e9545c4dde
Model: fpadovani/dan-latn-10mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
4079 lines
110 KiB
JSON
4079 lines
110 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.6736401673640167,
|
|
"eval_steps": 500,
|
|
"global_step": 2000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 7.010165643692017,
|
|
"epoch": 0.0041841004184100415,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 2e-06,
|
|
"loss": 6.5355,
|
|
"mean_token_accuracy": 0.10552035421133041,
|
|
"num_tokens": 9848.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 7.061803245544434,
|
|
"epoch": 0.008368200836820083,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 6.57,
|
|
"mean_token_accuracy": 0.10846093520522118,
|
|
"num_tokens": 18196.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 7.030461692810059,
|
|
"epoch": 0.012552301255230125,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 7e-06,
|
|
"loss": 6.4999,
|
|
"mean_token_accuracy": 0.10816544145345688,
|
|
"num_tokens": 26927.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 7.067274856567383,
|
|
"epoch": 0.016736401673640166,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 6.5822,
|
|
"mean_token_accuracy": 0.10798880383372307,
|
|
"num_tokens": 34987.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 7.056567573547364,
|
|
"epoch": 0.02092050209205021,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 6.5868,
|
|
"mean_token_accuracy": 0.1059174194931984,
|
|
"num_tokens": 44936.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 7.025119495391846,
|
|
"epoch": 0.02510460251046025,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 6.6011,
|
|
"mean_token_accuracy": 0.10448738262057304,
|
|
"num_tokens": 54562.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 7.036947107315063,
|
|
"epoch": 0.029288702928870293,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 6.6647,
|
|
"mean_token_accuracy": 0.10165500417351722,
|
|
"num_tokens": 63844.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 7.041130638122558,
|
|
"epoch": 0.03347280334728033,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 6.6492,
|
|
"mean_token_accuracy": 0.10998530834913253,
|
|
"num_tokens": 74331.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 7.070182371139526,
|
|
"epoch": 0.03765690376569038,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 6.6353,
|
|
"mean_token_accuracy": 0.10307188406586647,
|
|
"num_tokens": 83028.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 7.094835424423218,
|
|
"epoch": 0.04184100418410042,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 6.5676,
|
|
"mean_token_accuracy": 0.10705617740750313,
|
|
"num_tokens": 91395.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 7.037023973464966,
|
|
"epoch": 0.04602510460251046,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 6.4686,
|
|
"mean_token_accuracy": 0.11139860153198242,
|
|
"num_tokens": 101868.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 7.050775766372681,
|
|
"epoch": 0.0502092050209205,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 6.5234,
|
|
"mean_token_accuracy": 0.11568555235862732,
|
|
"num_tokens": 110403.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 7.024569654464722,
|
|
"epoch": 0.05439330543933055,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 6.52,
|
|
"mean_token_accuracy": 0.10725233480334281,
|
|
"num_tokens": 119557.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 7.0339525699615475,
|
|
"epoch": 0.058577405857740586,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 6.5532,
|
|
"mean_token_accuracy": 0.1088581494987011,
|
|
"num_tokens": 128903.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 7.077240371704102,
|
|
"epoch": 0.06276150627615062,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 6.5931,
|
|
"mean_token_accuracy": 0.10604406073689461,
|
|
"num_tokens": 137630.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 6.955969715118409,
|
|
"epoch": 0.06694560669456066,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 6.3729,
|
|
"mean_token_accuracy": 0.11078080907464027,
|
|
"num_tokens": 146335.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 6.9695604801177975,
|
|
"epoch": 0.07112970711297072,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 6.4896,
|
|
"mean_token_accuracy": 0.11115765944123268,
|
|
"num_tokens": 155766.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 6.9508363723754885,
|
|
"epoch": 0.07531380753138076,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 6.5297,
|
|
"mean_token_accuracy": 0.10771031454205512,
|
|
"num_tokens": 165844.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 6.9915656566619875,
|
|
"epoch": 0.0794979079497908,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 6.5913,
