|
"epoch": 3.4380746349378044,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00042671944492290277,
|
|
"loss": 5.3777,
|
|
"mean_token_accuracy": 0.18858571499586105,
|
|
"num_tokens": 7134161.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.6925208568573,
|
|
"epoch": 3.443482963764197,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004264024344384239,
|
|
"loss": 5.3949,
|
|
"mean_token_accuracy": 0.18640596717596053,
|
|
"num_tokens": 7146580.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.708399057388306,
|
|
"epoch": 3.4488912925905897,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00042608487375714847,
|
|
"loss": 5.3135,
|
|
"mean_token_accuracy": 0.18479917496442794,
|
|
"num_tokens": 7156969.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.615804290771484,
|
|
"epoch": 3.454299621416982,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00042576676403309265,
|
|
"loss": 5.2637,
|
|
"mean_token_accuracy": 0.2016796961426735,
|
|
"num_tokens": 7168687.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.740988874435425,
|
|
"epoch": 3.459707950243375,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004254481064222679,
|
|
"loss": 5.358,
|
|
"mean_token_accuracy": 0.19213685244321824,
|
|
"num_tokens": 7179996.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.7132988452911375,
|
|
"epoch": 3.4651162790697674,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00042512890208267677,
|
|
"loss": 5.3808,
|
|
"mean_token_accuracy": 0.18283187001943588,
|
|
"num_tokens": 7191214.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.670431280136109,
|
|
"epoch": 3.4705246078961602,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00042480915217430853,
|
|
"loss": 5.3967,
|
|
"mean_token_accuracy": 0.18734519183635712,
|
|
"num_tokens": 7203491.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.751726341247559,
|
|
"epoch": 3.4759329367225527,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00042448885785913507,
|
|
"loss": 5.3398,
|
|
"mean_token_accuracy": 0.1844811514019966,
|
|
"num_tokens": 7215018.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.672880029678344,
|
|
"epoch": 3.4813412655489455,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00042416802030110677,
|
|
"loss": 5.2811,
|
|
"mean_token_accuracy": 0.19865630120038985,
|
|
"num_tokens": 7226209.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.64788851737976,
|
|
"epoch": 3.486749594375338,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00042384664066614787,
|
|
"loss": 5.2942,
|
|
"mean_token_accuracy": 0.19473968148231507,
|
|
"num_tokens": 7236505.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.536511516571045,
|
|
"epoch": 3.492157923201731,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004235247201221528,
|
|
"loss": 5.1997,
|
|
"mean_token_accuracy": 0.19693764597177504,
|
|
"num_tokens": 7247414.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.707625102996826,
|
|
"epoch": 3.497566252028123,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004232022598389816,
|
|
"loss": 5.3992,
|
|
"mean_token_accuracy": 0.18702053725719453,
|
|
"num_tokens": 7258666.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.7081286907196045,
|
|
"epoch": 3.502974580854516,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004228792609884557,
|
|
"loss": 5.3236,
|
|
"mean_token_accuracy": 0.1975746273994446,
|
|
"num_tokens": 7270201.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.697741842269897,
|
|
"epoch": 3.5083829096809085,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00042255572474435364,
|
|
"loss": 5.3782,
|
|
"mean_token_accuracy": 0.19089791625738145,
|
|
"num_tokens": 7282254.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.70852313041687,
|
|
"epoch": 3.5137912385073014,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004222316522824069,
|
|
"loss": 5.4171,
|
|
"mean_token_accuracy": 0.1850898265838623,
|
|
"num_tokens": 7294579.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.617021465301514,
|
|
"epoch": 3.5191995673336938,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004219070447802955,
|
|
"loss": 5.361,
|
|
"mean_token_accuracy": 0.19242722243070604,
|
|
"num_tokens": 7305822.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.750279760360717,
|
|
"epoch": 3.5246078961600866,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00042158190341764384,
|
|
"loss": 5.4009,
|
|
"mean_token_accuracy": 0.18860818743705748,
|
