7a79b20bcf
Model: fpadovani/eng-latn-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed10 Source: Original Platform
10090 lines
274 KiB
JSON
10090 lines
274 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 5.405624661979449,
|
|
"eval_steps": 1000,
|
|
"global_step": 5000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 4.816491746902466,
|
|
"epoch": 0.005408328826392645,
|
|
"grad_norm": 12.0,
|
|
"learning_rate": 2e-06,
|
|
"loss": 14.3891,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 10948.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 4.827733516693115,
|
|
"epoch": 0.01081665765278529,
|
|
"grad_norm": 13.375,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 14.3509,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 21455.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 4.862413930892944,
|
|
"epoch": 0.016224986479177934,
|
|
"grad_norm": 17.875,
|
|
"learning_rate": 7e-06,
|
|
"loss": 14.1901,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 32377.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 5.008710432052612,
|
|
"epoch": 0.02163331530557058,
|
|
"grad_norm": 29.5,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 13.7041,
|
|
"mean_token_accuracy": 6.706907879561186e-05,
|
|
"num_tokens": 42276.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 5.984945297241211,
|
|
"epoch": 0.02704164413196322,
|
|
"grad_norm": 44.0,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 12.6083,
|
|
"mean_token_accuracy": 0.00013905229279771448,
|
|
"num_tokens": 55049.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 9.738362789154053,
|
|
"epoch": 0.03244997295835587,
|
|
"grad_norm": 4.125,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 11.0101,
|
|
"mean_token_accuracy": 9.293680195696652e-05,
|
|
"num_tokens": 66792.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.689375400543213,
|
|
"epoch": 0.03785830178474851,
|
|
"grad_norm": 3.53125,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.64,
|
|
"mean_token_accuracy": 0.004824416560586542,
|
|
"num_tokens": 78208.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.698863887786866,
|
|
"epoch": 0.04326663061114116,
|
|
"grad_norm": 2.84375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.2961,
|
|
"mean_token_accuracy": 0.023805590718984602,
|
|
"num_tokens": 88871.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.674648761749268,
|
|
"epoch": 0.048674959437533805,
|
|
"grad_norm": 2.65625,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.9755,
|
|
"mean_token_accuracy": 0.03232901841402054,
|
|
"num_tokens": 100118.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.597569847106934,
|
|
"epoch": 0.05408328826392644,
|
|
"grad_norm": 2.328125,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.7307,
|
|
"mean_token_accuracy": 0.03992431238293648,
|
|
"num_tokens": 113120.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.540758800506591,
|
|
"epoch": 0.05949161709031909,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.5288,
|
|
"mean_token_accuracy": 0.03650712165981531,
|
|
"num_tokens": 123220.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.527715396881103,
|
|
"epoch": 0.06489994591671173,
|
|
"grad_norm": 1.9921875,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.5023,
|
|
"mean_token_accuracy": 0.03586216233670712,
|
|
"num_tokens": 134479.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.531739139556885,
|
|
"epoch": 0.07030827474310439,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.3556,
|
|
"mean_token_accuracy": 0.04139025770127773,
|
|
"num_tokens": 145189.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.52338228225708,
|
|
"epoch": 0.07571660356949703,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.2769,
|
|
"mean_token_accuracy": 0.04220384769141674,
|
|
"num_tokens": 157354.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.495023918151855,
|
|
"epoch": 0.08112493239588967,
|
|
"grad_norm": 1.875,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.2309,
|
|
"mean_token_accuracy": 0.03997939024120569,
|
|
"num_tokens": 168821.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.466780185699463,
|
|
"epoch": 0.08653326122228232,
|
|
"grad_norm": 1.90625,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.1055,
|
|
"mean_token_accuracy": 0.03739558029919863,
|
|
"num_tokens": 180554.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.440871238708496,
|
|
"epoch": 0.09194159004867496,
|
|
"grad_norm": 2.171875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.0271,
|
|
"mean_token_accuracy": 0.03742563612759113,
|
|
"num_tokens": 191229.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.418688297271729,
|
|
"epoch": 0.09734991887506761,
|
|
"grad_norm": 1.9453125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 8.9282,
|
|
"mean_token_accuracy": 0.03555689416825771,
|
|
"num_tokens": 201903.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.375628566741943,
|
|
"epoch": 0.10275824770146025,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.8778,
|
|
"mean_token_accuracy": 0.03980072475969791,
|
|
"num_tokens": 214255.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.296640205383301,
|
|
"epoch": 0.10816657652785289,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.6959,
|
|
"mean_token_accuracy": 0.052101677656173705,
|
|
"num_tokens": 225423.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.19414758682251,
|
|
"epoch": 0.11357490535424554,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.594,
|
|
"mean_token_accuracy": 0.05231698267161846,
|
|
"num_tokens": 237801.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.054014301300048,
|
|
"epoch": 0.11898323418063818,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.3897,
|
|
"mean_token_accuracy": 0.051315633580088615,
|
|
"num_tokens": 249195.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 9.872079658508301,
|
|
"epoch": 0.12439156300703083,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.2393,
|
|
"mean_token_accuracy": 0.0522298626601696,
|
|
"num_tokens": 260129.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.810044193267823,
|
|
"epoch": 0.12979989183342347,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.2772,
|
|
"mean_token_accuracy": 0.05217186138033867,
|
|
"num_tokens": 271220.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.666457653045654,
|
|
"epoch": 0.1352082206598161,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.0237,
|
|
"mean_token_accuracy": 0.051312217861413954,
|
|
"num_tokens": 281560.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.364743328094482,
|
|
"epoch": 0.14061654948620878,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 7.8811,
|
|
"mean_token_accuracy": 0.053314308449625966,
|
|
"num_tokens": 292518.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.2109206199646,
|
|
"epoch": 0.1460248783126014,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.8517,
|
|
"mean_token_accuracy": 0.049220229312777516,
|
|
"num_tokens": 304865.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 8.980883979797364,
|
|
"epoch": 0.15143320713899405,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.6909,
|
|
"mean_token_accuracy": 0.055328205600380896,
|
|
"num_tokens": 315951.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 8.628364181518554,
|
|
"epoch": 0.1568415359653867,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.6116,
|
|
"mean_token_accuracy": 0.05774536021053791,
|
|
"num_tokens": 327094.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.470148944854737,
|
|
"epoch": 0.16224986479177933,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.5585,
|
|
"mean_token_accuracy": 0.058901453390717506,
|
|
"num_tokens": 337801.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.233556652069092,
|
|
"epoch": 0.167658193618172,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.5032,
|
|
"mean_token_accuracy": 0.061500222980976106,
|
|
"num_tokens": 349039.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.088182020187379,
|
|
"epoch": 0.17306652244456464,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.4874,
|
|
"mean_token_accuracy": 0.0693302620202303,
|
|
"num_tokens": 360639.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.017281103134156,
|
|
"epoch": 0.17847485127095727,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.5015,
|
|
"mean_token_accuracy": 0.06765063926577568,
|
|
"num_tokens": 371005.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 7.958820867538452,
|
|
"epoch": 0.1838831800973499,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.3887,
|
|
"mean_token_accuracy": 0.07020511478185654,
|
|
"num_tokens": 383340.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 7.844907283782959,
|
|
"epoch": 0.18929150892374255,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.3382,
|
|
"mean_token_accuracy": 0.07330376878380776,
|
|
"num_tokens": 394682.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 7.825045919418335,
|
|
"epoch": 0.19469983775013522,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.336,
|
|
"mean_token_accuracy": 0.07912327982485294,
|
|
"num_tokens": 405713.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 7.846845293045044,
|
|
"epoch": 0.20010816657652786,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.3066,
|
|
"mean_token_accuracy": 0.07909615188837052,
|
|
"num_tokens": 416208.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.744113254547119,
|
|
"epoch": 0.2055164954029205,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.2175,
|
|
"mean_token_accuracy": 0.0757235523313284,
|
|
"num_tokens": 427595.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.717983436584473,
|
|
"epoch": 0.21092482422931313,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.3271,
|
|
"mean_token_accuracy": 0.08375168889760971,
|
|
"num_tokens": 438273.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.786208295822144,
|
|
"epoch": 0.21633315305570577,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.2509,
|
|
"mean_token_accuracy": 0.07852847650647163,
|
|
"num_tokens": 448948.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.759875535964966,
|
|
"epoch": 0.22174148188209844,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.2839,
|
|
"mean_token_accuracy": 0.07927897274494171,
|
|
"num_tokens": 460838.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.683747482299805,
|
|
"epoch": 0.22714981070849108,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.2887,
|
|
"mean_token_accuracy": 0.082744500041008,
|
|
"num_tokens": 473555.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.688952970504761,
|
|
"epoch": 0.23255813953488372,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.2241,
|
|
"mean_token_accuracy": 0.08608492240309715,
|
|
"num_tokens": 485221.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.682677507400513,
|
|
"epoch": 0.23796646836127636,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.1214,
|
|
"mean_token_accuracy": 0.08501881062984466,
|
|
"num_tokens": 496552.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.638469791412353,
|
|
"epoch": 0.24337479718766902,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.1808,
|
|
"mean_token_accuracy": 0.08574967458844185,
|
|
"num_tokens": 509760.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.626894855499268,
|
|
"epoch": 0.24878312601406166,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.0246,
|
|
"mean_token_accuracy": 0.09106166735291481,
|
|
"num_tokens": 521175.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.549870872497559,
|
|
"epoch": 0.2541914548404543,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.1344,
|
|
"mean_token_accuracy": 0.08684690818190574,
|
|
"num_tokens": 532597.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.58236255645752,
|
|
"epoch": 0.25959978366684694,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.1753,
|
|
"mean_token_accuracy": 0.09069491475820542,
|
|
"num_tokens": 545661.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.5936479568481445,
|
|
"epoch": 0.2650081124932396,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.1579,
|
|
"mean_token_accuracy": 0.0852174036204815,
|
|
"num_tokens": 557336.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.536095857620239,
|
|
"epoch": 0.2704164413196322,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.0178,
|
|
"mean_token_accuracy": 0.0952389270067215,
|
|
"num_tokens": 567398.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.4575268745422365,
|
|
"epoch": 0.27582477014602486,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.0896,
|
|
"mean_token_accuracy": 0.08974430859088897,
|
|
"num_tokens": 579543.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.449991941452026,
|
|
"epoch": 0.28123309897241755,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 6.9161,
|
|
"mean_token_accuracy": 0.09924268275499344,
|
|
"num_tokens": 589218.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.458067178726196,
|
|
"epoch": 0.2866414277988102,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000132,
|
|
"loss": 6.9748,
|
|
"mean_token_accuracy": 0.09499456658959389,
|
|
"num_tokens": 600327.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.406368970870972,
|
|
"epoch": 0.2920497566252028,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 6.926,
|
|
"mean_token_accuracy": 0.09738428071141243,
|
|
"num_tokens": 610923.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.486721420288086,
|
|
"epoch": 0.29745808545159547,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.0091,
|
|
"mean_token_accuracy": 0.09363943114876747,
|
|
"num_tokens": 621138.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.314512109756469,
|
|
"epoch": 0.3028664142779881,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 6.948,
|
|
"mean_token_accuracy": 0.10305705517530442,
|
|
"num_tokens": 631440.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.410723161697388,
|
|
"epoch": 0.30827474310438074,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 6.9745,
|
|
"mean_token_accuracy": 0.10130783841013909,
|
|
"num_tokens": 642328.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.340037202835083,
|
|
"epoch": 0.3136830719307734,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.9318,
|
|
"mean_token_accuracy": 0.10102855190634727,
|
|
"num_tokens": 652875.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.346016311645508,
|
|
"epoch": 0.319091400757166,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 6.8677,
|
|
"mean_token_accuracy": 0.10212785899639129,
|
|
"num_tokens": 663726.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.447050857543945,
|
|
"epoch": 0.32449972958355866,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.0625,
|
|
"mean_token_accuracy": 0.09114478453993798,
|
|
"num_tokens": 675638.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.26244854927063,
|
|
"epoch": 0.3299080584099513,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.9195,
|
|
"mean_token_accuracy": 0.1022642120718956,
|
|
"num_tokens": 687239.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.387060499191284,
|
|
"epoch": 0.335316387236344,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.9658,
|
|
"mean_token_accuracy": 0.09624839648604393,
|
|
"num_tokens": 698905.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.278276062011718,
|
|
"epoch": 0.34072471606273663,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.9415,
|
|
"mean_token_accuracy": 0.10140556320548058,
|
|
"num_tokens": 710602.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.322119331359863,
|
|
"epoch": 0.34613304488912927,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.9135,
|
|
"mean_token_accuracy": 0.10094785168766976,
|
|
"num_tokens": 721326.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.384740495681763,
|
|
"epoch": 0.3515413737155219,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.894,
|
|
"mean_token_accuracy": 0.10391761958599091,
|
|
"num_tokens": 732972.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.19610185623169,
|
|
"epoch": 0.35694970254191455,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.8612,
|
|
"mean_token_accuracy": 0.10241201519966125,
|
|
"num_tokens": 743624.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.34399447441101,
|
|
"epoch": 0.3623580313683072,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.9483,
|
|
"mean_token_accuracy": 0.10304523780941963,
|
|
"num_tokens": 756384.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.241913795471191,
|
|
"epoch": 0.3677663601946998,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.9174,
|
|
"mean_token_accuracy": 0.10442400574684144,
|
|
"num_tokens": 767256.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.342946195602417,
|
|
"epoch": 0.37317468902109246,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.8438,
|
|
"mean_token_accuracy": 0.10971475169062614,
|
|
"num_tokens": 779059.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.177052688598633,
|
|
"epoch": 0.3785830178474851,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.9283,
|
|
"mean_token_accuracy": 0.11065090894699096,
|
|
"num_tokens": 790291.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.257277774810791,
|
|
"epoch": 0.3839913466738778,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.8328,
|
|
"mean_token_accuracy": 0.10423929244279861,
|
|
"num_tokens": 801450.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.145007610321045,
|
|
"epoch": 0.38939967550027044,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.7503,
|
|
"mean_token_accuracy": 0.11073209047317505,
|
|
"num_tokens": 812428.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.148734092712402,
|
|
"epoch": 0.3948080043266631,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.8508,
|
|
"mean_token_accuracy": 0.10246649086475372,
|
|
"num_tokens": 823644.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.288735342025757,
|
|
"epoch": 0.4002163331530557,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.8351,
|
|
"mean_token_accuracy": 0.11553217023611069,
|
|
"num_tokens": 834064.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.1784649848937985,
|
|
"epoch": 0.40562466197944835,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.7847,
|
|
"mean_token_accuracy": 0.10547300353646279,
|
|
"num_tokens": 845207.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.121596479415894,
|
|
"epoch": 0.411032990805841,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.843,
|
|
"mean_token_accuracy": 0.10698517560958862,
|
|
"num_tokens": 855564.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.189912986755371,
|
|
"epoch": 0.41644131963223363,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.8547,
|
|
"mean_token_accuracy": 0.11624484211206436,
|
|
"num_tokens": 866959.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.079056596755981,
|
|
"epoch": 0.42184964845862627,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.7793,
|
|
"mean_token_accuracy": 0.11480710729956627,
|
|
"num_tokens": 877319.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.181234741210938,
|
|
"epoch": 0.4272579772850189,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.866,
|
|
"mean_token_accuracy": 0.10589642673730851,
|
|
"num_tokens": 888037.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.0435555934906,
|
|
"epoch": 0.43266630611141155,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.6004,
|
|
"mean_token_accuracy": 0.1158117413520813,
|
|
"num_tokens": 898775.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.104381990432739,
|
|
"epoch": 0.43807463493780424,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.7622,
|
|
"mean_token_accuracy": 0.11117781847715377,
|
|
"num_tokens": 909513.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.088195323944092,
|
|
"epoch": 0.4434829637641969,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.8126,
|
|
"mean_token_accuracy": 0.10974752902984619,
|
|
"num_tokens": 920730.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.043075609207153,
|
|
"epoch": 0.4488912925905895,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.7279,
|
|
"mean_token_accuracy": 0.11817020252346992,
|
|
"num_tokens": 931457.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.0011475563049315,
|
|
"epoch": 0.45429962141698216,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.7087,
|
|
"mean_token_accuracy": 0.11752742975950241,
|
|
"num_tokens": 942572.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.129669713973999,
|
|
"epoch": 0.4597079502433748,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.7241,
|
|
"mean_token_accuracy": 0.10921066999435425,
|
|
"num_tokens": 953132.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.0359748840332035,
|
|
"epoch": 0.46511627906976744,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.7533,
|
|
"mean_token_accuracy": 0.11189360246062278,
|
|
"num_tokens": 963551.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.023745107650757,
|
|
"epoch": 0.4705246078961601,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.7177,
|
|
"mean_token_accuracy": 0.11238613426685333,
|
|
"num_tokens": 975297.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 6.960337781906128,
|
|
"epoch": 0.4759329367225527,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.7147,
|
|
"mean_token_accuracy": 0.11535659208893775,
|
|
"num_tokens": 986910.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.105753707885742,
|
|
"epoch": 0.48134126554894535,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.8383,
|
|
"mean_token_accuracy": 0.11518610119819642,
|
|
"num_tokens": 997058.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.0269519805908205,
|
|
"epoch": 0.48674959437533805,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.7099,
|
|
"mean_token_accuracy": 0.10921002328395843,
|
|
"num_tokens": 1008261.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.005563545227051,
|
|
"epoch": 0.4921579232017307,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.6246,
|
|
"mean_token_accuracy": 0.11358757168054581,
|
|
"num_tokens": 1019193.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 6.977084970474243,
|
|
"epoch": 0.4975662520281233,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.7447,
|
|
"mean_token_accuracy": 0.1139349453151226,
|
|
"num_tokens": 1031227.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 6.943169641494751,
|
|
"epoch": 0.502974580854516,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.5781,
|
|
"mean_token_accuracy": 0.1172634594142437,
|
|
"num_tokens": 1042963.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 6.869647264480591,
|
|
"epoch": 0.5083829096809086,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.6377,
|
|
"mean_token_accuracy": 0.1132569208741188,
|
|
"num_tokens": 1053709.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.069565773010254,
|
|
"epoch": 0.5137912385073012,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.6675,
|
|
"mean_token_accuracy": 0.12598269656300545,
|
|
"num_tokens": 1064527.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 6.852591276168823,
|
|
"epoch": 0.5191995673336939,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.7166,
|
|
"mean_token_accuracy": 0.11791721880435943,
|
|
"num_tokens": 1075127.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 6.959320735931397,
|
|
"epoch": 0.5246078961600865,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.7096,
|
|
"mean_token_accuracy": 0.11324949637055397,
|
|
"num_tokens": 1086581.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 6.855788326263427,
|
|
"epoch": 0.5300162249864792,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.6395,
|
|
"mean_token_accuracy": 0.12406286075711251,
|
|
"num_tokens": 1097818.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 6.895084428787231,
|
|
"epoch": 0.5354245538128718,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.62,
|
|
"mean_token_accuracy": 0.11416960656642913,
|
|
"num_tokens": 1108043.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 6.967974853515625,
|
|
"epoch": 0.5408328826392644,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.6878,
|
|
"mean_token_accuracy": 0.11347125247120857,
|
|
"num_tokens": 1119875.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 6.858421468734742,
|
|
"epoch": 0.5462412114656571,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.5843,
|
|
"mean_token_accuracy": 0.1187442697584629,
|
|
"num_tokens": 1132406.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.007649278640747,
|
|
"epoch": 0.5516495402920497,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.7718,
|
|
"mean_token_accuracy": 0.1115761086344719,
|
|
"num_tokens": 1143828.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 6.889253616333008,
|
|
"epoch": 0.5570578691184424,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.5729,
|
|
"mean_token_accuracy": 0.11763741374015808,
|
|
"num_tokens": 1155482.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 6.731452751159668,
|
|
"epoch": 0.5624661979448351,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.5244,
|
|
"mean_token_accuracy": 0.12875338792800903,
|
|
"num_tokens": 1166847.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 6.8908110618591305,
|
|
"epoch": 0.5678745267712277,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.5924,
|
|
"mean_token_accuracy": 0.12007522955536842,
|
|
"num_tokens": 1177419.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.906198453903198,
|
|
"epoch": 0.5732828555976204,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.7698,
|
|
"mean_token_accuracy": 0.11545055732131004,
|
|
"num_tokens": 1188975.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 6.836439752578736,
|
|
"epoch": 0.578691184424013,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.6162,
|
|
"mean_token_accuracy": 0.11988403424620628,
|
|
"num_tokens": 1200252.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.827834510803223,
|
|
"epoch": 0.5840995132504057,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.657,
|
|
"mean_token_accuracy": 0.12472999319434167,
|
|
"num_tokens": 1211864.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.788261890411377,
|
|
"epoch": 0.5895078420767983,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.524,
|
|
"mean_token_accuracy": 0.12068347483873368,
|
|
"num_tokens": 1222429.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.91283164024353,
|
|
"epoch": 0.5949161709031909,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.632,
|
|
"mean_token_accuracy": 0.11880290359258652,
|
|
"num_tokens": 1233782.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.792628812789917,
|
|
"epoch": 0.6003244997295836,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.5843,
|
|
"mean_token_accuracy": 0.11936102360486985,
|
|
"num_tokens": 1244309.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.7369684219360355,
|
|
"epoch": 0.6057328285559762,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.535,
|
|
"mean_token_accuracy": 0.12587283179163933,
|
|
"num_tokens": 1256403.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.848638105392456,
|
|
"epoch": 0.6111411573823688,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.6314,
|
|
"mean_token_accuracy": 0.11637956202030182,
|
|
"num_tokens": 1268758.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.7654791355133055,
|
|
"epoch": 0.6165494862087615,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.5245,
|
|
"mean_token_accuracy": 0.12454744204878807,
|
|
"num_tokens": 1279355.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.76825737953186,
|
|
"epoch": 0.6219578150351541,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.5023,
|
|
"mean_token_accuracy": 0.11673943027853965,
|
|
"num_tokens": 1290619.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.779633808135986,
|
|
"epoch": 0.6273661438615468,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.4739,
|
|
"mean_token_accuracy": 0.12195988073945045,
|
|
"num_tokens": 1301851.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.81631326675415,
|
|
"epoch": 0.6327744726879394,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.6271,
|
|
"mean_token_accuracy": 0.12264949083328247,
|
|
"num_tokens": 1313260.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.702813005447387,
|
|
"epoch": 0.638182801514332,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.614,
|
|
"mean_token_accuracy": 0.12174173891544342,
|
|
"num_tokens": 1324189.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.848341035842895,
|
|
"epoch": 0.6435911303407247,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.6393,
|
|
"mean_token_accuracy": 0.12240146771073342,
|
|
"num_tokens": 1335765.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.741462898254395,
|
|
"epoch": 0.6489994591671173,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.5711,
|
|
"mean_token_accuracy": 0.12267256677150726,
|
|
"num_tokens": 1347653.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.757457304000854,
|
|
"epoch": 0.65440778799351,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.565,
|
|
"mean_token_accuracy": 0.12561324387788772,
|
|
"num_tokens": 1360201.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.749183225631714,
|
|
"epoch": 0.6598161168199026,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.4399,
|
|
"mean_token_accuracy": 0.12647488713264465,
|
|
"num_tokens": 1370974.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.667228031158447,
|
|
"epoch": 0.6652244456462953,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.5024,
|
|
"mean_token_accuracy": 0.12279477640986443,
|
|
"num_tokens": 1383855.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.6716227531433105,
|
|
"epoch": 0.670632774472688,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.4609,
|
|
"mean_token_accuracy": 0.13709836080670357,
|
|
"num_tokens": 1395041.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.799122095108032,
|
|
"epoch": 0.6760411032990806,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.5701,
|
|
"mean_token_accuracy": 0.12464237436652184,
|
|
"num_tokens": 1406102.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.585179328918457,
|
|
"epoch": 0.6814494321254733,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.3918,
|
|
"mean_token_accuracy": 0.13039295971393586,
|
|
"num_tokens": 1417713.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.634333515167237,
|
|
"epoch": 0.6868577609518659,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.545,
|
|
"mean_token_accuracy": 0.12864828929305078,
|
|
"num_tokens": 1428526.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.689204120635987,
|
|
"epoch": 0.6922660897782585,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.4914,
|
|
"mean_token_accuracy": 0.130169028788805,
|
|
"num_tokens": 1438849.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.685839700698852,
|
|
"epoch": 0.6976744186046512,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.562,
|
|
"mean_token_accuracy": 0.1240805745124817,
|
|
"num_tokens": 1450776.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.58751163482666,
|
|
"epoch": 0.7030827474310438,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.4104,
|
|
"mean_token_accuracy": 0.12494389489293098,
|
|
"num_tokens": 1462552.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.5898017406463625,
|
|
"epoch": 0.7084910762574365,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.4285,
|
|
"mean_token_accuracy": 0.12942259460687638,
|
|
"num_tokens": 1473867.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.534844923019409,
|
|
"epoch": 0.7138994050838291,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.4365,
|
|
"mean_token_accuracy": 0.1297524243593216,
|
|
"num_tokens": 1485233.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.6757666110992435,
|
|
"epoch": 0.7193077339102217,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.5013,
|
|
"mean_token_accuracy": 0.1329445168375969,
|
|
"num_tokens": 1496919.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.615731143951416,
|
|
"epoch": 0.7247160627366144,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.4425,
|
|
"mean_token_accuracy": 0.1267342150211334,
|
|
"num_tokens": 1508695.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.712345457077026,
|
|
"epoch": 0.730124391563007,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.4929,
|
|
"mean_token_accuracy": 0.13088276386260986,
|
|
"num_tokens": 1519629.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.592201662063599,
|
|
"epoch": 0.7355327203893997,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.3946,
|
|
"mean_token_accuracy": 0.13325998485088347,
|
|
"num_tokens": 1530144.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.504761219024658,
|
|
"epoch": 0.7409410492157923,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.3993,
|
|
"mean_token_accuracy": 0.1304351307451725,
|
|
"num_tokens": 1541013.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.65454888343811,
|
|
"epoch": 0.7463493780421849,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.4689,
|
|