|
|
"mean_token_accuracy": 0.1018880434334278,
|
|
"num_tokens": 174858.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 7.071750593185425,
|
|
"epoch": 0.08368200836820083,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 6.5441,
|
|
"mean_token_accuracy": 0.11094660758972168,
|
|
"num_tokens": 184020.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 7.0119188785552975,
|
|
"epoch": 0.08786610878661087,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 6.4562,
|
|
"mean_token_accuracy": 0.11073266044259071,
|
|
"num_tokens": 193714.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 6.993580913543701,
|
|
"epoch": 0.09205020920502092,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 6.6389,
|
|
"mean_token_accuracy": 0.10732136517763138,
|
|
"num_tokens": 203139.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 7.036237573623657,
|
|
"epoch": 0.09623430962343096,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 6.5664,
|
|
"mean_token_accuracy": 0.10644949525594712,
|
|
"num_tokens": 212446.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 7.091158437728882,
|
|
"epoch": 0.100418410041841,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 6.5068,
|
|
"mean_token_accuracy": 0.11719691082835197,
|
|
"num_tokens": 220910.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 6.994235181808472,
|
|
"epoch": 0.10460251046025104,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 6.5415,
|
|
"mean_token_accuracy": 0.10888514444231986,
|
|
"num_tokens": 230522.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 7.030536556243897,
|
|
"epoch": 0.1087866108786611,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 6.5266,
|
|
"mean_token_accuracy": 0.11135743111371994,
|
|
"num_tokens": 240092.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 7.008419561386108,
|
|
"epoch": 0.11297071129707113,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 6.5977,
|
|
"mean_token_accuracy": 0.11229793429374695,
|
|
"num_tokens": 249478.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 6.962487030029297,
|
|
"epoch": 0.11715481171548117,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 6.4445,
|
|
"mean_token_accuracy": 0.12040354311466217,
|
|
"num_tokens": 257714.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 7.000783920288086,
|
|
"epoch": 0.12133891213389121,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 6.5263,
|
|
"mean_token_accuracy": 0.11713715940713883,
|
|
"num_tokens": 266569.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 6.985530710220337,
|
|
"epoch": 0.12552301255230125,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 6.4965,
|
|
"mean_token_accuracy": 0.11256701871752739,
|
|
"num_tokens": 276688.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 6.95616512298584,
|
|
"epoch": 0.1297071129707113,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 6.3899,
|
|
"mean_token_accuracy": 0.11239691302180291,
|
|
"num_tokens": 285430.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 6.88826355934143,
|
|
"epoch": 0.13389121338912133,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 6.5391,
|
|
"mean_token_accuracy": 0.10618084669113159,
|
|
"num_tokens": 295727.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 6.993566179275513,
|
|
"epoch": 0.13807531380753138,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 6.4894,
|
|
"mean_token_accuracy": 0.11518118530511856,
|
|
"num_tokens": 304854.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 7.000799751281738,
|
|
"epoch": 0.14225941422594143,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 6.5357,
|
|
"mean_token_accuracy": 0.11480183899402618,
|
|
"num_tokens": 313579.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 6.979376792907715,
|
|
"epoch": 0.14644351464435146,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 6.6224,
|
|
"mean_token_accuracy": 0.10865230411291123,
|
|
"num_tokens": 323310.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 7.12884750366211,
|
|
"epoch": 0.1506276150627615,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 6.4943,
|
|
"mean_token_accuracy": 0.1142081931233406,
|
|
"num_tokens": 331720.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 6.9554276943206785,
|
|
"epoch": 0.15481171548117154,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 6.4577,
|
|
"mean_token_accuracy": 0.11484430208802224,
|
|
"num_tokens": 340331.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 6.904936361312866,
|
|
"epoch": 0.1589958158995816,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 6.4996,
|
|
"mean_token_accuracy": 0.11663979142904282,
|
|
"num_tokens": 349677.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.076783514022827,
|
|