|
"num_tokens": 7317543.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.694949007034301,
|
|
"epoch": 3.530016224986479,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00042125622937601645,
|
|
"loss": 5.3214,
|
|
"mean_token_accuracy": 0.18934821486473083,
|
|
"num_tokens": 7327804.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.627478218078613,
|
|
"epoch": 3.535424553812872,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004209300238389135,
|
|
"loss": 5.2783,
|
|
"mean_token_accuracy": 0.1958963915705681,
|
|
"num_tokens": 7337045.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.6713035106658936,
|
|
"epoch": 3.5408328826392643,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004206032879917667,
|
|
"loss": 5.3705,
|
|
"mean_token_accuracy": 0.1957392007112503,
|
|
"num_tokens": 7348184.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.6514668464660645,
|
|
"epoch": 3.546241211465657,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004202760230219348,
|
|
"loss": 5.2725,
|
|
"mean_token_accuracy": 0.19847316145896912,
|
|
"num_tokens": 7359571.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.6079004287719725,
|
|
"epoch": 3.5516495402920496,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00041994823011869955,
|
|
"loss": 5.2913,
|
|
"mean_token_accuracy": 0.1934184744954109,
|
|
"num_tokens": 7370429.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.633288908004761,
|
|
"epoch": 3.5570578691184425,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00041961991047326116,
|
|
"loss": 5.2465,
|
|
"mean_token_accuracy": 0.19538926035165788,
|
|
"num_tokens": 7381569.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.610870790481568,
|
|
"epoch": 3.5624661979448353,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004192910652787338,
|
|
"loss": 5.2395,
|
|
"mean_token_accuracy": 0.19694219529628754,
|
|
"num_tokens": 7392746.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.6397631645202635,
|
|
"epoch": 3.5678745267712277,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004189616957301419,
|
|
"loss": 5.3558,
|
|
"mean_token_accuracy": 0.18522324413061142,
|
|
"num_tokens": 7405520.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.615693664550781,
|
|
"epoch": 3.57328285559762,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004186318030244149,
|
|
"loss": 5.3222,
|
|
"mean_token_accuracy": 0.18899650871753693,
|
|
"num_tokens": 7417788.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.699047708511353,
|
|
"epoch": 3.578691184424013,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00041830138836038383,
|
|
"loss": 5.3812,
|
|
"mean_token_accuracy": 0.18883986324071883,
|
|
"num_tokens": 7428912.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.651888990402222,
|
|
"epoch": 3.584099513250406,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004179704529387763,
|
|
"loss": 5.2791,
|
|
"mean_token_accuracy": 0.1944235935807228,
|
|
"num_tokens": 7439797.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.553241634368897,
|
|
"epoch": 3.5895078420767983,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004176389979622125,
|
|
"loss": 5.2519,
|
|
"mean_token_accuracy": 0.19207768887281418,
|
|
"num_tokens": 7451755.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.701638317108154,
|
|
"epoch": 3.5949161709031907,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004173070246352003,
|
|
"loss": 5.3589,
|
|
"mean_token_accuracy": 0.19077109396457673,
|
|
"num_tokens": 7463393.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.569928741455078,
|
|
"epoch": 3.6003244997295836,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00041697453416413184,
|
|
"loss": 5.2859,
|
|
"mean_token_accuracy": 0.19374929517507553,
|
|
"num_tokens": 7474404.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.672807455062866,
|
|
"epoch": 3.6057328285559764,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000416641527757278,
|
|
"loss": 5.281,
|
|
"mean_token_accuracy": 0.19363389462232589,
|
|
"num_tokens": 7485495.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.533652353286743,
|
|
"epoch": 3.611141157382369,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00041630800662478514,
|
|
"loss": 5.2567,
|
|
"mean_token_accuracy": 0.20005156695842743,
|
|
"num_tokens": 7497305.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.641712856292725,
|
|
"epoch": 3.6165494862087613,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00041597397197866957,