"mean_token_accuracy": 0.1292506165802479,
|
|
"num_tokens": 1553344.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.584673690795898,
|
|
"epoch": 0.7517577068685776,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.5258,
|
|
"mean_token_accuracy": 0.12957922369241714,
|
|
"num_tokens": 1565298.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.584612560272217,
|
|
"epoch": 0.7571660356949702,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.4559,
|
|
"mean_token_accuracy": 0.13932728618383408,
|
|
"num_tokens": 1577645.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.577200269699096,
|
|
"epoch": 0.7625743645213628,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.4113,
|
|
"mean_token_accuracy": 0.13460915088653563,
|
|
"num_tokens": 1588387.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.715465974807739,
|
|
"epoch": 0.7679826933477556,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.5222,
|
|
"mean_token_accuracy": 0.1325209140777588,
|
|
"num_tokens": 1599837.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.541527509689331,
|
|
"epoch": 0.7733910221741482,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.4538,
|
|
"mean_token_accuracy": 0.1247769333422184,
|
|
"num_tokens": 1610401.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.497765398025512,
|
|
"epoch": 0.7787993510005409,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.4115,
|
|
"mean_token_accuracy": 0.1288266658782959,
|
|
"num_tokens": 1622688.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.657041215896607,
|
|
"epoch": 0.7842076798269335,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.4753,
|
|
"mean_token_accuracy": 0.12559469044208527,
|
|
"num_tokens": 1633594.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.517681550979614,
|
|
"epoch": 0.7896160086533262,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.3937,
|
|
"mean_token_accuracy": 0.1302385449409485,
|
|
"num_tokens": 1645861.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.380044412612915,
|
|
"epoch": 0.7950243374797188,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.2871,
|
|
"mean_token_accuracy": 0.13864490687847136,
|
|
"num_tokens": 1656903.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.560413932800293,
|
|
"epoch": 0.8004326663061114,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.3941,
|
|
"mean_token_accuracy": 0.13524588868021964,
|
|
"num_tokens": 1668130.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.4031064987182615,
|
|
"epoch": 0.8058409951325041,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.2856,
|
|
"mean_token_accuracy": 0.1413346864283085,
|
|
"num_tokens": 1678228.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.61671633720398,
|
|
"epoch": 0.8112493239588967,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.4299,
|
|
"mean_token_accuracy": 0.1359289936721325,
|
|
"num_tokens": 1690339.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.550224161148071,
|
|
"epoch": 0.8166576527852893,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.4545,
|
|
"mean_token_accuracy": 0.1291230209171772,
|
|
"num_tokens": 1701067.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.591836929321289,
|
|
"epoch": 0.822065981611682,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.3858,
|
|
"mean_token_accuracy": 0.13830180689692498,
|
|
"num_tokens": 1712835.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.482460069656372,
|
|
"epoch": 0.8274743104380746,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.3431,
|
|
"mean_token_accuracy": 0.1339413307607174,
|
|
"num_tokens": 1723432.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.485507011413574,
|
|
"epoch": 0.8328826392644673,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.3816,
|
|
"mean_token_accuracy": 0.13742958456277848,
|
|
"num_tokens": 1734134.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.505664300918579,
|
|
"epoch": 0.8382909680908599,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.3263,
|
|
"mean_token_accuracy": 0.13703825324773788,
|
|
"num_tokens": 1744723.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.503205013275147,
|
|
"epoch": 0.8436992969172525,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.4692,
|
|
"mean_token_accuracy": 0.13435956612229347,
|
|
"num_tokens": 1755269.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.569448661804199,
|
|
"epoch": 0.8491076257436452,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.4695,
|
|
"mean_token_accuracy": 0.13084665387868882,
|
|
"num_tokens": 1766344.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.478793334960938,
|
|
"epoch": 0.8545159545700378,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.4252,
|
|
"mean_token_accuracy": 0.12769865095615388,
|
|
"num_tokens": 1779052.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.41169285774231,
|
|
"epoch": 0.8599242833964305,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.2906,
|
|
"mean_token_accuracy": 0.13171118646860122,
|
|
"num_tokens": 1789754.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.330496978759766,
|
|
"epoch": 0.8653326122228231,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.2156,
|
|
"mean_token_accuracy": 0.13964635729789734,
|
|
"num_tokens": 1801161.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.5482982158660885,
|
|
"epoch": 0.8707409410492158,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.3145,
|
|
"mean_token_accuracy": 0.13513469696044922,
|
|
"num_tokens": 1810892.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.267117166519165,
|
|
"epoch": 0.8761492698756085,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.3747,
|
|
"mean_token_accuracy": 0.1325235478579998,
|
|
"num_tokens": 1821574.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.579942226409912,
|
|
"epoch": 0.8815575987020011,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.4147,
|
|
"mean_token_accuracy": 0.13671720623970032,
|
|
"num_tokens": 1833342.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.4688621997833256,
|
|
"epoch": 0.8869659275283938,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.2885,
|
|
"mean_token_accuracy": 0.13777956888079643,
|
|
"num_tokens": 1843768.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.47753415107727,
|
|
"epoch": 0.8923742563547864,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.3655,
|
|
"mean_token_accuracy": 0.13368769288063048,
|
|
"num_tokens": 1855569.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.481969928741455,
|
|
"epoch": 0.897782585181179,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.3524,
|
|
"mean_token_accuracy": 0.12639456540346145,
|
|
"num_tokens": 1866684.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.346806287765503,
|
|
"epoch": 0.9031909140075717,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.2795,
|
|
"mean_token_accuracy": 0.13589733690023423,
|
|
"num_tokens": 1878700.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.458641433715821,
|
|
"epoch": 0.9085992428339643,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.3092,
|
|
"mean_token_accuracy": 0.13823288977146148,
|
|
"num_tokens": 1889970.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.346112823486328,
|
|
"epoch": 0.914007571660357,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.2211,
|
|
"mean_token_accuracy": 0.13803198635578157,
|
|
"num_tokens": 1902008.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.384577941894531,
|
|
"epoch": 0.9194159004867496,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.451,
|
|
"mean_token_accuracy": 0.13807493597269058,
|
|
"num_tokens": 1913253.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.430484962463379,
|
|
"epoch": 0.9248242293131422,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.2842,
|
|
"mean_token_accuracy": 0.14151261672377585,
|
|
"num_tokens": 1923408.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.387599515914917,
|
|
"epoch": 0.9302325581395349,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.3452,
|
|
"mean_token_accuracy": 0.1367344081401825,
|
|
"num_tokens": 1935315.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.312086009979248,
|
|
"epoch": 0.9356408869659275,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.1907,
|
|
"mean_token_accuracy": 0.13912060707807541,
|
|
"num_tokens": 1946524.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.40140061378479,
|
|
"epoch": 0.9410492157923201,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.3316,
|
|
"mean_token_accuracy": 0.13535718396306037,
|
|
"num_tokens": 1957561.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.32704906463623,
|
|
"epoch": 0.9464575446187128,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.2669,
|
|
"mean_token_accuracy": 0.14048450738191604,
|
|
"num_tokens": 1969089.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.425353097915649,
|
|
"epoch": 0.9518658734451054,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.3031,
|
|
"mean_token_accuracy": 0.13773690536618233,
|
|
"num_tokens": 1980341.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.304427719116211,
|
|
"epoch": 0.9572742022714981,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.2545,
|
|
"mean_token_accuracy": 0.14644995033740998,
|
|
"num_tokens": 1991220.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.370989036560059,
|
|
"epoch": 0.9626825310978907,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.3384,
|
|
"mean_token_accuracy": 0.13986072912812234,
|
|
"num_tokens": 2002650.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.394271993637085,
|
|
"epoch": 0.9680908599242833,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.245,
|
|
"mean_token_accuracy": 0.1418061763048172,
|
|
"num_tokens": 2012884.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.391276216506958,
|
|
"epoch": 0.9734991887506761,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.3143,
|
|
"mean_token_accuracy": 0.1372893489897251,
|
|
"num_tokens": 2023543.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.445915412902832,
|
|
"epoch": 0.9789075175770687,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.377,
|
|
"mean_token_accuracy": 0.13422725573182107,
|
|
"num_tokens": 2034789.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.311644124984741,
|
|
"epoch": 0.9843158464034614,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.2856,
|
|
"mean_token_accuracy": 0.1408335119485855,
|
|
"num_tokens": 2046719.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.227353477478028,
|
|
"epoch": 0.989724175229854,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.2471,
|
|
"mean_token_accuracy": 0.14040217846632003,
|
|
"num_tokens": 2057016.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.333889055252075,
|
|
"epoch": 0.9951325040562466,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.2208,
|
|
"mean_token_accuracy": 0.13608154356479646,
|
|
"num_tokens": 2067898.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.261037720574273,
|
|
"epoch": 1.0,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.2446,
|
|
"mean_token_accuracy": 0.14002363714906904,
|
|
"num_tokens": 2076889.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.367987632751465,
|
|
"epoch": 1.0054083288263926,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.1674,
|
|
"mean_token_accuracy": 0.13393356874585152,
|
|
"num_tokens": 2089228.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.271270132064819,
|
|
"epoch": 1.0108166576527853,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.1582,
|
|
"mean_token_accuracy": 0.14539326280355452,
|
|
"num_tokens": 2101098.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.24772367477417,
|
|
"epoch": 1.016224986479178,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.1042,
|
|
"mean_token_accuracy": 0.15308297127485276,
|
|
"num_tokens": 2113081.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.293310213088989,
|
|
"epoch": 1.0216333153055706,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.0795,
|
|
"mean_token_accuracy": 0.14808320105075837,
|
|
"num_tokens": 2123356.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.367437076568604,
|
|
"epoch": 1.0270416441319632,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.1807,
|
|
"mean_token_accuracy": 0.14733164012432098,
|
|
"num_tokens": 2134388.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.116773366928101,
|
|
"epoch": 1.0324499729583558,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.0548,
|
|
"mean_token_accuracy": 0.14803084284067153,
|
|
"num_tokens": 2145209.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.321672821044922,
|
|
"epoch": 1.0378583017847485,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.1102,
|
|
"mean_token_accuracy": 0.14471006840467454,
|
|
"num_tokens": 2157146.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.179027318954468,
|
|
"epoch": 1.043266630611141,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.1315,
|
|
"mean_token_accuracy": 0.1431720107793808,
|
|
"num_tokens": 2168721.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.207623815536499,
|
|
"epoch": 1.0486749594375337,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.1216,
|
|
"mean_token_accuracy": 0.14734558016061783,
|
|
"num_tokens": 2179213.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.2188718795776365,
|
|
"epoch": 1.0540832882639264,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.1466,
|
|
"mean_token_accuracy": 0.1441408984363079,
|
|
"num_tokens": 2191224.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.3757565975189205,
|
|
"epoch": 1.059491617090319,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.2076,
|
|
"mean_token_accuracy": 0.14605461582541465,
|
|
"num_tokens": 2203665.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.216717863082886,
|
|
"epoch": 1.0648999459167117,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.1271,
|
|
"mean_token_accuracy": 0.14381530210375787,
|
|
"num_tokens": 2215136.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.285787582397461,
|
|
"epoch": 1.0703082747431043,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.1576,
|
|
"mean_token_accuracy": 0.14630142003297805,
|
|
"num_tokens": 2226375.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.071659278869629,
|
|
"epoch": 1.075716603569497,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.0121,
|
|
"mean_token_accuracy": 0.15493061542510986,
|
|
"num_tokens": 2236097.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.3767985820770265,
|
|
"epoch": 1.0811249323958896,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.2663,
|
|
"mean_token_accuracy": 0.14112632349133492,
|
|
"num_tokens": 2247964.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 1.0811249323958896,
|
|
"eval_entropy": 6.043832895972512,
|
|
"eval_loss": 6.424855709075928,
|
|
"eval_mean_token_accuracy": 0.14066070311448792,
|
|
"eval_num_tokens": 2247964.0,
|
|
"eval_runtime": 2.5746,
|
|
"eval_samples_per_second": 1365.265,
|
|
"eval_steps_per_second": 170.901,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.229990291595459,
|
|
"epoch": 1.0865332612222822,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999997383518051,
|
|
"loss": 6.2027,
|
|
"mean_token_accuracy": 0.1477906174957752,
|
|
"num_tokens": 2260467.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.232913064956665,
|
|
"epoch": 1.0919415900486749,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999986754070562,
|
|
"loss": 6.0959,
|
|
"mean_token_accuracy": 0.14914645105600358,
|
|
"num_tokens": 2272176.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.20442566871643,
|
|
"epoch": 1.0973499188750675,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999967948166009,
|
|
"loss": 6.1396,
|
|
"mean_token_accuracy": 0.1481549024581909,
|
|
"num_tokens": 2283491.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.233330011367798,
|
|
"epoch": 1.1027582477014604,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999940965872734,
|
|
"loss": 6.0407,
|
|
"mean_token_accuracy": 0.1456727996468544,
|
|
"num_tokens": 2295131.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.085158205032348,
|
|
"epoch": 1.1081665765278528,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004999905807288788,
|
|
"loss": 6.0755,
|
|
"mean_token_accuracy": 0.14888137727975845,
|
|
"num_tokens": 2305832.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.158335256576538,
|
|
"epoch": 1.1135749053542456,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000499986247254194,
|
|
"loss": 6.0588,
|
|
"mean_token_accuracy": 0.15194551572203635,
|
|
"num_tokens": 2315874.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.308013772964477,
|
|
"epoch": 1.1189832341806383,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999810961789667,
|
|
"loss": 6.2591,
|
|
"mean_token_accuracy": 0.143500005453825,
|
|
"num_tokens": 2327580.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.206014204025268,
|
|
"epoch": 1.124391563007031,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000499975127521916,
|
|
"loss": 6.0977,
|
|
"mean_token_accuracy": 0.150868783891201,
|
|
"num_tokens": 2340426.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.178250312805176,
|
|
"epoch": 1.1297998918334236,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999683413047319,
|
|
"loss": 6.218,
|
|
"mean_token_accuracy": 0.14377831518650055,
|
|
"num_tokens": 2353228.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.260583209991455,
|
|
"epoch": 1.1352082206598162,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999607375520759,
|
|
"loss": 6.0363,
|
|
"mean_token_accuracy": 0.15142566561698914,
|
|
"num_tokens": 2364238.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.168040084838867,
|
|
"epoch": 1.1406165494862088,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999523162915794,
|
|
"loss": 6.1824,
|
|
"mean_token_accuracy": 0.14444541782140732,
|
|
"num_tokens": 2375847.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.131425571441651,
|
|
"epoch": 1.1460248783126015,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999430775538459,
|
|
"loss": 6.0675,
|
|
"mean_token_accuracy": 0.15268574059009551,
|
|
"num_tokens": 2386489.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.163880252838135,
|
|
"epoch": 1.151433207138994,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999330213724487,
|
|
"loss": 6.1005,
|
|
"mean_token_accuracy": 0.1485110491514206,
|
|
"num_tokens": 2398839.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.247439813613892,
|
|
"epoch": 1.1568415359653867,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999221477839319,
|
|
"loss": 6.0623,
|
|
"mean_token_accuracy": 0.1486197292804718,
|
|
"num_tokens": 2410016.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.108935403823852,
|
|
"epoch": 1.1622498647917794,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999104568278104,
|
|
"loss": 6.1025,
|
|
"mean_token_accuracy": 0.14894724488258362,
|
|
"num_tokens": 2421202.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.264668560028076,
|
|
"epoch": 1.167658193618172,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000499897948546569,
|
|
"loss": 6.1788,
|
|
"mean_token_accuracy": 0.1392741844058037,
|
|
"num_tokens": 2431943.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.151232147216797,
|
|
"epoch": 1.1730665224445647,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998846229856629,
|
|
"loss": 6.0074,
|
|
"mean_token_accuracy": 0.14810821115970613,
|
|
"num_tokens": 2443454.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.190730094909668,
|
|
"epoch": 1.1784748512709573,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004998704801935171,
|
|
"loss": 6.1985,
|
|
"mean_token_accuracy": 0.13861391544342042,
|
|
"num_tokens": 2455909.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.099481296539307,
|
|
"epoch": 1.18388318009735,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004998555202215267,
|
|
"loss": 6.0634,
|
|
"mean_token_accuracy": 0.15428348630666733,
|
|
"num_tokens": 2467486.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.141802978515625,
|
|
"epoch": 1.1892915089237426,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004998397431240561,
|
|
"loss": 6.0136,
|
|
"mean_token_accuracy": 0.15860070437192916,
|
|
"num_tokens": 2477560.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.134382534027099,
|
|
"epoch": 1.1946998377501352,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004998231489584395,
|
|
"loss": 6.0603,
|
|
"mean_token_accuracy": 0.15650582909584046,
|
|
"num_tokens": 2487222.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.098001194000244,
|
|
"epoch": 1.2001081665765279,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00049980573778498,
|
|
"loss": 6.0827,
|
|
"mean_token_accuracy": 0.1502831093966961,
|
|
"num_tokens": 2497474.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.204444503784179,
|
|
"epoch": 1.2055164954029205,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004997875096669498,
|
|
"loss": 6.1461,
|
|
"mean_token_accuracy": 0.14976901710033416,
|
|
"num_tokens": 2508185.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.155468559265136,
|
|
"epoch": 1.2109248242293131,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004997684646705901,
|
|
"loss": 6.0862,
|
|
"mean_token_accuracy": 0.1465001329779625,
|
|
"num_tokens": 2518876.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.141084146499634,
|
|
"epoch": 1.2163331530557058,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004997486028651103,
|
|
"loss": 6.077,
|
|
"mean_token_accuracy": 0.1541600450873375,
|
|
"num_tokens": 2530554.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.042589950561523,
|
|
"epoch": 1.2217414818820984,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004997279243226884,
|
|
"loss": 5.9542,
|
|
"mean_token_accuracy": 0.15289611518383026,
|
|
"num_tokens": 2541331.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.214589309692383,
|
|
"epoch": 1.227149810708491,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004997064291184701,
|
|
"loss": 6.185,
|
|
"mean_token_accuracy": 0.14048560559749604,
|
|
"num_tokens": 2554225.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.11103253364563,
|
|
"epoch": 1.2325581395348837,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004996841173305691,
|
|
"loss": 6.0709,
|
|
"mean_token_accuracy": 0.15029430389404297,
|
|
"num_tokens": 2565050.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.197625541687012,
|
|
"epoch": 1.2379664683612763,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004996609890400665,
|
|
"loss": 6.139,
|
|
"mean_token_accuracy": 0.1492899015545845,
|
|
"num_tokens": 2576926.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.239265632629395,
|
|
"epoch": 1.243374797187669,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004996370443310107,
|
|
"loss": 6.2099,
|
|
"mean_token_accuracy": 0.14150406569242477,
|
|
"num_tokens": 2588775.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.0373786926269535,
|
|
"epoch": 1.2487831260140616,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004996122832904165,
|
|
"loss": 6.0548,
|
|
"mean_token_accuracy": 0.15066525042057038,
|
|
"num_tokens": 2600534.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.0793595790863035,
|
|
"epoch": 1.2541914548404542,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004995867060082658,
|
|
"loss": 5.8936,
|
|
"mean_token_accuracy": 0.15576101765036582,
|
|
"num_tokens": 2612350.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.0462251663208,
|
|
"epoch": 1.2595997836668469,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004995603125775065,
|
|
"loss": 6.0797,
|
|
"mean_token_accuracy": 0.1474672242999077,
|
|
"num_tokens": 2623871.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.06495213508606,
|
|
"epoch": 1.2650081124932395,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004995331030940523,
|
|
"loss": 6.0103,
|
|
"mean_token_accuracy": 0.1476856827735901,
|
|
"num_tokens": 2636371.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.206931161880493,
|
|
"epoch": 1.2704164413196322,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004995050776567826,
|
|
"loss": 6.0505,
|
|
"mean_token_accuracy": 0.15328404754400254,
|
|
"num_tokens": 2646794.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.004800415039062,
|
|
"epoch": 1.2758247701460248,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004994762363675421,
|
|
"loss": 6.0645,
|
|
"mean_token_accuracy": 0.15141474902629853,
|
|
"num_tokens": 2657105.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.092851257324218,
|
|
"epoch": 1.2812330989724177,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004994465793311396,
|
|
"loss": 6.0061,
|
|
"mean_token_accuracy": 0.15085877180099488,
|
|
"num_tokens": 2668297.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.058598804473877,
|
|
"epoch": 1.28664142779881,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004994161066553493,
|
|
"loss": 5.87,
|
|
"mean_token_accuracy": 0.15484228432178498,
|
|
"num_tokens": 2678834.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.071593379974365,
|
|
"epoch": 1.292049756625203,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004993848184509086,
|
|
"loss": 6.1074,
|
|
"mean_token_accuracy": 0.15071124136447905,
|
|
"num_tokens": 2690392.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 5.9669997215271,
|
|
"epoch": 1.2974580854515954,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004993527148315193,
|
|
"loss": 5.9083,
|
|
"mean_token_accuracy": 0.15376149863004684,
|
|
"num_tokens": 2701319.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.13273663520813,
|
|
"epoch": 1.3028664142779882,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004993197959138458,
|
|
"loss": 6.0755,
|
|
"mean_token_accuracy": 0.14556824415922165,
|
|
"num_tokens": 2713018.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.115740489959717,
|
|
"epoch": 1.3082747431043806,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004992860618175156,
|
|
"loss": 6.1823,
|
|
"mean_token_accuracy": 0.14297083020210266,
|
|
"num_tokens": 2724995.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.086751890182495,
|
|
"epoch": 1.3136830719307735,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004992515126651184,
|
|
"loss": 6.0153,
|
|
"mean_token_accuracy": 0.1531880423426628,
|
|
"num_tokens": 2736733.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.020717763900757,
|
|
"epoch": 1.319091400757166,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004992161485822059,
|
|
"loss": 5.9837,
|
|
"mean_token_accuracy": 0.15751376152038574,
|
|
"num_tokens": 2748436.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 5.990799331665039,
|
|
"epoch": 1.3244997295835588,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004991799696972913,
|
|
"loss": 6.0117,
|
|
"mean_token_accuracy": 0.15618861317634583,
|
|
"num_tokens": 2758924.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.203850746154785,
|
|
"epoch": 1.3299080584099512,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004991429761418488,
|
|
"loss": 6.0534,
|
|
"mean_token_accuracy": 0.15221881717443467,
|
|
"num_tokens": 2769773.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.185452604293824,
|
|
"epoch": 1.335316387236344,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.000499105168050313,
|
|
"loss": 6.2072,
|
|
"mean_token_accuracy": 0.15274276584386826,
|
|
"num_tokens": 2782414.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 5.985244703292847,
|
|
"epoch": 1.3407247160627367,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004990665455600787,
|
|
"loss": 5.9669,
|
|
"mean_token_accuracy": 0.15554977357387542,
|
|
"num_tokens": 2794325.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 5.946992588043213,
|
|
"epoch": 1.3461330448891293,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004990271088114999,
|
|
"loss": 5.8137,
|
|
"mean_token_accuracy": 0.1607433706521988,
|
|
"num_tokens": 2805391.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 5.921567630767822,
|
|
"epoch": 1.351541373715522,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004989868579478901,
|
|
"loss": 5.9462,
|
|
"mean_token_accuracy": 0.15325599461793898,
|
|
"num_tokens": 2816633.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.093569993972778,
|
|
"epoch": 1.3569497025419146,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004989457931155209,
|
|
"loss": 6.0643,
|
|
"mean_token_accuracy": 0.15207150727510452,
|
|
"num_tokens": 2828395.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.117164230346679,
|
|
"epoch": 1.3623580313683072,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004989039144636219,
|
|
"loss": 6.0333,
|
|
"mean_token_accuracy": 0.16129040718078613,
|
|
"num_tokens": 2839592.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.035921382904053,
|
|
"epoch": 1.3677663601946999,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004988612221443804,
|
|
"loss": 6.0125,
|
|
"mean_token_accuracy": 0.16414364278316498,
|
|
"num_tokens": 2849557.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.132221460342407,
|
|
"epoch": 1.3731746890210925,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004988177163129403,
|
|
"loss": 6.0556,
|
|
"mean_token_accuracy": 0.15453781485557555,
|
|
"num_tokens": 2862030.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 5.983911514282227,
|
|
"epoch": 1.3785830178474852,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004987733971274018,
|
|
"loss": 5.9904,
|
|
"mean_token_accuracy": 0.15506967157125473,
|
|
"num_tokens": 2872379.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.115880632400513,
|
|
"epoch": 1.3839913466738778,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004987282647488211,
|
|
"loss": 6.0727,
|
|
"mean_token_accuracy": 0.15147668570280076,
|
|
"num_tokens": 2883170.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.004611968994141,
|
|
"epoch": 1.3893996755002704,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004986823193412093,
|
|
"loss": 5.9904,
|
|
"mean_token_accuracy": 0.15152569562196733,
|
|
"num_tokens": 2894013.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.062729454040527,
|
|
"epoch": 1.394808004326663,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000498635561071532,
|
|
"loss": 6.0839,
|
|
"mean_token_accuracy": 0.1524193838238716,
|
|
"num_tokens": 2905073.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.174641942977905,
|
|
"epoch": 1.4002163331530557,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000498587990109709,
|
|
"loss": 6.0491,
|
|
"mean_token_accuracy": 0.15236106663942336,
|
|
"num_tokens": 2916579.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 5.917173624038696,
|
|
"epoch": 1.4056246619794484,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004985396066286133,
|
|
"loss": 5.9333,
|
|
"mean_token_accuracy": 0.15752357393503189,
|
|
"num_tokens": 2928522.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.14077205657959,
|
|
"epoch": 1.411032990805841,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004984904108040706,
|
|
"loss": 6.028,
|
|
"mean_token_accuracy": 0.15925881862640381,
|
|
"num_tokens": 2938754.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 5.8736213684082035,
|
|
"epoch": 1.4164413196322336,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004984404028148585,
|
|
"loss": 5.8263,
|
|
"mean_token_accuracy": 0.15991710126399994,
|
|
"num_tokens": 2949696.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.006898593902588,
|
|
"epoch": 1.4218496484586263,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004983895828427062,
|
|
"loss": 6.0224,
|
|
"mean_token_accuracy": 0.14963622838258744,
|
|
"num_tokens": 2961144.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.118567180633545,
|
|
"epoch": 1.427257977285019,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004983379510722935,
|
|
"loss": 6.0121,
|
|
"mean_token_accuracy": 0.1556963086128235,
|
|
"num_tokens": 2972368.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 5.935350704193115,
|
|
"epoch": 1.4326663061114115,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004982855076912506,
|
|
"loss": 5.9328,
|
|
"mean_token_accuracy": 0.1555210143327713,
|
|
"num_tokens": 2984853.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.173395776748658,
|
|
"epoch": 1.4380746349378042,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004982322528901568,
|
|
"loss": 6.0042,
|
|
"mean_token_accuracy": 0.15539158880710602,
|
|
"num_tokens": 2995095.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 5.931844806671142,
|
|
"epoch": 1.4434829637641968,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00049817818686254,
|
|
"loss": 6.0473,
|
|
"mean_token_accuracy": 0.15262697488069535,
|
|
"num_tokens": 3005926.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.168377923965454,
|
|
"epoch": 1.4488912925905895,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004981233098048765,
|
|
