99770894f8
Model: fpadovani/eng-latn-10mb-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": 4.91689715385437,
|
|
"epoch": 0.005408328826392645,
|
|
"grad_norm": 19.875,
|
|
"learning_rate": 2e-06,
|
|
"loss": 14.1759,
|
|
"mean_token_accuracy": 9.372071363031865e-05,
|
|
"num_tokens": 10948.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 4.94269962310791,
|
|
"epoch": 0.01081665765278529,
|
|
"grad_norm": 23.75,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 14.1132,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 21455.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 4.975670337677002,
|
|
"epoch": 0.016224986479177934,
|
|
"grad_norm": 27.625,
|
|
"learning_rate": 7e-06,
|
|
"loss": 14.0429,
|
|
"mean_token_accuracy": 8.873114711605012e-05,
|
|
"num_tokens": 32377.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 5.118604516983032,
|
|
"epoch": 0.02163331530557058,
|
|
"grad_norm": 32.75,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 13.6272,
|
|
"mean_token_accuracy": 6.706907879561186e-05,
|
|
"num_tokens": 42276.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 5.543013238906861,
|
|
"epoch": 0.02704164413196322,
|
|
"grad_norm": 40.75,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 13.0208,
|
|
"mean_token_accuracy": 0.00013905229279771448,
|
|
"num_tokens": 55049.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 7.297787666320801,
|
|
"epoch": 0.03244997295835587,
|
|
"grad_norm": 26.375,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 11.8384,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 66792.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.221694660186767,
|
|
"epoch": 0.03785830178474851,
|
|
"grad_norm": 3.640625,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.8659,
|
|
"mean_token_accuracy": 0.004666937288129702,
|
|
"num_tokens": 78208.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.73062868118286,
|
|
"epoch": 0.04326663061114116,
|
|
"grad_norm": 3.359375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.6432,
|
|
"mean_token_accuracy": 0.013734118174761533,
|
|
"num_tokens": 88871.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.73009204864502,
|
|
"epoch": 0.048674959437533805,
|
|
"grad_norm": 3.140625,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.5305,
|
|
"mean_token_accuracy": 0.0145380774512887,
|
|
"num_tokens": 100118.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.722947692871093,
|
|
"epoch": 0.05408328826392644,
|
|
"grad_norm": 2.4375,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 10.362,
|
|
"mean_token_accuracy": 0.012977494951337576,
|
|
"num_tokens": 113120.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.69485158920288,
|
|
"epoch": 0.05949161709031909,
|
|
"grad_norm": 2.890625,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 10.146,
|
|
"mean_token_accuracy": 0.016121995542198418,
|
|
"num_tokens": 123220.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.705347919464112,
|
|
"epoch": 0.06489994591671173,
|
|
"grad_norm": 1.984375,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 10.053,
|
|
"mean_token_accuracy": 0.01566679133102298,
|
|
"num_tokens": 134479.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.732111358642578,
|
|
"epoch": 0.07030827474310439,
|
|
"grad_norm": 1.953125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.9286,
|
|
"mean_token_accuracy": 0.030279947444796562,
|
|
"num_tokens": 145189.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.72783441543579,
|
|
"epoch": 0.07571660356949703,
|
|
"grad_norm": 1.96875,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.8452,
|
|
"mean_token_accuracy": 0.03374573942273855,
|
|
"num_tokens": 157354.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.714603519439697,
|
|
"epoch": 0.08112493239588967,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.7908,
|
|
"mean_token_accuracy": 0.03534814547747374,
|
|
"num_tokens": 168821.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.72258472442627,
|
|
"epoch": 0.08653326122228232,
|
|
"grad_norm": 1.8984375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.7017,
|
|
"mean_token_accuracy": 0.04154699631035328,
|
|
"num_tokens": 180554.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.723107147216798,
|
|
"epoch": 0.09194159004867496,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.6177,
|
|
"mean_token_accuracy": 0.04423647858202458,
|
|
"num_tokens": 191229.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.715191745758057,
|
|
"epoch": 0.09734991887506761,
|
|
"grad_norm": 1.9296875,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.5371,
|
|
"mean_token_accuracy": 0.03949474133551121,
|
|
"num_tokens": 201903.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.705009365081787,
|
|
"epoch": 0.10275824770146025,
|
|
"grad_norm": 1.9140625,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.4731,
|
|
"mean_token_accuracy": 0.03862293735146523,
|
|
"num_tokens": 214255.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.693695735931396,
|
|
"epoch": 0.10816657652785289,
|
|
"grad_norm": 1.859375,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.3346,
|
|
"mean_token_accuracy": 0.04093401432037354,
|
|
"num_tokens": 225423.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.675817108154297,
|
|
"epoch": 0.11357490535424554,
|
|
"grad_norm": 1.859375,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.2358,
|
|
"mean_token_accuracy": 0.046506138145923616,
|
|
"num_tokens": 237801.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.651765060424804,
|
|
"epoch": 0.11898323418063818,
|
|
"grad_norm": 1.9765625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.0789,
|
|
"mean_token_accuracy": 0.05033866539597511,
|
|
"num_tokens": 249195.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.622098636627197,
|
|
"epoch": 0.12439156300703083,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.9519,
|
|
"mean_token_accuracy": 0.046295545622706415,
|
|
"num_tokens": 260129.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.583237266540527,
|
|
"epoch": 0.12979989183342347,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.9331,
|
|
"mean_token_accuracy": 0.044216564670205113,
|
|
"num_tokens": 271220.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.535919380187988,
|
|
"epoch": 0.1352082206598161,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.7308,
|
|
"mean_token_accuracy": 0.05066610909998417,
|
|
"num_tokens": 281560.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.480502033233643,
|
|
"epoch": 0.14061654948620878,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.6033,
|
|
"mean_token_accuracy": 0.04720585718750954,
|
|
"num_tokens": 292518.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.400840473175048,
|
|
"epoch": 0.1460248783126014,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.537,
|
|
"mean_token_accuracy": 0.04684793278574943,
|
|
"num_tokens": 304865.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.312588691711426,
|
|
"epoch": 0.15143320713899405,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.3772,
|
|
"mean_token_accuracy": 0.04892168343067169,
|
|
"num_tokens": 315951.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.193289756774902,
|
|
"epoch": 0.1568415359653867,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.2559,
|
|
"mean_token_accuracy": 0.047295660525560376,
|
|
"num_tokens": 327094.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 10.057239818572999,
|
|
"epoch": 0.16224986479177933,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.1645,
|
|
"mean_token_accuracy": 0.04913058504462242,
|
|
"num_tokens": 337801.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.920266246795654,
|
|
"epoch": 0.167658193618172,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.0478,
|
|
"mean_token_accuracy": 0.04764632135629654,
|
|
"num_tokens": 349039.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.760813426971435,
|
|
"epoch": 0.17306652244456464,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.9465,
|
|
"mean_token_accuracy": 0.049764279276132584,
|
|
"num_tokens": 360639.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.56991138458252,
|
|
"epoch": 0.17847485127095727,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.8804,
|
|
"mean_token_accuracy": 0.05512551851570606,
|
|
"num_tokens": 371005.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.318969631195069,
|
|
"epoch": 0.1838831800973499,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.7379,
|
|
"mean_token_accuracy": 0.05374485366046429,
|
|
"num_tokens": 383340.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 9.042248249053955,
|
|
"epoch": 0.18929150892374255,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.6303,
|
|
"mean_token_accuracy": 0.04950361587107181,
|
|
"num_tokens": 394682.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.764017868041993,
|
|
"epoch": 0.19469983775013522,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.5815,
|
|
"mean_token_accuracy": 0.059978223592042926,
|
|
"num_tokens": 405713.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.552699661254882,
|
|
"epoch": 0.20010816657652786,
|
|
"grad_norm": 0.703125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.5243,
|
|
"mean_token_accuracy": 0.06490898504853249,
|
|
"num_tokens": 416208.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.323922729492187,
|
|
"epoch": 0.2055164954029205,
|
|
"grad_norm": 0.73828125,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.4179,
|
|
"mean_token_accuracy": 0.05761545822024346,
|
|
"num_tokens": 427595.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.167845249176025,
|
|
"epoch": 0.21092482422931313,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.506,
|
|
"mean_token_accuracy": 0.06468431949615479,
|
|
"num_tokens": 438273.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.073835182189942,
|
|
"epoch": 0.21633315305570577,
|
|
"grad_norm": 0.6875,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.4195,
|
|
"mean_token_accuracy": 0.06935816742479801,
|
|
"num_tokens": 448948.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.998777389526367,
|
|
"epoch": 0.22174148188209844,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.456,
|
|
"mean_token_accuracy": 0.06588310226798058,
|
|
"num_tokens": 460838.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.015505456924439,
|
|
"epoch": 0.22714981070849108,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.4476,
|
|
"mean_token_accuracy": 0.06993156149983407,
|
|
"num_tokens": 473555.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.928825998306275,
|
|
"epoch": 0.23255813953488372,
|
|
"grad_norm": 0.6875,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.3803,
|
|
"mean_token_accuracy": 0.07670803219079972,
|
|
"num_tokens": 485221.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.916201114654541,
|
|
"epoch": 0.23796646836127636,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.2914,
|
|
"mean_token_accuracy": 0.07322146371006966,
|
|
"num_tokens": 496552.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.902019739151001,
|
|
"epoch": 0.24337479718766902,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.3551,
|
|
"mean_token_accuracy": 0.07495715916156769,
|
|
"num_tokens": 509760.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.87739930152893,
|
|
"epoch": 0.24878312601406166,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.2003,
|
|
"mean_token_accuracy": 0.07767140977084637,
|
|
"num_tokens": 521175.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.810295057296753,
|
|
"epoch": 0.2541914548404543,
|
|
"grad_norm": 0.66796875,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.2954,
|
|
"mean_token_accuracy": 0.07847636789083481,
|
|
"num_tokens": 532597.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.815588092803955,
|
|
"epoch": 0.25959978366684694,
|
|
"grad_norm": 0.69140625,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.3371,
|
|
"mean_token_accuracy": 0.07920736819505692,
|
|
"num_tokens": 545661.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.830226469039917,
|
|
"epoch": 0.2650081124932396,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.3186,
|
|
"mean_token_accuracy": 0.07971370816230774,
|
|
"num_tokens": 557336.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.839285278320313,
|
|
"epoch": 0.2704164413196322,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.1789,
|
|
"mean_token_accuracy": 0.08502659052610398,
|
|
"num_tokens": 567398.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.752034616470337,
|
|
"epoch": 0.27582477014602486,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.254,
|
|
"mean_token_accuracy": 0.07750963866710663,
|
|
"num_tokens": 579543.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.797254037857056,
