7a79b20bcf
Model: fpadovani/eng-latn-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed10 Source: Original Platform
4057 lines
109 KiB
JSON
4057 lines
109 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 2.162249864791779,
|
|
"eval_steps": 1000,
|
|
"global_step": 2000,
|
|
"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
|
|
}
|
|
],
|
|
"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": 7052141297664000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|