"loss": 6.0131,
|
|
"mean_token_accuracy": 0.15233441442251205,
|
|
"num_tokens": 3017647.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 5.899349927902222,
|
|
"epoch": 1.454299621416982,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004980676219165894,
|
|
"loss": 5.9181,
|
|
"mean_token_accuracy": 0.16069033443927766,
|
|
"num_tokens": 3028904.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.105036830902099,
|
|
"epoch": 1.4597079502433747,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000498011123400049,
|
|
"loss": 5.9608,
|
|
"mean_token_accuracy": 0.15869410187005997,
|
|
"num_tokens": 3039347.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 5.974474143981934,
|
|
"epoch": 1.4651162790697674,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004979538144605707,
|
|
"loss": 5.9358,
|
|
"mean_token_accuracy": 0.15985441207885742,
|
|
"num_tokens": 3049951.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.081779146194458,
|
|
"epoch": 1.47052460789616,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004978956953064154,
|
|
"loss": 5.9839,
|
|
"mean_token_accuracy": 0.15747965574264527,
|
|
"num_tokens": 3061894.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 5.99430022239685,
|
|
"epoch": 1.4759329367225527,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004978367661487883,
|
|
"loss": 5.9512,
|
|
"mean_token_accuracy": 0.16217859983444213,
|
|
"num_tokens": 3072393.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.011140203475952,
|
|
"epoch": 1.4813412655489453,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004977770272018379,
|
|
"loss": 5.9812,
|
|
"mean_token_accuracy": 0.1571822300553322,
|
|
"num_tokens": 3083541.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 5.916278409957886,
|
|
"epoch": 1.486749594375338,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000497716478682656,
|
|
"loss": 5.8185,
|
|
"mean_token_accuracy": 0.16223051249980927,
|
|
"num_tokens": 3093851.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 5.925797653198242,
|
|
"epoch": 1.4921579232017308,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004976551208112759,
|
|
"loss": 5.9093,
|
|
"mean_token_accuracy": 0.15642808079719545,
|
|
"num_tokens": 3105703.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.1236663341522215,
|
|
"epoch": 1.4975662520281232,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004975929538106721,
|
|
"loss": 6.0368,
|
|
"mean_token_accuracy": 0.1538296341896057,
|
|
"num_tokens": 3117063.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 5.93872652053833,
|
|
"epoch": 1.502974580854516,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004975299779067598,
|
|
"loss": 6.0016,
|
|
"mean_token_accuracy": 0.15524751394987107,
|
|
"num_tokens": 3129020.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.069183444976806,
|
|
"epoch": 1.5083829096809085,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004974661933283936,
|
|
"loss": 5.9968,
|
|
"mean_token_accuracy": 0.15477502197027207,
|
|
"num_tokens": 3140627.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 5.9552083015441895,
|
|
"epoch": 1.5137912385073014,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004974016003073667,
|
|
"loss": 6.0085,
|
|
"mean_token_accuracy": 0.15283810198307038,
|
|
"num_tokens": 3152294.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.043582010269165,
|
|
"epoch": 1.5191995673336938,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004973361990784104,
|
|
"loss": 6.0009,
|
|
"mean_token_accuracy": 0.15213296562433243,
|
|
"num_tokens": 3163770.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.062752628326416,
|
|
"epoch": 1.5246078961600866,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004972699898791927,
|
|
"loss": 5.954,
|
|
"mean_token_accuracy": 0.1555445075035095,
|
|
"num_tokens": 3175804.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.047053480148316,
|
|
"epoch": 1.530016224986479,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004972029729503182,
|
|
"loss": 5.9709,
|
|
"mean_token_accuracy": 0.15807701796293258,
|
|
"num_tokens": 3187233.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 5.909609460830689,
|
|
"epoch": 1.535424553812872,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004971351485353265,
|
|
"loss": 5.9326,
|
|
"mean_token_accuracy": 0.1669146314263344,
|
|
"num_tokens": 3198884.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 5.947835254669189,
|
|
"epoch": 1.5408328826392643,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004970665168806917,
|
|
"loss": 5.9024,
|
|
"mean_token_accuracy": 0.15887657999992372,
|
|
"num_tokens": 3209836.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 5.91135311126709,
|
|
"epoch": 1.5462412114656572,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004969970782358214,
|
|
"loss": 5.9287,
|
|
"mean_token_accuracy": 0.15876529216766358,
|
|
"num_tokens": 3221295.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 5.992637395858765,
|
|
"epoch": 1.5516495402920496,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004969268328530557,
|
|
"loss": 5.8827,
|
|
"mean_token_accuracy": 0.16646618098020555,
|
|
"num_tokens": 3231832.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 5.981706285476685,
|
|
"epoch": 1.5570578691184425,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004968557809876666,
|
|
"loss": 5.9673,
|
|
"mean_token_accuracy": 0.15562295764684678,
|
|
"num_tokens": 3243483.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 5.975745582580567,
|
|
"epoch": 1.562466197944835,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004967839228978568,
|
|
"loss": 6.0008,
|
|
"mean_token_accuracy": 0.15327105075120925,
|
|
"num_tokens": 3255785.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 5.991179180145264,
|
|
"epoch": 1.5678745267712277,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004967112588447585,
|
|
"loss": 5.899,
|
|
"mean_token_accuracy": 0.15747871696949006,
|
|
"num_tokens": 3267035.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.089357900619507,
|
|
"epoch": 1.5732828555976204,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004966377890924332,
|
|
"loss": 5.9754,
|
|
"mean_token_accuracy": 0.15453036725521088,
|
|
"num_tokens": 3277848.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 5.888553237915039,
|
|
"epoch": 1.578691184424013,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004965635139078704,
|
|
"loss": 5.8899,
|
|
"mean_token_accuracy": 0.16223473101854324,
|
|
"num_tokens": 3289864.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.076331996917725,
|
|
"epoch": 1.5840995132504057,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004964884335609857,
|
|
"loss": 6.0237,
|
|
"mean_token_accuracy": 0.15636843517422677,
|
|
"num_tokens": 3301441.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 5.889344882965088,
|
|
"epoch": 1.5895078420767983,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004964125483246218,
|
|
"loss": 5.9735,
|
|
"mean_token_accuracy": 0.15142715126276016,
|
|
"num_tokens": 3313216.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 5.914896535873413,
|
|
"epoch": 1.594916170903191,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004963358584745457,
|
|
"loss": 5.7908,
|
|
"mean_token_accuracy": 0.15975363105535506,
|
|
"num_tokens": 3323760.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 5.854299402236938,
|
|
"epoch": 1.6003244997295836,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004962583642894482,
|
|
"loss": 5.8661,
|
|
"mean_token_accuracy": 0.1571563720703125,
|
|
"num_tokens": 3334785.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.029454088211059,
|
|
"epoch": 1.6057328285559762,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004961800660509437,
|
|
"loss": 5.9356,
|
|
"mean_token_accuracy": 0.1598338857293129,
|
|
"num_tokens": 3345741.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 5.872513961791992,
|
|
"epoch": 1.6111411573823688,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000496100964043568,
|
|
"loss": 5.8937,
|
|
"mean_token_accuracy": 0.16037271469831466,
|
|
"num_tokens": 3356118.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 5.9037333011627195,
|
|
"epoch": 1.6165494862087615,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000496021058554778,
|
|
"loss": 5.8227,
|
|
"mean_token_accuracy": 0.15988235622644426,
|
|
"num_tokens": 3368107.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 5.955432462692261,
|
|
"epoch": 1.6219578150351541,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004959403498749505,
|
|
"loss": 5.8301,
|
|
"mean_token_accuracy": 0.1608538195490837,
|
|
"num_tokens": 3379367.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 5.827828025817871,
|
|
"epoch": 1.6273661438615468,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004958588382973809,
|
|
"loss": 5.9629,
|
|
"mean_token_accuracy": 0.15254009068012236,
|
|
"num_tokens": 3390597.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.047002983093262,
|
|
"epoch": 1.6327744726879394,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004957765241182826,
|
|
"loss": 5.8847,
|
|
"mean_token_accuracy": 0.16022928431630135,
|
|
"num_tokens": 3402618.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 5.813410902023316,
|
|
"epoch": 1.638182801514332,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004956934076367854,
|
|
"loss": 5.7949,
|
|
"mean_token_accuracy": 0.16659812033176422,
|
|
"num_tokens": 3413162.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 5.892169380187989,
|
|
"epoch": 1.6435911303407247,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000495609489154935,
|
|
"loss": 5.8696,
|
|
"mean_token_accuracy": 0.1670006960630417,
|
|
"num_tokens": 3424331.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 5.980246305465698,
|
|
"epoch": 1.6489994591671173,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004955247689776911,
|
|
"loss": 5.9862,
|
|
"mean_token_accuracy": 0.15606614649295808,
|
|
"num_tokens": 3434739.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.0311259746551515,
|
|
"epoch": 1.65440778799351,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000495439247412927,
|
|
"loss": 6.0628,
|
|
"mean_token_accuracy": 0.15121788680553436,
|
|
"num_tokens": 3445984.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 5.82039361000061,
|
|
"epoch": 1.6598161168199026,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004953529247714285,
|
|
"loss": 5.7785,
|
|
"mean_token_accuracy": 0.16936214864253998,
|
|
"num_tokens": 3456955.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.046815347671509,
|
|
"epoch": 1.6652244456462952,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004952658013668922,
|
|
"loss": 5.9455,
|
|
"mean_token_accuracy": 0.15965672582387924,
|
|
"num_tokens": 3468895.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 5.789427375793457,
|
|
"epoch": 1.670632774472688,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004951778775159246,
|
|
"loss": 5.834,
|
|
"mean_token_accuracy": 0.16249921619892121,
|
|
"num_tokens": 3480083.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.049218416213989,
|
|
"epoch": 1.6760411032990805,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004950891535380414,
|
|
"loss": 5.9877,
|
|
"mean_token_accuracy": 0.15260062366724014,
|
|
"num_tokens": 3492242.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 5.882490396499634,
|
|
"epoch": 1.6814494321254734,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004949996297556656,
|
|
"loss": 5.8069,
|
|
"mean_token_accuracy": 0.15946704596281053,
|
|
"num_tokens": 3501926.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 5.865342473983764,
|
|
"epoch": 1.6868577609518658,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004949093064941268,
|
|
"loss": 5.8727,
|
|
"mean_token_accuracy": 0.16740100383758544,
|
|
"num_tokens": 3511421.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 5.967023801803589,
|
|
"epoch": 1.6922660897782587,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004948181840816601,
|
|
"loss": 5.8197,
|
|
"mean_token_accuracy": 0.16478042155504227,
|
|
"num_tokens": 3522033.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 5.864649152755737,
|
|
"epoch": 1.697674418604651,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004947262628494045,
|
|
"loss": 5.8663,
|
|
"mean_token_accuracy": 0.16560869961977004,
|
|
"num_tokens": 3533232.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 5.953764629364014,
|
|
"epoch": 1.703082747431044,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004946335431314018,
|
|
"loss": 5.8824,
|
|
"mean_token_accuracy": 0.16067960858345032,
|
|
"num_tokens": 3544904.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 5.916748809814453,
|
|
"epoch": 1.7084910762574363,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004945400252645959,
|
|
"loss": 5.9541,
|
|
"mean_token_accuracy": 0.1579635426402092,
|
|
"num_tokens": 3556545.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 5.92917685508728,
|
|
"epoch": 1.7138994050838292,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004944457095888308,
|
|
"loss": 5.832,
|
|
"mean_token_accuracy": 0.1624446004629135,
|
|
"num_tokens": 3567361.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 5.948728799819946,
|
|
"epoch": 1.7193077339102216,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00049435059644685,
|
|
"loss": 5.8611,
|
|
"mean_token_accuracy": 0.16280075013637543,
|
|
"num_tokens": 3578943.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 5.911354494094849,
|
|
"epoch": 1.7247160627366145,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004942546861842947,
|
|
"loss": 5.841,
|
|
"mean_token_accuracy": 0.1613238736987114,
|
|
"num_tokens": 3589868.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 5.8840008735656735,
|
|
"epoch": 1.730124391563007,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004941579791497032,
|
|
"loss": 5.8367,
|
|
"mean_token_accuracy": 0.16888057589530944,
|
|
"num_tokens": 3601245.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 5.912482261657715,
|
|
"epoch": 1.7355327203893998,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004940604756945091,
|
|
"loss": 5.8903,
|
|
"mean_token_accuracy": 0.16273186802864076,
|
|
"num_tokens": 3612168.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 5.821870422363281,
|
|
"epoch": 1.7409410492157922,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004939621761730399,
|
|
"loss": 5.8132,
|
|
"mean_token_accuracy": 0.16740759909152986,
|
|
"num_tokens": 3624699.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 5.993101644515991,
|
|
"epoch": 1.746349378042185,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004938630809425166,
|
|
"loss": 5.9623,
|
|
"mean_token_accuracy": 0.16454029828310013,
|
|
"num_tokens": 3635554.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 5.804750585556031,
|
|
"epoch": 1.7517577068685775,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004937631903630513,
|
|
"loss": 5.8473,
|
|
"mean_token_accuracy": 0.16223347187042236,
|
|
"num_tokens": 3646776.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 5.927067852020263,
|
|
"epoch": 1.7571660356949703,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004936625047976468,
|
|
"loss": 5.7819,
|
|
"mean_token_accuracy": 0.16398447155952453,
|
|
"num_tokens": 3657976.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 5.910075044631958,
|
|
"epoch": 1.7625743645213627,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004935610246121944,
|
|
"loss": 5.9229,
|
|
"mean_token_accuracy": 0.1553028106689453,
|
|
"num_tokens": 3668242.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 5.9981360912323,
|
|
"epoch": 1.7679826933477556,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004934587501754736,
|
|
"loss": 5.9479,
|
|
"mean_token_accuracy": 0.15933531075716018,
|
|
"num_tokens": 3679200.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 5.878190422058106,
|
|
"epoch": 1.7733910221741482,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004933556818591498,
|
|
"loss": 5.8575,
|
|
"mean_token_accuracy": 0.16800351589918136,
|
|
"num_tokens": 3690227.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 5.968534517288208,
|
|
"epoch": 1.7787993510005409,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004932518200377736,
|
|
"loss": 5.8778,
|
|
"mean_token_accuracy": 0.16286956816911696,
|
|
"num_tokens": 3701821.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 5.8273231983184814,
|
|
"epoch": 1.7842076798269335,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004931471650887789,
|
|
"loss": 5.8132,
|
|
"mean_token_accuracy": 0.15990324765443803,
|
|
"num_tokens": 3712420.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 5.953809309005737,
|
|
"epoch": 1.7896160086533262,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004930417173924822,
|
|
"loss": 5.8714,
|
|
"mean_token_accuracy": 0.16492005437612534,
|
|
"num_tokens": 3722211.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 5.8503564357757565,
|
|
"epoch": 1.7950243374797188,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004929354773320806,
|
|
"loss": 5.8493,
|
|
"mean_token_accuracy": 0.15776370763778685,
|
|
"num_tokens": 3732822.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 5.932883930206299,
|
|
"epoch": 1.8004326663061114,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004928284452936506,
|
|
"loss": 5.9055,
|
|
"mean_token_accuracy": 0.16148029118776322,
|
|
"num_tokens": 3744262.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 5.863234043121338,
|
|
"epoch": 1.805840995132504,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004927206216661473,
|
|
"loss": 5.7948,
|
|
"mean_token_accuracy": 0.16473406851291655,
|
|
"num_tokens": 3755713.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 5.815153503417969,
|
|
"epoch": 1.8112493239588967,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004926120068414015,
|
|
"loss": 5.8093,
|
|
"mean_token_accuracy": 0.16667513251304628,
|
|
"num_tokens": 3767179.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 5.911891984939575,
|
|
"epoch": 1.8166576527852893,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004925026012141201,
|
|
"loss": 5.8399,
|
|
"mean_token_accuracy": 0.16335198730230333,
|
|
"num_tokens": 3777714.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 5.9336024761199955,
|
|
"epoch": 1.822065981611682,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004923924051818832,
|
|
"loss": 5.8169,
|
|
"mean_token_accuracy": 0.16608950793743132,
|
|
"num_tokens": 3788092.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 5.831817102432251,
|
|
"epoch": 1.8274743104380746,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004922814191451435,
|
|
"loss": 5.788,
|
|
"mean_token_accuracy": 0.15890535041689874,
|
|
"num_tokens": 3798080.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 5.8690248966217045,
|
|
"epoch": 1.8328826392644673,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004921696435072245,
|
|
"loss": 5.8303,
|
|
"mean_token_accuracy": 0.16325227320194244,
|
|
"num_tokens": 3809547.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 5.913704919815063,
|
|
"epoch": 1.83829096809086,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004920570786743191,
|
|
"loss": 5.7978,
|
|
"mean_token_accuracy": 0.1684496209025383,
|
|
"num_tokens": 3820681.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 5.85491247177124,
|
|
"epoch": 1.8436992969172525,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004919437250554882,
|
|
"loss": 5.8169,
|
|
"mean_token_accuracy": 0.16609950959682465,
|
|
"num_tokens": 3831598.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 5.876249885559082,
|
|
"epoch": 1.8491076257436452,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004918295830626592,
|
|
"loss": 5.8679,
|
|
"mean_token_accuracy": 0.16632142812013626,
|
|
"num_tokens": 3843437.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 5.820015001296997,
|
|
"epoch": 1.8545159545700378,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004917146531106242,
|
|
"loss": 5.7286,
|
|
"mean_token_accuracy": 0.16300380676984788,
|
|
"num_tokens": 3855387.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 5.875252771377563,
|
|
"epoch": 1.8599242833964305,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004915989356170389,
|
|
"loss": 5.8909,
|
|
"mean_token_accuracy": 0.15887996852397918,
|
|
"num_tokens": 3866999.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 5.875708103179932,
|
|
"epoch": 1.865332612222823,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004914824310024213,
|
|
"loss": 5.8203,
|
|
"mean_token_accuracy": 0.16333778202533722,
|
|
"num_tokens": 3877721.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 5.892795133590698,
|
|
"epoch": 1.870740941049216,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004913651396901489,
|
|
"loss": 5.904,
|
|
"mean_token_accuracy": 0.15644253939390182,
|
|
"num_tokens": 3888831.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 5.862974405288696,
|
|
"epoch": 1.8761492698756084,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000491247062106459,
|
|
"loss": 5.8355,
|
|
"mean_token_accuracy": 0.16764473170042038,
|
|
"num_tokens": 3900389.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 5.816805601119995,
|
|
"epoch": 1.8815575987020012,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004911281986804455,
|
|
"loss": 5.7994,
|
|
"mean_token_accuracy": 0.16391095370054246,
|
|
"num_tokens": 3911641.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 5.739569568634034,
|
|
"epoch": 1.8869659275283936,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004910085498440587,
|
|
"loss": 5.6951,
|
|
"mean_token_accuracy": 0.17215199172496795,
|
|
"num_tokens": 3922877.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 5.873764610290527,
|
|
"epoch": 1.8923742563547865,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004908881160321025,
|
|
"loss": 5.7648,
|
|
"mean_token_accuracy": 0.16693327724933624,
|
|
"num_tokens": 3933573.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.813666725158692,
|
|
"epoch": 1.897782585181179,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004907668976822338,
|
|
"loss": 5.7148,
|
|
"mean_token_accuracy": 0.17644793689250945,
|
|
"num_tokens": 3944339.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 5.917697858810425,
|
|
"epoch": 1.9031909140075718,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004906448952349603,
|
|
"loss": 5.8139,
|
|
"mean_token_accuracy": 0.17634026259183883,
|
|
"num_tokens": 3955459.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.686581373214722,
|
|
"epoch": 1.9085992428339642,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004905221091336392,
|
|
"loss": 5.7258,
|
|
"mean_token_accuracy": 0.17054691165685654,
|
|
"num_tokens": 3966782.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.892930173873902,
|
|
"epoch": 1.914007571660357,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004903985398244753,
|
|
"loss": 5.8449,
|
|
"mean_token_accuracy": 0.16712595522403717,
|
|
"num_tokens": 3978794.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 5.912125539779663,
|
|
"epoch": 1.9194159004867495,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004902741877565201,
|
|
"loss": 5.8571,
|
|
"mean_token_accuracy": 0.166175577044487,
|
|
"num_tokens": 3990838.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.915352249145508,
|
|
"epoch": 1.9248242293131423,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000490149053381669,
|
|
"loss": 5.8847,
|
|
"mean_token_accuracy": 0.16127822399139405,
|
|
"num_tokens": 4000630.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.844710636138916,
|
|
"epoch": 1.9302325581395348,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004900231371546608,
|
|
"loss": 5.7494,
|
|
"mean_token_accuracy": 0.1685309126973152,
|
|
"num_tokens": 4011433.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.857338094711304,
|
|
"epoch": 1.9356408869659276,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004898964395330753,
|
|
"loss": 5.9263,
|
|
"mean_token_accuracy": 0.16273758560419083,
|
|
"num_tokens": 4023671.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 5.827757453918457,
|
|
"epoch": 1.94104921579232,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004897689609773318,
|
|
"loss": 5.6931,
|
|
"mean_token_accuracy": 0.17371835857629775,
|
|
"num_tokens": 4034686.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 5.910330009460449,
|
|
"epoch": 1.946457544618713,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004896407019506879,
|
|
"loss": 5.8134,
|
|
"mean_token_accuracy": 0.1701274260878563,
|
|
"num_tokens": 4046387.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.763359785079956,
|
|
"epoch": 1.9518658734451053,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004895116629192371,
|
|
"loss": 5.7232,
|
|
"mean_token_accuracy": 0.16949945241212844,
|
|
"num_tokens": 4056103.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.772169733047486,
|
|
"epoch": 1.9572742022714982,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004893818443519074,
|
|
"loss": 5.6889,
|
|
"mean_token_accuracy": 0.1775139167904854,
|
|
"num_tokens": 4066940.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.768274974822998,
|
|
"epoch": 1.9626825310978906,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048925124672046,
|
|
"loss": 5.8384,
|
|
"mean_token_accuracy": 0.17425315380096434,
|
|
"num_tokens": 4077232.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.860826015472412,
|
|
"epoch": 1.9680908599242835,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004891198704994868,
|
|
"loss": 5.8813,
|
|
"mean_token_accuracy": 0.16128703355789184,
|
|
"num_tokens": 4089322.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.878625535964966,
|
|
"epoch": 1.973499188750676,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004889877161664096,
|
|
"loss": 5.7291,
|
|
"mean_token_accuracy": 0.1727048560976982,
|
|
"num_tokens": 4099737.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.752609014511108,
|
|
"epoch": 1.9789075175770687,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004888547842014771,
|
|
"loss": 5.729,
|
|
"mean_token_accuracy": 0.1681004837155342,
|
|
"num_tokens": 4110793.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.841432809829712,
|
|
"epoch": 1.9843158464034614,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004887210750877648,
|
|
"loss": 5.7709,
|
|
"mean_token_accuracy": 0.17114366441965104,
|
|
"num_tokens": 4121481.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.7735851287841795,
|
|
"epoch": 1.989724175229854,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004885865893111716,
|
|
"loss": 5.7205,
|
|
"mean_token_accuracy": 0.1756225749850273,
|
|
"num_tokens": 4132701.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.835188293457032,
|
|
"epoch": 1.9951325040562466,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004884513273604195,
|
|
"loss": 5.8101,
|
|
"mean_token_accuracy": 0.16929660588502884,
|
|
"num_tokens": 4143764.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 5.829058435228136,
|
|
"epoch": 2.0,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 0.0004883152897270502,
|
|
"loss": 5.7603,
|
|
"mean_token_accuracy": 0.16963467664188808,
|
|
"num_tokens": 4153778.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.856991720199585,
|
|
"epoch": 2.005408328826393,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004881784769054253,
|
|
"loss": 5.486,
|
|
"mean_token_accuracy": 0.18024396151304245,
|
|
"num_tokens": 4165099.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.778524732589721,
|
|
"epoch": 2.0108166576527853,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004880408893927226,
|
|
"loss": 5.5556,
|
|
"mean_token_accuracy": 0.1812357112765312,
|
|
"num_tokens": 4178051.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.7850189208984375,
|
|
"epoch": 2.016224986479178,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004879025276889355,
|
|
"loss": 5.4706,
|
|
"mean_token_accuracy": 0.18264889866113662,
|
|
"num_tokens": 4189987.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.793973731994629,
|
|
"epoch": 2.0216333153055706,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048776339229687077,
|
|
"loss": 5.5044,
|
|
"mean_token_accuracy": 0.17972788512706755,
|
|
"num_tokens": 4200999.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.673948478698731,
|
|
"epoch": 2.0270416441319634,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048762348372214683,
|
|
"loss": 5.4365,
|
|
"mean_token_accuracy": 0.1799351304769516,
|
|
"num_tokens": 4212122.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.7424413681030275,
|
|
"epoch": 2.032449972958356,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00048748280247319165,
|
|
"loss": 5.5044,
|
|
"mean_token_accuracy": 0.17869459688663483,
|
|
"num_tokens": 4222580.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.69903712272644,
|
|
"epoch": 2.0378583017847487,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004873413490612412,
|
|
"loss": 5.37,
|
|
"mean_token_accuracy": 0.18325228095054627,
|
|
"num_tokens": 4233756.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.661659097671508,
|
|
"epoch": 2.043266630611141,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048719912400033757,
|
|
"loss": 5.4877,
|
|
"mean_token_accuracy": 0.1774522915482521,
|
|
"num_tokens": 4245625.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.853540754318237,
|
|
"epoch": 2.048674959437534,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000487056127807327,
|
|
"loss": 5.702,
|
|
"mean_token_accuracy": 0.17503489851951598,
|
|
"num_tokens": 4257788.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.652139139175415,
|
|
"epoch": 2.0540832882639264,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048691236100185795,
|
|
"loss": 5.4438,
|
|
"mean_token_accuracy": 0.18356238156557084,
|
|
"num_tokens": 4269641.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.770523118972778,
|
|
"epoch": 2.0594916170903192,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004867678241063792,
|
|
"loss": 5.5013,
|
|
"mean_token_accuracy": 0.18512098640203475,
|
|
"num_tokens": 4280952.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.735342502593994,
|
|
"epoch": 2.0648999459167117,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00048662251764613833,
|
|
"loss": 5.4501,
|
|
"mean_token_accuracy": 0.18542484045028687,
|
|
"num_tokens": 4294371.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.59959306716919,
|
|
"epoch": 2.0703082747431045,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004864764421491793,
|
|
"loss": 5.4496,
|
|
"mean_token_accuracy": 0.1797779008746147,
|
|
"num_tokens": 4307112.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.744608974456787,
|
|
"epoch": 2.075716603569497,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004863295981463408,
|
|
"loss": 5.4332,
|
|
"mean_token_accuracy": 0.184219653904438,
|
|
"num_tokens": 4317199.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.5987160205841064,
|
|
"epoch": 2.08112493239589,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004861819861712543,
|
|
"loss": 5.4411,
|
|
"mean_token_accuracy": 0.18531058728694916,
|
|
"num_tokens": 4328546.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.83209547996521,
|
|
"epoch": 2.086533261222282,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048603360676034203,
|
|
"loss": 5.5093,
|
|
"mean_token_accuracy": 0.1795883372426033,
|
|
"num_tokens": 4338795.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.602935171127319,
|
|
"epoch": 2.091941590048675,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048588446045281505,
|
|
"loss": 5.4799,
|
|
"mean_token_accuracy": 0.18391085267066956,
|
|
"num_tokens": 4350373.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.669620132446289,
|
|
"epoch": 2.0973499188750675,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004857345477906714,
|
|
"loss": 5.4479,
|
|
"mean_token_accuracy": 0.1810387909412384,
|
|
"num_tokens": 4362801.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.751356840133667,
|
|
"epoch": 2.1027582477014604,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00048558386931869417,
|
|
"loss": 5.5632,
|
|
"mean_token_accuracy": 0.18413745760917663,
|
|
"num_tokens": 4375782.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.705285263061524,
|
|
"epoch": 2.1081665765278528,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004854324255844492,
|
|
"loss": 5.464,
|
|
"mean_token_accuracy": 0.18208637982606887,
|
|
"num_tokens": 4385580.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.681854009628296,
|
|
"epoch": 2.1135749053542456,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048528021713828336,
|
|
"loss": 5.4699,