|
|
"epoch": 0.28123309897241755,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.0797,
|
|
"mean_token_accuracy": 0.0938566841185093,
|
|
"num_tokens": 589218.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.631022548675537,
|
|
"epoch": 0.2866414277988102,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.1516,
|
|
"mean_token_accuracy": 0.0841096393764019,
|
|
"num_tokens": 600327.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.694751644134522,
|
|
"epoch": 0.2920497566252028,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.0968,
|
|
"mean_token_accuracy": 0.08569389209151268,
|
|
"num_tokens": 610923.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.742217016220093,
|
|
"epoch": 0.29745808545159547,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.1711,
|
|
"mean_token_accuracy": 0.08022493869066238,
|
|
"num_tokens": 621138.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.701586723327637,
|
|
"epoch": 0.3028664142779881,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.0946,
|
|
"mean_token_accuracy": 0.0921371579170227,
|
|
"num_tokens": 631440.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.630160140991211,
|
|
"epoch": 0.30827474310438074,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.1312,
|
|
"mean_token_accuracy": 0.08813719153404236,
|
|
"num_tokens": 642328.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.665129518508911,
|
|
"epoch": 0.3136830719307734,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.0909,
|
|
"mean_token_accuracy": 0.0894947700202465,
|
|
"num_tokens": 652875.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.610320234298706,
|
|
"epoch": 0.319091400757166,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.031,
|
|
"mean_token_accuracy": 0.08914472088217736,
|
|
"num_tokens": 663726.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.622774505615235,
|
|
"epoch": 0.32449972958355866,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.2158,
|
|
"mean_token_accuracy": 0.08153997585177422,
|
|
"num_tokens": 675638.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.5613326072692875,
|
|
"epoch": 0.3299080584099513,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.0719,
|
|
"mean_token_accuracy": 0.09202613458037376,
|
|
"num_tokens": 687239.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.59527850151062,
|
|
"epoch": 0.335316387236344,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.0996,
|
|
"mean_token_accuracy": 0.08664119839668274,
|
|
"num_tokens": 698905.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.607007694244385,
|
|
"epoch": 0.34072471606273663,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.09,
|
|
"mean_token_accuracy": 0.0928929626941681,
|
|
"num_tokens": 710602.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.484040689468384,
|
|
"epoch": 0.34613304488912927,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.0519,
|
|
"mean_token_accuracy": 0.08713712617754936,
|
|
"num_tokens": 721326.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.593839359283447,
|
|
"epoch": 0.3515413737155219,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.0427,
|
|
"mean_token_accuracy": 0.0913507841527462,
|
|
"num_tokens": 732972.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.492053413391114,
|
|
"epoch": 0.35694970254191455,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.0028,
|
|
"mean_token_accuracy": 0.09586582258343697,
|
|
"num_tokens": 743624.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.524333477020264,
|
|
"epoch": 0.3623580313683072,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.076,
|
|
"mean_token_accuracy": 0.09233946427702903,
|
|
"num_tokens": 756384.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.58307991027832,
|
|
"epoch": 0.3677663601946998,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.0545,
|
|
"mean_token_accuracy": 0.09635318964719772,
|
|
"num_tokens": 767256.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.474336338043213,
|
|
"epoch": 0.37317468902109246,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.9979,
|
|
"mean_token_accuracy": 0.09708881303668022,
|
|
"num_tokens": 779059.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.505778360366821,
|
|
"epoch": 0.3785830178474851,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.0686,
|
|
"mean_token_accuracy": 0.09691090956330299,
|
|
"num_tokens": 790291.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.510104370117188,
|
|
"epoch": 0.3839913466738778,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.9605,
|
|
"mean_token_accuracy": 0.09804000407457351,
|
|
"num_tokens": 801450.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.427967882156372,
|
|
"epoch": 0.38939967550027044,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.8869,
|
|
"mean_token_accuracy": 0.10146211236715316,
|
|
"num_tokens": 812428.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.357033348083496,
|
|
"epoch": 0.3948080043266631,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.9743,
|
|
"mean_token_accuracy": 0.09577520415186883,
|
|
"num_tokens": 823644.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.539126396179199,
|
|
"epoch": 0.4002163331530557,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.967,
|
|
"mean_token_accuracy": 0.10204331800341607,
|
|
"num_tokens": 834064.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.464277219772339,
|
|
"epoch": 0.40562466197944835,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.9062,
|
|
"mean_token_accuracy": 0.1017673671245575,
|
|
"num_tokens": 845207.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.3919721126556395,
|
|
"epoch": 0.411032990805841,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.9524,
|
|
"mean_token_accuracy": 0.09975179508328438,
|
|
"num_tokens": 855564.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.475247764587403,
|
|
"epoch": 0.41644131963223363,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.9807,
|
|
"mean_token_accuracy": 0.10713295936584473,
|
|
"num_tokens": 866959.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.331955623626709,
|
|
"epoch": 0.42184964845862627,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.9058,
|
|
"mean_token_accuracy": 0.10511352196335792,
|
|
"num_tokens": 877319.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.372516441345215,
|
|
"epoch": 0.4272579772850189,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.9939,
|
|
"mean_token_accuracy": 0.10005753636360168,
|
|
"num_tokens": 888037.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.404991912841797,
|
|
"epoch": 0.43266630611141155,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.7495,
|
|
"mean_token_accuracy": 0.10937805995345115,
|
|
"num_tokens": 898775.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.364098739624024,
|
|
"epoch": 0.43807463493780424,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.8821,
|
|
"mean_token_accuracy": 0.10221289917826652,
|
|
"num_tokens": 909513.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.383790636062622,
|
|
"epoch": 0.4434829637641969,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.923,
|
|
"mean_token_accuracy": 0.10452766120433807,
|
|
"num_tokens": 920730.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.36325511932373,
|
|
"epoch": 0.4488912925905895,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.8456,
|
|
"mean_token_accuracy": 0.11156300008296967,
|
|
"num_tokens": 931457.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.295261812210083,
|
|
"epoch": 0.45429962141698216,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.8372,
|
|
"mean_token_accuracy": 0.10695499628782272,
|
|
"num_tokens": 942572.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.35825834274292,
|
|
"epoch": 0.4597079502433748,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.8416,
|
|
"mean_token_accuracy": 0.09994131103157997,
|
|
"num_tokens": 953132.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.296279668807983,
|
|
"epoch": 0.46511627906976744,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.8596,
|
|
"mean_token_accuracy": 0.10478808656334877,
|
|
"num_tokens": 963551.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.320891809463501,
|
|
"epoch": 0.4705246078961601,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.8273,
|
|
"mean_token_accuracy": 0.10656212195754051,
|
|
"num_tokens": 975297.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.322855615615845,
|
|
"epoch": 0.4759329367225527,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.8267,
|
|
"mean_token_accuracy": 0.10611714869737625,
|
|
"num_tokens": 986910.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.355953693389893,
|
|
"epoch": 0.48134126554894535,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.94,
|
|
"mean_token_accuracy": 0.11208969727158546,
|
|
"num_tokens": 997058.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.25784215927124,
|
|
"epoch": 0.48674959437533805,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.8025,
|
|
"mean_token_accuracy": 0.10399467796087265,
|
|
"num_tokens": 1008261.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.249159240722657,
|
|
"epoch": 0.4921579232017307,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.7439,
|
|
"mean_token_accuracy": 0.11101565584540367,
|
|
"num_tokens": 1019193.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.18016037940979,
|
|
"epoch": 0.4975662520281233,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.853,
|
|
"mean_token_accuracy": 0.10952779948711396,
|
|
"num_tokens": 1031227.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.201208257675171,
|
|
"epoch": 0.502974580854516,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.6978,
|
|
"mean_token_accuracy": 0.10652377605438232,
|
|
"num_tokens": 1042963.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.195361566543579,
|
|
"epoch": 0.5083829096809086,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.7466,
|
|
"mean_token_accuracy": 0.1075796753168106,
|
|
"num_tokens": 1053709.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.22407546043396,
|
|
"epoch": 0.5137912385073012,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.7614,
|
|
"mean_token_accuracy": 0.11968052312731743,
|
|
"num_tokens": 1064527.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.201431703567505,
|
|
"epoch": 0.5191995673336939,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.8083,
|
|
"mean_token_accuracy": 0.11001280769705772,
|
|
"num_tokens": 1075127.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.159610843658447,
|
|
"epoch": 0.5246078961600865,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.8036,
|
|
"mean_token_accuracy": 0.11006302908062934,
|
|
"num_tokens": 1086581.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.195751047134399,
|
|
"epoch": 0.5300162249864792,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.7321,
|
|
"mean_token_accuracy": 0.1197355069220066,
|
|
"num_tokens": 1097818.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.115999031066894,
|
|
"epoch": 0.5354245538128718,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.7213,
|
|
"mean_token_accuracy": 0.11185307502746582,
|
|
"num_tokens": 1108043.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.117218828201294,
|
|
"epoch": 0.5408328826392644,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.7845,
|
|
"mean_token_accuracy": 0.11339325457811356,
|
|
"num_tokens": 1119875.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.161368274688721,
|
|
"epoch": 0.5462412114656571,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.6836,
|
|
"mean_token_accuracy": 0.11546274945139885,
|
|
"num_tokens": 1132406.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.2924041748046875,
|
|
"epoch": 0.5516495402920497,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.8578,
|
|
"mean_token_accuracy": 0.10715171545743943,
|
|
"num_tokens": 1143828.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.111097812652588,
|
|
"epoch": 0.5570578691184424,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.6629,
|
|
"mean_token_accuracy": 0.1149873249232769,
|
|
"num_tokens": 1155482.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.082928657531738,
|
|
"epoch": 0.5624661979448351,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.6192,
|
|
"mean_token_accuracy": 0.12244254648685456,
|
|
"num_tokens": 1166847.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.048942422866821,
|
|
"epoch": 0.5678745267712277,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.6899,
|
|
"mean_token_accuracy": 0.11682257875800132,
|
|
"num_tokens": 1177419.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.181053352355957,
|
|