"epoch": 0.16317991631799164,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 6.4853,
|
|
"mean_token_accuracy": 0.10861474201083184,
|
|
"num_tokens": 358674.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 6.883452844619751,
|
|
"epoch": 0.16736401673640167,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 6.5506,
|
|
"mean_token_accuracy": 0.11058205664157868,
|
|
"num_tokens": 368258.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.004560375213623,
|
|
"epoch": 0.17154811715481172,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000102,
|
|
"loss": 6.4364,
|
|
"mean_token_accuracy": 0.1253327950835228,
|
|
"num_tokens": 377083.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.0509617805480955,
|
|
"epoch": 0.17573221757322174,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 6.5112,
|
|
"mean_token_accuracy": 0.10879353061318398,
|
|
"num_tokens": 385456.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 6.877067422866821,
|
|
"epoch": 0.1799163179916318,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000107,
|
|
"loss": 6.4228,
|
|
"mean_token_accuracy": 0.11805413141846657,
|
|
"num_tokens": 395295.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 6.991623020172119,
|
|
"epoch": 0.18410041841004185,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 6.4963,
|
|
"mean_token_accuracy": 0.1142610065639019,
|
|
"num_tokens": 405023.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 6.963863325119019,
|
|
"epoch": 0.18828451882845187,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 6.5422,
|
|
"mean_token_accuracy": 0.11503933966159821,
|
|
"num_tokens": 414224.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 6.947235012054444,
|
|
"epoch": 0.19246861924686193,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 6.416,
|
|
"mean_token_accuracy": 0.11018036603927613,
|
|
"num_tokens": 423951.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 6.950005388259887,
|
|
"epoch": 0.19665271966527198,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 6.4792,
|
|
"mean_token_accuracy": 0.10819886773824691,
|
|
"num_tokens": 432427.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 6.893841791152954,
|
|
"epoch": 0.200836820083682,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 6.412,
|
|
"mean_token_accuracy": 0.11521083116531372,
|
|
"num_tokens": 441560.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.0014019966125485,
|
|
"epoch": 0.20502092050209206,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000122,
|
|
"loss": 6.5156,
|
|
"mean_token_accuracy": 0.10784649103879929,
|
|
"num_tokens": 450364.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 6.933096218109131,
|
|
"epoch": 0.20920502092050208,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 6.4641,
|
|
"mean_token_accuracy": 0.11214551329612732,
|
|
"num_tokens": 460076.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 6.953990316390991,
|
|
"epoch": 0.21338912133891214,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 6.4217,
|
|
"mean_token_accuracy": 0.1185675323009491,
|
|
"num_tokens": 468718.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.013789224624634,
|
|
"epoch": 0.2175732217573222,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 6.5185,
|
|
"mean_token_accuracy": 0.1128389596939087,
|
|
"num_tokens": 478436.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 6.9821484088897705,
|
|
"epoch": 0.2217573221757322,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000132,
|
|
"loss": 6.5132,
|
|
"mean_token_accuracy": 0.10592028349637986,
|
|
"num_tokens": 486833.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 6.969932222366333,
|
|
"epoch": 0.22594142259414227,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 6.4085,
|
|
"mean_token_accuracy": 0.11758552938699722,
|
|
"num_tokens": 495119.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 6.981298875808716,
|
|
"epoch": 0.2301255230125523,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 6.5819,
|
|
"mean_token_accuracy": 0.1129891075193882,
|
|
"num_tokens": 505396.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 6.901115369796753,
|
|
"epoch": 0.23430962343096234,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 6.458,
|
|
"mean_token_accuracy": 0.11121175289154053,
|
|
"num_tokens": 515259.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.048445796966552,
|
|
"epoch": 0.2384937238493724,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 6.4627,
|
|
"mean_token_accuracy": 0.11158816590905189,
|
|
"num_tokens": 523608.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 6.87994794845581,
|
|
"epoch": 0.24267782426778242,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.4489,
|
|
"mean_token_accuracy": 0.11322836801409722,
|
|