|
|
"mean_token_accuracy": 0.17763171941041947,
|
|
"num_tokens": 4397434.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.700517749786377,
|
|
"epoch": 2.118983234180638,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004851272445333225,
|
|
"loss": 5.5247,
|
|
"mean_token_accuracy": 0.17917665839195251,
|
|
"num_tokens": 4407888.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.655262994766235,
|
|
"epoch": 2.124391563007031,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00048497350832546937,
|
|
"loss": 5.4313,
|
|
"mean_token_accuracy": 0.1841058224439621,
|
|
"num_tokens": 4420010.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.627641201019287,
|
|
"epoch": 2.1297998918334233,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004848190090734018,
|
|
"loss": 5.4768,
|
|
"mean_token_accuracy": 0.17749007493257524,
|
|
"num_tokens": 4430730.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.67692461013794,
|
|
"epoch": 2.135208220659816,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00048466374733857035,
|
|
"loss": 5.4562,
|
|
"mean_token_accuracy": 0.19095246493816376,
|
|
"num_tokens": 4442296.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.707493305206299,
|
|
"epoch": 2.1406165494862086,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004845077236851966,
|
|
"loss": 5.4981,
|
|
"mean_token_accuracy": 0.1773895278573036,
|
|
"num_tokens": 4453996.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.618096017837525,
|
|
"epoch": 2.1460248783126015,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00048435093868027076,
|
|
"loss": 5.4857,
|
|
"mean_token_accuracy": 0.18061366826295852,
|
|
"num_tokens": 4465779.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.600418424606323,
|
|
"epoch": 2.151433207138994,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00048419339289354995,
|
|
"loss": 5.4293,
|
|
"mean_token_accuracy": 0.18757479786872863,
|
|
"num_tokens": 4477113.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.671384191513061,
|
|
"epoch": 2.1568415359653867,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000484035086897556,
|
|
"loss": 5.3672,
|
|
"mean_token_accuracy": 0.184905806183815,
|
|
"num_tokens": 4487920.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.807044839859008,
|
|
"epoch": 2.162249864791779,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004838760212675732,
|
|
"loss": 5.5772,
|
|
"mean_token_accuracy": 0.17598584443330764,
|
|
"num_tokens": 4499049.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 2.162249864791779,
|
|
"eval_entropy": 5.628456380150535,
|
|
"eval_loss": 6.069297790527344,
|
|
"eval_mean_token_accuracy": 0.1626782597127286,
|
|
"eval_num_tokens": 4499049.0,
|
|
"eval_runtime": 2.7514,
|
|
"eval_samples_per_second": 1277.509,
|
|
"eval_steps_per_second": 159.916,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.6429602146148685,
|
|
"epoch": 2.167658193618172,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004837161965816464,
|
|
"loss": 5.4098,
|
|
"mean_token_accuracy": 0.18215955942869186,
|
|
"num_tokens": 4508944.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.63345422744751,
|
|
"epoch": 2.1730665224445644,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048355561342057886,
|
|
"loss": 5.4382,
|
|
"mean_token_accuracy": 0.17879818826913835,
|
|
"num_tokens": 4518845.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.54923300743103,
|
|
"epoch": 2.1784748512709573,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048339427236793024,
|
|
"loss": 5.4814,
|
|
"mean_token_accuracy": 0.1791173204779625,
|
|
"num_tokens": 4530416.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.702573633193969,
|
|
"epoch": 2.1838831800973497,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004832321740100143,
|
|
"loss": 5.4608,
|
|
"mean_token_accuracy": 0.17992149740457536,
|
|
"num_tokens": 4542630.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.578261470794677,
|
|
"epoch": 2.1892915089237426,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048306931893589666,
|
|
"loss": 5.4536,
|
|
"mean_token_accuracy": 0.17908935695886613,
|
|
"num_tokens": 4553270.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.562483406066894,
|
|
"epoch": 2.194699837750135,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048290570773739327,
|
|
"loss": 5.3244,
|
|
"mean_token_accuracy": 0.18711396157741547,
|
|
"num_tokens": 4564647.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.697152614593506,
|
|
"epoch": 2.200108166576528,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00048274134100906747,
|
|
"loss": 5.5094,
|
|
"mean_token_accuracy": 0.17573244422674178,
|
|
"num_tokens": 4575507.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.6559234142303465,
|
|
"epoch": 2.2055164954029207,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00048257621934822835,
|
|
"loss": 5.5022,
|
|
"mean_token_accuracy": 0.18520522266626357,
|
|
"num_tokens": 4586761.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.638587379455567,
|
|
"epoch": 2.210924824229313,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004824103433549285,
|
|
"loss": 5.4214,
|
|
"mean_token_accuracy": 0.18610170632600784,
|
|
"num_tokens": 4597689.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.637989330291748,
|
|
"epoch": 2.2163331530557056,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004822437136319616,
|
|
"loss": 5.4532,
|
|
"mean_token_accuracy": 0.1798434376716614,
|
|
"num_tokens": 4608886.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.64138617515564,
|
|
"epoch": 2.2217414818820984,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004820763307848604,
|
|
"loss": 5.5111,
|
|
"mean_token_accuracy": 0.17216380089521408,
|
|
"num_tokens": 4619989.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.634832429885864,
|
|
"epoch": 2.2271498107084913,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048190819542189456,
|
|
"loss": 5.3853,
|
|
"mean_token_accuracy": 0.18451308310031891,
|
|
"num_tokens": 4630218.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.589566421508789,
|
|
"epoch": 2.2325581395348837,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00048173930815406847,
|
|
"loss": 5.491,
|
|
"mean_token_accuracy": 0.17246077656745912,
|
|
"num_tokens": 4642024.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.67005763053894,
|
|
"epoch": 2.2379664683612766,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004815696695951187,
|
|
"loss": 5.486,
|
|
"mean_token_accuracy": 0.1855996698141098,
|
|
"num_tokens": 4653370.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.643704175949097,
|
|
"epoch": 2.243374797187669,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004813992803615123,
|
|
"loss": 5.4836,
|
|
"mean_token_accuracy": 0.18200527578592302,
|
|
"num_tokens": 4664522.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.648378610610962,
|
|
"epoch": 2.248783126014062,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048122814107244414,
|
|
"loss": 5.5186,
|
|
"mean_token_accuracy": 0.18285816609859468,
|
|
"num_tokens": 4676542.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.590424489974976,
|
|
"epoch": 2.2541914548404542,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048105625234983466,
|
|
"loss": 5.451,
|
|
"mean_token_accuracy": 0.17852285504341125,
|
|
"num_tokens": 4687810.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.68006272315979,
|
|
"epoch": 2.259599783666847,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048088361481832807,
|
|
"loss": 5.4326,
|
|
"mean_token_accuracy": 0.1851690724492073,
|
|
"num_tokens": 4697661.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.5374799251556395,
|
|
"epoch": 2.2650081124932395,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048071022910528934,
|
|
"loss": 5.3807,
|
|
"mean_token_accuracy": 0.18470762223005294,
|
|
"num_tokens": 4708585.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.733445072174073,
|
|
"epoch": 2.2704164413196324,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00048053609584080274,
|
|
"loss": 5.5451,
|
|
"mean_token_accuracy": 0.18038288205862046,
|
|
"num_tokens": 4719848.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.6617279052734375,
|
|
"epoch": 2.275824770146025,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048036121565766876,
|
|
"loss": 5.5152,
|
|
"mean_token_accuracy": 0.1745545193552971,
|
|
"num_tokens": 4731474.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.642642641067505,
|
|
"epoch": 2.2812330989724177,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004801855891914026,
|
|
"loss": 5.4799,
|
|
"mean_token_accuracy": 0.18345258384943008,
|
|
"num_tokens": 4743081.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.664417552947998,
|
|
"epoch": 2.28664142779881,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048000921708023117,
|
|
"loss": 5.4689,
|
|
"mean_token_accuracy": 0.1781705603003502,
|
|
"num_tokens": 4754829.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.574846839904785,
|
|
"epoch": 2.292049756625203,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004798320999650911,
|
|
"loss": 5.4233,
|
|
"mean_token_accuracy": 0.18039131462574004,
|
|
"num_tokens": 4765522.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.601164245605469,
|
|
"epoch": 2.2974580854515954,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047965423848962636,
|
|
"loss": 5.4171,
|
|
"mean_token_accuracy": 0.18348476886749268,
|
|
"num_tokens": 4776246.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.6312909603118895,
|
|
"epoch": 2.302866414277988,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000479475633300186,
|
|
"loss": 5.4929,
|
|
"mean_token_accuracy": 0.17429995983839036,
|
|
"num_tokens": 4787595.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.706297969818115,
|
|
"epoch": 2.3082747431043806,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047929628504582173,
|
|
"loss": 5.5313,
|
|
"mean_token_accuracy": 0.1716928079724312,
|
|
"num_tokens": 4798877.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.582214784622193,
|
|
"epoch": 2.3136830719307735,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004791161943782856,
|
|
"loss": 5.5364,
|
|
"mean_token_accuracy": 0.1815814644098282,
|
|
"num_tokens": 4809542.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.790981912612915,
|
|
"epoch": 2.319091400757166,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047893536195202726,
|
|
"loss": 5.6481,
|
|
"mean_token_accuracy": 0.17538082748651504,
|
|
"num_tokens": 4820982.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.528646087646484,
|
|
"epoch": 2.3244997295835588,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004787537884241923,
|
|
"loss": 5.477,
|
|
"mean_token_accuracy": 0.17634275555610657,
|
|
"num_tokens": 4831543.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.841570663452148,
|
|
"epoch": 2.329908058409951,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047857147445461943,
|
|
"loss": 5.5496,
|
|
"mean_token_accuracy": 0.17713170051574706,
|
|
"num_tokens": 4843221.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.542124271392822,
|
|
"epoch": 2.335316387236344,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004783884207058379,
|
|
"loss": 5.4088,
|
|
"mean_token_accuracy": 0.1824680134654045,
|
|
"num_tokens": 4855201.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.695831632614135,
|
|
"epoch": 2.3407247160627365,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047820462784306555,
|
|
"loss": 5.5099,
|
|
"mean_token_accuracy": 0.18193862736225128,
|
|
"num_tokens": 4866387.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.5891622543334964,
|
|
"epoch": 2.3461330448891293,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000478020096534206,
|
|
"loss": 5.415,
|
|
"mean_token_accuracy": 0.18357495367527008,
|
|
"num_tokens": 4877702.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.584392738342285,
|
|
"epoch": 2.3515413737155217,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047783482744984644,
|
|
"loss": 5.4943,
|
|
"mean_token_accuracy": 0.18229788690805435,
|
|
"num_tokens": 4888652.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.713996124267578,
|
|
"epoch": 2.3569497025419146,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00047764882126325527,
|
|
"loss": 5.588,
|
|
"mean_token_accuracy": 0.1755495101213455,
|
|
"num_tokens": 4900995.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.578362941741943,
|
|
"epoch": 2.362358031368307,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004774620786503793,
|
|
"loss": 5.3624,
|
|
"mean_token_accuracy": 0.1858825668692589,
|
|
"num_tokens": 4912173.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.512315082550049,
|
|
"epoch": 2.3677663601947,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004772746002898417,
|
|
"loss": 5.4069,
|
|
"mean_token_accuracy": 0.1850082829594612,
|
|
"num_tokens": 4923149.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.559813117980957,
|
|
"epoch": 2.3731746890210923,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00047708638686293903,
|
|
"loss": 5.3973,
|
|
"mean_token_accuracy": 0.1855332985520363,
|
|
"num_tokens": 4934156.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.686374473571777,
|
|
"epoch": 2.378583017847485,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047689743905363963,
|
|
"loss": 5.5933,
|
|
"mean_token_accuracy": 0.17512153834104538,
|
|
"num_tokens": 4946690.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.654055118560791,
|
|
"epoch": 2.383991346673878,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004767077575485801,
|
|
"loss": 5.497,
|
|
"mean_token_accuracy": 0.18948966264724731,
|
|
"num_tokens": 4958208.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.5970635414123535,
|
|
"epoch": 2.3893996755002704,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004765173430370635,
|
|
"loss": 5.4445,
|
|
"mean_token_accuracy": 0.1833168238401413,
|
|
"num_tokens": 4968815.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.595647621154785,
|
|
"epoch": 2.394808004326663,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004763261962110567,
|
|
"loss": 5.4336,
|
|
"mean_token_accuracy": 0.1873726263642311,
|
|
"num_tokens": 4979468.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.691272306442261,
|
|
"epoch": 2.4002163331530557,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004761343177651875,
|
|
"loss": 5.5261,
|
|
"mean_token_accuracy": 0.17809998840093613,
|
|
"num_tokens": 4990841.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.678408813476563,
|
|
"epoch": 2.4056246619794486,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004759417083967428,
|
|
"loss": 5.5493,
|
|
"mean_token_accuracy": 0.18093332946300505,
|
|
"num_tokens": 5001567.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.563894844055175,
|
|
"epoch": 2.411032990805841,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047574836880566544,
|
|
"loss": 5.428,
|
|
"mean_token_accuracy": 0.18067546039819718,
|
|
"num_tokens": 5013219.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.633818292617798,
|
|
"epoch": 2.4164413196322334,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004755542996945519,
|
|
"loss": 5.5546,
|
|
"mean_token_accuracy": 0.17981842160224915,
|
|
"num_tokens": 5025771.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.616273212432861,
|
|
"epoch": 2.4218496484586263,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047535950176864976,
|
|
"loss": 5.5071,
|
|
"mean_token_accuracy": 0.17907274514436722,
|
|
"num_tokens": 5038205.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.669723320007324,
|
|
"epoch": 2.427257977285019,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047516397573585525,
|
|
"loss": 5.4747,
|
|
"mean_token_accuracy": 0.18287517577409745,
|
|
"num_tokens": 5048428.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.602979421615601,
|
|
"epoch": 2.4326663061114115,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047496772230671034,
|
|
"loss": 5.4459,
|
|
"mean_token_accuracy": 0.1822010099887848,
|
|
"num_tokens": 5058391.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.579944515228272,
|
|
"epoch": 2.4380746349378044,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004747707421944004,
|
|
"loss": 5.4748,
|
|
"mean_token_accuracy": 0.17697897404432297,
|
|
"num_tokens": 5069832.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.668799543380738,
|
|
"epoch": 2.443482963764197,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004745730361147517,
|
|
"loss": 5.4587,
|
|
"mean_token_accuracy": 0.18840455412864685,
|
|
"num_tokens": 5080198.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.574374055862426,
|
|
"epoch": 2.4488912925905897,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004743746047862286,
|
|
"loss": 5.5206,
|
|
"mean_token_accuracy": 0.18029489666223525,
|
|
"num_tokens": 5090817.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.607488918304443,
|
|
"epoch": 2.454299621416982,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004741754489299309,
|
|
"loss": 5.4145,
|
|
"mean_token_accuracy": 0.17971412986516952,
|
|
"num_tokens": 5102699.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.553569746017456,
|
|
"epoch": 2.459707950243375,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047397556926959155,
|
|
"loss": 5.4153,
|
|
"mean_token_accuracy": 0.18327369838953017,
|
|
"num_tokens": 5113922.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.656823253631591,
|
|
"epoch": 2.4651162790697674,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047377496653157374,
|
|
"loss": 5.5118,
|
|
"mean_token_accuracy": 0.18528055250644684,
|
|
"num_tokens": 5124920.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.573303079605102,
|
|
"epoch": 2.4705246078961602,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047357364144486806,
|
|
"loss": 5.5047,
|
|
"mean_token_accuracy": 0.17224933803081513,
|
|
"num_tokens": 5136304.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.627311325073242,
|
|
"epoch": 2.4759329367225527,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00047337159474109057,
|
|
"loss": 5.606,
|
|
"mean_token_accuracy": 0.17537178844213486,
|
|
"num_tokens": 5148101.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.717777252197266,
|
|
"epoch": 2.4813412655489455,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00047316882715447934,
|
|
"loss": 5.5067,
|
|
"mean_token_accuracy": 0.1809531882405281,
|
|
"num_tokens": 5159186.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.55581955909729,
|
|
"epoch": 2.486749594375338,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004729653394218922,
|
|
"loss": 5.4792,
|
|
"mean_token_accuracy": 0.18029095977544785,
|
|
"num_tokens": 5170542.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.630924320220947,
|
|
"epoch": 2.492157923201731,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00047276113228280427,
|
|
"loss": 5.5537,
|
|
"mean_token_accuracy": 0.17482594549655914,
|
|
"num_tokens": 5182685.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.652374315261841,
|
|
"epoch": 2.497566252028123,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004725562064793045,
|
|
"loss": 5.3968,
|
|
"mean_token_accuracy": 0.18380746692419053,
|
|
"num_tokens": 5192622.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.642563915252685,
|
|
"epoch": 2.502974580854516,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047235056275609397,
|
|
"loss": 5.5646,
|
|
"mean_token_accuracy": 0.17235066294670104,
|
|
"num_tokens": 5205068.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.571609210968018,
|
|
"epoch": 2.5083829096809085,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004721442018604822,
|
|
"loss": 5.4589,
|
|
"mean_token_accuracy": 0.18188732266426086,
|
|
"num_tokens": 5216049.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.528693628311157,
|
|
"epoch": 2.5137912385073014,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004719371245423853,
|
|
"loss": 5.4418,
|
|
"mean_token_accuracy": 0.18215959370136262,
|
|
"num_tokens": 5227024.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.661283922195435,
|
|
"epoch": 2.5191995673336938,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047172933155432273,
|
|
"loss": 5.5513,
|
|
"mean_token_accuracy": 0.1805453360080719,
|
|
"num_tokens": 5238862.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.678373098373413,
|
|
"epoch": 2.5246078961600866,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047152082365141466,
|
|
"loss": 5.5835,
|
|
"mean_token_accuracy": 0.1714266359806061,
|
|
"num_tokens": 5250020.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.648456621170044,
|
|
"epoch": 2.530016224986479,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004713116015913794,
|
|
"loss": 5.5016,
|
|
"mean_token_accuracy": 0.1778024211525917,
|
|
"num_tokens": 5260281.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.643498516082763,
|
|
"epoch": 2.535424553812872,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00047110166613453016,
|
|
"loss": 5.607,
|
|
"mean_token_accuracy": 0.1801880493760109,
|
|
"num_tokens": 5272909.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.594450569152832,
|
|
"epoch": 2.5408328826392643,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00047089101804377305,
|
|
"loss": 5.4292,
|
|
"mean_token_accuracy": 0.18523310571908952,
|
|
"num_tokens": 5285629.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.638476610183716,
|
|
"epoch": 2.546241211465657,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004706796580846037,
|
|
"loss": 5.4865,
|
|
"mean_token_accuracy": 0.1825991988182068,
|
|
"num_tokens": 5296104.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.754293918609619,
|
|
"epoch": 2.5516495402920496,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00047046758702510465,
|
|
"loss": 5.6274,
|
|
"mean_token_accuracy": 0.1725674033164978,
|
|
"num_tokens": 5307148.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.529038953781128,
|
|
"epoch": 2.5570578691184425,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047025480563594264,
|
|
"loss": 5.4423,
|
|
"mean_token_accuracy": 0.17616584300994872,
|
|
"num_tokens": 5317878.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.662910509109497,
|
|
"epoch": 2.5624661979448353,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00047004131469036576,
|
|
"loss": 5.5413,
|
|
"mean_token_accuracy": 0.17493427097797393,
|
|
"num_tokens": 5329603.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.575104904174805,
|
|
"epoch": 2.5678745267712277,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046982711496420057,
|
|
"loss": 5.5091,
|
|
"mean_token_accuracy": 0.1768939509987831,
|
|
"num_tokens": 5340611.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.619778776168824,
|
|
"epoch": 2.57328285559762,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046961220723584925,
|
|
"loss": 5.3973,
|
|
"mean_token_accuracy": 0.18415542244911193,
|
|
"num_tokens": 5352017.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.539876508712768,
|
|
"epoch": 2.578691184424013,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046939659228628725,
|
|
"loss": 5.4864,
|
|
"mean_token_accuracy": 0.18267493546009064,
|
|
"num_tokens": 5362482.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.572665691375732,
|
|
"epoch": 2.584099513250406,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00046918027089905956,
|
|
"loss": 5.4629,
|
|
"mean_token_accuracy": 0.18541775941848754,
|
|
"num_tokens": 5374074.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.5728918552398685,
|
|
"epoch": 2.5895078420767983,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046896324386027873,
|
|
"loss": 5.2844,
|
|
"mean_token_accuracy": 0.19210532903671265,
|
|
"num_tokens": 5384650.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.395759534835816,
|
|
"epoch": 2.5949161709031907,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046874551195862145,
|
|
"loss": 5.5216,
|
|
"mean_token_accuracy": 0.17698657065629958,
|
|
"num_tokens": 5395184.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.611190891265869,
|
|
"epoch": 2.6003244997295836,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000468527075985326,
|
|
"loss": 5.3385,
|
|
"mean_token_accuracy": 0.18617893904447555,
|
|
"num_tokens": 5405947.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.508720779418946,
|
|
"epoch": 2.6057328285559764,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00046830793673418924,
|
|
"loss": 5.4378,
|
|
"mean_token_accuracy": 0.1834009811282158,
|
|
"num_tokens": 5418237.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.547865533828736,
|
|
"epoch": 2.611141157382369,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046808809500156355,
|
|
"loss": 5.4799,
|
|
"mean_token_accuracy": 0.17972009032964706,
|
|
"num_tokens": 5429336.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.6402891159057615,
|
|
"epoch": 2.6165494862087613,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004678675515863545,
|
|
"loss": 5.4562,
|
|
"mean_token_accuracy": 0.1773098051548004,
|
|
"num_tokens": 5440257.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.6730303287506105,
|
|
"epoch": 2.621957815035154,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004676463072900173,
|
|
"loss": 5.5822,
|
|
"mean_token_accuracy": 0.17927723973989487,
|
|
"num_tokens": 5453079.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.555848503112793,
|
|
"epoch": 2.627366143861547,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046742436291655415,
|
|
"loss": 5.4517,
|
|
"mean_token_accuracy": 0.18331391513347625,
|
|
"num_tokens": 5464551.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.645699596405029,
|
|
"epoch": 2.6327744726879394,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046720171927251155,
|
|
"loss": 5.4931,
|
|
"mean_token_accuracy": 0.18631495535373688,
|
|
"num_tokens": 5474797.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.5558301448822025,
|
|
"epoch": 2.638182801514332,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004669783771669769,
|
|
"loss": 5.4466,
|
|
"mean_token_accuracy": 0.1874983102083206,
|
|
"num_tokens": 5485760.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.635391092300415,
|
|
"epoch": 2.6435911303407247,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000466754337411576,
|
|
"loss": 5.4326,
|
|
"mean_token_accuracy": 0.18973675966262818,
|
|
"num_tokens": 5496182.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.580751609802246,
|
|
"epoch": 2.6489994591671175,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046652960082046985,
|
|
"loss": 5.5154,
|
|
"mean_token_accuracy": 0.18605446368455886,
|
|
"num_tokens": 5506848.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.66952919960022,
|
|
"epoch": 2.65440778799351,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046630416821035164,
|
|
"loss": 5.6342,
|
|
"mean_token_accuracy": 0.17808265686035157,
|
|
"num_tokens": 5518430.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.60065279006958,
|
|
"epoch": 2.6598161168199024,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004660780404004441,
|
|
"loss": 5.4874,
|
|
"mean_token_accuracy": 0.17583505660295487,
|
|
"num_tokens": 5529741.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.498660850524902,
|
|
"epoch": 2.6652244456462952,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046585121821249623,
|
|
"loss": 5.4853,
|
|
"mean_token_accuracy": 0.18564430624246597,
|
|
"num_tokens": 5541835.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.646913051605225,
|
|
"epoch": 2.670632774472688,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046562370247078025,
|
|
"loss": 5.3832,
|
|
"mean_token_accuracy": 0.18622685372829437,
|
|
"num_tokens": 5552177.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.597685670852661,
|
|
"epoch": 2.6760411032990805,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000465395494002089,
|
|
"loss": 5.594,
|
|
"mean_token_accuracy": 0.1729603961110115,
|
|
"num_tokens": 5564440.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.623934268951416,
|
|
"epoch": 2.6814494321254734,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004651665936357325,
|
|
"loss": 5.3954,
|
|
"mean_token_accuracy": 0.18756800293922424,
|
|
"num_tokens": 5574905.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.563076639175415,
|
|
"epoch": 2.686857760951866,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004649370022035354,
|
|
"loss": 5.4757,
|
|
"mean_token_accuracy": 0.1770721659064293,
|
|
"num_tokens": 5585583.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.577158069610595,
|
|
"epoch": 2.6922660897782587,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004647067205398332,
|
|
"loss": 5.4657,
|
|
"mean_token_accuracy": 0.18533270061016083,
|
|
"num_tokens": 5597077.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.499568510055542,
|
|
"epoch": 2.697674418604651,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046447574948147027,
|
|
"loss": 5.3101,
|
|
"mean_token_accuracy": 0.18903629779815673,
|
|
"num_tokens": 5607904.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.565848779678345,
|
|
"epoch": 2.703082747431044,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00046424408986779587,
|
|
"loss": 5.4943,
|
|
"mean_token_accuracy": 0.18183189928531646,
|
|
"num_tokens": 5619611.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.666162109375,
|
|
"epoch": 2.7084910762574363,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004640117425406615,
|
|
"loss": 5.5117,
|
|
"mean_token_accuracy": 0.18600348979234696,
|
|
"num_tokens": 5630365.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.619974040985108,
|
|
"epoch": 2.713899405083829,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000463778708344418,
|
|
"loss": 5.4621,
|
|
"mean_token_accuracy": 0.1833752691745758,
|
|
"num_tokens": 5641008.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.526811027526856,
|
|
"epoch": 2.7193077339102216,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046354498812591207,
|
|
"loss": 5.4814,
|
|
"mean_token_accuracy": 0.1793915495276451,
|
|
"num_tokens": 5652633.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.590511894226074,
|
|
"epoch": 2.7247160627366145,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046331058273448364,
|
|
"loss": 5.4956,
|
|
"mean_token_accuracy": 0.17486928552389144,
|
|
"num_tokens": 5664108.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.583278512954712,
|
|
"epoch": 2.730124391563007,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004630754930219624,
|
|
"loss": 5.4742,
|
|
"mean_token_accuracy": 0.18829027712345123,
|
|
"num_tokens": 5675701.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.496996641159058,
|
|
"epoch": 2.7355327203893998,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004628397198426648,
|
|
"loss": 5.3902,
|
|
"mean_token_accuracy": 0.18832722008228303,
|
|
"num_tokens": 5687292.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.568372297286987,
|
|
"epoch": 2.740941049215792,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046260326405339126,
|
|
"loss": 5.4481,
|
|
"mean_token_accuracy": 0.18286903202533722,
|
|
"num_tokens": 5698905.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.493212270736694,
|
|
"epoch": 2.746349378042185,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004623661265134226,
|
|
"loss": 5.3313,
|
|
"mean_token_accuracy": 0.18769839257001877,
|
|
"num_tokens": 5709678.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.5725525379180905,
|
|
"epoch": 2.7517577068685775,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004621283080845172,
|
|
"loss": 5.4445,
|
|
"mean_token_accuracy": 0.1916872888803482,
|
|
"num_tokens": 5720460.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.522984266281128,
|
|
"epoch": 2.7571660356949703,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046188980963090783,
|
|
"loss": 5.4776,
|
|
"mean_token_accuracy": 0.18121275454759597,
|
|
"num_tokens": 5731701.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.589510536193847,
|
|
"epoch": 2.7625743645213627,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046165063201929825,
|
|
"loss": 5.4953,
|
|
"mean_token_accuracy": 0.17924681007862092,
|
|
"num_tokens": 5741512.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.556698369979858,
|
|
"epoch": 2.7679826933477556,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00046141077611886074,
|
|
"loss": 5.4144,
|
|
"mean_token_accuracy": 0.1847005307674408,
|
|
"num_tokens": 5751882.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.548605346679688,
|
|
"epoch": 2.773391022174148,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004611702428012319,
|
|
"loss": 5.5083,
|
|
"mean_token_accuracy": 0.1785637453198433,
|