"epoch": 0.5732828555976204,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.8558,
|
|
"mean_token_accuracy": 0.10907636061310769,
|
|
"num_tokens": 1188975.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.074787044525147,
|
|
"epoch": 0.578691184424013,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.6899,
|
|
"mean_token_accuracy": 0.11405783966183662,
|
|
"num_tokens": 1200252.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.110227489471436,
|
|
"epoch": 0.5840995132504057,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.7279,
|
|
"mean_token_accuracy": 0.12044090032577515,
|
|
"num_tokens": 1211864.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.007605695724488,
|
|
"epoch": 0.5895078420767983,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.6089,
|
|
"mean_token_accuracy": 0.11882655844092369,
|
|
"num_tokens": 1222429.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.159776020050049,
|
|
"epoch": 0.5949161709031909,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.7062,
|
|
"mean_token_accuracy": 0.11563858911395072,
|
|
"num_tokens": 1233782.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.999762058258057,
|
|
"epoch": 0.6003244997295836,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.6624,
|
|
"mean_token_accuracy": 0.11662511229515075,
|
|
"num_tokens": 1244309.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.080610895156861,
|
|
"epoch": 0.6057328285559762,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.6242,
|
|
"mean_token_accuracy": 0.1214139811694622,
|
|
"num_tokens": 1256403.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.026603126525879,
|
|
"epoch": 0.6111411573823688,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.7025,
|
|
"mean_token_accuracy": 0.11629552468657493,
|
|
"num_tokens": 1268758.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.049094152450562,
|
|
"epoch": 0.6165494862087615,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.5898,
|
|
"mean_token_accuracy": 0.12565962076187134,
|
|
"num_tokens": 1279355.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.986866807937622,
|
|
"epoch": 0.6219578150351541,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.5681,
|
|
"mean_token_accuracy": 0.118294008821249,
|
|
"num_tokens": 1290619.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.035441446304321,
|
|
"epoch": 0.6273661438615468,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.5748,
|
|
"mean_token_accuracy": 0.12222455516457557,
|
|
"num_tokens": 1301851.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.028501653671265,
|
|
"epoch": 0.6327744726879394,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.6876,
|
|
"mean_token_accuracy": 0.12039072066545486,
|
|
"num_tokens": 1313260.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.0265075206756595,
|
|
"epoch": 0.638182801514332,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.6719,
|
|
"mean_token_accuracy": 0.11919900625944138,
|
|
"num_tokens": 1324189.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.09400782585144,
|
|
"epoch": 0.6435911303407247,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.7163,
|
|
"mean_token_accuracy": 0.11836740374565125,
|
|
"num_tokens": 1335765.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.97562804222107,
|
|
"epoch": 0.6489994591671173,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.6305,
|
|
"mean_token_accuracy": 0.12355265021324158,
|
|
"num_tokens": 1347653.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.013906860351563,
|
|
"epoch": 0.65440778799351,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.627,
|
|
"mean_token_accuracy": 0.12648534998297692,
|
|
"num_tokens": 1360201.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.952250146865845,
|
|
"epoch": 0.6598161168199026,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.5073,
|
|
"mean_token_accuracy": 0.12420706525444984,
|
|
"num_tokens": 1370974.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.93790454864502,
|
|
"epoch": 0.6652244456462953,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.5714,
|
|
"mean_token_accuracy": 0.12407453581690789,
|
|
"num_tokens": 1383855.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.938852977752686,
|
|
"epoch": 0.670632774472688,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.536,
|
|
"mean_token_accuracy": 0.12525033503770827,
|
|
"num_tokens": 1395041.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.976763200759888,
|
|
"epoch": 0.6760411032990806,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.6361,
|
|
"mean_token_accuracy": 0.12184548899531364,
|
|
"num_tokens": 1406102.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.916235494613647,
|
|
"epoch": 0.6814494321254733,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.4695,
|
|
"mean_token_accuracy": 0.1252748616039753,
|
|
"num_tokens": 1417713.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.897937250137329,
|
|
"epoch": 0.6868577609518659,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.6022,
|
|
"mean_token_accuracy": 0.1269610993564129,
|
|
"num_tokens": 1428526.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.923327493667602,
|
|
"epoch": 0.6922660897782585,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.5567,
|
|
"mean_token_accuracy": 0.12470588013529778,
|
|
"num_tokens": 1438849.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.979862880706787,
|
|
"epoch": 0.6976744186046512,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.6268,
|
|
"mean_token_accuracy": 0.12327171638607978,
|
|
"num_tokens": 1450776.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.835441255569458,
|
|
"epoch": 0.7030827474310438,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.4933,
|
|
"mean_token_accuracy": 0.12600184082984925,
|
|
"num_tokens": 1462552.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.796398162841797,
|
|
"epoch": 0.7084910762574365,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.4924,
|
|
"mean_token_accuracy": 0.12828927859663963,
|
|
"num_tokens": 1473867.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.799835538864135,
|
|
"epoch": 0.7138994050838291,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.497,
|
|
"mean_token_accuracy": 0.12288376614451409,
|
|
"num_tokens": 1485233.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.901881504058838,
|
|
"epoch": 0.7193077339102217,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.5643,
|
|
"mean_token_accuracy": 0.12888247296214103,
|
|
"num_tokens": 1496919.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.90382194519043,
|
|
"epoch": 0.7247160627366144,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.5039,
|
|
"mean_token_accuracy": 0.12288212552666664,
|
|
"num_tokens": 1508695.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.869405269622803,
|
|
"epoch": 0.730124391563007,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.5593,
|
|
"mean_token_accuracy": 0.1286173418164253,
|
|
"num_tokens": 1519629.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.8630880355834964,
|
|
"epoch": 0.7355327203893997,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.4569,
|
|
"mean_token_accuracy": 0.12953948080539704,
|
|
"num_tokens": 1530144.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.725225353240967,
|
|
"epoch": 0.7409410492157923,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.4683,
|
|
"mean_token_accuracy": 0.12884586453437805,
|
|
"num_tokens": 1541013.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.818743658065796,
|
|
"epoch": 0.7463493780421849,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.54,
|
|
"mean_token_accuracy": 0.1279229611158371,
|
|
"num_tokens": 1553344.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.90196681022644,
|
|
"epoch": 0.7517577068685776,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.5904,
|
|
"mean_token_accuracy": 0.123768300563097,
|
|
"num_tokens": 1565298.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.776127195358276,
|
|
"epoch": 0.7571660356949702,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.5125,
|
|
"mean_token_accuracy": 0.1363504111766815,
|
|
"num_tokens": 1577645.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.815530920028687,
|
|
"epoch": 0.7625743645213628,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.4674,
|
|
"mean_token_accuracy": 0.13412229791283609,
|
|
"num_tokens": 1588387.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.863557958602906,
|
|
"epoch": 0.7679826933477556,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.5756,
|
|
"mean_token_accuracy": 0.13169802352786064,
|
|
"num_tokens": 1599837.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.80879316329956,
|
|
"epoch": 0.7733910221741482,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.5179,
|
|
"mean_token_accuracy": 0.12488113269209862,
|
|
"num_tokens": 1610401.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.792469310760498,
|
|
"epoch": 0.7787993510005409,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.4423,
|
|
"mean_token_accuracy": 0.12702938616275788,
|
|
"num_tokens": 1622688.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.785536813735962,
|
|
"epoch": 0.7842076798269335,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.5322,
|
|
"mean_token_accuracy": 0.11961603239178657,
|
|
"num_tokens": 1633594.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.796718645095825,
|
|
"epoch": 0.7896160086533262,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.431,
|
|
"mean_token_accuracy": 0.1275559477508068,
|
|
"num_tokens": 1645861.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.67658257484436,
|
|
"epoch": 0.7950243374797188,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.3343,
|
|
"mean_token_accuracy": 0.13718428909778596,
|
|
"num_tokens": 1656903.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.801560592651367,
|
|
"epoch": 0.8004326663061114,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.441,
|
|
"mean_token_accuracy": 0.13460942655801772,
|
|
"num_tokens": 1668130.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.625500011444092,
|
|
"epoch": 0.8058409951325041,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.3347,
|
|
"mean_token_accuracy": 0.14152220115065575,
|
|
"num_tokens": 1678228.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.857194662094116,
|
|
"epoch": 0.8112493239588967,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.487,
|
|
"mean_token_accuracy": 0.1327496737241745,
|
|
"num_tokens": 1690339.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.740129566192627,
|
|
"epoch": 0.8166576527852893,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.4986,
|
|
"mean_token_accuracy": 0.12592008486390113,
|
|
"num_tokens": 1701067.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.824430894851685,
|
|
"epoch": 0.822065981611682,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.4343,
|
|
"mean_token_accuracy": 0.13577242717146873,
|
|
"num_tokens": 1712835.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.693302774429322,
|
|
"epoch": 0.8274743104380746,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.3978,
|
|
"mean_token_accuracy": 0.13282729983329772,
|
|
"num_tokens": 1723432.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.772776031494141,
|
|
"epoch": 0.8328826392644673,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.4324,
|
|
"mean_token_accuracy": 0.13451724424958228,
|
|
"num_tokens": 1734134.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.688350820541382,
|
|
"epoch": 0.8382909680908599,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.3783,
|
|
"mean_token_accuracy": 0.13390962406992912,
|
|
"num_tokens": 1744723.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.730376195907593,
|
|
"epoch": 0.8436992969172525,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.511,
|
|
"mean_token_accuracy": 0.131271693110466,
|
|
"num_tokens": 1755269.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.77961688041687,
|
|
"epoch": 0.8491076257436452,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.5187,
|
|
"mean_token_accuracy": 0.13170363679528235,
|
|
"num_tokens": 1766344.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.774808263778686,
|
|
"epoch": 0.8545159545700378,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.4812,
|
|
"mean_token_accuracy": 0.12957761734724044,
|
|
"num_tokens": 1779052.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.616818809509278,
|
|
"epoch": 0.8599242833964305,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.32,
|
|
"mean_token_accuracy": 0.13625366762280464,
|
|
"num_tokens": 1789754.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.559850549697876,
|
|
"epoch": 0.8653326122228231,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.2474,
|
|
"mean_token_accuracy": 0.1390355110168457,