"num_tokens": 534613.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 6.989324522018433,
|
|
"epoch": 0.24686192468619247,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 6.5155,
|
|
"mean_token_accuracy": 0.11460626497864723,
|
|
"num_tokens": 543347.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 6.947943115234375,
|
|
"epoch": 0.2510460251046025,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.4637,
|
|
"mean_token_accuracy": 0.11149389520287514,
|
|
"num_tokens": 552371.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 6.888587379455567,
|
|
"epoch": 0.25523012552301255,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.4233,
|
|
"mean_token_accuracy": 0.1118509091436863,
|
|
"num_tokens": 560742.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 6.991633749008178,
|
|
"epoch": 0.2594142259414226,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.549,
|
|
"mean_token_accuracy": 0.11592641770839691,
|
|
"num_tokens": 569960.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.021643543243409,
|
|
"epoch": 0.26359832635983266,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.4939,
|
|
"mean_token_accuracy": 0.11037873178720474,
|
|
"num_tokens": 578859.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 6.992051315307617,
|
|
"epoch": 0.26778242677824265,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.4976,
|
|
"mean_token_accuracy": 0.11440840438008308,
|
|
"num_tokens": 588146.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 6.80839729309082,
|
|
"epoch": 0.2719665271966527,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.4457,
|
|
"mean_token_accuracy": 0.10799639150500298,
|
|
"num_tokens": 597268.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 6.877410840988159,
|
|
"epoch": 0.27615062761506276,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.3857,
|
|
"mean_token_accuracy": 0.11793641448020935,
|
|
"num_tokens": 606199.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.094079399108887,
|
|
"epoch": 0.2803347280334728,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.5184,
|
|
"mean_token_accuracy": 0.10466537326574325,
|
|
"num_tokens": 615966.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 6.915050363540649,
|
|
"epoch": 0.28451882845188287,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.4998,
|
|
"mean_token_accuracy": 0.11134347319602966,
|
|
"num_tokens": 625689.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 6.929769897460938,
|
|
"epoch": 0.28870292887029286,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.528,
|
|
"mean_token_accuracy": 0.10973049625754357,
|
|
"num_tokens": 634877.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.018622636795044,
|
|
"epoch": 0.2928870292887029,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.4516,
|
|
"mean_token_accuracy": 0.11552532091736793,
|
|
"num_tokens": 644042.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 6.833831834793091,
|
|
"epoch": 0.29707112970711297,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.4695,
|
|
"mean_token_accuracy": 0.1079812467098236,
|
|
"num_tokens": 653072.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 6.952274990081787,
|
|
"epoch": 0.301255230125523,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.4749,
|
|
"mean_token_accuracy": 0.11333532109856606,
|
|
"num_tokens": 663686.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 6.91602087020874,
|
|
"epoch": 0.3054393305439331,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.4151,
|
|
"mean_token_accuracy": 0.11366223692893981,
|
|
"num_tokens": 672293.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 6.774839878082275,
|
|
"epoch": 0.30962343096234307,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.3793,
|
|
"mean_token_accuracy": 0.11558515504002571,
|
|
"num_tokens": 681758.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 6.897351408004761,
|
|
"epoch": 0.3138075313807531,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.4087,
|
|
"mean_token_accuracy": 0.1156994417309761,
|
|
"num_tokens": 690727.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 6.964544582366943,
|
|
"epoch": 0.3179916317991632,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.3579,
|
|
"mean_token_accuracy": 0.11470428928732872,
|
|
"num_tokens": 701210.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 6.701919221878052,
|
|
"epoch": 0.32217573221757323,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.3836,
|
|
"mean_token_accuracy": 0.11825001761317253,
|
|
"num_tokens": 710869.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 6.931270742416382,
|
|
"epoch": 0.3263598326359833,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.503,
|
|
"mean_token_accuracy": 0.10896454304456711,
|
|
"num_tokens": 721455.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 6.858742380142212,