|
"num_tokens": 5761989.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.581991815567017,
|
|
"epoch": 2.778799351000541,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00046092903294051057,
|
|
"loss": 5.4194,
|
|
"mean_token_accuracy": 0.1797608584165573,
|
|
"num_tokens": 5772532.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.607827043533325,
|
|
"epoch": 2.7842076798269337,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046068714741325385,
|
|
"loss": 5.5312,
|
|
"mean_token_accuracy": 0.18338500708341599,
|
|
"num_tokens": 5784152.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.603776979446411,
|
|
"epoch": 2.789616008653326,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004604445870984743,
|
|
"loss": 5.486,
|
|
"mean_token_accuracy": 0.1790586903691292,
|
|
"num_tokens": 5793748.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.5170189380645756,
|
|
"epoch": 2.7950243374797186,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004602013528776367,
|
|
"loss": 5.4335,
|
|
"mean_token_accuracy": 0.18346816897392274,
|
|
"num_tokens": 5804371.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.56022539138794,
|
|
"epoch": 2.8004326663061114,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004599574456346548,
|
|
"loss": 5.4388,
|
|
"mean_token_accuracy": 0.18434640616178513,
|
|
"num_tokens": 5815470.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.621416425704956,
|
|
"epoch": 2.8058409951325043,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045971286625588796,
|
|
"loss": 5.5262,
|
|
"mean_token_accuracy": 0.17809488177299498,
|
|
"num_tokens": 5826037.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.510598087310791,
|
|
"epoch": 2.8112493239588967,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004594676156301383,
|
|
"loss": 5.3831,
|
|
"mean_token_accuracy": 0.18959970474243165,
|
|
"num_tokens": 5836468.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.532654523849487,
|
|
"epoch": 2.816657652785289,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045922169464864725,
|
|
"loss": 5.3102,
|
|
"mean_token_accuracy": 0.19008912593126298,
|
|
"num_tokens": 5846371.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.373010635375977,
|
|
"epoch": 2.822065981611682,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000458975104205092,
|
|
"loss": 5.3987,
|
|
"mean_token_accuracy": 0.18497667461633682,
|
|
"num_tokens": 5856890.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.6032470703125,
|
|
"epoch": 2.827474310438075,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045872784519558304,
|
|
"loss": 5.4587,
|
|
"mean_token_accuracy": 0.18357819467782974,
|
|
"num_tokens": 5868075.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.645890569686889,
|
|
"epoch": 2.8328826392644673,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004584799185186599,
|
|
"loss": 5.5357,
|
|
"mean_token_accuracy": 0.18506620302796364,
|
|
"num_tokens": 5879579.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.511896085739136,
|
|
"epoch": 2.8382909680908597,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00045823132507528903,
|
|
"loss": 5.3861,
|
|
"mean_token_accuracy": 0.18361688554286956,
|
|
"num_tokens": 5890648.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.443297290802002,
|
|
"epoch": 2.8436992969172525,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004579820657688593,
|
|
"loss": 5.4239,
|
|
"mean_token_accuracy": 0.18471520394086838,
|
|
"num_tokens": 5902914.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.618084287643432,
|
|
"epoch": 2.8491076257436454,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004577321415051798,
|
|
"loss": 5.458,
|
|
"mean_token_accuracy": 0.17708844542503357,
|
|
"num_tokens": 5915801.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.631968927383423,
|
|
"epoch": 2.854515954570038,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00045748155319247564,
|
|
"loss": 5.535,
|
|
"mean_token_accuracy": 0.17567279785871506,
|
|
"num_tokens": 5928973.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.552738237380981,
|
|
"epoch": 2.8599242833964302,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004572303017413856,
|
|
"loss": 5.5523,
|
|
"mean_token_accuracy": 0.17950572967529296,
|
|
"num_tokens": 5939720.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.627747678756714,
|
|
"epoch": 2.865332612222823,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004569783880649578,
|
|
"loss": 5.4964,
|
|
"mean_token_accuracy": 0.1777748703956604,
|
|
"num_tokens": 5951956.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.6360191822052,
|
|
"epoch": 2.870740941049216,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004567258130786474,
|
|
"loss": 5.6302,
|
|
"mean_token_accuracy": 0.17279365807771682,
|
|
"num_tokens": 5963735.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.639206218719482,
|
|
"epoch": 2.8761492698756084,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045647257770031215,
|
|
"loss": 5.3675,
|
|
"mean_token_accuracy": 0.18931113481521605,
|
|
"num_tokens": 5974567.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.526039075851441,
|
|
"epoch": 2.8815575987020012,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045621868285021023,
|
|
"loss": 5.5087,
|
|
"mean_token_accuracy": 0.1847616985440254,
|
|
"num_tokens": 5985800.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.522658729553223,
|
|
"epoch": 2.8869659275283936,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045596412945099615,
|
|
"loss": 5.3634,
|
|
"mean_token_accuracy": 0.18461358100175856,
|
|
"num_tokens": 5997388.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.6060858249664305,
|
|
"epoch": 2.8923742563547865,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045570891842771727,
|
|
"loss": 5.4792,
|
|
"mean_token_accuracy": 0.1850425735116005,
|
|
"num_tokens": 6008739.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.486096715927124,
|
|
"epoch": 2.897782585181179,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045545305070781135,
|
|
"loss": 5.363,
|
|
"mean_token_accuracy": 0.18809035420417786,
|
|
"num_tokens": 6019490.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.484896326065064,
|
|
"epoch": 2.903190914007572,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045519652722110206,
|
|
"loss": 5.4792,
|
|
"mean_token_accuracy": 0.1885268583893776,
|
|
"num_tokens": 6030712.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.574739122390747,
|
|
"epoch": 2.908599242833964,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00045493934889979655,
|
|
"loss": 5.3709,
|
|
"mean_token_accuracy": 0.19513980895280839,
|
|
"num_tokens": 6042161.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.489770793914795,
|
|
"epoch": 2.914007571660357,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045468151667848116,
|
|
"loss": 5.4084,
|
|
"mean_token_accuracy": 0.17882926911115646,
|
|
"num_tokens": 6054107.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.5454610824584964,
|
|
"epoch": 2.9194159004867495,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000454423031494119,
|
|
"loss": 5.4756,
|
|
"mean_token_accuracy": 0.18480944782495498,
|
|
"num_tokens": 6065232.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.531529569625855,
|
|
"epoch": 2.9248242293131423,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004541638942860458,
|
|
"loss": 5.4395,
|
|
"mean_token_accuracy": 0.18232246488332748,
|
|
"num_tokens": 6076294.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.468943977355957,
|
|
"epoch": 2.9302325581395348,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004539041059959667,
|
|
"loss": 5.3378,
|
|
"mean_token_accuracy": 0.19850929975509643,
|
|
"num_tokens": 6087700.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.552276563644409,
|
|
"epoch": 2.9356408869659276,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045364366756795295,
|
|
"loss": 5.4297,
|
|
"mean_token_accuracy": 0.19124325066804887,
|
|
"num_tokens": 6098561.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.610854625701904,
|
|
"epoch": 2.94104921579232,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045338257994843845,
|
|
"loss": 5.4712,
|
|
"mean_token_accuracy": 0.18611805737018586,
|
|
"num_tokens": 6109729.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.52247052192688,
|
|
"epoch": 2.946457544618713,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004531208440862162,
|
|
"loss": 5.4309,
|
|
"mean_token_accuracy": 0.18569640219211578,
|
|
"num_tokens": 6120468.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.47803635597229,
|
|
"epoch": 2.9518658734451053,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00045285846093243485,
|
|
"loss": 5.3606,
|
|
"mean_token_accuracy": 0.19005629271268845,
|
|
"num_tokens": 6132457.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.493604946136474,
|
|
"epoch": 2.957274202271498,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045259543144059557,
|
|
"loss": 5.4347,
|
|
"mean_token_accuracy": 0.1807558938860893,
|
|
"num_tokens": 6142289.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.570860958099365,
|
|
"epoch": 2.9626825310978906,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004523317565665479,
|
|
"loss": 5.4325,
|
|
"mean_token_accuracy": 0.18680479973554612,
|
|
"num_tokens": 6153786.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.555897235870361,
|
|
"epoch": 2.9680908599242835,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00045206743726848716,
|
|
"loss": 5.423,
|
|
"mean_token_accuracy": 0.1865623652935028,
|
|
"num_tokens": 6163988.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.458732604980469,
|
|
"epoch": 2.973499188750676,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004518024745069502,
|
|
"loss": 5.5109,
|
|
"mean_token_accuracy": 0.1849767670035362,
|
|
"num_tokens": 6175859.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.784033346176147,
|
|
"epoch": 2.9789075175770687,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004515368692448123,
|
|
"loss": 5.6927,
|
|
"mean_token_accuracy": 0.16950009167194366,
|
|
"num_tokens": 6188116.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.541548204421997,
|
|
"epoch": 2.9843158464034616,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045127062244728353,
|
|
"loss": 5.5002,
|
|
"mean_token_accuracy": 0.18853752613067626,
|
|
"num_tokens": 6198606.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.632490587234497,
|
|
"epoch": 2.989724175229854,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004510037350819055,
|
|
"loss": 5.5198,
|
|
"mean_token_accuracy": 0.18157044053077698,
|
|
"num_tokens": 6210366.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.490554475784302,
|
|
"epoch": 2.9951325040562464,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045073620811854744,
|
|
"loss": 5.4074,
|
|
"mean_token_accuracy": 0.18180620819330215,
|
|
"num_tokens": 6221376.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.627877182430691,
|
|
"epoch": 3.0,
|
|
"grad_norm": 1.8203125,
|
|
"learning_rate": 0.00045046804252940294,
|
|
"loss": 5.3936,
|
|
"mean_token_accuracy": 0.18600195480717552,
|
|
"num_tokens": 6230667.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.468294143676758,
|
|
"epoch": 3.005408328826393,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045019923928898644,
|
|
"loss": 5.0319,
|
|
"mean_token_accuracy": 0.20169661343097686,
|
|
"num_tokens": 6241158.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.511208438873291,
|
|
"epoch": 3.0108166576527853,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004499297993741294,
|
|
"loss": 5.1237,
|
|
"mean_token_accuracy": 0.19160998463630677,
|
|
"num_tokens": 6253811.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.497205972671509,
|
|
"epoch": 3.016224986479178,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004496597237639773,
|
|
"loss": 5.1048,
|
|
"mean_token_accuracy": 0.1983844220638275,
|
|
"num_tokens": 6265066.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.437740659713745,
|
|
"epoch": 3.0216333153055706,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044938901343998536,
|
|
"loss": 5.0983,
|
|
"mean_token_accuracy": 0.1972705602645874,
|
|
"num_tokens": 6275889.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.4698113918304445,
|
|
"epoch": 3.0270416441319634,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004491176693859156,
|
|
"loss": 5.0305,
|
|
"mean_token_accuracy": 0.2037506118416786,
|
|
"num_tokens": 6287576.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.45487117767334,
|
|
"epoch": 3.032449972958356,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004488456925878329,
|
|
"loss": 5.1093,
|
|
"mean_token_accuracy": 0.19570358395576476,
|
|
"num_tokens": 6299419.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.452791738510132,
|
|
"epoch": 3.0378583017847487,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044857308403410163,
|
|
"loss": 5.0801,
|
|
"mean_token_accuracy": 0.20720178186893462,
|
|
"num_tokens": 6310464.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.5076933860778805,
|
|
"epoch": 3.043266630611141,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00044829984471538183,
|
|
"loss": 5.0839,
|
|
"mean_token_accuracy": 0.2033734768629074,
|
|
"num_tokens": 6321442.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.454289722442627,
|
|
"epoch": 3.048674959437534,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004480259756246258,
|
|
"loss": 5.1363,
|
|
"mean_token_accuracy": 0.20159937888383866,
|
|
"num_tokens": 6332868.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.543545246124268,
|
|
"epoch": 3.0540832882639264,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00044775147775707465,
|
|
"loss": 5.1325,
|
|
"mean_token_accuracy": 0.2005767598748207,
|
|
"num_tokens": 6345232.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.448202657699585,
|
|
"epoch": 3.0594916170903192,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000447476352110254,
|
|
"loss": 5.1682,
|
|
"mean_token_accuracy": 0.19499213993549347,
|
|
"num_tokens": 6356524.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.509788656234742,
|
|
"epoch": 3.0648999459167117,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004472005996839712,
|
|
"loss": 5.114,
|
|
"mean_token_accuracy": 0.20154579281806945,
|
|
"num_tokens": 6367590.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.558334064483643,
|
|
"epoch": 3.0703082747431045,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004469242214803112,
|
|
"loss": 5.1485,
|
|
"mean_token_accuracy": 0.1895488902926445,
|
|
"num_tokens": 6378211.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.410256004333496,
|
|
"epoch": 3.075716603569497,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004466472185036331,
|
|
"loss": 5.1536,
|
|
"mean_token_accuracy": 0.20190663635730743,
|
|
"num_tokens": 6389595.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.5060454368591305,
|
|
"epoch": 3.08112493239589,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004463695917605663,
|
|
"loss": 5.1434,
|
|
"mean_token_accuracy": 0.19547284841537477,
|
|
"num_tokens": 6401286.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.422776985168457,
|
|
"epoch": 3.086533261222282,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000446091342260007,
|
|
"loss": 5.0902,
|
|
"mean_token_accuracy": 0.2042829394340515,
|
|
"num_tokens": 6411946.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.427499389648437,
|
|
"epoch": 3.091941590048675,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004458124710131145,
|
|
"loss": 5.0588,
|
|
"mean_token_accuracy": 0.20717511475086212,
|
|
"num_tokens": 6422463.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.366190195083618,
|
|
"epoch": 3.0973499188750675,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00044553297903330766,
|
|
"loss": 5.0575,
|
|
"mean_token_accuracy": 0.20171311646699905,
|
|
"num_tokens": 6433576.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.412723064422607,
|
|
"epoch": 3.1027582477014604,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044525286733626085,
|
|
"loss": 5.0423,
|
|
"mean_token_accuracy": 0.20086741596460342,
|
|
"num_tokens": 6444058.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.491685676574707,
|
|
"epoch": 3.1081665765278528,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004449721369399008,
|
|
"loss": 5.2088,
|
|
"mean_token_accuracy": 0.1972980335354805,
|
|
"num_tokens": 6453872.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.414568328857422,
|
|
"epoch": 3.1135749053542456,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00044469078886440227,
|
|
"loss": 5.1201,
|
|
"mean_token_accuracy": 0.19709280133247375,
|
|
"num_tokens": 6465367.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.408377408981323,
|
|
"epoch": 3.118983234180638,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00044440882413218484,
|
|
"loss": 5.0494,
|
|
"mean_token_accuracy": 0.2074790671467781,
|
|
"num_tokens": 6476655.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.3520111560821535,
|
|
"epoch": 3.124391563007031,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044412624376790905,
|
|
"loss": 5.1231,
|
|
"mean_token_accuracy": 0.1953948676586151,
|
|
"num_tokens": 6488348.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.480614566802979,
|
|
"epoch": 3.1297998918334233,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004438430487984726,
|
|
"loss": 5.1497,
|
|
"mean_token_accuracy": 0.19856213629245759,
|
|
"num_tokens": 6500127.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.335084629058838,
|
|
"epoch": 3.135208220659816,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004435592402530066,
|
|
"loss": 5.0934,
|
|
"mean_token_accuracy": 0.1986870363354683,
|
|
"num_tokens": 6512772.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.445497846603393,
|
|
"epoch": 3.1406165494862086,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000443274819162872,
|
|
"loss": 5.1279,
|
|
"mean_token_accuracy": 0.20131367295980454,
|
|
"num_tokens": 6523359.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.405496311187744,
|
|
"epoch": 3.1460248783126015,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004429897865616557,
|
|
"loss": 5.0977,
|
|
"mean_token_accuracy": 0.19828550070524215,
|
|
"num_tokens": 6533679.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.410326433181763,
|
|
"epoch": 3.151433207138994,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004427041434851668,
|
|
"loss": 5.0543,
|
|
"mean_token_accuracy": 0.20287300348281861,
|
|
"num_tokens": 6544694.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.384902906417847,
|
|
"epoch": 3.1568415359653867,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044241789097143293,
|
|
"loss": 5.1072,
|
|
"mean_token_accuracy": 0.2010298490524292,
|
|
"num_tokens": 6555858.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.397913837432862,
|
|
"epoch": 3.162249864791779,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00044213103006069635,
|
|
"loss": 5.1327,
|
|
"mean_token_accuracy": 0.19846443235874175,
|
|
"num_tokens": 6568214.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.337772274017334,
|
|
"epoch": 3.167658193618172,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00044184356179541035,
|
|
"loss": 5.0847,
|
|
"mean_token_accuracy": 0.20672140419483184,
|
|
"num_tokens": 6579646.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.4350510597229,
|
|
"epoch": 3.1730665224445644,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004415554872202352,
|
|
"loss": 5.0471,
|
|
"mean_token_accuracy": 0.2057007685303688,
|
|
"num_tokens": 6590683.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.474175310134887,
|
|
"epoch": 3.1784748512709573,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00044126680738203466,
|
|
"loss": 5.1792,
|
|
"mean_token_accuracy": 0.20006656646728516,
|
|
"num_tokens": 6601197.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.40438404083252,
|
|
"epoch": 3.1838831800973497,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004409775233298718,
|
|
"loss": 5.2307,
|
|
"mean_token_accuracy": 0.1858530282974243,
|
|
"num_tokens": 6613564.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.462791013717651,
|
|
"epoch": 3.1892915089237426,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004406876361150056,
|
|
"loss": 5.0813,
|
|
"mean_token_accuracy": 0.19810996502637862,
|
|
"num_tokens": 6624696.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.352666664123535,
|
|
"epoch": 3.194699837750135,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004403971467908869,
|
|
"loss": 5.0945,
|
|
"mean_token_accuracy": 0.20819108039140702,
|
|
"num_tokens": 6636302.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.4936810493469235,
|
|
"epoch": 3.200108166576528,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004401060564131545,
|
|
"loss": 5.1454,
|
|
"mean_token_accuracy": 0.20026006698608398,
|
|
"num_tokens": 6646517.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.329576587677002,
|
|
"epoch": 3.2055164954029207,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004398143660396315,
|
|
"loss": 5.1212,
|
|
"mean_token_accuracy": 0.1982803151011467,
|
|
"num_tokens": 6657768.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.32588415145874,
|
|
"epoch": 3.210924824229313,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004395220767303215,
|
|
"loss": 5.0409,
|
|
"mean_token_accuracy": 0.20511585474014282,
|
|
"num_tokens": 6669110.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.42305212020874,
|
|
"epoch": 3.2163331530557056,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004392291895474045,
|
|
"loss": 5.1291,
|
|
"mean_token_accuracy": 0.19366252273321152,
|
|
"num_tokens": 6680925.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.43647346496582,
|
|
"epoch": 3.2217414818820984,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004389357055552331,
|
|
"loss": 5.2081,
|
|
"mean_token_accuracy": 0.19939788728952407,
|
|
"num_tokens": 6692923.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.443479394912719,
|
|
"epoch": 3.2271498107084913,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00043864162582032897,
|
|
"loss": 5.0935,
|
|
"mean_token_accuracy": 0.201037235558033,
|
|
"num_tokens": 6702692.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.416078042984009,
|
|
"epoch": 3.2325581395348837,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004383469514113784,
|
|
"loss": 5.1746,
|
|
"mean_token_accuracy": 0.19405930787324904,
|
|
"num_tokens": 6713830.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.457262992858887,
|
|
"epoch": 3.2379664683612766,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004380516833992289,
|
|
"loss": 5.1648,
|
|
"mean_token_accuracy": 0.19341373443603516,
|
|
"num_tokens": 6725564.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.3240556716918945,
|
|
"epoch": 3.243374797187669,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000437755822856885,
|
|
"loss": 5.1144,
|
|
"mean_token_accuracy": 0.1982475072145462,
|
|
"num_tokens": 6736563.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 3.243374797187669,
|
|
"eval_entropy": 5.3295986381444065,
|
|
"eval_loss": 5.9673051834106445,
|
|
"eval_mean_token_accuracy": 0.17137166922065344,
|
|
"eval_num_tokens": 6736563.0,
|
|
"eval_runtime": 2.5403,
|
|
"eval_samples_per_second": 1383.691,
|
|
"eval_steps_per_second": 173.207,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.360402679443359,
|
|
"epoch": 3.248783126014062,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004374593708595049,
|
|
"loss": 5.0952,
|
|
"mean_token_accuracy": 0.20260146856307984,
|
|
"num_tokens": 6745845.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.3957164764404295,
|
|
"epoch": 3.2541914548404542,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00043716232848439555,
|
|
"loss": 5.0295,
|
|
"mean_token_accuracy": 0.20676433593034743,
|
|
"num_tokens": 6757450.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.481287479400635,
|
|
"epoch": 3.259599783666847,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00043686469681100985,
|
|
"loss": 5.1921,
|
|
"mean_token_accuracy": 0.19729999750852584,
|
|
"num_tokens": 6768542.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.383357143402099,
|
|
"epoch": 3.2650081124932395,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00043656647692094186,
|
|
"loss": 5.104,
|
|
"mean_token_accuracy": 0.21509281545877457,
|
|
"num_tokens": 6779873.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.409872150421142,
|
|
"epoch": 3.2704164413196324,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004362676698979233,
|
|
"loss": 5.1705,
|
|
"mean_token_accuracy": 0.20248037129640578,
|
|
"num_tokens": 6790898.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.460791063308716,
|
|
"epoch": 3.275824770146025,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00043596827682781974,
|
|
"loss": 5.1269,
|
|
"mean_token_accuracy": 0.19579315185546875,
|
|
"num_tokens": 6802175.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.449886417388916,
|
|
"epoch": 3.2812330989724177,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004356682987986262,
|
|
"loss": 5.2657,
|
|
"mean_token_accuracy": 0.19761422872543336,
|
|
"num_tokens": 6813675.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.402672624588012,
|
|
"epoch": 3.28664142779881,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004353677369004635,
|
|
"loss": 5.1183,
|
|
"mean_token_accuracy": 0.20132410824298858,
|
|
"num_tokens": 6825020.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.3562298774719235,
|
|
"epoch": 3.292049756625203,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00043506659222557426,
|
|
"loss": 5.0065,
|
|
"mean_token_accuracy": 0.22016204446554183,
|
|
"num_tokens": 6834954.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.428651523590088,
|
|
"epoch": 3.2974580854515954,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000434764865868319,
|
|
"loss": 5.2501,
|
|
"mean_token_accuracy": 0.19206423312425613,
|
|
"num_tokens": 6847958.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.429607200622558,
|
|
"epoch": 3.302866414277988,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000434462558925172,
|
|
"loss": 5.0645,
|
|
"mean_token_accuracy": 0.2126267820596695,
|
|
"num_tokens": 6858782.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.267402839660645,
|
|
"epoch": 3.3082747431043806,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00043415967249471734,
|
|
"loss": 5.0949,
|
|
"mean_token_accuracy": 0.19767530113458634,
|
|
"num_tokens": 6869807.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.4389395236969,
|
|
"epoch": 3.3136830719307735,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000433856207677645,
|
|
"loss": 5.122,
|
|
"mean_token_accuracy": 0.2048116371035576,
|
|
"num_tokens": 6880642.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.520693635940551,
|
|
"epoch": 3.319091400757166,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000433552165576747,
|
|
"loss": 5.2303,
|
|
"mean_token_accuracy": 0.18994351476430893,
|
|
"num_tokens": 6891677.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.401134061813354,
|
|
"epoch": 3.3244997295835588,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00043324754729691275,
|
|
"loss": 5.1102,
|
|
"mean_token_accuracy": 0.20367221385240555,
|
|
"num_tokens": 6903383.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.453644275665283,
|
|
"epoch": 3.329908058409951,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00043294235394512594,
|
|
"loss": 5.1738,
|
|
"mean_token_accuracy": 0.20837322622537613,
|
|
"num_tokens": 6915874.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.330303478240967,
|
|
"epoch": 3.335316387236344,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004326365866304599,
|
|
"loss": 4.9806,
|
|
"mean_token_accuracy": 0.20401023477315902,
|
|
"num_tokens": 6926356.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.411937189102173,
|
|
"epoch": 3.3407247160627365,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004323302464640738,
|
|
"loss": 5.1269,
|
|
"mean_token_accuracy": 0.20709482580423355,
|
|
"num_tokens": 6937678.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.3702147006988525,
|
|
"epoch": 3.3461330448891293,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00043202333455920843,
|
|
"loss": 5.1093,
|
|
"mean_token_accuracy": 0.20375309139490128,
|
|
"num_tokens": 6949447.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.425435018539429,
|
|
"epoch": 3.3515413737155217,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004317158520311824,
|
|
"loss": 5.0479,
|
|
"mean_token_accuracy": 0.19849031269550324,
|
|
"num_tokens": 6960347.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.299737071990966,
|
|
"epoch": 3.3569497025419146,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004314077999973879,
|
|
"loss": 5.0673,
|
|
"mean_token_accuracy": 0.20092922896146775,
|
|
"num_tokens": 6971353.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.404045057296753,
|
|
"epoch": 3.362358031368307,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004310991795772869,
|
|
"loss": 5.199,
|
|
"mean_token_accuracy": 0.19329828768968582,
|
|
"num_tokens": 6982519.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.328901720046997,
|
|
"epoch": 3.3677663601947,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004307899918924065,
|
|
"loss": 5.0587,
|
|
"mean_token_accuracy": 0.2079089805483818,
|
|
"num_tokens": 6994021.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.340012502670288,
|
|
"epoch": 3.3731746890210923,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004304802380663356,
|
|
"loss": 5.1503,
|
|
"mean_token_accuracy": 0.2011229932308197,
|
|
"num_tokens": 7005505.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.420154809951782,
|
|
"epoch": 3.378583017847485,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00043016991922472015,
|
|
"loss": 5.1797,
|
|
"mean_token_accuracy": 0.20180473178625108,
|
|
"num_tokens": 7016123.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.447136163711548,
|
|
"epoch": 3.383991346673878,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00042985903649525977,
|
|
"loss": 5.1242,
|
|
"mean_token_accuracy": 0.20436165928840638,
|
|
"num_tokens": 7027083.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.438285684585571,
|
|
"epoch": 3.3893996755002704,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00042954759100770275,
|
|
"loss": 5.2231,
|
|
"mean_token_accuracy": 0.1902200162410736,
|
|
"num_tokens": 7037892.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.42435655593872,
|
|
"epoch": 3.394808004326663,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00042923558389384303,
|
|
"loss": 5.1859,
|
|
"mean_token_accuracy": 0.1983151316642761,
|
|
"num_tokens": 7048996.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.420599699020386,
|
|
"epoch": 3.4002163331530557,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004289230162875149,
|
|
"loss": 5.1967,
|
|
"mean_token_accuracy": 0.19115833938121796,
|
|
"num_tokens": 7060884.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.459965229034424,
|
|
"epoch": 3.4056246619794486,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00042860988932458987,
|
|
"loss": 5.2078,
|
|
"mean_token_accuracy": 0.1965504989027977,
|
|
"num_tokens": 7072550.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.350359773635864,
|
|
"epoch": 3.411032990805841,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004282962041429721,
|
|
"loss": 5.1322,
|
|
"mean_token_accuracy": 0.20643099248409272,
|
|
"num_tokens": 7085325.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.3334959030151365,
|
|
"epoch": 3.4164413196322334,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00042798196188259427,
|
|
"loss": 5.0023,
|
|
"mean_token_accuracy": 0.2047032430768013,
|
|
"num_tokens": 7095691.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.2706080913543705,
|
|
"epoch": 3.4218496484586263,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004276671636854133,
|
|
"loss": 4.9804,
|
|
"mean_token_accuracy": 0.21151768267154694,
|
|
"num_tokens": 7106552.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.251733016967774,
|
|
"epoch": 3.427257977285019,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00042735181069540677,
|
|
"loss": 5.1225,
|
|
"mean_token_accuracy": 0.19668723940849303,
|
|
"num_tokens": 7119708.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.378653573989868,
|
|
"epoch": 3.4326663061114115,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004270359040585681,
|
|
"loss": 5.1283,
|
|
"mean_token_accuracy": 0.19715497642755508,
|
|