|
|
"loss": 5.1944,
|
|
"mean_token_accuracy": 0.19611842334270477,
|
|
"num_tokens": 7509008.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.538056993484497,
|
|
"epoch": 3.621957815035154,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004156394250328141,
|
|
"loss": 5.2704,
|
|
"mean_token_accuracy": 0.20210754871368408,
|
|
"num_tokens": 7520079.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.5933324813842775,
|
|
"epoch": 3.627366143861547,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000415304367002963,
|
|
"loss": 5.2936,
|
|
"mean_token_accuracy": 0.1978504702448845,
|
|
"num_tokens": 7531294.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.638067388534546,
|
|
"epoch": 3.6327744726879394,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00041496879910671796,
|
|
"loss": 5.3024,
|
|
"mean_token_accuracy": 0.19586898535490035,
|
|
"num_tokens": 7542508.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.632113265991211,
|
|
"epoch": 3.638182801514332,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004146327225635336,
|
|
"loss": 5.2686,
|
|
"mean_token_accuracy": 0.20014205127954482,
|
|
"num_tokens": 7553288.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.698447942733765,
|
|
"epoch": 3.6435911303407247,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004142961385947127,
|
|
"loss": 5.452,
|
|
"mean_token_accuracy": 0.1902895078063011,
|
|
"num_tokens": 7564532.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.607742166519165,
|
|
"epoch": 3.6489994591671175,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00041395904842340215,
|
|
"loss": 5.2699,
|
|
"mean_token_accuracy": 0.19600602835416794,
|
|
"num_tokens": 7575109.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.57494888305664,
|
|
"epoch": 3.65440778799351,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004136214532745887,
|
|
"loss": 5.2705,
|
|
"mean_token_accuracy": 0.20107389390468597,
|
|
"num_tokens": 7585871.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.711588287353516,
|
|
"epoch": 3.6598161168199024,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004132833543750936,
|
|
"loss": 5.3574,
|
|
"mean_token_accuracy": 0.1989874556660652,
|
|
"num_tokens": 7596821.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.5790708541870115,
|
|
"epoch": 3.6652244456462952,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00041294475295356917,
|
|
"loss": 5.2744,
|
|
"mean_token_accuracy": 0.19830991178750992,
|
|
"num_tokens": 7608101.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.598383140563965,
|
|
"epoch": 3.670632774472688,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004126056502404937,
|
|
"loss": 5.2741,
|
|
"mean_token_accuracy": 0.19256848692893982,
|
|
"num_tokens": 7618722.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.621268606185913,
|
|
"epoch": 3.6760411032990805,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004122660474681672,
|
|
"loss": 5.286,
|
|
"mean_token_accuracy": 0.20010304003953933,
|
|
"num_tokens": 7630061.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.621401929855347,
|
|
"epoch": 3.6814494321254734,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00041192594587070706,
|
|
"loss": 5.2878,
|
|
"mean_token_accuracy": 0.19454421997070312,
|
|
"num_tokens": 7641830.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.5925281047821045,
|
|
"epoch": 3.686857760951866,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00041158534668404325,
|
|
"loss": 5.315,
|
|
"mean_token_accuracy": 0.19491345584392547,
|
|
"num_tokens": 7652750.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.604039287567138,
|
|
"epoch": 3.6922660897782587,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00041124425114591386,
|
|
"loss": 5.227,
|
|
"mean_token_accuracy": 0.2018936112523079,
|
|
"num_tokens": 7663405.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.508400392532349,
|
|
"epoch": 3.697674418604651,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00041090266049586107,
|
|
"loss": 5.2461,
|
|
"mean_token_accuracy": 0.1907604977488518,
|
|
"num_tokens": 7673958.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.612121295928955,
|
|
"epoch": 3.703082747431044,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004105605759752258,
|
|
"loss": 5.2439,
|
|
"mean_token_accuracy": 0.19162216186523437,
|
|
"num_tokens": 7685475.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.521811580657959,