|
|
"epoch": 0.3305439330543933,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.3847,
|
|
"mean_token_accuracy": 0.11934179365634918,
|
|
"num_tokens": 730917.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 6.929948425292968,
|
|
"epoch": 0.33472803347280333,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.4168,
|
|
"mean_token_accuracy": 0.1181195355951786,
|
|
"num_tokens": 739302.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 6.811960315704345,
|
|
"epoch": 0.3389121338912134,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.3778,
|
|
"mean_token_accuracy": 0.11517146825790406,
|
|
"num_tokens": 748785.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 6.896191072463989,
|
|
"epoch": 0.34309623430962344,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.3118,
|
|
"mean_token_accuracy": 0.12097605615854264,
|
|
"num_tokens": 757215.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 6.799113607406616,
|
|
"epoch": 0.3472803347280335,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.2572,
|
|
"mean_token_accuracy": 0.12601873874664307,
|
|
"num_tokens": 765682.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 6.882309913635254,
|
|
"epoch": 0.3514644351464435,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.4683,
|
|
"mean_token_accuracy": 0.11330038234591484,
|
|
"num_tokens": 774814.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 6.807965230941773,
|
|
"epoch": 0.35564853556485354,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.3905,
|
|
"mean_token_accuracy": 0.1169172041118145,
|
|
"num_tokens": 784019.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 6.758611345291138,
|
|
"epoch": 0.3598326359832636,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.2659,
|
|
"mean_token_accuracy": 0.12330161556601524,
|
|
"num_tokens": 793237.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 6.809417343139648,
|
|
"epoch": 0.36401673640167365,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.3273,
|
|
"mean_token_accuracy": 0.11897384449839592,
|
|
"num_tokens": 801548.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 6.806345653533936,
|
|
"epoch": 0.3682008368200837,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.3845,
|
|
"mean_token_accuracy": 0.12210910469293594,
|
|
"num_tokens": 810974.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 6.756527662277222,
|
|
"epoch": 0.3723849372384937,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.2695,
|
|
"mean_token_accuracy": 0.12554480582475663,
|
|
"num_tokens": 820023.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 6.874709272384644,
|
|
"epoch": 0.37656903765690375,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.3588,
|
|
"mean_token_accuracy": 0.12106196507811547,
|
|
"num_tokens": 828297.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 6.802942419052124,
|
|
"epoch": 0.3807531380753138,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.4073,
|
|
"mean_token_accuracy": 0.11363389864563941,
|
|
"num_tokens": 838268.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 6.882838296890259,
|
|
"epoch": 0.38493723849372385,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.3277,
|
|
"mean_token_accuracy": 0.1192656472325325,
|
|
"num_tokens": 846491.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 6.7082360744476315,
|
|
"epoch": 0.3891213389121339,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.2891,
|
|
"mean_token_accuracy": 0.12571795657277107,
|
|
"num_tokens": 856501.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 6.777575159072876,
|
|
"epoch": 0.39330543933054396,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.334,
|
|
"mean_token_accuracy": 0.11964721754193305,
|
|
"num_tokens": 867054.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 6.719767951965332,
|
|
"epoch": 0.39748953974895396,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.367,
|
|
"mean_token_accuracy": 0.12227437198162079,
|
|
"num_tokens": 875956.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 6.690861368179322,
|
|
"epoch": 0.401673640167364,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.2799,
|
|
"mean_token_accuracy": 0.12380229756236076,
|
|
"num_tokens": 884427.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 6.913175106048584,
|
|
"epoch": 0.40585774058577406,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.3865,
|
|
"mean_token_accuracy": 0.11986539214849472,
|
|
"num_tokens": 894181.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 6.629990673065185,
|
|
"epoch": 0.4100418410041841,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.2187,
|
|
"mean_token_accuracy": 0.12581772431731225,
|
|
"num_tokens": 904076.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 6.789698457717895,