|
|
"num_tokens": 1801161.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.705098056793213,
|
|
"epoch": 0.8707409410492158,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.3624,
|
|
"mean_token_accuracy": 0.13253424242138861,
|
|
"num_tokens": 1810892.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.548418378829956,
|
|
"epoch": 0.8761492698756085,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.403,
|
|
"mean_token_accuracy": 0.13478510975837707,
|
|
"num_tokens": 1821574.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.720035552978516,
|
|
"epoch": 0.8815575987020011,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.4552,
|
|
"mean_token_accuracy": 0.13616764321923255,
|
|
"num_tokens": 1833342.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.621445178985596,
|
|
"epoch": 0.8869659275283938,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.3237,
|
|
"mean_token_accuracy": 0.1389230526983738,
|
|
"num_tokens": 1843768.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.765406370162964,
|
|
"epoch": 0.8923742563547864,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.4073,
|
|
"mean_token_accuracy": 0.1308816060423851,
|
|
"num_tokens": 1855569.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.6211669921875,
|
|
"epoch": 0.897782585181179,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.3909,
|
|
"mean_token_accuracy": 0.1304692007601261,
|
|
"num_tokens": 1866684.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.672160291671753,
|
|
"epoch": 0.9031909140075717,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.3193,
|
|
"mean_token_accuracy": 0.1392674833536148,
|
|
"num_tokens": 1878700.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.6165376663208,
|
|
"epoch": 0.9085992428339643,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.3703,
|
|
"mean_token_accuracy": 0.13358763232827187,
|
|
"num_tokens": 1889970.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.5742271900177,
|
|
"epoch": 0.914007571660357,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.2577,
|
|
"mean_token_accuracy": 0.13828584030270577,
|
|
"num_tokens": 1902008.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.615272140502929,
|
|
"epoch": 0.9194159004867496,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.4757,
|
|
"mean_token_accuracy": 0.13929339945316316,
|
|
"num_tokens": 1913253.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.614001512527466,
|
|
"epoch": 0.9248242293131422,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.3162,
|
|
"mean_token_accuracy": 0.143621264398098,
|
|
"num_tokens": 1923408.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.615392589569092,
|
|
"epoch": 0.9302325581395349,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.3774,
|
|
"mean_token_accuracy": 0.13835460394620896,
|
|
"num_tokens": 1935315.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.567425060272217,
|
|
"epoch": 0.9356408869659275,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.2219,
|
|
"mean_token_accuracy": 0.14048045724630356,
|
|
"num_tokens": 1946524.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.63021936416626,
|
|
"epoch": 0.9410492157923201,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.3635,
|
|
"mean_token_accuracy": 0.13752378821372985,
|
|
"num_tokens": 1957561.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.573499870300293,
|
|
"epoch": 0.9464575446187128,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.2889,
|
|
"mean_token_accuracy": 0.14111032485961914,
|
|
"num_tokens": 1969089.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.578218793869018,
|
|
"epoch": 0.9518658734451054,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.3304,
|
|
"mean_token_accuracy": 0.13999469056725503,
|
|
"num_tokens": 1980341.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.540131092071533,
|
|
"epoch": 0.9572742022714981,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.2821,
|
|
"mean_token_accuracy": 0.14466761127114297,
|
|
"num_tokens": 1991220.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.561265563964843,
|
|
"epoch": 0.9626825310978907,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.3599,
|
|
"mean_token_accuracy": 0.1384817786514759,
|
|
"num_tokens": 2002650.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.561820602416992,
|
|
"epoch": 0.9680908599242833,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.274,
|
|
"mean_token_accuracy": 0.14329540133476257,
|
|
"num_tokens": 2012884.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.60991587638855,
|
|
"epoch": 0.9734991887506761,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.354,
|
|
"mean_token_accuracy": 0.1367944747209549,
|
|
"num_tokens": 2023543.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.675815534591675,
|
|
"epoch": 0.9789075175770687,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.4228,
|
|
"mean_token_accuracy": 0.1348619595170021,
|
|
"num_tokens": 2034789.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.4962972640991214,
|
|
"epoch": 0.9843158464034614,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.3091,
|
|
"mean_token_accuracy": 0.13746569231152533,
|
|
"num_tokens": 2046719.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.470368957519531,
|
|
"epoch": 0.989724175229854,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.2586,
|
|
"mean_token_accuracy": 0.13995983302593232,
|
|
"num_tokens": 2057016.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.526668930053711,
|
|
"epoch": 0.9951325040562466,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.2582,
|
|
"mean_token_accuracy": 0.1368989862501621,
|
|
"num_tokens": 2067898.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.4760607613457575,
|
|
"epoch": 1.0,
|
|
"grad_norm": 2.078125,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.2648,
|
|
"mean_token_accuracy": 0.14498859478367698,
|
|
"num_tokens": 2076889.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.640408563613891,
|
|
"epoch": 1.0054083288263926,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.2126,
|
|
"mean_token_accuracy": 0.13421827852725982,
|
|
"num_tokens": 2089228.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.453851366043091,
|
|
"epoch": 1.0108166576527853,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.2329,
|
|
"mean_token_accuracy": 0.1465141922235489,
|
|
"num_tokens": 2101098.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.46987681388855,
|
|
"epoch": 1.016224986479178,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.1718,
|
|
"mean_token_accuracy": 0.14940173774957657,
|
|
"num_tokens": 2113081.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.516440105438233,
|
|
"epoch": 1.0216333153055706,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.1389,
|
|
"mean_token_accuracy": 0.14872755408287047,
|
|
"num_tokens": 2123356.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.582454013824463,
|
|
"epoch": 1.0270416441319632,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.2387,
|
|
"mean_token_accuracy": 0.14650474786758422,
|
|
"num_tokens": 2134388.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.41712703704834,
|
|
"epoch": 1.0324499729583558,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.1166,
|
|
"mean_token_accuracy": 0.14716140180826187,
|
|
"num_tokens": 2145209.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.449684047698975,
|
|
"epoch": 1.0378583017847485,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.1591,
|
|
"mean_token_accuracy": 0.14986583292484285,
|
|
"num_tokens": 2157146.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.381077623367309,
|
|
"epoch": 1.043266630611141,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.1832,
|
|
"mean_token_accuracy": 0.14549879282712935,
|
|
"num_tokens": 2168721.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.512086629867554,
|
|
"epoch": 1.0486749594375337,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.2002,
|
|
"mean_token_accuracy": 0.14564864486455917,
|
|
"num_tokens": 2179213.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.524026775360108,
|
|
"epoch": 1.0540832882639264,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.1874,
|
|
"mean_token_accuracy": 0.1458301618695259,
|
|
"num_tokens": 2191224.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.455999326705933,
|
|
"epoch": 1.059491617090319,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.2661,
|
|
"mean_token_accuracy": 0.14348524808883667,
|
|
"num_tokens": 2203665.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.5328833103179935,
|
|
"epoch": 1.0648999459167117,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.1686,
|
|
"mean_token_accuracy": 0.1457217141985893,
|
|
"num_tokens": 2215136.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.491800928115845,
|
|
"epoch": 1.0703082747431043,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.2022,
|
|
"mean_token_accuracy": 0.14653969779610634,
|
|
"num_tokens": 2226375.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.3262214183807375,
|
|
"epoch": 1.075716603569497,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.0592,
|
|
"mean_token_accuracy": 0.15307881981134414,
|
|
"num_tokens": 2236097.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.592307758331299,
|
|
"epoch": 1.0811249323958896,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.3171,
|
|
"mean_token_accuracy": 0.13785527274012566,
|
|
"num_tokens": 2247964.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 1.0811249323958896,
|
|
"eval_entropy": 6.237357024712996,
|
|
"eval_loss": 6.420225620269775,
|
|
"eval_mean_token_accuracy": 0.1424645613455637,
|
|
"eval_num_tokens": 2247964.0,
|
|
"eval_runtime": 2.0378,
|
|
"eval_samples_per_second": 1724.891,
|
|
"eval_steps_per_second": 215.918,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.435177373886108,
|
|
"epoch": 1.0865332612222822,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999997383518051,
|
|
"loss": 6.2381,
|
|
"mean_token_accuracy": 0.14582235515117645,
|
|
"num_tokens": 2260467.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.423948192596436,
|
|
"epoch": 1.0919415900486749,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999986754070562,
|
|
"loss": 6.1489,
|
|
"mean_token_accuracy": 0.14645124971866608,
|
|
"num_tokens": 2272176.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.449360799789429,
|
|
"epoch": 1.0973499188750675,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004999967948166009,
|
|
"loss": 6.1928,
|
|
"mean_token_accuracy": 0.15050890892744065,
|
|
"num_tokens": 2283491.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.479728126525879,
|
|
"epoch": 1.1027582477014604,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004999940965872734,
|
|
"loss": 6.0795,
|
|
"mean_token_accuracy": 0.14409494027495384,
|
|
"num_tokens": 2295131.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.365811204910278,
|
|
"epoch": 1.1081665765278528,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999905807288788,
|
|
"loss": 6.0943,
|
|
"mean_token_accuracy": 0.151952263712883,
|
|
"num_tokens": 2305832.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.331813812255859,
|
|
"epoch": 1.1135749053542456,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000499986247254194,
|
|
"loss": 6.0956,
|
|
"mean_token_accuracy": 0.15410929024219513,
|
|
"num_tokens": 2315874.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.534620904922486,
|
|
"epoch": 1.1189832341806383,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999810961789667,
|
|
"loss": 6.3005,
|
|
"mean_token_accuracy": 0.14081365540623664,
|
|
"num_tokens": 2327580.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.438720798492431,
|
|
"epoch": 1.124391563007031,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000499975127521916,
|
|
"loss": 6.1331,
|
|
"mean_token_accuracy": 0.1487702637910843,
|
|
"num_tokens": 2340426.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.418537998199463,
|
|
"epoch": 1.1297998918334236,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999683413047319,
|
|
"loss": 6.2465,
|
|
"mean_token_accuracy": 0.14431385397911073,
|
|
"num_tokens": 2353228.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.436538887023926,
|
|
"epoch": 1.1352082206598162,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999607375520759,
|
|
"loss": 6.0737,
|
|
"mean_token_accuracy": 0.1525431215763092,
|
|
"num_tokens": 2364238.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.368298864364624,
|
|
"epoch": 1.1406165494862088,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004999523162915794,
|
|
"loss": 6.2128,
|
|
"mean_token_accuracy": 0.14406751096248627,
|
|
"num_tokens": 2375847.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.355217170715332,