|
|
"epoch": 3.7084910762574363,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00041021799882714436,
|
|
"loss": 5.2277,
|
|
"mean_token_accuracy": 0.19989577680826187,
|
|
"num_tokens": 7697296.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.6373435974121096,
|
|
"epoch": 3.713899405083829,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004098749302965426,
|
|
"loss": 5.3847,
|
|
"mean_token_accuracy": 0.19401603937149048,
|
|
"num_tokens": 7709414.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.698872137069702,
|
|
"epoch": 3.7193077339102216,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004095313716301324,
|
|
"loss": 5.3533,
|
|
"mean_token_accuracy": 0.18348875939846038,
|
|
"num_tokens": 7720965.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.656430053710937,
|
|
"epoch": 3.7247160627366145,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004091873240764069,
|
|
"loss": 5.3124,
|
|
"mean_token_accuracy": 0.18447436392307281,
|
|
"num_tokens": 7733044.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.663830423355103,
|
|
"epoch": 3.730124391563007,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00040884278888563566,
|
|
"loss": 5.3098,
|
|
"mean_token_accuracy": 0.1942358359694481,
|
|
"num_tokens": 7743878.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.555906343460083,
|
|
"epoch": 3.7355327203893998,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00040849776730986014,
|
|
"loss": 5.2909,
|
|
"mean_token_accuracy": 0.19837642908096315,
|
|
"num_tokens": 7755445.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.690651750564575,
|
|
"epoch": 3.740941049215792,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004081522606028899,
|
|
"loss": 5.3543,
|
|
"mean_token_accuracy": 0.19153521507978438,
|
|
"num_tokens": 7765990.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.557598400115967,
|
|
"epoch": 3.746349378042185,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00040780627002029674,
|
|
"loss": 5.1605,
|
|
"mean_token_accuracy": 0.20882601290941238,
|
|
"num_tokens": 7777064.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.489257574081421,
|
|
"epoch": 3.7517577068685775,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004074597968194116,
|
|
"loss": 5.1945,
|
|
"mean_token_accuracy": 0.19225298762321472,
|
|
"num_tokens": 7788618.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.609942436218262,
|
|
"epoch": 3.7571660356949703,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004071128422593186,
|
|
"loss": 5.3247,
|
|
"mean_token_accuracy": 0.18635189831256865,
|
|
"num_tokens": 7798484.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.621126508712768,
|
|
"epoch": 3.7625743645213627,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004067654076008515,
|
|
"loss": 5.3055,
|
|
"mean_token_accuracy": 0.19010080248117447,
|
|
"num_tokens": 7810270.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.580523586273193,
|
|
"epoch": 3.7679826933477556,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00040641749410658877,
|
|
"loss": 5.3762,
|
|
"mean_token_accuracy": 0.1904643341898918,
|
|
"num_tokens": 7822616.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.674696016311645,
|
|
"epoch": 3.773391022174148,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004060691030408487,
|
|
"loss": 5.3043,
|
|
"mean_token_accuracy": 0.19476685672998428,
|
|
"num_tokens": 7834933.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.6469498634338375,
|
|
"epoch": 3.778799351000541,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00040572023566968536,
|
|
"loss": 5.3151,
|
|
"mean_token_accuracy": 0.19442807585000993,
|
|
"num_tokens": 7847686.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.58784031867981,
|
|
"epoch": 3.7842076798269337,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00040537089326088345,
|
|
"loss": 5.2943,
|
|
"mean_token_accuracy": 0.1878056839108467,
|
|
"num_tokens": 7859087.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 3.7842076798269337,
|
|
"eval_entropy": 5.538707366856662,
|
|
"eval_loss": 5.896342754364014,
|
|
"eval_mean_token_accuracy": 0.17500235853208737,
|
|
"eval_num_tokens": 7859087.0,
|
|
"eval_runtime": 15.416,
|
|
"eval_samples_per_second": 228.009,
|
|
"eval_steps_per_second": 28.542,
|
|
"step": 3500
|
|
}
|
|
],
|
|
"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": 1822957332480000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|