"num_tokens": 7130663.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.350176668167114,
|
|
"epoch": 3.4380746349378044,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00042671944492290277,
|
|
"loss": 5.058,
|
|
"mean_token_accuracy": 0.19353861659765242,
|
|
"num_tokens": 7141513.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.30481333732605,
|
|
"epoch": 3.443482963764197,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004264024344384239,
|
|
"loss": 5.137,
|
|
"mean_token_accuracy": 0.2001711145043373,
|
|
"num_tokens": 7153387.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.437187337875367,
|
|
"epoch": 3.4488912925905897,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00042608487375714847,
|
|
"loss": 5.1131,
|
|
"mean_token_accuracy": 0.20553285628557205,
|
|
"num_tokens": 7164905.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.360888195037842,
|
|
"epoch": 3.454299621416982,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00042576676403309265,
|
|
"loss": 5.1646,
|
|
"mean_token_accuracy": 0.2038762614130974,
|
|
"num_tokens": 7175636.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.427288293838501,
|
|
"epoch": 3.459707950243375,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004254481064222679,
|
|
"loss": 5.1451,
|
|
"mean_token_accuracy": 0.20136284530162812,
|
|
"num_tokens": 7185903.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.28635835647583,
|
|
"epoch": 3.4651162790697674,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00042512890208267677,
|
|
"loss": 5.0644,
|
|
"mean_token_accuracy": 0.2008620247244835,
|
|
"num_tokens": 7198257.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.353400754928589,
|
|
"epoch": 3.4705246078961602,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00042480915217430853,
|
|
"loss": 5.0544,
|
|
"mean_token_accuracy": 0.21452305614948272,
|
|
"num_tokens": 7208839.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.385164546966553,
|
|
"epoch": 3.4759329367225527,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00042448885785913507,
|
|
"loss": 5.164,
|
|
"mean_token_accuracy": 0.19736500829458237,
|
|
"num_tokens": 7221078.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.3973085403442385,
|
|
"epoch": 3.4813412655489455,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00042416802030110677,
|
|
"loss": 5.0755,
|
|
"mean_token_accuracy": 0.20369009524583817,
|
|
"num_tokens": 7230287.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.322775983810425,
|
|
"epoch": 3.486749594375338,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00042384664066614787,
|
|
"loss": 5.1647,
|
|
"mean_token_accuracy": 0.19989106208086013,
|
|
"num_tokens": 7242253.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.392164659500122,
|
|
"epoch": 3.492157923201731,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004235247201221528,
|
|
"loss": 5.1168,
|
|
"mean_token_accuracy": 0.21092349886894227,
|
|
"num_tokens": 7254098.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.286330461502075,
|
|
"epoch": 3.497566252028123,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004232022598389816,
|
|
"loss": 5.0297,
|
|
"mean_token_accuracy": 0.205195152759552,
|
|
"num_tokens": 7264153.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.295954895019531,
|
|
"epoch": 3.502974580854516,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004228792609884557,
|
|
"loss": 5.1495,
|
|
"mean_token_accuracy": 0.1964465171098709,
|
|
"num_tokens": 7276431.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.347540760040284,
|
|
"epoch": 3.5083829096809085,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00042255572474435364,
|
|
"loss": 5.0882,
|
|
"mean_token_accuracy": 0.20361680984497071,
|
|
"num_tokens": 7287882.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.402281427383423,
|
|
"epoch": 3.5137912385073014,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004222316522824069,
|
|
"loss": 5.2156,
|
|
"mean_token_accuracy": 0.18984193950891495,
|
|
"num_tokens": 7298846.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.409946346282959,
|
|
"epoch": 3.5191995673336938,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004219070447802955,
|
|
"loss": 5.1013,
|
|
"mean_token_accuracy": 0.2012368381023407,
|
|
"num_tokens": 7310372.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.382275009155274,
|
|
"epoch": 3.5246078961600866,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00042158190341764384,
|
|
"loss": 5.1883,
|
|
"mean_token_accuracy": 0.1983090490102768,
|
|
"num_tokens": 7322200.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.3486621379852295,
|
|
"epoch": 3.530016224986479,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00042125622937601645,
|
|
"loss": 5.0652,
|
|
"mean_token_accuracy": 0.20590971261262894,
|
|
"num_tokens": 7331824.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.364713382720947,
|
|
"epoch": 3.535424553812872,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004209300238389135,
|
|
"loss": 5.076,
|
|
"mean_token_accuracy": 0.20500284135341645,
|
|
"num_tokens": 7341522.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.308546924591065,
|
|
"epoch": 3.5408328826392643,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004206032879917667,
|
|
"loss": 5.1382,
|
|
"mean_token_accuracy": 0.20895732939243317,
|
|
"num_tokens": 7353312.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.4163555145263675,
|
|
"epoch": 3.546241211465657,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004202760230219348,
|
|
"loss": 5.1491,
|
|
"mean_token_accuracy": 0.20819591730833054,
|
|
"num_tokens": 7363985.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.392454195022583,
|
|
"epoch": 3.5516495402920496,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00041994823011869955,
|
|
"loss": 5.1345,
|
|
"mean_token_accuracy": 0.20077044069766997,
|
|
"num_tokens": 7373872.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.359751749038696,
|
|
"epoch": 3.5570578691184425,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00041961991047326116,
|
|
"loss": 5.1453,
|
|
"mean_token_accuracy": 0.20554754734039307,
|
|
"num_tokens": 7386759.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.328323221206665,
|
|
"epoch": 3.5624661979448353,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004192910652787338,
|
|
"loss": 5.0837,
|
|
"mean_token_accuracy": 0.21025900095701217,
|
|
"num_tokens": 7398001.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.426917791366577,
|
|
"epoch": 3.5678745267712277,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004189616957301419,
|
|
"loss": 5.1566,
|
|
"mean_token_accuracy": 0.20049830973148347,
|
|
"num_tokens": 7409159.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.301196193695068,
|
|
"epoch": 3.57328285559762,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004186318030244149,
|
|
"loss": 5.0768,
|
|
"mean_token_accuracy": 0.203865085542202,
|
|
"num_tokens": 7421028.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.370200681686401,
|
|
"epoch": 3.578691184424013,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00041830138836038383,
|
|
"loss": 5.177,
|
|
"mean_token_accuracy": 0.19776048362255097,
|
|
"num_tokens": 7432321.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.424042654037476,
|
|
"epoch": 3.584099513250406,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004179704529387763,
|
|
"loss": 5.1936,
|
|
"mean_token_accuracy": 0.20398965626955032,
|
|
"num_tokens": 7443607.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.370753812789917,
|
|
"epoch": 3.5895078420767983,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004176389979622125,
|
|
"loss": 5.0566,
|
|
"mean_token_accuracy": 0.20633578300476074,
|
|
"num_tokens": 7455222.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.37438998222351,
|
|
"epoch": 3.5949161709031907,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004173070246352003,
|
|
"loss": 5.2097,
|
|
"mean_token_accuracy": 0.19915072172880172,
|
|
"num_tokens": 7468511.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.371615552902222,
|
|
"epoch": 3.6003244997295836,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00041697453416413184,
|
|
"loss": 5.2082,
|
|
"mean_token_accuracy": 0.2001425877213478,
|
|
"num_tokens": 7479810.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.30455846786499,
|
|
"epoch": 3.6057328285559764,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000416641527757278,
|
|
"loss": 5.019,
|
|
"mean_token_accuracy": 0.20204450339078903,
|
|
"num_tokens": 7491405.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.4047504425048825,
|
|
"epoch": 3.611141157382369,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00041630800662478514,
|
|
"loss": 5.1751,
|
|
"mean_token_accuracy": 0.19143394380807877,
|
|
"num_tokens": 7501820.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.387286138534546,
|
|
"epoch": 3.6165494862087613,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00041597397197866957,
|
|
"loss": 5.1203,
|
|
"mean_token_accuracy": 0.2006533220410347,
|
|
"num_tokens": 7512165.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.284640789031982,
|
|
"epoch": 3.621957815035154,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004156394250328141,
|
|
"loss": 5.1075,
|
|
"mean_token_accuracy": 0.20234882235527038,
|
|
"num_tokens": 7523755.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.322055768966675,
|
|
"epoch": 3.627366143861547,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000415304367002963,
|
|
"loss": 5.1128,
|
|
"mean_token_accuracy": 0.19949207901954652,
|
|
"num_tokens": 7535674.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.3532407760620115,
|
|
"epoch": 3.6327744726879394,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00041496879910671796,
|
|
"loss": 5.0892,
|
|
"mean_token_accuracy": 0.20500377118587493,
|
|
"num_tokens": 7546885.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.298316383361817,
|
|
"epoch": 3.638182801514332,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004146327225635336,
|
|
"loss": 5.0816,
|
|
"mean_token_accuracy": 0.2053226500749588,
|
|
"num_tokens": 7557718.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.314270782470703,
|
|
"epoch": 3.6435911303407247,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004142961385947127,
|
|
"loss": 5.0543,
|
|
"mean_token_accuracy": 0.2027818188071251,
|
|
"num_tokens": 7569472.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.41190390586853,
|
|
"epoch": 3.6489994591671175,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00041395904842340215,
|
|
"loss": 5.092,
|
|
"mean_token_accuracy": 0.19857909977436067,
|
|
"num_tokens": 7579954.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.280649137496948,
|
|
"epoch": 3.65440778799351,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004136214532745887,
|
|
"loss": 5.0361,
|
|
"mean_token_accuracy": 0.2091826856136322,
|
|
"num_tokens": 7590717.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.4638982772827145,
|
|
"epoch": 3.6598161168199024,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004132833543750936,
|
|
"loss": 5.3549,
|
|
"mean_token_accuracy": 0.18796722292900087,
|
|
"num_tokens": 7602802.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.4786789894104,
|
|
"epoch": 3.6652244456462952,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00041294475295356917,
|
|
"loss": 5.2792,
|
|
"mean_token_accuracy": 0.19338245689868927,
|
|
"num_tokens": 7614944.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.37835693359375,
|
|
"epoch": 3.670632774472688,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004126056502404937,
|
|
"loss": 5.1735,
|
|
"mean_token_accuracy": 0.19305696040391923,
|
|
"num_tokens": 7626449.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.342289066314697,
|
|
"epoch": 3.6760411032990805,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004122660474681672,
|
|
"loss": 5.0435,
|
|
"mean_token_accuracy": 0.20813355445861817,
|
|
"num_tokens": 7637878.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.400849056243897,
|
|
"epoch": 3.6814494321254734,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00041192594587070706,
|
|
"loss": 5.2765,
|
|
"mean_token_accuracy": 0.19228302389383317,
|
|
"num_tokens": 7649428.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.462759685516358,
|
|
"epoch": 3.686857760951866,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00041158534668404325,
|
|
"loss": 5.1918,
|
|
"mean_token_accuracy": 0.19401995241641998,
|
|
"num_tokens": 7660527.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.302572727203369,
|
|
"epoch": 3.6922660897782587,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00041124425114591386,
|
|
"loss": 5.1224,
|
|
"mean_token_accuracy": 0.20590430349111558,
|
|
"num_tokens": 7671249.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.443627834320068,
|
|
"epoch": 3.697674418604651,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00041090266049586107,
|
|
"loss": 5.1283,
|
|
"mean_token_accuracy": 0.19254487603902817,
|
|
"num_tokens": 7681824.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.340479469299316,
|
|
"epoch": 3.703082747431044,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004105605759752258,
|
|
"loss": 5.1673,
|
|
"mean_token_accuracy": 0.19476546347141266,
|
|
"num_tokens": 7691916.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.405886936187744,
|
|
"epoch": 3.7084910762574363,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00041021799882714436,
|
|
"loss": 5.2133,
|
|
"mean_token_accuracy": 0.1952840730547905,
|
|
"num_tokens": 7703355.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.359181928634643,
|
|
"epoch": 3.713899405083829,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004098749302965426,
|
|
"loss": 5.1506,
|
|
"mean_token_accuracy": 0.202774016559124,
|
|
"num_tokens": 7714731.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.327584552764892,
|
|
"epoch": 3.7193077339102216,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004095313716301324,
|
|
"loss": 5.0756,
|
|
"mean_token_accuracy": 0.20912852883338928,
|
|
"num_tokens": 7725696.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.324631309509277,
|
|
"epoch": 3.7247160627366145,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004091873240764069,
|
|
"loss": 5.0897,
|
|
"mean_token_accuracy": 0.19863477498292922,
|
|
"num_tokens": 7737121.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.455405330657959,
|
|
"epoch": 3.730124391563007,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00040884278888563566,
|
|
"loss": 5.1817,
|
|
"mean_token_accuracy": 0.20030153095722197,
|
|
"num_tokens": 7749088.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.4676737785339355,
|
|
"epoch": 3.7355327203893998,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00040849776730986014,
|
|
"loss": 5.278,
|
|
"mean_token_accuracy": 0.19652654379606246,
|
|
"num_tokens": 7760676.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.364774465560913,
|
|
"epoch": 3.740941049215792,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004081522606028899,
|
|
"loss": 5.1943,
|
|
"mean_token_accuracy": 0.19928837120532988,
|
|
"num_tokens": 7772560.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.378541755676269,
|
|
"epoch": 3.746349378042185,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00040780627002029674,
|
|
"loss": 5.122,
|
|
"mean_token_accuracy": 0.20663439929485322,
|
|
"num_tokens": 7782206.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.371955299377442,
|
|
"epoch": 3.7517577068685775,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004074597968194116,
|
|
"loss": 5.2127,
|
|
"mean_token_accuracy": 0.19306979030370713,
|
|
"num_tokens": 7793862.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.408205556869507,
|
|
"epoch": 3.7571660356949703,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004071128422593186,
|
|
"loss": 5.2422,
|
|
"mean_token_accuracy": 0.1988579049706459,
|
|
"num_tokens": 7805291.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.410107231140136,
|
|
"epoch": 3.7625743645213627,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004067654076008515,
|
|
"loss": 5.2088,
|
|
"mean_token_accuracy": 0.19674475789070128,
|
|
"num_tokens": 7816141.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.402823877334595,
|
|
"epoch": 3.7679826933477556,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00040641749410658877,
|
|
"loss": 5.1967,
|
|
"mean_token_accuracy": 0.18460217714309693,
|
|
"num_tokens": 7826209.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.381054639816284,
|
|
"epoch": 3.773391022174148,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004060691030408487,
|
|
"loss": 5.1423,
|
|
"mean_token_accuracy": 0.19456969797611237,
|
|
"num_tokens": 7836734.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.388647413253784,
|
|
"epoch": 3.778799351000541,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00040572023566968536,
|
|
"loss": 5.1973,
|
|
"mean_token_accuracy": 0.19378932565450668,
|
|
"num_tokens": 7848283.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.405528020858765,
|
|
"epoch": 3.7842076798269337,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00040537089326088345,
|
|
"loss": 5.0904,
|
|
"mean_token_accuracy": 0.2019022896885872,
|
|
"num_tokens": 7861226.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.2667375087738035,
|
|
"epoch": 3.789616008653326,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004050210770839541,
|
|
"loss": 5.086,
|
|
"mean_token_accuracy": 0.2090648129582405,
|
|
"num_tokens": 7872426.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.419815540313721,
|
|
"epoch": 3.7950243374797186,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004046707884101303,
|
|
"loss": 5.1572,
|
|
"mean_token_accuracy": 0.20294291526079178,
|
|
"num_tokens": 7883884.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.2453395366668705,
|
|
"epoch": 3.8004326663061114,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00040432002851236164,
|
|
"loss": 5.0923,
|
|
"mean_token_accuracy": 0.20555179119110106,
|
|
"num_tokens": 7895415.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.343411064147949,
|
|
"epoch": 3.8058409951325043,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004039687986653106,
|
|
"loss": 5.1891,
|
|
"mean_token_accuracy": 0.19917936623096466,
|
|
"num_tokens": 7907444.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.510727262496948,
|
|
"epoch": 3.8112493239588967,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00040361710014534714,
|
|
"loss": 5.1789,
|
|
"mean_token_accuracy": 0.19772082865238189,
|
|
"num_tokens": 7918811.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.372313737869263,
|
|
"epoch": 3.816657652785289,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004032649342305445,
|
|
"loss": 5.2419,
|
|
"mean_token_accuracy": 0.1921569287776947,
|
|
"num_tokens": 7930657.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.249884796142578,
|
|
"epoch": 3.822065981611682,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004029123022006745,
|
|
"loss": 5.0887,
|
|
"mean_token_accuracy": 0.19494519084692002,
|
|
"num_tokens": 7942469.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.464029169082641,
|
|
"epoch": 3.827474310438075,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004025592053372027,
|
|
"loss": 5.2293,
|
|
"mean_token_accuracy": 0.19938496500253677,
|
|
"num_tokens": 7954034.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.344295024871826,
|
|
"epoch": 3.8328826392644673,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004022056449232839,
|
|
"loss": 5.1304,
|
|
"mean_token_accuracy": 0.2021975964307785,
|
|
"num_tokens": 7966079.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.2508948802947994,
|
|
"epoch": 3.8382909680908597,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004018516222437575,
|
|
"loss": 5.1021,
|
|
"mean_token_accuracy": 0.20688396096229553,
|
|
"num_tokens": 7977785.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.473352336883545,
|
|
"epoch": 3.8436992969172525,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004014971385851428,
|
|
"loss": 5.2744,
|
|
"mean_token_accuracy": 0.19294358491897584,
|
|
"num_tokens": 7988874.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.259639596939087,
|
|
"epoch": 3.8491076257436454,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000401142195235634,
|
|
"loss": 5.0543,
|
|
"mean_token_accuracy": 0.20376185923814774,
|
|
"num_tokens": 7999438.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.396929931640625,
|
|
"epoch": 3.854515954570038,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004007867934850963,
|
|
"loss": 5.1423,
|
|
"mean_token_accuracy": 0.20456838458776475,
|
|
"num_tokens": 8009658.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.415792226791382,
|
|
"epoch": 3.8599242833964302,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004004309346250603,
|
|
"loss": 5.1072,
|
|
"mean_token_accuracy": 0.19776116609573363,
|
|
"num_tokens": 8020988.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.254811143875122,
|
|
"epoch": 3.865332612222823,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004000746199487181,
|
|
"loss": 5.0038,
|
|
"mean_token_accuracy": 0.20381066501140593,
|
|
"num_tokens": 8031329.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.280174827575683,
|
|
"epoch": 3.870740941049216,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000399717850750918,
|
|
"loss": 5.0974,
|
|
"mean_token_accuracy": 0.2014472782611847,
|
|
"num_tokens": 8041693.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.344280099868774,
|
|
"epoch": 3.8761492698756084,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00039936062832816,
|
|
"loss": 5.1511,
|
|
"mean_token_accuracy": 0.20884522050619125,
|
|
"num_tokens": 8053123.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.424546575546264,
|
|
"epoch": 3.8815575987020012,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003990029539785913,
|
|
"loss": 5.1492,
|
|
"mean_token_accuracy": 0.19642920643091202,
|
|
"num_tokens": 8064270.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.334866285324097,
|
|
"epoch": 3.8869659275283936,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00039864482900200135,
|
|
"loss": 5.0679,
|
|
"mean_token_accuracy": 0.20613076388835908,
|
|
"num_tokens": 8074631.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.31110029220581,
|
|
"epoch": 3.8923742563547865,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003982862546998169,
|
|
"loss": 5.1427,
|
|
"mean_token_accuracy": 0.20392687916755675,
|
|
"num_tokens": 8086794.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.390283107757568,
|
|
"epoch": 3.897782585181179,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003979272323750981,
|
|
"loss": 5.1572,
|
|
"mean_token_accuracy": 0.19975161850452422,
|
|
"num_tokens": 8097595.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.341314506530762,
|
|
"epoch": 3.903190914007572,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003975677633325327,
|
|
"loss": 5.1215,
|
|
"mean_token_accuracy": 0.2041239008307457,
|
|
"num_tokens": 8107174.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.320006370544434,
|
|
"epoch": 3.908599242833964,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003972078488784321,
|
|
"loss": 5.1279,
|
|
"mean_token_accuracy": 0.20672304034233094,
|
|
"num_tokens": 8118065.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.284181880950928,
|
|
"epoch": 3.914007571660357,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003968474903207263,
|
|
"loss": 5.1168,
|
|
"mean_token_accuracy": 0.19935960620641707,
|
|
"num_tokens": 8130604.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.388709688186646,
|
|
"epoch": 3.9194159004867495,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003964866889689592,
|
|
"loss": 5.1793,
|
|
"mean_token_accuracy": 0.2032311424612999,
|
|
"num_tokens": 8141808.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.409710884094238,
|
|
"epoch": 3.9248242293131423,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003961254461342834,
|
|
"loss": 5.1402,
|
|
"mean_token_accuracy": 0.19552997648715972,
|
|
"num_tokens": 8152101.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.369632053375244,
|
|
"epoch": 3.9302325581395348,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003957637631294566,
|
|
"loss": 5.1737,
|
|
"mean_token_accuracy": 0.20101014971733094,
|
|
"num_tokens": 8163963.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.358480834960938,
|
|
"epoch": 3.9356408869659276,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00039540164126883554,
|
|
"loss": 5.0833,
|
|
"mean_token_accuracy": 0.2070899173617363,
|
|
"num_tokens": 8174063.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.267921733856201,
|
|
"epoch": 3.94104921579232,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003950390818683718,
|
|
"loss": 5.0868,
|
|
"mean_token_accuracy": 0.20732434540987016,
|
|
"num_tokens": 8186800.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.353105878829956,
|
|
"epoch": 3.946457544618713,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003946760862456071,
|
|
"loss": 5.179,
|
|
"mean_token_accuracy": 0.19387145340442657,
|
|
"num_tokens": 8198420.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.426562786102295,
|
|
"epoch": 3.9518658734451053,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003943126557196685,
|
|
"loss": 5.2141,
|
|
"mean_token_accuracy": 0.19274465888738632,
|
|
"num_tokens": 8209384.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.342242336273193,
|
|
"epoch": 3.957274202271498,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00039394879161126333,
|
|
"loss": 5.1577,
|
|
"mean_token_accuracy": 0.19894379526376724,
|
|
"num_tokens": 8220356.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.289758968353271,
|
|
"epoch": 3.9626825310978906,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003935844952426745,
|
|
"loss": 5.0946,
|
|
"mean_token_accuracy": 0.20172848254442216,
|
|
"num_tokens": 8231597.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.28520245552063,
|
|
"epoch": 3.9680908599242835,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00039321976793775586,
|
|
"loss": 5.0617,
|
|
"mean_token_accuracy": 0.20787712931632996,
|
|
"num_tokens": 8242554.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.445166969299317,
|
|
"epoch": 3.973499188750676,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003928546110219275,
|
|
"loss": 5.2891,
|
|
"mean_token_accuracy": 0.1908406674861908,
|
|
"num_tokens": 8253396.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.2753277778625485,
|
|
"epoch": 3.9789075175770687,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00039248902582217024,
|
|
"loss": 5.0613,
|
|
"mean_token_accuracy": 0.20754480361938477,
|
|
"num_tokens": 8265193.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.277149820327759,
|
|
"epoch": 3.9843158464034616,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00039212301366702153,
|
|
"loss": 5.1742,
|
|
"mean_token_accuracy": 0.19748681783676147,
|
|
"num_tokens": 8276343.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.429461860656739,
|
|
"epoch": 3.989724175229854,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00039175657588657063,
|
|
"loss": 5.1595,
|
|
"mean_token_accuracy": 0.2013916090130806,
|
|
"num_tokens": 8288069.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.310443496704101,
|
|
"epoch": 3.9951325040562464,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003913897138124531,
|
|
"loss": 5.1294,
|
|
"mean_token_accuracy": 0.20000299513339997,
|
|
"num_tokens": 8299382.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.343001259697808,
|
|
"epoch": 4.0,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 0.0003910224287778466,
|
|
"loss": 5.0553,
|
|
"mean_token_accuracy": 0.20640678041511112,
|
|
"num_tokens": 8307556.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.359926223754883,
|
|
"epoch": 4.005408328826393,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00039065472211746586,
|
|
"loss": 4.843,
|
|
"mean_token_accuracy": 0.22658650428056717,
|
|
"num_tokens": 8317851.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.207902908325195,
|
|
"epoch": 4.010816657652786,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003902865951675576,
|
|
"loss": 4.7503,
|
|
"mean_token_accuracy": 0.2283554971218109,
|
|
"num_tokens": 8328485.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.275643968582154,
|
|
"epoch": 4.016224986479178,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00038991804926589616,
|
|
"loss": 4.7076,
|
|
"mean_token_accuracy": 0.22811219841241837,
|
|
"num_tokens": 8339625.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.299853277206421,
|
|
"epoch": 4.0216333153055706,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000389549085751778,
|
|
"loss": 4.7707,
|
|
"mean_token_accuracy": 0.2248838096857071,
|
|
"num_tokens": 8350667.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.304503488540649,
|
|
"epoch": 4.027041644131963,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00038917970596601765,
|
|
"loss": 4.8172,
|
|
"mean_token_accuracy": 0.21980251669883727,
|
|
"num_tokens": 8362521.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.285195016860962,
|
|
"epoch": 4.032449972958356,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003888099112509419,
|
|
"loss": 4.8742,
|
|
"mean_token_accuracy": 0.2134528249502182,
|
|
"num_tokens": 8372703.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.351338911056518,
|
|
"epoch": 4.037858301784748,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00038843970295038564,
|
|
"loss": 4.8374,
|
|
"mean_token_accuracy": 0.21921467334032058,
|
|
"num_tokens": 8383345.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.239619302749634,
|
|
"epoch": 4.043266630611141,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00038806908240968666,
|
|
"loss": 4.7624,
|
|
"mean_token_accuracy": 0.22690549343824387,
|
|
"num_tokens": 8394056.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.2638486385345455,
|
|
"epoch": 4.048674959437534,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000387698050975681,
|
|
"loss": 4.8077,
|
|
"mean_token_accuracy": 0.2155721440911293,
|
|
"num_tokens": 8403889.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.224754476547242,
|
|
"epoch": 4.054083288263927,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003873266099966976,
|
|
"loss": 4.8132,
|
|
"mean_token_accuracy": 0.21631087064743043,
|
|
"num_tokens": 8415169.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.250479793548584,
|
|
"epoch": 4.059491617090319,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00038695476082255395,
|
|
"loss": 4.8285,
|
|
"mean_token_accuracy": 0.2161998987197876,
|
|
"num_tokens": 8427056.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.242709922790527,
|
|
"epoch": 4.064899945916712,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003865825048045507,
|
|
"loss": 4.7994,
|
|
"mean_token_accuracy": 0.21993166953325272,
|
|
"num_tokens": 8438900.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.269728946685791,
|
|
"epoch": 4.0703082747431045,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003862098432954672,
|
|
"loss": 4.7824,
|
|
"mean_token_accuracy": 0.22001834362745284,
|
|
"num_tokens": 8449037.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.178552722930908,
|
|
"epoch": 4.075716603569497,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003858367776495561,
|
|
"loss": 4.7652,
|
|
"mean_token_accuracy": 0.21764655262231827,
|
|
"num_tokens": 8459409.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.280428457260132,
|
|
"epoch": 4.081124932395889,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00038546330922253886,
|
|
"loss": 4.8593,
|
|
"mean_token_accuracy": 0.217335844039917,
|
|
"num_tokens": 8470190.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.311921834945679,
|
|
"epoch": 4.086533261222282,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003850894393716007,
|
|
"loss": 4.8364,
|
|
"mean_token_accuracy": 0.21542377322912215,
|
|
"num_tokens": 8481559.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.307945966720581,
|
|
"epoch": 4.091941590048675,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003847151694553855,
|
|
"loss": 4.8655,
|
|
"mean_token_accuracy": 0.21167138069868088,
|
|
"num_tokens": 8492607.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.291337728500366,
|
|
"epoch": 4.097349918875068,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003843405008339908,
|
|
"loss": 4.9612,
|
|
"mean_token_accuracy": 0.21009797751903533,
|
|
"num_tokens": 8504954.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.394812154769897,
|
|