|
|
"epoch": 1.1460248783126015,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999430775538459,
|
|
"loss": 6.0978,
|
|
"mean_token_accuracy": 0.1541084788739681,
|
|
"num_tokens": 2386489.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.399798774719239,
|
|
"epoch": 1.151433207138994,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999330213724487,
|
|
"loss": 6.1412,
|
|
"mean_token_accuracy": 0.14941838160157203,
|
|
"num_tokens": 2398839.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.436131286621094,
|
|
"epoch": 1.1568415359653867,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999221477839319,
|
|
"loss": 6.1034,
|
|
"mean_token_accuracy": 0.1503540799021721,
|
|
"num_tokens": 2410016.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.342180395126343,
|
|
"epoch": 1.1622498647917794,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999104568278104,
|
|
"loss": 6.1401,
|
|
"mean_token_accuracy": 0.14826380759477614,
|
|
"num_tokens": 2421202.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.498359870910645,
|
|
"epoch": 1.167658193618172,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000499897948546569,
|
|
"loss": 6.2367,
|
|
"mean_token_accuracy": 0.13952652588486672,
|
|
"num_tokens": 2431943.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.358457422256469,
|
|
"epoch": 1.1730665224445647,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004998846229856629,
|
|
"loss": 6.0432,
|
|
"mean_token_accuracy": 0.14694100618362427,
|
|
"num_tokens": 2443454.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.39977617263794,
|
|
"epoch": 1.1784748512709573,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998704801935171,
|
|
"loss": 6.2352,
|
|
"mean_token_accuracy": 0.1409311354160309,
|
|
"num_tokens": 2455909.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.27590274810791,
|
|
"epoch": 1.18388318009735,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004998555202215267,
|
|
"loss": 6.0927,
|
|
"mean_token_accuracy": 0.1536378711462021,
|
|
"num_tokens": 2467486.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.408985948562622,
|
|
"epoch": 1.1892915089237426,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998397431240561,
|
|
"loss": 6.0573,
|
|
"mean_token_accuracy": 0.15721485614776612,
|
|
"num_tokens": 2477560.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.319623804092407,
|
|
"epoch": 1.1946998377501352,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004998231489584395,
|
|
"loss": 6.0898,
|
|
"mean_token_accuracy": 0.15928600281476973,
|
|
"num_tokens": 2487222.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.336186408996582,
|
|
"epoch": 1.2001081665765279,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00049980573778498,
|
|
"loss": 6.1226,
|
|
"mean_token_accuracy": 0.1514808341860771,
|
|
"num_tokens": 2497474.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.386902284622193,
|
|
"epoch": 1.2055164954029205,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004997875096669498,
|
|
"loss": 6.1757,
|
|
"mean_token_accuracy": 0.1496924042701721,
|
|
"num_tokens": 2508185.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.38016037940979,
|
|
"epoch": 1.2109248242293131,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004997684646705901,
|
|
"loss": 6.1276,
|
|
"mean_token_accuracy": 0.14406513720750808,
|
|
"num_tokens": 2518876.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.339206218719482,
|
|
"epoch": 1.2163331530557058,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004997486028651103,
|
|
"loss": 6.1181,
|
|
"mean_token_accuracy": 0.1543666183948517,
|
|
"num_tokens": 2530554.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.226878881454468,
|
|
"epoch": 1.2217414818820984,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004997279243226884,
|
|
"loss": 5.9914,
|
|
"mean_token_accuracy": 0.14919330775737763,
|
|
"num_tokens": 2541331.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.428718852996826,
|
|
"epoch": 1.227149810708491,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004997064291184701,
|
|
"loss": 6.2128,
|
|
"mean_token_accuracy": 0.14138779938220977,
|
|
"num_tokens": 2554225.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.325518941879272,
|
|
"epoch": 1.2325581395348837,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004996841173305691,
|
|
"loss": 6.089,
|
|
"mean_token_accuracy": 0.15215825885534287,
|
|
"num_tokens": 2565050.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.39612340927124,
|
|
"epoch": 1.2379664683612763,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004996609890400665,
|
|
"loss": 6.1638,
|
|
"mean_token_accuracy": 0.14989278465509415,
|
|
"num_tokens": 2576926.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.47265625,
|
|
"epoch": 1.243374797187669,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004996370443310107,
|
|
"loss": 6.25,
|
|
"mean_token_accuracy": 0.14251136481761933,
|
|
"num_tokens": 2588775.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.229281759262085,
|
|
"epoch": 1.2487831260140616,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004996122832904165,
|
|
"loss": 6.0869,
|
|
"mean_token_accuracy": 0.1530088022351265,
|
|
"num_tokens": 2600534.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.241557168960571,
|
|
"epoch": 1.2541914548404542,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004995867060082658,
|
|
"loss": 5.9202,
|
|
"mean_token_accuracy": 0.15806380808353424,
|
|
"num_tokens": 2612350.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.271128797531128,
|
|
"epoch": 1.2595997836668469,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004995603125775065,
|
|
"loss": 6.1085,
|
|
"mean_token_accuracy": 0.1509605959057808,
|
|
"num_tokens": 2623871.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.281409692764282,
|
|
"epoch": 1.2650081124932395,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004995331030940523,
|
|
"loss": 6.0328,
|
|
"mean_token_accuracy": 0.1470465123653412,
|
|
"num_tokens": 2636371.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.366229057312012,
|
|
"epoch": 1.2704164413196322,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004995050776567826,
|
|
"loss": 6.0853,
|
|
"mean_token_accuracy": 0.15386933013796805,
|
|
"num_tokens": 2646794.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.328421592712402,
|
|
"epoch": 1.2758247701460248,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004994762363675421,
|
|
"loss": 6.101,
|
|
"mean_token_accuracy": 0.1538513943552971,
|
|
"num_tokens": 2657105.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.261244678497315,
|
|
"epoch": 1.2812330989724177,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004994465793311396,
|
|
"loss": 6.0358,
|
|
"mean_token_accuracy": 0.1510448545217514,
|
|
"num_tokens": 2668297.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.218826055526733,
|
|
"epoch": 1.28664142779881,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004994161066553493,
|
|
"loss": 5.8786,
|
|
"mean_token_accuracy": 0.1550386816263199,
|
|
"num_tokens": 2678834.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.323922491073608,
|
|
"epoch": 1.292049756625203,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004993848184509086,
|
|
"loss": 6.1272,
|
|
"mean_token_accuracy": 0.14862247109413146,
|
|
"num_tokens": 2690392.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.249981784820557,
|
|
"epoch": 1.2974580854515954,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004993527148315193,
|
|
"loss": 5.9429,
|
|
"mean_token_accuracy": 0.1543548472225666,
|
|
"num_tokens": 2701319.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.296136903762817,
|
|
"epoch": 1.3028664142779882,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004993197959138458,
|
|
"loss": 6.1028,
|
|
"mean_token_accuracy": 0.14995811581611634,
|
|
"num_tokens": 2713018.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.341501522064209,
|
|
"epoch": 1.3082747431043806,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0004992860618175156,
|
|
"loss": 6.1861,
|
|
"mean_token_accuracy": 0.1437143176794052,
|
|
"num_tokens": 2724995.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.296142959594727,
|
|
"epoch": 1.3136830719307735,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004992515126651184,
|
|
"loss": 6.0388,
|
|
"mean_token_accuracy": 0.15356030091643333,
|
|
"num_tokens": 2736733.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.1639915943145756,
|
|
"epoch": 1.319091400757166,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004992161485822059,
|
|
"loss": 6.0109,
|
|
"mean_token_accuracy": 0.15306128412485123,
|
|
"num_tokens": 2748436.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.265667676925659,
|
|
"epoch": 1.3244997295835588,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004991799696972913,
|
|
"loss": 6.0276,
|
|
"mean_token_accuracy": 0.1567312479019165,
|
|
"num_tokens": 2758924.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.281225633621216,
|
|
"epoch": 1.3299080584099512,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004991429761418488,
|
|
"loss": 6.0892,
|
|
"mean_token_accuracy": 0.14596257656812667,
|
|
"num_tokens": 2769773.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.410810232162476,
|
|
"epoch": 1.335316387236344,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000499105168050313,
|
|
"loss": 6.2244,
|
|
"mean_token_accuracy": 0.1527717739343643,
|
|
"num_tokens": 2782414.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.173236465454101,
|
|
"epoch": 1.3407247160627367,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004990665455600787,
|
|
"loss": 5.9954,
|
|
"mean_token_accuracy": 0.15391278266906738,
|
|
"num_tokens": 2794325.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.122881603240967,
|
|
"epoch": 1.3461330448891293,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004990271088114999,
|
|
"loss": 5.8397,
|
|
"mean_token_accuracy": 0.1607421964406967,
|
|
"num_tokens": 2805391.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.106888484954834,
|
|
"epoch": 1.351541373715522,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004989868579478901,
|
|
"loss": 5.9587,
|
|
"mean_token_accuracy": 0.15657583773136138,
|
|
"num_tokens": 2816633.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.261765575408935,
|
|
"epoch": 1.3569497025419146,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004989457931155209,
|
|
"loss": 6.0819,
|
|
"mean_token_accuracy": 0.15534812957048416,
|
|
"num_tokens": 2828395.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.284375190734863,
|
|
"epoch": 1.3623580313683072,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004989039144636219,
|
|
"loss": 6.0659,
|
|
"mean_token_accuracy": 0.15927724540233612,
|
|
"num_tokens": 2839592.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.2426642894744875,
|
|
"epoch": 1.3677663601946999,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004988612221443804,
|
|
"loss": 6.0407,
|
|
"mean_token_accuracy": 0.16757878810167312,
|
|
"num_tokens": 2849557.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.320202732086182,
|
|
"epoch": 1.3731746890210925,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004988177163129403,
|
|
"loss": 6.0789,
|
|
"mean_token_accuracy": 0.15343549251556396,
|
|
"num_tokens": 2862030.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.190796136856079,
|
|
"epoch": 1.3785830178474852,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004987733971274018,
|
|
"loss": 6.0134,
|
|
"mean_token_accuracy": 0.1602019637823105,
|
|
"num_tokens": 2872379.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.323750352859497,
|
|
"epoch": 1.3839913466738778,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004987282647488211,
|
|
"loss": 6.0854,
|
|
"mean_token_accuracy": 0.15255604684352875,
|
|
"num_tokens": 2883170.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.206164073944092,
|
|
"epoch": 1.3893996755002704,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004986823193412093,
|
|
"loss": 6.0281,
|
|
"mean_token_accuracy": 0.1504069373011589,
|
|
"num_tokens": 2894013.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.265843105316162,
|
|
"epoch": 1.394808004326663,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000498635561071532,
|
|
"loss": 6.0965,
|
|
"mean_token_accuracy": 0.15722659826278687,
|
|
"num_tokens": 2905073.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.322227430343628,
|
|
"epoch": 1.4002163331530557,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000498587990109709,
|
|
"loss": 6.0611,
|
|
"mean_token_accuracy": 0.15372281968593599,
|
|
"num_tokens": 2916579.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.202920436859131,
|
|
"epoch": 1.4056246619794484,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004985396066286133,
|
|
"loss": 5.9529,
|
|
"mean_token_accuracy": 0.15599865168333055,
|
|
"num_tokens": 2928522.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.246823740005493,
|
|
"epoch": 1.411032990805841,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004984904108040706,
|
|
"loss": 6.0418,
|
|
"mean_token_accuracy": 0.16009896695613862,
|
|
"num_tokens": 2938754.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.096659517288208,
|
|
"epoch": 1.4164413196322336,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004984404028148585,
|
|
"loss": 5.8621,
|
|
"mean_token_accuracy": 0.16269333213567733,
|
|
"num_tokens": 2949696.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.224114513397216,
|
|
"epoch": 1.4218496484586263,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004983895828427062,
|
|
"loss": 6.0548,
|
|
"mean_token_accuracy": 0.14663359969854356,
|
|
"num_tokens": 2961144.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.282317972183227,
|
|
"epoch": 1.427257977285019,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004983379510722935,
|
|
"loss": 6.0389,
|
|
"mean_token_accuracy": 0.15705436319112778,
|
|
"num_tokens": 2972368.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.181911420822144,
|
|
"epoch": 1.4326663061114115,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004982855076912506,
|
|
"loss": 5.9638,
|
|
"mean_token_accuracy": 0.1542641267180443,
|
|
"num_tokens": 2984853.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.2893272876739506,
|
|
"epoch": 1.4380746349378042,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004982322528901568,
|
|
"loss": 6.0196,
|
|
"mean_token_accuracy": 0.15734567493200302,
|
|
"num_tokens": 2995095.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.136825799942017,
|
|
"epoch": 1.4434829637641968,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00049817818686254,
|
|
"loss": 6.0786,
|
|
"mean_token_accuracy": 0.15213681608438492,
|
|
"num_tokens": 3005926.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.327638864517212,
|
|
"epoch": 1.4488912925905895,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004981233098048765,
|
|
"loss": 6.029,
|
|
"mean_token_accuracy": 0.1540302723646164,
|
|
"num_tokens": 3017647.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.090318870544434,
|
|
"epoch": 1.454299621416982,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004980676219165894,
|
|
"loss": 5.9315,
|
|
"mean_token_accuracy": 0.16002180129289628,
|
|
"num_tokens": 3028904.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.2678004741668705,
|
|
"epoch": 1.4597079502433747,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000498011123400049,
|
|
"loss": 5.9859,
|
|
"mean_token_accuracy": 0.15976762771606445,
|
|
"num_tokens": 3039347.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.134640789031982,
|
|
"epoch": 1.4651162790697674,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004979538144605707,
|
|
"loss": 5.9547,
|
|
"mean_token_accuracy": 0.15985055714845658,
|
|
"num_tokens": 3049951.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.237525272369385,
|
|
"epoch": 1.47052460789616,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004978956953064154,
|
|
"loss": 6.0,
|
|
"mean_token_accuracy": 0.1584839180111885,
|
|
"num_tokens": 3061894.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.22474308013916,
|
|
"epoch": 1.4759329367225527,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004978367661487883,
|
|
"loss": 5.9786,
|
|
"mean_token_accuracy": 0.16255034655332565,
|
|
"num_tokens": 3072393.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.14780969619751,
|
|
"epoch": 1.4813412655489453,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004977770272018379,
|
|
"loss": 6.016,
|
|
"mean_token_accuracy": 0.15890694856643678,
|
|
"num_tokens": 3083541.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.193307638168335,
|
|
"epoch": 1.486749594375338,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000497716478682656,
|
|
"loss": 5.8382,
|
|
"mean_token_accuracy": 0.15965883284807206,
|
|
"num_tokens": 3093851.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.02174940109253,
|
|
"epoch": 1.4921579232017308,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004976551208112759,
|
|
"loss": 5.9241,
|
|
"mean_token_accuracy": 0.1565147563815117,
|
|
"num_tokens": 3105703.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.326010656356812,
|
|
"epoch": 1.4975662520281232,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004975929538106721,
|
|
"loss": 6.0652,
|
|
"mean_token_accuracy": 0.15564267784357072,
|
|
"num_tokens": 3117063.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.1744105339050295,
|
|
"epoch": 1.502974580854516,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004975299779067598,
|
|
"loss": 6.0253,
|
|
"mean_token_accuracy": 0.15938320606946946,
|
|
"num_tokens": 3129020.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.207565689086914,
|
|
"epoch": 1.5083829096809085,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004974661933283936,
|
|
"loss": 6.0008,
|
|
"mean_token_accuracy": 0.15763620287179947,
|
|
"num_tokens": 3140627.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.17385630607605,
|
|
"epoch": 1.5137912385073014,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004974016003073667,
|
|
"loss": 6.0328,
|
|
"mean_token_accuracy": 0.15484507232904435,
|
|
"num_tokens": 3152294.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.2234649658203125,
|
|
"epoch": 1.5191995673336938,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004973361990784104,
|
|
"loss": 6.0313,
|
|
"mean_token_accuracy": 0.15227337181568146,
|
|
"num_tokens": 3163770.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.2478063106536865,
|
|
"epoch": 1.5246078961600866,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004972699898791927,
|
|
"loss": 5.9855,
|
|
"mean_token_accuracy": 0.1566041514277458,
|
|
"num_tokens": 3175804.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.2702436447143555,
|
|
"epoch": 1.530016224986479,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004972029729503182,
|
|
"loss": 5.9948,
|
|
"mean_token_accuracy": 0.15855323225259782,
|
|
"num_tokens": 3187233.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.095852327346802,
|
|
"epoch": 1.535424553812872,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004971351485353265,
|
|
"loss": 5.9551,
|
|
"mean_token_accuracy": 0.16486137360334396,
|
|
"num_tokens": 3198884.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.124053001403809,
|
|
"epoch": 1.5408328826392643,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004970665168806917,
|
|
"loss": 5.9263,
|
|
"mean_token_accuracy": 0.15676868557929993,
|
|
"num_tokens": 3209836.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.1572350025177,
|
|
"epoch": 1.5462412114656572,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004969970782358214,
|
|
"loss": 5.9607,
|
|
"mean_token_accuracy": 0.15960703641176224,
|
|
"num_tokens": 3221295.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.123327684402466,
|
|
"epoch": 1.5516495402920496,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004969268328530557,
|
|
"loss": 5.9088,
|
|
"mean_token_accuracy": 0.16821087300777435,
|
|
"num_tokens": 3231832.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.160953330993652,
|
|
"epoch": 1.5570578691184425,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004968557809876666,
|
|
"loss": 5.9881,
|
|
"mean_token_accuracy": 0.15660501420497894,
|
|
"num_tokens": 3243483.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.189469814300537,
|
|
"epoch": 1.562466197944835,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004967839228978568,
|
|
"loss": 6.0146,
|
|
"mean_token_accuracy": 0.15640730857849122,
|
|
"num_tokens": 3255785.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.14083776473999,
|
|
"epoch": 1.5678745267712277,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004967112588447585,
|
|
"loss": 5.9098,
|
|
"mean_token_accuracy": 0.16025495976209642,
|
|
"num_tokens": 3267035.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.290738296508789,
|
|
"epoch": 1.5732828555976204,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004966377890924332,
|
|
"loss": 5.9932,
|
|
"mean_token_accuracy": 0.16156057864427567,
|
|
"num_tokens": 3277848.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.070386457443237,
|
|
"epoch": 1.578691184424013,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004965635139078704,
|
|
"loss": 5.9082,
|
|
"mean_token_accuracy": 0.16141002327203752,
|
|
"num_tokens": 3289864.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.239854574203491,
|
|
"epoch": 1.5840995132504057,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004964884335609857,
|
|
"loss": 6.0348,
|
|
"mean_token_accuracy": 0.16052431836724282,
|
|
"num_tokens": 3301441.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.124419498443603,
|
|
"epoch": 1.5895078420767983,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004964125483246218,
|
|
"loss": 5.9979,
|
|
"mean_token_accuracy": 0.15076059475541115,
|
|
"num_tokens": 3313216.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.070195484161377,
|
|
"epoch": 1.594916170903191,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004963358584745457,
|
|
"loss": 5.8015,
|
|
"mean_token_accuracy": 0.16063344478607178,
|
|
"num_tokens": 3323760.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.0324774265289305,
|
|
"epoch": 1.6003244997295836,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004962583642894482,
|
|
"loss": 5.8812,
|
|
"mean_token_accuracy": 0.16131513714790344,
|
|
"num_tokens": 3334785.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.173135423660279,
|
|
"epoch": 1.6057328285559762,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004961800660509437,
|
|
"loss": 5.9541,
|
|
"mean_token_accuracy": 0.16318012028932571,
|
|
"num_tokens": 3345741.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.118630647659302,
|
|
"epoch": 1.6111411573823688,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000496100964043568,
|
|
"loss": 5.8857,
|
|
"mean_token_accuracy": 0.16484943330287932,
|
|
"num_tokens": 3356118.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.00065393447876,
|
|
"epoch": 1.6165494862087615,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000496021058554778,
|
|
"loss": 5.8363,
|
|
"mean_token_accuracy": 0.16194700598716735,
|
|
"num_tokens": 3368107.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.143065738677978,
|
|
"epoch": 1.6219578150351541,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004959403498749505,
|
|
"loss": 5.8521,
|
|
"mean_token_accuracy": 0.16149836629629136,
|
|
"num_tokens": 3379367.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.014459323883057,
|
|
"epoch": 1.6273661438615468,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004958588382973809,
|
|
"loss": 5.9791,
|
|
"mean_token_accuracy": 0.1566162794828415,
|
|
"num_tokens": 3390597.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.243663692474366,
|
|
"epoch": 1.6327744726879394,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004957765241182826,
|
|
"loss": 5.8886,
|
|
"mean_token_accuracy": 0.1644733265042305,
|
|
"num_tokens": 3402618.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 5.919700384140015,
|
|
"epoch": 1.638182801514332,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004956934076367854,
|
|
"loss": 5.8038,
|
|
"mean_token_accuracy": 0.1675911948084831,
|
|
"num_tokens": 3413162.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.054862642288208,
|
|
"epoch": 1.6435911303407247,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000495609489154935,
|
|
"loss": 5.8933,
|
|
"mean_token_accuracy": 0.1644514724612236,
|
|
"num_tokens": 3424331.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.142803049087524,
|
|
"epoch": 1.6489994591671173,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004955247689776911,
|
|
"loss": 5.9979,
|
|
"mean_token_accuracy": 0.15501589924097062,
|
|
"num_tokens": 3434739.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.229400968551635,
|
|
"epoch": 1.65440778799351,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000495439247412927,
|
|
"loss": 6.0683,
|
|
"mean_token_accuracy": 0.14968138113617896,
|
|
"num_tokens": 3445984.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 5.973660135269165,
|
|
"epoch": 1.6598161168199026,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004953529247714285,