"epoch": 4.10275824770146,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00038396543486896354,
|
|
"loss": 4.8245,
|
|
"mean_token_accuracy": 0.21956509202718735,
|
|
"num_tokens": 8516250.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.212130975723267,
|
|
"epoch": 4.108166576527853,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00038358997292329406,
|
|
"loss": 4.8087,
|
|
"mean_token_accuracy": 0.22222018241882324,
|
|
"num_tokens": 8527220.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.283579683303833,
|
|
"epoch": 4.113574905354246,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00038321411636141214,
|
|
"loss": 4.8999,
|
|
"mean_token_accuracy": 0.21446193307638167,
|
|
"num_tokens": 8538037.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.265017557144165,
|
|
"epoch": 4.1189832341806385,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003828378665491811,
|
|
"loss": 4.863,
|
|
"mean_token_accuracy": 0.2218945801258087,
|
|
"num_tokens": 8548684.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.164436197280883,
|
|
"epoch": 4.1243915630070305,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000382461224853894,
|
|
"loss": 4.7784,
|
|
"mean_token_accuracy": 0.21653792411088943,
|
|
"num_tokens": 8559756.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.306323003768921,
|
|
"epoch": 4.129799891833423,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003820841926442674,
|
|
"loss": 4.8142,
|
|
"mean_token_accuracy": 0.22202768176794052,
|
|
"num_tokens": 8570640.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.270534706115723,
|
|
"epoch": 4.135208220659816,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003817067712904372,
|
|
"loss": 4.837,
|
|
"mean_token_accuracy": 0.21775893718004227,
|
|
"num_tokens": 8583127.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.218170261383056,
|
|
"epoch": 4.140616549486209,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00038132896216395355,
|
|
"loss": 4.8082,
|
|
"mean_token_accuracy": 0.22074409127235411,
|
|
"num_tokens": 8595213.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.303803300857544,
|
|
"epoch": 4.146024878312601,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003809507666377758,
|
|
"loss": 4.82,
|
|
"mean_token_accuracy": 0.2326643794775009,
|
|
"num_tokens": 8606057.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.293350505828857,
|
|
"epoch": 4.151433207138994,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00038057218608626714,
|
|
"loss": 4.8684,
|
|
"mean_token_accuracy": 0.2255711928009987,
|
|
"num_tokens": 8617660.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.211659240722656,
|
|
"epoch": 4.156841535965387,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003801932218851903,
|
|
"loss": 4.815,
|
|
"mean_token_accuracy": 0.22232111990451814,
|
|
"num_tokens": 8628595.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.286400556564331,
|
|
"epoch": 4.16224986479178,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000379813875411702,
|
|
"loss": 4.8567,
|
|
"mean_token_accuracy": 0.21733250021934508,
|
|
"num_tokens": 8639831.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.222960090637207,
|
|
"epoch": 4.167658193618172,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00037943414804434826,
|
|
"loss": 4.7967,
|
|
"mean_token_accuracy": 0.21822197884321212,
|
|
"num_tokens": 8650470.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.280391597747803,
|
|
"epoch": 4.173066522444564,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003790540411630592,
|
|
"loss": 4.912,
|
|
"mean_token_accuracy": 0.21849605292081833,
|
|
"num_tokens": 8660969.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.1879082202911375,
|
|
"epoch": 4.178474851270957,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00037867355614914404,
|
|
"loss": 4.7881,
|
|
"mean_token_accuracy": 0.22661226391792297,
|
|
"num_tokens": 8670913.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.258349657058716,
|
|
"epoch": 4.18388318009735,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003782926943852864,
|
|
"loss": 4.9249,
|
|
"mean_token_accuracy": 0.21352409571409225,
|
|
"num_tokens": 8683028.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.363445615768432,
|
|
"epoch": 4.189291508923742,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00037791145725553853,
|
|
"loss": 4.9867,
|
|
"mean_token_accuracy": 0.2116711527109146,
|
|
"num_tokens": 8696454.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.333937883377075,
|
|
"epoch": 4.194699837750135,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003775298461453172,
|
|
"loss": 4.8027,
|
|
"mean_token_accuracy": 0.2220937728881836,
|
|
"num_tokens": 8707112.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.133482885360718,
|
|
"epoch": 4.200108166576528,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000377147862441398,
|
|
"loss": 4.8086,
|
|
"mean_token_accuracy": 0.22494329363107682,
|
|
"num_tokens": 8717696.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.218525838851929,
|
|
"epoch": 4.205516495402921,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00037676550753191075,
|
|
"loss": 4.7421,
|
|
"mean_token_accuracy": 0.22583360522985457,
|
|
"num_tokens": 8728571.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.2811414241790775,
|
|
"epoch": 4.210924824229313,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003763827828063339,
|
|
"loss": 4.8556,
|
|
"mean_token_accuracy": 0.22195011824369432,
|
|
"num_tokens": 8739582.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.267028570175171,
|
|
"epoch": 4.2163331530557056,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00037599968965549,
|
|
"loss": 4.8102,
|
|
"mean_token_accuracy": 0.22244389802217485,
|
|
"num_tokens": 8750527.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.238075637817383,
|
|
"epoch": 4.221741481882098,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003756162294715405,
|
|
"loss": 4.7992,
|
|
"mean_token_accuracy": 0.22126877605915068,
|
|
"num_tokens": 8763140.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.221660470962524,
|
|
"epoch": 4.227149810708491,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00037523240364798063,
|
|
"loss": 4.8479,
|
|
"mean_token_accuracy": 0.21473010778427123,
|
|
"num_tokens": 8774328.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.268135356903076,
|
|
"epoch": 4.232558139534884,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00037484821357963423,
|
|
"loss": 4.8719,
|
|
"mean_token_accuracy": 0.21630001217126846,
|
|
"num_tokens": 8785119.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.260245847702026,
|
|
"epoch": 4.237966468361276,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000374463660662649,
|
|
"loss": 4.8329,
|
|
"mean_token_accuracy": 0.21931773126125337,
|
|
"num_tokens": 8795833.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.239442968368531,
|
|
"epoch": 4.243374797187669,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000374078746294491,
|
|
"loss": 4.9394,
|
|
"mean_token_accuracy": 0.2120554953813553,
|
|
"num_tokens": 8807098.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.244119548797608,
|
|
"epoch": 4.248783126014062,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00037369347187393996,
|
|
"loss": 4.749,
|
|
"mean_token_accuracy": 0.2268562287092209,
|
|
"num_tokens": 8818096.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.2393505573272705,
|
|
"epoch": 4.254191454840455,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003733078388010842,
|
|
"loss": 4.9105,
|
|
"mean_token_accuracy": 0.20613012313842774,
|
|
"num_tokens": 8829451.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.304215049743652,
|
|
"epoch": 4.259599783666847,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003729218484773149,
|
|
"loss": 4.8522,
|
|
"mean_token_accuracy": 0.2258807674050331,
|
|
"num_tokens": 8840099.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.1575202465057375,
|
|
"epoch": 4.2650081124932395,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003725355023053221,
|
|
"loss": 4.7564,
|
|
"mean_token_accuracy": 0.22680438607931136,
|
|
"num_tokens": 8851269.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.21615195274353,
|
|
"epoch": 4.270416441319632,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003721488016890884,
|
|
"loss": 4.7837,
|
|
"mean_token_accuracy": 0.2269557699561119,
|
|
"num_tokens": 8862871.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.301479005813599,
|
|
"epoch": 4.275824770146025,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00037176174803388496,
|
|
"loss": 4.8177,
|
|
"mean_token_accuracy": 0.2245062917470932,
|
|
"num_tokens": 8874328.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.250480890274048,
|
|
"epoch": 4.281233098972417,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00037137434274626545,
|
|
"loss": 4.8478,
|
|
"mean_token_accuracy": 0.2159382551908493,
|
|
"num_tokens": 8885133.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.241719770431518,
|
|
"epoch": 4.28664142779881,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003709865872340618,
|
|
"loss": 4.856,
|
|
"mean_token_accuracy": 0.22154630571603776,
|
|
"num_tokens": 8896479.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.239784240722656,
|
|
"epoch": 4.292049756625203,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003705984829063782,
|
|
"loss": 4.7548,
|
|
"mean_token_accuracy": 0.22353190779685975,
|
|
"num_tokens": 8907966.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.153713750839233,
|
|
"epoch": 4.297458085451596,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003702100311735869,
|
|
"loss": 4.8434,
|
|
"mean_token_accuracy": 0.2146587073802948,
|
|
"num_tokens": 8919042.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.213410997390747,
|
|
"epoch": 4.302866414277988,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00036982123344732223,
|
|
"loss": 4.8636,
|
|
"mean_token_accuracy": 0.21314563006162643,
|
|
"num_tokens": 8930456.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.223792314529419,
|
|
"epoch": 4.308274743104381,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00036943209114047613,
|
|
"loss": 4.9079,
|
|
"mean_token_accuracy": 0.21624339371919632,
|
|
"num_tokens": 8941473.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.274001741409302,
|
|
"epoch": 4.3136830719307735,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003690426056671926,
|
|
"loss": 4.9224,
|
|
"mean_token_accuracy": 0.21655383110046386,
|
|
"num_tokens": 8952927.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.263446712493897,
|
|
"epoch": 4.319091400757166,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003686527784428627,
|
|
"loss": 4.8757,
|
|
"mean_token_accuracy": 0.21662932336330415,
|
|
"num_tokens": 8963831.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.222810077667236,
|
|
"epoch": 4.324499729583558,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00036826261088411955,
|
|
"loss": 4.8245,
|
|
"mean_token_accuracy": 0.21820705980062485,
|
|
"num_tokens": 8977147.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 4.324499729583558,
|
|
"eval_entropy": 5.212938701022755,
|
|
"eval_loss": 5.915652751922607,
|
|
"eval_mean_token_accuracy": 0.17569336684590037,
|
|
"eval_num_tokens": 8977147.0,
|
|
"eval_runtime": 2.7171,
|
|
"eval_samples_per_second": 1293.65,
|
|
"eval_steps_per_second": 161.936,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 5.284003448486328,
|
|
"epoch": 4.329908058409951,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003678721044088329,
|
|
"loss": 4.856,
|
|
"mean_token_accuracy": 0.22199487090110778,
|
|
"num_tokens": 8990031.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 5.185904455184937,
|
|
"epoch": 4.335316387236344,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003674812604361043,
|
|
"loss": 4.7592,
|
|
"mean_token_accuracy": 0.22557856291532516,
|
|
"num_tokens": 9000555.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 5.161986923217773,
|
|
"epoch": 4.340724716062737,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003670900803862615,
|
|
"loss": 4.7828,
|
|
"mean_token_accuracy": 0.22235632091760635,
|
|
"num_tokens": 9011631.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 5.224789571762085,
|
|
"epoch": 4.346133044889129,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003666985656808537,
|
|
"loss": 4.8136,
|
|
"mean_token_accuracy": 0.22236808836460115,
|
|
"num_tokens": 9023073.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 5.22872486114502,
|
|
"epoch": 4.351541373715522,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00036630671774264635,
|
|
"loss": 4.8648,
|
|
"mean_token_accuracy": 0.21899616420269014,
|
|
"num_tokens": 9033360.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 5.173056936264038,
|
|
"epoch": 4.356949702541915,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003659145379956158,
|
|
"loss": 4.8102,
|
|
"mean_token_accuracy": 0.2183179572224617,
|
|
"num_tokens": 9044404.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 5.2909932136535645,
|
|
"epoch": 4.3623580313683075,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00036552202786494393,
|
|
"loss": 4.8098,
|
|
"mean_token_accuracy": 0.2234760493040085,
|
|
"num_tokens": 9054826.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 5.214385557174682,
|
|
"epoch": 4.367766360194699,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00036512918877701367,
|
|
"loss": 4.8703,
|
|
"mean_token_accuracy": 0.218776935338974,
|
|
"num_tokens": 9065363.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 5.20676817893982,
|
|
"epoch": 4.373174689021092,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00036473602215940316,
|
|
"loss": 4.8206,
|
|
"mean_token_accuracy": 0.22543764412403106,
|
|
"num_tokens": 9075692.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 5.284393405914306,
|
|
"epoch": 4.378583017847485,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003643425294408807,
|
|
"loss": 4.9171,
|
|
"mean_token_accuracy": 0.21689888834953308,
|
|
"num_tokens": 9087618.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 5.261648321151734,
|
|
"epoch": 4.383991346673878,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00036394871205139984,
|
|
"loss": 4.8215,
|
|
"mean_token_accuracy": 0.21469242572784425,
|
|
"num_tokens": 9098508.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 5.173863506317138,
|
|
"epoch": 4.38939967550027,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003635545714220938,
|
|
"loss": 4.8208,
|
|
"mean_token_accuracy": 0.2154018446803093,
|
|
"num_tokens": 9109003.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 5.221544027328491,
|
|
"epoch": 4.394808004326663,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003631601089852706,
|
|
"loss": 4.8486,
|
|
"mean_token_accuracy": 0.21496025323867798,
|
|
"num_tokens": 9120615.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 5.303488302230835,
|
|
"epoch": 4.400216333153056,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003627653261744075,
|
|
"loss": 4.9629,
|
|
"mean_token_accuracy": 0.21106828153133392,
|
|
"num_tokens": 9131814.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 5.272027015686035,
|
|
"epoch": 4.405624661979449,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00036237022442414636,
|
|
"loss": 4.8659,
|
|
"mean_token_accuracy": 0.21529262363910676,
|
|
"num_tokens": 9142443.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 5.279282283782959,
|
|
"epoch": 4.411032990805841,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003619748051702878,
|
|
"loss": 4.8777,
|
|
"mean_token_accuracy": 0.21672908812761307,
|
|
"num_tokens": 9154433.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 5.170158195495605,
|
|
"epoch": 4.416441319632233,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003615790698497862,
|
|
"loss": 4.8503,
|
|
"mean_token_accuracy": 0.2149826854467392,
|
|
"num_tokens": 9165153.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 5.261667966842651,
|
|
"epoch": 4.421849648458626,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00036118301990074463,
|
|
"loss": 4.9329,
|
|
"mean_token_accuracy": 0.2128029778599739,
|
|
"num_tokens": 9176568.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 5.198450899124145,
|
|
"epoch": 4.427257977285019,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003607866567624097,
|
|
"loss": 4.8879,
|
|
"mean_token_accuracy": 0.21580330282449722,
|
|
"num_tokens": 9187205.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 5.3185113906860355,
|
|
"epoch": 4.432666306111411,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00036038998187516583,
|
|
"loss": 4.9412,
|
|
"mean_token_accuracy": 0.21081258952617646,
|
|
"num_tokens": 9198486.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 5.171548748016358,
|
|
"epoch": 4.438074634937804,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00035999299668053043,
|
|
"loss": 4.8378,
|
|
"mean_token_accuracy": 0.22099575251340867,
|
|
"num_tokens": 9209714.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.2236875057220455,
|
|
"epoch": 4.443482963764197,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003595957026211487,
|
|
"loss": 4.8163,
|
|
"mean_token_accuracy": 0.21579581052064895,
|
|
"num_tokens": 9221968.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 5.221615171432495,
|
|
"epoch": 4.44889129259059,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00035919810114078815,
|
|
"loss": 4.8882,
|
|
"mean_token_accuracy": 0.2084411174058914,
|
|
"num_tokens": 9234437.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 5.156426477432251,
|
|
"epoch": 4.4542996214169825,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00035880019368433354,
|
|
"loss": 4.7716,
|
|
"mean_token_accuracy": 0.22343935817480087,
|
|
"num_tokens": 9246181.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 5.283278560638427,
|
|
"epoch": 4.4597079502433745,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003584019816977813,
|
|
"loss": 4.8852,
|
|
"mean_token_accuracy": 0.2204149693250656,
|
|
"num_tokens": 9256951.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 5.219197797775268,
|
|
"epoch": 4.465116279069767,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00035800346662823493,
|
|
"loss": 4.831,
|
|
"mean_token_accuracy": 0.2290960133075714,
|
|
"num_tokens": 9268783.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 5.219265651702881,
|
|
"epoch": 4.47052460789616,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000357604649923899,
|
|
"loss": 4.8826,
|
|
"mean_token_accuracy": 0.21168054193258284,
|
|
"num_tokens": 9279541.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 5.201429986953736,
|
|
"epoch": 4.475932936722553,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003572055330340744,
|
|
"loss": 4.9031,
|
|
"mean_token_accuracy": 0.214005309343338,
|
|
"num_tokens": 9291041.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 5.172361278533936,
|
|
"epoch": 4.481341265548945,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00035680611740915264,
|
|
"loss": 4.8311,
|
|
"mean_token_accuracy": 0.21395229399204255,
|
|
"num_tokens": 9303451.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 5.263820934295654,
|
|
"epoch": 4.486749594375338,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003564064045006113,
|
|
"loss": 4.8696,
|
|
"mean_token_accuracy": 0.22041955292224885,
|
|
"num_tokens": 9315275.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 5.257158803939819,
|
|
"epoch": 4.492157923201731,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00035600639576100786,
|
|
"loss": 4.917,
|
|
"mean_token_accuracy": 0.2131795972585678,
|
|
"num_tokens": 9326987.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 5.189502859115601,
|
|
"epoch": 4.497566252028124,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00035560609264397497,
|
|
"loss": 4.8907,
|
|
"mean_token_accuracy": 0.21561261862516404,
|
|
"num_tokens": 9337885.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 5.173984575271606,
|
|
"epoch": 4.502974580854516,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00035520549660421515,
|
|
"loss": 4.8962,
|
|
"mean_token_accuracy": 0.215626323223114,
|
|
"num_tokens": 9348881.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 5.219392251968384,
|
|
"epoch": 4.5083829096809085,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003548046090974954,
|
|
"loss": 4.8479,
|
|
"mean_token_accuracy": 0.22406667470932007,
|
|
"num_tokens": 9361066.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 5.242049503326416,
|
|
"epoch": 4.513791238507301,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00035440343158064176,
|
|
"loss": 4.8032,
|
|
"mean_token_accuracy": 0.2318715989589691,
|
|
"num_tokens": 9371985.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 5.318117237091064,
|
|
"epoch": 4.519199567333694,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003540019655115341,
|
|
"loss": 4.9242,
|
|
"mean_token_accuracy": 0.21440426856279374,
|
|
"num_tokens": 9383756.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 5.17117223739624,
|
|
"epoch": 4.524607896160086,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00035360021234910133,
|
|
"loss": 4.7911,
|
|
"mean_token_accuracy": 0.2210676670074463,
|
|
"num_tokens": 9394892.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 5.201945972442627,
|
|
"epoch": 4.530016224986479,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00035319817355331537,
|
|
"loss": 4.8876,
|
|
"mean_token_accuracy": 0.2120958998799324,
|
|
"num_tokens": 9406646.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 5.268325567245483,
|
|
"epoch": 4.535424553812872,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003527958505851858,
|
|
"loss": 4.8444,
|
|
"mean_token_accuracy": 0.2203937754034996,
|
|
"num_tokens": 9417548.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 5.2219689846038815,
|
|
"epoch": 4.540832882639265,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00035239324490675546,
|
|
"loss": 4.8154,
|
|
"mean_token_accuracy": 0.22027368545532228,
|
|
"num_tokens": 9428481.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 5.196447086334229,
|
|
"epoch": 4.546241211465657,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003519903579810943,
|
|
"loss": 4.8214,
|
|
"mean_token_accuracy": 0.2232879653573036,
|
|
"num_tokens": 9439451.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 5.226025819778442,
|
|
"epoch": 4.55164954029205,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003515871912722944,
|
|
"loss": 4.8577,
|
|
"mean_token_accuracy": 0.2119873821735382,
|
|
"num_tokens": 9450141.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 5.257914733886719,
|
|
"epoch": 4.5570578691184425,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003511837462454643,
|
|
"loss": 4.9192,
|
|
"mean_token_accuracy": 0.21548717916011811,
|
|
"num_tokens": 9461309.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 5.243169164657592,
|
|
"epoch": 4.562466197944835,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00035078002436672443,
|
|
"loss": 4.8493,
|
|
"mean_token_accuracy": 0.22207145392894745,
|
|
"num_tokens": 9473147.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 5.151864624023437,
|
|
"epoch": 4.567874526771227,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003503760271032007,
|
|
"loss": 4.7687,
|
|
"mean_token_accuracy": 0.2189494013786316,
|
|
"num_tokens": 9483106.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 5.233551025390625,
|
|
"epoch": 4.57328285559762,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003499717559230205,
|
|
"loss": 4.8185,
|
|
"mean_token_accuracy": 0.21824068278074266,
|
|
"num_tokens": 9493735.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 5.177496290206909,
|
|
"epoch": 4.578691184424013,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00034956721229530586,
|
|
"loss": 4.8988,
|
|
"mean_token_accuracy": 0.21518220156431198,
|
|
"num_tokens": 9505385.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 5.346733570098877,
|
|
"epoch": 4.584099513250406,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003491623976901695,
|
|
"loss": 4.9815,
|
|
"mean_token_accuracy": 0.20884381532669066,
|
|
"num_tokens": 9518170.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 5.240568685531616,
|
|
"epoch": 4.589507842076799,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003487573135787085,
|
|
"loss": 4.9228,
|
|
"mean_token_accuracy": 0.2190106675028801,
|
|
"num_tokens": 9530416.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 5.167938852310181,
|
|
"epoch": 4.594916170903191,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00034835196143299956,
|
|
"loss": 4.7757,
|
|
"mean_token_accuracy": 0.22761004120111467,
|
|
"num_tokens": 9542365.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 5.255697774887085,
|
|
"epoch": 4.600324499729584,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00034794634272609313,
|
|
"loss": 4.913,
|
|
"mean_token_accuracy": 0.2066371262073517,
|
|
"num_tokens": 9552903.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 5.236261510848999,
|
|
"epoch": 4.605732828555976,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003475404589320089,
|
|
"loss": 4.9102,
|
|
"mean_token_accuracy": 0.2165045663714409,
|
|
"num_tokens": 9563311.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 5.344020509719849,
|
|
"epoch": 4.611141157382368,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003471343115257291,
|
|
"loss": 4.9745,
|
|
"mean_token_accuracy": 0.20958462059497834,
|
|
"num_tokens": 9574673.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 5.247348642349243,
|
|
"epoch": 4.616549486208761,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00034672790198319453,
|
|
"loss": 4.898,
|
|
"mean_token_accuracy": 0.20761748999357224,
|
|
"num_tokens": 9584572.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 5.1242717742919925,
|
|
"epoch": 4.621957815035154,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003463212317812985,
|
|
"loss": 4.7516,
|
|
"mean_token_accuracy": 0.22929577976465226,
|
|
"num_tokens": 9594828.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 5.1937650680542,
|
|
"epoch": 4.627366143861547,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003459143023978813,
|
|
"loss": 4.91,
|
|
"mean_token_accuracy": 0.2139766737818718,
|
|
"num_tokens": 9606456.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 5.271865177154541,
|
|
"epoch": 4.63277447268794,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003455071153117252,
|
|
"loss": 4.8158,
|
|
"mean_token_accuracy": 0.21743298768997193,
|
|
"num_tokens": 9618092.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 5.1861162185668945,
|
|
"epoch": 4.638182801514332,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003450996720025493,
|
|
"loss": 4.8719,
|
|
"mean_token_accuracy": 0.22271620482206345,
|
|
"num_tokens": 9629344.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 5.231480407714844,
|
|
"epoch": 4.643591130340725,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00034469197395100323,
|
|
"loss": 4.8842,
|
|
"mean_token_accuracy": 0.21185634285211563,
|
|
"num_tokens": 9639956.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 5.249001359939575,
|
|
"epoch": 4.6489994591671175,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00034428402263866285,
|
|
"loss": 4.8234,
|
|
"mean_token_accuracy": 0.21541695445775985,
|
|
"num_tokens": 9649528.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 5.248314237594604,
|
|
"epoch": 4.6544077879935095,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00034387581954802406,
|
|
"loss": 4.8862,
|
|
"mean_token_accuracy": 0.21263853162527085,
|
|
"num_tokens": 9661374.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 5.230815505981445,
|
|
"epoch": 4.659816116819902,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00034346736616249786,
|
|
"loss": 4.8954,
|
|
"mean_token_accuracy": 0.21234040409326554,
|
|
"num_tokens": 9672359.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 5.269152402877808,
|
|
"epoch": 4.665224445646295,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00034305866396640473,
|
|
"loss": 4.9423,
|
|
"mean_token_accuracy": 0.21107330471277236,
|
|
"num_tokens": 9683748.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 5.304159355163574,
|
|
"epoch": 4.670632774472688,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00034264971444496954,
|
|
"loss": 4.8897,
|
|
"mean_token_accuracy": 0.21753455847501754,
|
|
"num_tokens": 9696049.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 5.202760171890259,
|
|
"epoch": 4.676041103299081,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00034224051908431574,
|
|
"loss": 4.8659,
|
|
"mean_token_accuracy": 0.22080424726009368,
|
|
"num_tokens": 9707358.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 5.220529174804687,
|
|
"epoch": 4.681449432125473,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003418310793714602,
|
|
"loss": 4.9011,
|
|
"mean_token_accuracy": 0.21544400006532669,
|
|
"num_tokens": 9718890.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 5.205951786041259,
|
|
"epoch": 4.686857760951866,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003414213967943077,
|
|
"loss": 4.7268,
|
|
"mean_token_accuracy": 0.2349315583705902,
|
|
"num_tokens": 9730395.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 5.254697322845459,
|
|
"epoch": 4.692266089778259,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00034101147284164583,
|
|
"loss": 4.844,
|
|
"mean_token_accuracy": 0.21442699581384658,
|
|
"num_tokens": 9741846.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 5.181782341003418,
|
|
"epoch": 4.6976744186046515,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00034060130900313905,
|
|
"loss": 4.867,
|
|
"mean_token_accuracy": 0.21829884946346284,
|
|
"num_tokens": 9752936.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 5.1658162593841555,
|
|
"epoch": 4.7030827474310435,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00034019090676932383,
|
|
"loss": 4.8758,
|
|
"mean_token_accuracy": 0.22126574665308,
|
|
"num_tokens": 9763913.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 5.256973648071289,
|
|
"epoch": 4.708491076257436,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003397802676316027,
|
|
"loss": 4.9336,
|
|
"mean_token_accuracy": 0.21468402147293092,
|
|
"num_tokens": 9775682.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 5.237012147903442,
|
|
"epoch": 4.713899405083829,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00033936939308223935,
|
|
"loss": 4.9506,
|
|
"mean_token_accuracy": 0.21234848946332932,
|
|
"num_tokens": 9787403.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 5.300046491622925,
|
|
"epoch": 4.719307733910222,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00033895828461435275,
|
|
"loss": 4.8637,
|
|
"mean_token_accuracy": 0.219329135119915,
|
|
"num_tokens": 9798166.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 5.243693923950195,
|
|
"epoch": 4.724716062736614,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003385469437219121,
|
|
"loss": 4.8334,
|
|
"mean_token_accuracy": 0.21857866495847703,
|
|
"num_tokens": 9809405.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 5.228618860244751,
|
|
"epoch": 4.730124391563007,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00033813537189973105,
|
|
"loss": 4.8734,
|
|
"mean_token_accuracy": 0.2202115312218666,
|
|
"num_tokens": 9820953.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 5.293294429779053,
|
|
"epoch": 4.7355327203894,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00033772357064346266,
|
|
"loss": 4.917,
|
|
"mean_token_accuracy": 0.22025146782398225,
|
|
"num_tokens": 9831916.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 5.202264261245728,
|
|
"epoch": 4.740941049215793,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003373115414495935,
|
|
"loss": 4.8534,
|
|
"mean_token_accuracy": 0.22620849162340165,
|
|
"num_tokens": 9841800.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 5.26312780380249,
|
|
"epoch": 4.746349378042185,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003368992858154389,
|
|
"loss": 4.9496,
|
|
"mean_token_accuracy": 0.2213471382856369,
|
|
"num_tokens": 9854713.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 5.30489296913147,
|
|
"epoch": 4.7517577068685775,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00033648680523913643,
|
|
"loss": 4.8618,
|
|
"mean_token_accuracy": 0.2187336876988411,
|
|
"num_tokens": 9865730.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 5.2923425197601315,
|
|
"epoch": 4.75716603569497,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003360741012196417,
|
|
"loss": 4.8572,
|
|
"mean_token_accuracy": 0.2203948453068733,
|
|