|
|
"loss": 5.8081,
|
|
"mean_token_accuracy": 0.16547674387693406,
|
|
"num_tokens": 3456955.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.215045833587647,
|
|
"epoch": 1.6652244456462952,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004952658013668922,
|
|
"loss": 5.9572,
|
|
"mean_token_accuracy": 0.16140533089637757,
|
|
"num_tokens": 3468895.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.016715955734253,
|
|
"epoch": 1.670632774472688,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004951778775159246,
|
|
"loss": 5.8493,
|
|
"mean_token_accuracy": 0.16565765887498857,
|
|
"num_tokens": 3480083.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.1815954685211185,
|
|
"epoch": 1.6760411032990805,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004950891535380414,
|
|
"loss": 5.9857,
|
|
"mean_token_accuracy": 0.1545367419719696,
|
|
"num_tokens": 3492242.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.1077052593231205,
|
|
"epoch": 1.6814494321254734,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004949996297556656,
|
|
"loss": 5.8076,
|
|
"mean_token_accuracy": 0.16661108732223512,
|
|
"num_tokens": 3501926.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 5.967837524414063,
|
|
"epoch": 1.6868577609518658,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004949093064941268,
|
|
"loss": 5.8846,
|
|
"mean_token_accuracy": 0.16590456813573837,
|
|
"num_tokens": 3511421.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.1532501697540285,
|
|
"epoch": 1.6922660897782587,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004948181840816601,
|
|
"loss": 5.8242,
|
|
"mean_token_accuracy": 0.16331643760204315,
|
|
"num_tokens": 3522033.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.050137948989868,
|
|
"epoch": 1.697674418604651,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004947262628494045,
|
|
"loss": 5.8848,
|
|
"mean_token_accuracy": 0.16303865164518355,
|
|
"num_tokens": 3533232.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.181738662719726,
|
|
"epoch": 1.703082747431044,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004946335431314018,
|
|
"loss": 5.8914,
|
|
"mean_token_accuracy": 0.16298120319843293,
|
|
"num_tokens": 3544904.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.067026996612549,
|
|
"epoch": 1.7084910762574363,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004945400252645959,
|
|
"loss": 5.9643,
|
|
"mean_token_accuracy": 0.16130709350109101,
|
|
"num_tokens": 3556545.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.050175046920776,
|
|
"epoch": 1.7138994050838292,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004944457095888308,
|
|
"loss": 5.8467,
|
|
"mean_token_accuracy": 0.16689251065254213,
|
|
"num_tokens": 3567361.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.096579885482788,
|
|
"epoch": 1.7193077339102216,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00049435059644685,
|
|
"loss": 5.8759,
|
|
"mean_token_accuracy": 0.16191123872995378,
|
|
"num_tokens": 3578943.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.071146059036255,
|
|
"epoch": 1.7247160627366145,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004942546861842947,
|
|
"loss": 5.8558,
|
|
"mean_token_accuracy": 0.16385841220617295,
|
|
"num_tokens": 3589868.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.076867294311524,
|
|
"epoch": 1.730124391563007,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004941579791497032,
|
|
"loss": 5.8472,
|
|
"mean_token_accuracy": 0.1693348065018654,
|
|
"num_tokens": 3601245.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.083900833129883,
|
|
"epoch": 1.7355327203893998,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004940604756945091,
|
|
"loss": 5.9021,
|
|
"mean_token_accuracy": 0.16264761984348297,
|
|
"num_tokens": 3612168.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.028712034225464,
|
|
"epoch": 1.7409410492157922,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004939621761730399,
|
|
"loss": 5.8361,
|
|
"mean_token_accuracy": 0.16627179235219955,
|
|
"num_tokens": 3624699.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.148156976699829,
|
|
"epoch": 1.746349378042185,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004938630809425166,
|
|
"loss": 5.9773,
|
|
"mean_token_accuracy": 0.16007167398929595,
|
|
"num_tokens": 3635554.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 5.953314685821534,
|
|
"epoch": 1.7517577068685775,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004937631903630513,
|
|
"loss": 5.8586,
|
|
"mean_token_accuracy": 0.16080379039049147,
|
|
"num_tokens": 3646776.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.064091157913208,
|
|
"epoch": 1.7571660356949703,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004936625047976468,
|
|
"loss": 5.8028,
|
|
"mean_token_accuracy": 0.1684928059577942,
|
|
"num_tokens": 3657976.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.161013555526734,
|
|
"epoch": 1.7625743645213627,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004935610246121944,
|
|
"loss": 5.9253,
|
|
"mean_token_accuracy": 0.15793002545833587,
|
|
"num_tokens": 3668242.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.1408998489379885,
|
|
"epoch": 1.7679826933477556,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004934587501754736,
|
|
"loss": 5.9519,
|
|
"mean_token_accuracy": 0.16056527346372604,
|
|
"num_tokens": 3679200.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.017309331893921,
|
|
"epoch": 1.7733910221741482,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004933556818591498,
|
|
"loss": 5.8448,
|
|
"mean_token_accuracy": 0.16986124962568283,
|
|
"num_tokens": 3690227.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.095331621170044,
|
|
"epoch": 1.7787993510005409,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004932518200377736,
|
|
"loss": 5.8789,
|
|
"mean_token_accuracy": 0.16581207513809204,
|
|
"num_tokens": 3701821.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 5.968796920776367,
|
|
"epoch": 1.7842076798269335,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004931471650887789,
|
|
"loss": 5.8202,
|
|
"mean_token_accuracy": 0.160505910217762,
|
|
"num_tokens": 3712420.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.156483602523804,
|
|
"epoch": 1.7896160086533262,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004930417173924822,
|
|
"loss": 5.8718,
|
|
"mean_token_accuracy": 0.1640415072441101,
|
|
"num_tokens": 3722211.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.002517366409302,
|
|
"epoch": 1.7950243374797188,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004929354773320806,
|
|
"loss": 5.8544,
|
|
"mean_token_accuracy": 0.159525004029274,
|
|
"num_tokens": 3732822.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.117061996459961,
|
|
"epoch": 1.8004326663061114,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004928284452936506,
|
|
"loss": 5.9121,
|
|
"mean_token_accuracy": 0.16279875338077546,
|
|
"num_tokens": 3744262.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 5.985845041275025,
|
|
"epoch": 1.805840995132504,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004927206216661473,
|
|
"loss": 5.811,
|
|
"mean_token_accuracy": 0.16297463476657867,
|
|
"num_tokens": 3755713.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.010738372802734,
|
|
"epoch": 1.8112493239588967,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004926120068414015,
|
|
"loss": 5.8269,
|
|
"mean_token_accuracy": 0.16570561081171037,
|
|
"num_tokens": 3767179.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.005726718902588,
|
|
"epoch": 1.8166576527852893,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004925026012141201,
|
|
"loss": 5.8377,
|
|
"mean_token_accuracy": 0.1643834114074707,
|
|
"num_tokens": 3777714.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.100140476226807,
|
|
"epoch": 1.822065981611682,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004923924051818832,
|
|
"loss": 5.8391,
|
|
"mean_token_accuracy": 0.16618706285953522,
|
|
"num_tokens": 3788092.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.018685340881348,
|
|
"epoch": 1.8274743104380746,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004922814191451435,
|
|
"loss": 5.7929,
|
|
"mean_token_accuracy": 0.16196684390306473,
|
|
"num_tokens": 3798080.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.0061867237091064,
|
|
"epoch": 1.8328826392644673,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004921696435072245,
|
|
"loss": 5.8373,
|
|
"mean_token_accuracy": 0.16848887354135514,
|
|
"num_tokens": 3809547.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.0457484245300295,
|
|
"epoch": 1.83829096809086,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004920570786743191,
|
|
"loss": 5.8173,
|
|
"mean_token_accuracy": 0.17119550853967666,
|
|
"num_tokens": 3820681.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.0401458740234375,
|
|
"epoch": 1.8436992969172525,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004919437250554882,
|
|
"loss": 5.8286,
|
|
"mean_token_accuracy": 0.16929454803466798,
|
|
"num_tokens": 3831598.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.023636150360107,
|
|
"epoch": 1.8491076257436452,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004918295830626592,
|
|
"loss": 5.8708,
|
|
"mean_token_accuracy": 0.16968689113855362,
|
|
"num_tokens": 3843437.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 5.988752841949463,
|
|
"epoch": 1.8545159545700378,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004917146531106242,
|
|
"loss": 5.7428,
|
|
"mean_token_accuracy": 0.1648212730884552,
|
|
"num_tokens": 3855387.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.023560953140259,
|
|
"epoch": 1.8599242833964305,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004915989356170389,
|
|
"loss": 5.8914,
|
|
"mean_token_accuracy": 0.1611024931073189,
|
|
"num_tokens": 3866999.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.036889457702637,
|
|
"epoch": 1.865332612222823,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004914824310024213,
|
|
"loss": 5.8236,
|
|
"mean_token_accuracy": 0.16607335805892945,
|
|
"num_tokens": 3877721.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.054341411590576,
|
|
"epoch": 1.870740941049216,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004913651396901489,
|
|
"loss": 5.9087,
|
|
"mean_token_accuracy": 0.16150028258562088,
|
|
"num_tokens": 3888831.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.0500908374786375,
|
|
"epoch": 1.8761492698756084,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000491247062106459,
|
|
"loss": 5.8339,
|
|
"mean_token_accuracy": 0.16762833893299103,
|
|
"num_tokens": 3900389.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 5.942811632156372,
|
|
"epoch": 1.8815575987020012,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004911281986804455,
|
|
"loss": 5.8029,
|
|
"mean_token_accuracy": 0.16817400902509688,
|
|
"num_tokens": 3911641.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 5.86643352508545,
|
|
"epoch": 1.8869659275283936,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004910085498440587,
|
|
"loss": 5.7026,
|
|
"mean_token_accuracy": 0.16996440291404724,
|
|
"num_tokens": 3922877.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.069116258621216,
|
|
"epoch": 1.8923742563547865,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004908881160321025,
|
|
"loss": 5.7807,
|
|
"mean_token_accuracy": 0.16565157771110534,
|
|
"num_tokens": 3933573.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.921838235855103,
|
|
"epoch": 1.897782585181179,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004907668976822338,
|
|
"loss": 5.7278,
|
|
"mean_token_accuracy": 0.17627133280038834,
|
|
"num_tokens": 3944339.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.119243955612182,
|
|
"epoch": 1.9031909140075718,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004906448952349603,
|
|
"loss": 5.8294,
|
|
"mean_token_accuracy": 0.17736046612262726,
|
|
"num_tokens": 3955459.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.896986055374145,
|
|
"epoch": 1.9085992428339642,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004905221091336392,
|
|
"loss": 5.7274,
|
|
"mean_token_accuracy": 0.1731481984257698,
|
|
"num_tokens": 3966782.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.9977803230285645,
|
|
"epoch": 1.914007571660357,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004903985398244753,
|
|
"loss": 5.8518,
|
|
"mean_token_accuracy": 0.16875911355018616,
|
|