"num_tokens": 9875771.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 5.262787532806397,
|
|
"epoch": 4.762574364521363,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000335661175256722,
|
|
"loss": 4.9603,
|
|
"mean_token_accuracy": 0.21174326539039612,
|
|
"num_tokens": 9888839.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 5.25744662284851,
|
|
"epoch": 4.767982693347756,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003352480288509513,
|
|
"loss": 4.8528,
|
|
"mean_token_accuracy": 0.21356087177991867,
|
|
"num_tokens": 9899150.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 5.293668794631958,
|
|
"epoch": 4.773391022174148,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00033483466350370445,
|
|
"loss": 4.9984,
|
|
"mean_token_accuracy": 0.2105510339140892,
|
|
"num_tokens": 9912083.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 5.262047386169433,
|
|
"epoch": 4.778799351000541,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00033442108071715215,
|
|
"loss": 4.9548,
|
|
"mean_token_accuracy": 0.2120837762951851,
|
|
"num_tokens": 9924217.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 5.389695119857788,
|
|
"epoch": 4.784207679826934,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00033400728199425525,
|
|
"loss": 4.9915,
|
|
"mean_token_accuracy": 0.21072959452867507,
|
|
"num_tokens": 9936298.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 5.159387016296387,
|
|
"epoch": 4.789616008653326,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00033359326883875917,
|
|
"loss": 4.8097,
|
|
"mean_token_accuracy": 0.22412396222352982,
|
|
"num_tokens": 9946752.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 5.189387226104737,
|
|
"epoch": 4.795024337479719,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00033317904275518873,
|
|
"loss": 4.8466,
|
|
"mean_token_accuracy": 0.22212548851966857,
|
|
"num_tokens": 9958445.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 5.316000604629517,
|
|
"epoch": 4.800432666306111,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003327646052488425,
|
|
"loss": 4.8859,
|
|
"mean_token_accuracy": 0.2146509885787964,
|
|
"num_tokens": 9970021.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 5.1887969970703125,
|
|
"epoch": 4.805840995132504,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003323499578257873,
|
|
"loss": 4.8338,
|
|
"mean_token_accuracy": 0.22547864466905593,
|
|
"num_tokens": 9979703.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 5.14729118347168,
|
|
"epoch": 4.811249323958897,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00033193510199285286,
|
|
"loss": 4.8708,
|
|
"mean_token_accuracy": 0.22148658484220504,
|
|
"num_tokens": 9991574.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 5.18608283996582,
|
|
"epoch": 4.816657652785289,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00033152003925762615,
|
|
"loss": 4.8131,
|
|
"mean_token_accuracy": 0.2199581190943718,
|
|
"num_tokens": 10002410.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 5.214183187484741,
|
|
"epoch": 4.822065981611682,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003311047711284462,
|
|
"loss": 4.7857,
|
|
"mean_token_accuracy": 0.22130955308675765,
|
|
"num_tokens": 10012576.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 5.25377631187439,
|
|
"epoch": 4.827474310438075,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003306892991143983,
|
|
"loss": 4.8767,
|
|
"mean_token_accuracy": 0.2185003563761711,
|
|
"num_tokens": 10024298.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 5.1521131038665775,
|
|
"epoch": 4.832882639264467,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003302736247253087,
|
|
"loss": 4.7927,
|
|
"mean_token_accuracy": 0.22591112852096557,
|
|
"num_tokens": 10035835.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 5.286689186096192,
|
|
"epoch": 4.83829096809086,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003298577494717391,
|
|
"loss": 4.9248,
|
|
"mean_token_accuracy": 0.21669082641601561,
|
|
"num_tokens": 10046880.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 5.277845048904419,
|
|
"epoch": 4.8436992969172525,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003294416748649812,
|
|
"loss": 4.8887,
|
|
"mean_token_accuracy": 0.2165769636631012,
|
|
"num_tokens": 10057761.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 5.0738983154296875,
|
|
"epoch": 4.849107625743645,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00032902540241705106,
|
|
"loss": 4.7458,
|
|
"mean_token_accuracy": 0.22820984423160554,
|
|
"num_tokens": 10069714.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 5.151675796508789,
|
|
"epoch": 4.854515954570038,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003286089336406837,
|
|
"loss": 4.7877,
|
|
"mean_token_accuracy": 0.22735073864459993,
|
|
"num_tokens": 10080292.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 5.19876503944397,
|
|
"epoch": 4.85992428339643,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00032819227004932745,
|
|
"loss": 4.9298,
|
|
"mean_token_accuracy": 0.21047663539648057,
|
|
"num_tokens": 10091707.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 5.226641225814819,
|
|
"epoch": 4.865332612222823,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000327775413157139,
|
|
"loss": 4.8224,
|
|
"mean_token_accuracy": 0.2169093444943428,
|
|
"num_tokens": 10103268.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 5.191351699829101,
|
|
"epoch": 4.870740941049216,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003273583644789772,
|
|
"loss": 4.8513,
|
|
"mean_token_accuracy": 0.21752614080905913,
|
|
"num_tokens": 10114690.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 5.235945081710815,
|
|
"epoch": 4.876149269875609,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00032694112553039795,
|
|
"loss": 4.7801,
|
|
"mean_token_accuracy": 0.2293578252196312,
|
|
"num_tokens": 10125126.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 5.16181206703186,
|
|
"epoch": 4.881557598702001,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003265236978276484,
|
|
"loss": 4.8136,
|
|
"mean_token_accuracy": 0.2172074317932129,
|
|
"num_tokens": 10137093.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 5.1497221946716305,
|
|
"epoch": 4.886965927528394,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00032610608288766185,
|
|
"loss": 4.7574,
|
|
"mean_token_accuracy": 0.22725875079631805,
|
|
"num_tokens": 10147487.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 5.280003833770752,
|
|
"epoch": 4.8923742563547865,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00032568828222805216,
|
|
"loss": 5.0106,
|
|
"mean_token_accuracy": 0.21792767941951752,
|
|
"num_tokens": 10160351.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 5.196026802062988,
|
|
"epoch": 4.897782585181179,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003252702973671075,
|
|
"loss": 4.9518,
|
|
"mean_token_accuracy": 0.20679229497909546,
|
|
"num_tokens": 10172059.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 5.260401773452759,
|
|
"epoch": 4.903190914007571,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00032485212982378615,
|
|
"loss": 4.9581,
|
|
"mean_token_accuracy": 0.2080315813422203,
|
|
"num_tokens": 10182758.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 5.29714183807373,
|
|
"epoch": 4.908599242833964,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003244337811177098,
|
|
"loss": 4.8905,
|
|
"mean_token_accuracy": 0.21965555399656295,
|
|
"num_tokens": 10193391.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 5.16299352645874,
|
|
"epoch": 4.914007571660357,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003240152527691586,
|
|
"loss": 4.8224,
|
|
"mean_token_accuracy": 0.22473730891942978,
|
|
"num_tokens": 10204298.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 5.264654016494751,
|
|
"epoch": 4.91941590048675,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00032359654629906555,
|
|
"loss": 4.9257,
|
|
"mean_token_accuracy": 0.21384968757629394,
|
|
"num_tokens": 10215115.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 5.267312002182007,
|
|
"epoch": 4.924824229313142,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00032317766322901106,
|
|
"loss": 4.8809,
|
|
"mean_token_accuracy": 0.21506910473108293,
|
|
"num_tokens": 10227649.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 5.148030090332031,
|
|
"epoch": 4.930232558139535,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000322758605081217,
|
|
"loss": 4.8557,
|
|
"mean_token_accuracy": 0.22122482657432557,
|
|
"num_tokens": 10238830.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 5.143020582199097,
|
|
"epoch": 4.935640886965928,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00032233937337854167,
|
|
"loss": 4.8656,
|
|
"mean_token_accuracy": 0.21658693850040436,
|
|
"num_tokens": 10250431.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 5.339710140228272,
|
|
"epoch": 4.9410492157923205,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003219199696444741,
|
|
"loss": 5.009,
|
|
"mean_token_accuracy": 0.21164430677890778,
|
|
"num_tokens": 10263070.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 5.188231134414673,
|
|
"epoch": 4.9464575446187125,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00032150039540312843,
|
|
"loss": 4.9129,
|
|
"mean_token_accuracy": 0.2143416628241539,
|
|
"num_tokens": 10275647.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 5.184282636642456,
|
|
"epoch": 4.951865873445105,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00032108065217923836,
|
|
"loss": 4.8935,
|
|
"mean_token_accuracy": 0.21629296541213988,
|
|
"num_tokens": 10286651.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 5.2287853240966795,
|
|
"epoch": 4.957274202271498,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00032066074149815184,
|
|
"loss": 4.8774,
|
|
"mean_token_accuracy": 0.21249490827322007,
|
|
"num_tokens": 10296692.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 5.197828388214111,
|
|
"epoch": 4.962682531097891,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00032024066488582517,
|
|
"loss": 4.8908,
|
|
"mean_token_accuracy": 0.21573451459407805,
|
|
"num_tokens": 10307936.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 5.199468517303467,
|
|
"epoch": 4.968090859924283,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003198204238688178,
|
|
"loss": 4.7867,
|
|
"mean_token_accuracy": 0.22113231420516968,
|
|
"num_tokens": 10318637.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 5.195602035522461,
|
|
"epoch": 4.973499188750676,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00031940001997428645,
|
|
"loss": 4.7193,
|
|
"mean_token_accuracy": 0.2265659675002098,
|
|
"num_tokens": 10328432.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 5.088107538223267,
|
|
"epoch": 4.978907517577069,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00031897945472997996,
|
|
"loss": 4.7674,
|
|
"mean_token_accuracy": 0.22346219122409822,
|
|
"num_tokens": 10338971.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 5.148088216781616,
|
|
"epoch": 4.984315846403462,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003185587296642333,
|
|
"loss": 4.8432,
|
|
"mean_token_accuracy": 0.21601056009531022,
|
|
"num_tokens": 10350586.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 5.257126426696777,
|
|
"epoch": 4.9897241752298545,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003181378463059625,
|
|
"loss": 4.8157,
|
|
"mean_token_accuracy": 0.2199622243642807,
|
|
"num_tokens": 10363150.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 5.1718888759613035,
|
|
"epoch": 4.995132504056246,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003177168061846584,
|
|
"loss": 4.8113,
|
|
"mean_token_accuracy": 0.2159444734454155,
|
|
"num_tokens": 10374431.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 5.192457516988118,
|
|
"epoch": 5.0,
|
|
"grad_norm": 2.609375,
|
|
"learning_rate": 0.0003172956108303819,
|
|
"loss": 4.8567,
|
|
"mean_token_accuracy": 0.23210355142752329,
|
|
"num_tokens": 10384445.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 5.269854879379272,
|
|
"epoch": 5.005408328826393,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003168742617737579,
|
|
"loss": 4.6531,
|
|
"mean_token_accuracy": 0.2253870338201523,
|
|
"num_tokens": 10395995.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 5.200144290924072,
|
|
"epoch": 5.010816657652786,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00031645276054597006,
|
|
"loss": 4.5823,
|
|
"mean_token_accuracy": 0.22997160851955414,
|
|
"num_tokens": 10408385.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 5.125444126129151,
|
|
"epoch": 5.016224986479178,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00031603110867875435,
|
|
"loss": 4.5319,
|
|
"mean_token_accuracy": 0.24703483730554582,
|
|
"num_tokens": 10419629.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 5.262092447280883,
|
|
"epoch": 5.0216333153055706,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000315609307704395,
|
|
"loss": 4.6821,
|
|
"mean_token_accuracy": 0.22896433621644974,
|
|
"num_tokens": 10431040.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 5.14223952293396,
|
|
"epoch": 5.027041644131963,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003151873591557176,
|
|
"loss": 4.5728,
|
|
"mean_token_accuracy": 0.23871285915374757,
|
|
"num_tokens": 10442909.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 5.11332859992981,
|
|
"epoch": 5.032449972958356,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00031476526456608415,
|
|
"loss": 4.52,
|
|
"mean_token_accuracy": 0.24671415984630585,
|
|
"num_tokens": 10454008.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 5.2240842342376705,
|
|
"epoch": 5.037858301784748,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003143430254693873,
|
|
"loss": 4.587,
|
|
"mean_token_accuracy": 0.23459565192461013,
|
|
"num_tokens": 10465037.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 5.178358888626098,
|
|
"epoch": 5.043266630611141,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003139206434000451,
|
|
"loss": 4.5959,
|
|
"mean_token_accuracy": 0.23333267867565155,
|
|
"num_tokens": 10475654.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 5.173251914978027,
|
|
"epoch": 5.048674959437534,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003134981198929948,
|
|
"loss": 4.6381,
|
|
"mean_token_accuracy": 0.23443091958761214,
|
|
"num_tokens": 10486700.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 5.182063245773316,
|
|
"epoch": 5.054083288263927,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00031307545648368777,
|
|
"loss": 4.6372,
|
|
"mean_token_accuracy": 0.2260841593146324,
|
|
"num_tokens": 10498055.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 5.156552314758301,
|
|
"epoch": 5.059491617090319,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00031265265470808393,
|
|
"loss": 4.6728,
|
|
"mean_token_accuracy": 0.23410805612802504,
|
|
"num_tokens": 10511418.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 5.109159326553344,
|
|
"epoch": 5.064899945916712,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00031222971610264595,
|
|
"loss": 4.4365,
|
|
"mean_token_accuracy": 0.2557953104376793,
|
|
"num_tokens": 10522498.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 5.179452276229858,
|
|
"epoch": 5.0703082747431045,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00031180664220433347,
|
|
"loss": 4.6067,
|
|
"mean_token_accuracy": 0.23463710248470307,
|
|
"num_tokens": 10534366.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 5.319882249832153,
|
|
"epoch": 5.075716603569497,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00031138343455059823,
|
|
"loss": 4.693,
|
|
"mean_token_accuracy": 0.23739415109157563,
|
|
"num_tokens": 10545256.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 5.134870481491089,
|
|
"epoch": 5.081124932395889,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003109600946793778,
|
|
"loss": 4.5574,
|
|
"mean_token_accuracy": 0.24140521585941316,
|
|
"num_tokens": 10555927.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 5.108152246475219,
|
|
"epoch": 5.086533261222282,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00031053662412909024,
|
|
"loss": 4.5618,
|
|
"mean_token_accuracy": 0.2384909838438034,
|
|
"num_tokens": 10567383.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 5.260105276107788,
|
|
"epoch": 5.091941590048675,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003101130244386283,
|
|
"loss": 4.6452,
|
|
"mean_token_accuracy": 0.23423611521720886,
|
|
"num_tokens": 10580257.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 5.095328664779663,
|
|
"epoch": 5.097349918875068,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00030968929714735456,
|
|
"loss": 4.6221,
|
|
"mean_token_accuracy": 0.24258344769477844,
|
|
"num_tokens": 10590838.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 5.170976018905639,
|
|
"epoch": 5.10275824770146,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003092654437950947,
|
|
"loss": 4.5659,
|
|
"mean_token_accuracy": 0.2385046288371086,
|
|
"num_tokens": 10602254.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 5.252436637878418,
|
|
"epoch": 5.108166576527853,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003088414659221331,
|
|
"loss": 4.6864,
|
|
"mean_token_accuracy": 0.23316697180271148,
|
|
"num_tokens": 10613630.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 5.113118314743042,
|
|
"epoch": 5.113574905354246,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000308417365069206,
|
|
"loss": 4.4767,
|
|
"mean_token_accuracy": 0.24711329042911528,
|
|
"num_tokens": 10623754.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 5.1272046089172365,
|
|
"epoch": 5.1189832341806385,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00030799314277749697,
|
|
"loss": 4.6071,
|
|
"mean_token_accuracy": 0.2341666504740715,
|
|
"num_tokens": 10633071.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 5.185282373428345,
|
|
"epoch": 5.1243915630070305,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00030756880058863085,
|
|
"loss": 4.5555,
|
|
"mean_token_accuracy": 0.2438496619462967,
|
|
"num_tokens": 10643963.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 5.173864841461182,
|
|
"epoch": 5.129799891833423,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00030714434004466823,
|
|
"loss": 4.587,
|
|
"mean_token_accuracy": 0.23955532163381577,
|
|
"num_tokens": 10655716.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 5.186474752426148,
|
|
"epoch": 5.135208220659816,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00030671976268809947,
|
|
"loss": 4.6535,
|
|
"mean_token_accuracy": 0.23002523928880692,
|
|
"num_tokens": 10666231.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 5.15754451751709,
|
|
"epoch": 5.140616549486209,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00030629507006183995,
|
|
"loss": 4.5929,
|
|
"mean_token_accuracy": 0.2359611839056015,
|
|
"num_tokens": 10677672.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 5.15077314376831,
|
|
"epoch": 5.146024878312601,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003058702637092234,
|
|
"loss": 4.654,
|
|
"mean_token_accuracy": 0.22554715126752853,
|
|
"num_tokens": 10689968.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 5.192187833786011,
|
|
"epoch": 5.151433207138994,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00030544534517399724,
|
|
"loss": 4.639,
|
|
"mean_token_accuracy": 0.23661068379878997,
|
|
"num_tokens": 10699705.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 5.175465440750122,
|
|
"epoch": 5.156841535965387,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003050203160003163,
|
|
"loss": 4.5707,
|
|
"mean_token_accuracy": 0.23488177508115768,
|
|
"num_tokens": 10710464.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 5.158905458450318,
|
|
"epoch": 5.16224986479178,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003045951777327377,
|
|
"loss": 4.5543,
|
|
"mean_token_accuracy": 0.24192606508731843,
|
|
"num_tokens": 10720592.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 5.1494465351104735,
|
|
"epoch": 5.167658193618172,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003041699319162148,
|
|
"loss": 4.6413,
|
|
"mean_token_accuracy": 0.23382354527711868,
|
|
"num_tokens": 10731670.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 5.1562597274780275,
|
|
"epoch": 5.173066522444564,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00030374458009609193,
|
|
"loss": 4.5898,
|
|
"mean_token_accuracy": 0.24109318256378173,
|
|
"num_tokens": 10743123.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 5.090175914764404,
|
|
"epoch": 5.178474851270957,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003033191238180985,
|
|
"loss": 4.6157,
|
|
"mean_token_accuracy": 0.2355712577700615,
|
|
"num_tokens": 10753855.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 5.117549991607666,
|
|
"epoch": 5.18388318009735,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003028935646283438,
|
|
"loss": 4.602,
|
|
"mean_token_accuracy": 0.23887638747692108,
|
|
"num_tokens": 10765275.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 5.114529848098755,
|
|
"epoch": 5.189291508923742,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00030246790407331063,
|
|
"loss": 4.6153,
|
|
"mean_token_accuracy": 0.22960194945335388,
|
|
"num_tokens": 10776563.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 5.1570152759552,
|
|
"epoch": 5.194699837750135,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00030204214369985066,
|
|
"loss": 4.5679,
|
|
"mean_token_accuracy": 0.24049003571271896,
|
|
"num_tokens": 10786989.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 5.115324401855469,
|
|
"epoch": 5.200108166576528,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00030161628505517797,
|
|
"loss": 4.5509,
|
|
"mean_token_accuracy": 0.23526498228311538,
|
|
"num_tokens": 10798460.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 5.135961675643921,
|
|
"epoch": 5.205516495402921,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00030119032968686397,
|
|
"loss": 4.6427,
|
|
"mean_token_accuracy": 0.23722067773342131,
|
|
"num_tokens": 10810100.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 5.179385566711426,
|
|
"epoch": 5.210924824229313,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00030076427914283127,
|
|
"loss": 4.6194,
|
|
"mean_token_accuracy": 0.2371158555150032,
|
|
"num_tokens": 10821019.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 5.110940265655517,
|
|
"epoch": 5.2163331530557056,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00030033813497134864,
|
|
"loss": 4.5492,
|
|
"mean_token_accuracy": 0.23882660567760466,
|
|
"num_tokens": 10831583.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 5.045322895050049,
|
|
"epoch": 5.221741481882098,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000299911898721025,
|
|
"loss": 4.5893,
|
|
"mean_token_accuracy": 0.2328151360154152,
|
|
"num_tokens": 10844117.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 5.106189203262329,
|
|
"epoch": 5.227149810708491,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00029948557194080385,
|
|
"loss": 4.5823,
|
|
"mean_token_accuracy": 0.24504978060722352,
|
|
"num_tokens": 10855577.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 5.145179891586304,
|
|
"epoch": 5.232558139534884,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002990591561799576,
|
|
"loss": 4.612,
|
|
"mean_token_accuracy": 0.23598862141370774,
|
|
"num_tokens": 10867153.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 5.210226058959961,
|
|
"epoch": 5.237966468361276,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002986326529880823,
|
|
"loss": 4.714,
|
|
"mean_token_accuracy": 0.2309194639325142,
|
|
"num_tokens": 10878666.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 5.185020065307617,
|
|
"epoch": 5.243374797187669,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002982060639150913,
|
|
"loss": 4.6453,
|
|
"mean_token_accuracy": 0.23511116653680803,
|
|
"num_tokens": 10890197.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 5.152313232421875,
|
|
"epoch": 5.248783126014062,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002977793905112106,
|
|
"loss": 4.6177,
|
|
"mean_token_accuracy": 0.22962196469306945,
|
|
"num_tokens": 10901423.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 5.120763778686523,
|
|
"epoch": 5.254191454840455,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00029735263432697183,
|
|
"loss": 4.6226,
|
|
"mean_token_accuracy": 0.23838821649551392,
|
|
"num_tokens": 10913023.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 5.211461687088013,
|
|
"epoch": 5.259599783666847,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002969257969132085,
|
|
"loss": 4.6669,
|
|
"mean_token_accuracy": 0.22829086929559708,
|
|
"num_tokens": 10924404.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 5.203697919845581,
|
|
"epoch": 5.2650081124932395,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002964988798210485,
|
|
"loss": 4.6874,
|
|
"mean_token_accuracy": 0.23138597309589387,
|
|
"num_tokens": 10935001.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 5.146004152297974,
|
|
"epoch": 5.270416441319632,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002960718846019095,
|
|
"loss": 4.6974,
|
|
"mean_token_accuracy": 0.2272897720336914,
|
|
"num_tokens": 10947035.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 5.250476884841919,
|
|
"epoch": 5.275824770146025,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002956448128074931,
|
|
"loss": 4.652,
|
|
"mean_token_accuracy": 0.2330253779888153,
|
|
"num_tokens": 10960387.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 5.061264610290527,
|
|
"epoch": 5.281233098972417,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002952176659897794,
|
|
"loss": 4.4932,
|
|
"mean_token_accuracy": 0.25233867168426516,
|
|
"num_tokens": 10972949.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 5.042494058609009,
|
|
"epoch": 5.28664142779881,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002947904457010207,
|
|
"loss": 4.5795,
|
|
"mean_token_accuracy": 0.23271046131849288,
|
|
"num_tokens": 10983488.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 5.110189819335938,
|
|
"epoch": 5.292049756625203,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002943631534937366,
|
|
"loss": 4.6056,
|
|
"mean_token_accuracy": 0.22957539707422256,
|
|
"num_tokens": 10995453.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 5.237284469604492,
|
|
"epoch": 5.297458085451596,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002939357909207079,
|
|
"loss": 4.7618,
|
|
"mean_token_accuracy": 0.23337202966213227,
|
|
"num_tokens": 11007945.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 5.208136510848999,
|
|
"epoch": 5.302866414277988,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002935083595349713,
|
|
"loss": 4.6707,
|
|
"mean_token_accuracy": 0.23304233849048614,
|
|
"num_tokens": 11020574.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 5.137352466583252,
|
|
"epoch": 5.308274743104381,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00029308086088981335,
|
|
"loss": 4.5924,
|
|
"mean_token_accuracy": 0.2387629196047783,
|
|
"num_tokens": 11030893.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 5.09926118850708,
|
|
"epoch": 5.3136830719307735,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00029265329653876515,
|
|
"loss": 4.656,
|
|
"mean_token_accuracy": 0.22831930965185165,
|
|
"num_tokens": 11042104.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 5.237166786193848,
|
|
"epoch": 5.319091400757166,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002922256680355965,
|
|
"loss": 4.6776,
|
|
"mean_token_accuracy": 0.2212482437491417,
|
|
"num_tokens": 11051961.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 5.202642202377319,
|
|
"epoch": 5.324499729583558,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00029179797693431046,
|
|
"loss": 4.7458,
|
|
"mean_token_accuracy": 0.22569220066070556,
|
|
"num_tokens": 11063877.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 5.112744379043579,
|
|
"epoch": 5.329908058409951,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002913702247891374,
|
|
"loss": 4.5714,
|
|
"mean_token_accuracy": 0.247721266746521,
|
|
"num_tokens": 11074085.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 5.137529611587524,
|
|
"epoch": 5.335316387236344,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002909424131545296,
|
|
"loss": 4.6102,
|
|
"mean_token_accuracy": 0.23508343398571013,
|
|
"num_tokens": 11085689.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 5.202236032485962,
|
|
"epoch": 5.340724716062737,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00029051454358515553,
|
|
"loss": 4.6971,
|
|
"mean_token_accuracy": 0.2297946736216545,
|
|
"num_tokens": 11096635.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 5.09616231918335,
|
|
"epoch": 5.346133044889129,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00029008661763589436,
|
|
"loss": 4.6322,
|
|
"mean_token_accuracy": 0.2315840855240822,
|
|
"num_tokens": 11107646.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 5.167543983459472,
|
|
"epoch": 5.351541373715522,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002896586368618297,
|
|
"loss": 4.6218,
|
|
"mean_token_accuracy": 0.23649637550115585,
|
|
"num_tokens": 11117785.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 5.111290740966797,
|
|
"epoch": 5.356949702541915,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00028923060281824483,
|
|
"loss": 4.597,
|
|
"mean_token_accuracy": 0.23767180740833282,
|
|
"num_tokens": 11129668.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 5.155202198028564,
|
|
"epoch": 5.3623580313683075,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002888025170606165,
|
|
"loss": 4.6868,
|
|
"mean_token_accuracy": 0.23212609589099883,
|
|
"num_tokens": 11141549.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 5.1573858737945555,
|
|
"epoch": 5.367766360194699,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00028837438114460924,
|
|
"loss": 4.638,
|
|
"mean_token_accuracy": 0.2347456619143486,
|
|
"num_tokens": 11153599.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 5.175124597549439,
|
|
"epoch": 5.373174689021092,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002879461966260699,
|
|
"loss": 4.7109,
|
|
"mean_token_accuracy": 0.22406936585903167,
|
|
"num_tokens": 11164392.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 5.107207202911377,
|
|
"epoch": 5.378583017847485,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00028751796506102224,
|
|
"loss": 4.5833,
|
|
"mean_token_accuracy": 0.23927603513002396,
|
|
"num_tokens": 11175707.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 5.175127649307251,
|
|
"epoch": 5.383991346673878,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00028708968800566056,
|
|
"loss": 4.697,
|
|
"mean_token_accuracy": 0.2315484493970871,
|
|
"num_tokens": 11186921.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 5.19052267074585,
|
|
"epoch": 5.38939967550027,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00028666136701634474,
|
|
"loss": 4.6424,
|
|
"mean_token_accuracy": 0.22656507343053817,
|
|
"num_tokens": 11197917.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 5.18916506767273,
|
|
"epoch": 5.394808004326663,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002862330036495942,
|
|
"loss": 4.6548,
|
|
"mean_token_accuracy": 0.24126090109348297,
|
|
"num_tokens": 11209001.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 5.149725484848022,
|
|
"epoch": 5.400216333153056,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002858045994620825,
|
|
"loss": 4.6549,
|
|
"mean_token_accuracy": 0.2314258873462677,
|
|
"num_tokens": 11220661.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 5.158996963500977,
|
|
"epoch": 5.405624661979449,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00028537615601063133,
|
|
"loss": 4.6545,
|
|
"mean_token_accuracy": 0.2340800017118454,
|
|
"num_tokens": 11232418.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 5.405624661979449,
|
|
"eval_entropy": 5.124407329342582,
|
|
"eval_loss": 5.898890018463135,
|
|
"eval_mean_token_accuracy": 0.1787239752879197,
|
|
"eval_num_tokens": 11232418.0,
|
|
"eval_runtime": 2.5259,
|
|
"eval_samples_per_second": 1391.573,
|
|
"eval_steps_per_second": 174.194,
|
|
"step": 5000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 9240,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 1000,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 1.7549115918336e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|