"num_tokens": 3978794.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.09336519241333,
|
|
"epoch": 1.9194159004867495,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004902741877565201,
|
|
"loss": 5.8626,
|
|
"mean_token_accuracy": 0.16668010503053665,
|
|
"num_tokens": 3990838.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.055723190307617,
|
|
"epoch": 1.9248242293131423,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000490149053381669,
|
|
"loss": 5.8951,
|
|
"mean_token_accuracy": 0.1674518883228302,
|
|
"num_tokens": 4000630.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.027546167373657,
|
|
"epoch": 1.9302325581395348,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004900231371546608,
|
|
"loss": 5.7717,
|
|
"mean_token_accuracy": 0.17280880063772203,
|
|
"num_tokens": 4011433.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.9891197681427,
|
|
"epoch": 1.9356408869659276,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004898964395330753,
|
|
"loss": 5.9187,
|
|
"mean_token_accuracy": 0.166441847383976,
|
|
"num_tokens": 4023671.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 5.98225588798523,
|
|
"epoch": 1.94104921579232,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004897689609773318,
|
|
"loss": 5.705,
|
|
"mean_token_accuracy": 0.16938872784376144,
|
|
"num_tokens": 4034686.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.053580093383789,
|
|
"epoch": 1.946457544618713,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004896407019506879,
|
|
"loss": 5.8215,
|
|
"mean_token_accuracy": 0.17409656196832657,
|
|
"num_tokens": 4046387.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.924436092376709,
|
|
"epoch": 1.9518658734451053,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004895116629192371,
|
|
"loss": 5.7349,
|
|
"mean_token_accuracy": 0.17091045677661895,
|
|
"num_tokens": 4056103.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.991981029510498,
|
|
"epoch": 1.9572742022714982,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004893818443519074,
|
|
"loss": 5.7,
|
|
"mean_token_accuracy": 0.18084933310747148,
|
|
"num_tokens": 4066940.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.908279466629028,
|
|
"epoch": 1.9626825310978906,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00048925124672046,
|
|
"loss": 5.8338,
|
|
"mean_token_accuracy": 0.17757519632577895,
|
|
"num_tokens": 4077232.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.0143650531768795,
|
|
"epoch": 1.9680908599242835,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004891198704994868,
|
|
"loss": 5.8963,
|
|
"mean_token_accuracy": 0.16193630546331406,
|
|
"num_tokens": 4089322.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.993407154083252,
|
|
"epoch": 1.973499188750676,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004889877161664096,
|
|
"loss": 5.7359,
|
|
"mean_token_accuracy": 0.1731564626097679,
|
|
"num_tokens": 4099737.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.926445531845093,
|
|
"epoch": 1.9789075175770687,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004888547842014771,
|
|
"loss": 5.7443,
|
|
"mean_token_accuracy": 0.16823544651269912,
|
|
"num_tokens": 4110793.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.006844568252563,
|
|
"epoch": 1.9843158464034614,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004887210750877648,
|
|
"loss": 5.781,
|
|
"mean_token_accuracy": 0.17114681750535965,
|
|
"num_tokens": 4121481.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.902514410018921,
|
|
"epoch": 1.989724175229854,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004885865893111716,
|
|
"loss": 5.7411,
|
|
"mean_token_accuracy": 0.17603601664304733,
|
|
"num_tokens": 4132701.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.985061120986939,
|
|
"epoch": 1.9951325040562466,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004884513273604195,
|
|
"loss": 5.8062,
|
|
"mean_token_accuracy": 0.17002770006656648,
|
|
"num_tokens": 4143764.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.0021616617838545,
|
|
"epoch": 2.0,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 0.0004883152897270502,
|
|
"loss": 5.7771,
|
|
"mean_token_accuracy": 0.170842116077741,
|
|
"num_tokens": 4153778.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.96401014328003,
|
|
"epoch": 2.005408328826393,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004881784769054253,
|
|
"loss": 5.5318,
|
|
"mean_token_accuracy": 0.18372500836849212,
|
|
"num_tokens": 4165099.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.953923511505127,
|
|
"epoch": 2.0108166576527853,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004880408893927226,
|
|
"loss": 5.6148,
|
|
"mean_token_accuracy": 0.1810345396399498,
|
|
"num_tokens": 4178051.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.986582851409912,
|
|
"epoch": 2.016224986479178,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004879025276889355,
|
|
"loss": 5.5256,
|
|
"mean_token_accuracy": 0.18339935392141343,
|
|
"num_tokens": 4189987.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.020156669616699,
|
|
"epoch": 2.0216333153055706,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048776339229687077,
|
|
"loss": 5.5614,
|
|
"mean_token_accuracy": 0.18038318008184434,
|
|
"num_tokens": 4200999.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.838561725616455,
|
|
"epoch": 2.0270416441319634,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048762348372214683,
|
|
"loss": 5.488,
|
|
"mean_token_accuracy": 0.17765122652053833,
|
|
"num_tokens": 4212122.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.974667358398437,
|
|
"epoch": 2.032449972958356,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00048748280247319165,
|
|
"loss": 5.5611,
|
|
"mean_token_accuracy": 0.18277785331010818,
|
|
"num_tokens": 4222580.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.839652013778687,
|
|
"epoch": 2.0378583017847487,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004873413490612412,
|
|
"loss": 5.4009,
|
|
"mean_token_accuracy": 0.18563185334205629,
|
|
"num_tokens": 4233756.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.89198784828186,
|
|
"epoch": 2.043266630611141,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048719912400033757,
|
|
"loss": 5.53,
|
|
"mean_token_accuracy": 0.17903512865304946,
|
|
"num_tokens": 4245625.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.023007154464722,
|
|
"epoch": 2.048674959437534,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000487056127807327,
|
|
"loss": 5.7544,
|
|
"mean_token_accuracy": 0.176625756919384,
|
|
"num_tokens": 4257788.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.884478378295898,
|
|
"epoch": 2.0540832882639264,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00048691236100185795,
|
|
"loss": 5.4863,
|
|
"mean_token_accuracy": 0.18627301156520842,
|
|
"num_tokens": 4269641.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.923391056060791,
|
|
"epoch": 2.0594916170903192,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004867678241063792,
|
|
"loss": 5.5542,
|
|
"mean_token_accuracy": 0.18891489654779434,
|
|
"num_tokens": 4280952.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.90982141494751,
|
|
"epoch": 2.0648999459167117,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00048662251764613833,
|
|
"loss": 5.4983,
|
|
"mean_token_accuracy": 0.18720743656158448,
|
|
"num_tokens": 4294371.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.810564851760864,
|
|
"epoch": 2.0703082747431045,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004864764421491793,
|
|
"loss": 5.5078,
|
|
"mean_token_accuracy": 0.18103248476982117,
|
|
"num_tokens": 4307112.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.906817483901977,
|
|
"epoch": 2.075716603569497,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004863295981463408,
|
|
"loss": 5.4927,
|
|
"mean_token_accuracy": 0.18527511060237883,
|
|
"num_tokens": 4317199.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.839090013504029,
|
|
"epoch": 2.08112493239589,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004861819861712543,
|
|
"loss": 5.492,
|
|
"mean_token_accuracy": 0.1836896702647209,
|
|
"num_tokens": 4328546.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.978621196746826,
|
|
"epoch": 2.086533261222282,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048603360676034203,
|
|
"loss": 5.545,
|
|
"mean_token_accuracy": 0.18170068711042403,
|
|
"num_tokens": 4338795.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.801554346084595,
|
|
"epoch": 2.091941590048675,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048588446045281505,
|
|
"loss": 5.535,
|
|
"mean_token_accuracy": 0.17918962985277176,
|
|
"num_tokens": 4350373.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.860919666290283,
|
|
"epoch": 2.0973499188750675,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004857345477906714,
|
|
"loss": 5.5141,
|
|
"mean_token_accuracy": 0.17788466066122055,
|
|
"num_tokens": 4362801.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.988518285751343,
|
|
"epoch": 2.1027582477014604,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00048558386931869417,
|
|
"loss": 5.5982,
|
|
"mean_token_accuracy": 0.18577716499567032,
|
|
"num_tokens": 4375782.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.7955482006073,
|
|
"epoch": 2.1081665765278528,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004854324255844492,
|
|
"loss": 5.5003,
|
|
"mean_token_accuracy": 0.18517474532127381,
|
|
"num_tokens": 4385580.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.883208179473877,
|
|
"epoch": 2.1135749053542456,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048528021713828336,
|
|
"loss": 5.5169,
|
|
"mean_token_accuracy": 0.17969103306531906,
|
|
"num_tokens": 4397434.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.881035614013672,
|
|
"epoch": 2.118983234180638,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004851272445333225,
|
|
"loss": 5.5583,
|
|
"mean_token_accuracy": 0.17854398339986802,
|
|
"num_tokens": 4407888.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.829453849792481,
|
|
"epoch": 2.124391563007031,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00048497350832546937,
|
|
"loss": 5.4845,
|
|
"mean_token_accuracy": 0.1851338788866997,
|
|
"num_tokens": 4420010.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.817228889465332,
|
|
"epoch": 2.1297998918334233,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004848190090734018,
|
|
"loss": 5.5211,
|
|
"mean_token_accuracy": 0.17922184765338897,
|
|
"num_tokens": 4430730.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.825629997253418,
|
|
"epoch": 2.135208220659816,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00048466374733857035,
|
|
"loss": 5.4969,
|
|
"mean_token_accuracy": 0.18972273021936417,
|
|
"num_tokens": 4442296.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.90300760269165,
|
|
"epoch": 2.1406165494862086,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004845077236851966,
|
|
"loss": 5.5284,
|
|
"mean_token_accuracy": 0.18410815596580504,
|
|
"num_tokens": 4453996.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.782866811752319,
|
|
"epoch": 2.1460248783126015,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00048435093868027076,
|
|
"loss": 5.5212,
|
|
"mean_token_accuracy": 0.18278288543224336,
|
|
"num_tokens": 4465779.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.725567245483399,
|
|
"epoch": 2.151433207138994,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00048419339289354995,
|
|
"loss": 5.4559,
|
|
"mean_token_accuracy": 0.1890066847205162,
|
|
"num_tokens": 4477113.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.857697677612305,
|
|
"epoch": 2.1568415359653867,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000484035086897556,
|
|
"loss": 5.4231,
|
|
"mean_token_accuracy": 0.1865213468670845,
|
|
"num_tokens": 4487920.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.944146585464478,
|
|
"epoch": 2.162249864791779,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004838760212675732,
|
|
"loss": 5.6188,
|
|
"mean_token_accuracy": 0.17914327830076218,
|
|
"num_tokens": 4499049.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 2.162249864791779,
|
|
"eval_entropy": 5.714854424650019,
|
|
"eval_loss": 6.0214762687683105,
|
|
"eval_mean_token_accuracy": 0.1667367118156769,
|
|
"eval_num_tokens": 4499049.0,
|
|
"eval_runtime": 1.9484,
|
|
"eval_samples_per_second": 1804.032,
|
|
"eval_steps_per_second": 225.825,
|
|
"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": 1045563522416640.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|