|
|
"epoch": 0.41422594142259417,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.3516,
|
|
"mean_token_accuracy": 0.12482998818159104,
|
|
"num_tokens": 913915.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 6.840354251861572,
|
|
"epoch": 0.41841004184100417,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.4418,
|
|
"mean_token_accuracy": 0.11700730696320534,
|
|
"num_tokens": 922876.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.41841004184100417,
|
|
"eval_entropy": 6.618946797033517,
|
|
"eval_loss": 6.314803600311279,
|
|
"eval_mean_token_accuracy": 0.12852142196613342,
|
|
"eval_num_tokens": 922876.0,
|
|
"eval_runtime": 2.1866,
|
|
"eval_samples_per_second": 1786.781,
|
|
"eval_steps_per_second": 223.633,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 6.682973623275757,
|
|
"epoch": 0.4225941422594142,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.2227,
|
|
"mean_token_accuracy": 0.13042852878570557,
|
|
"num_tokens": 931998.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 6.769066524505615,
|
|
"epoch": 0.42677824267782427,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.3005,
|
|
"mean_token_accuracy": 0.12563085854053496,
|
|
"num_tokens": 941387.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 6.687266302108765,
|
|
"epoch": 0.4309623430962343,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.3447,
|
|
"mean_token_accuracy": 0.11680261939764022,
|
|
"num_tokens": 951866.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 6.677826023101806,
|
|
"epoch": 0.4351464435146444,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.2818,
|
|
"mean_token_accuracy": 0.11924662739038468,
|
|
"num_tokens": 961252.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 6.823406028747558,
|
|
"epoch": 0.4393305439330544,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.28,
|
|
"mean_token_accuracy": 0.12683919668197632,
|
|
"num_tokens": 969857.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.668636322021484,
|
|
"epoch": 0.4435146443514644,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.2752,
|
|
"mean_token_accuracy": 0.12249023914337158,
|
|
"num_tokens": 978754.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 6.767824792861939,
|
|
"epoch": 0.4476987447698745,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.2675,
|
|
"mean_token_accuracy": 0.12591633051633835,
|
|
"num_tokens": 987935.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.742959976196289,
|
|
"epoch": 0.45188284518828453,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.2393,
|
|
"mean_token_accuracy": 0.1245250605046749,
|
|
"num_tokens": 996873.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.697275876998901,
|
|
"epoch": 0.4560669456066946,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.3406,
|
|
"mean_token_accuracy": 0.12331385985016823,
|
|
"num_tokens": 1006487.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.597655963897705,
|
|
"epoch": 0.4602510460251046,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.1744,
|
|
"mean_token_accuracy": 0.13314277157187462,
|
|
"num_tokens": 1015715.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.776229286193848,
|
|
"epoch": 0.46443514644351463,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.2913,
|
|
"mean_token_accuracy": 0.12320854589343071,
|
|
"num_tokens": 1025091.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.589776659011841,
|
|
"epoch": 0.4686192468619247,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.2137,
|
|
"mean_token_accuracy": 0.12773831263184549,
|
|
"num_tokens": 1034174.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.811848163604736,
|
|
"epoch": 0.47280334728033474,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.423,
|
|
"mean_token_accuracy": 0.1170275293290615,
|
|
"num_tokens": 1043777.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.752657032012939,
|
|
"epoch": 0.4769874476987448,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.2635,
|
|
"mean_token_accuracy": 0.12874307632446289,
|
|
"num_tokens": 1052749.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.689546823501587,
|
|
"epoch": 0.4811715481171548,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.2866,
|
|
"mean_token_accuracy": 0.12773908227682113,
|
|
"num_tokens": 1061564.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.665796375274658,
|
|
"epoch": 0.48535564853556484,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.2014,
|
|
"mean_token_accuracy": 0.12742372676730157,
|
|
"num_tokens": 1070006.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.594886112213135,
|
|
"epoch": 0.4895397489539749,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.2688,
|
|
"mean_token_accuracy": 0.12371853217482567,
|
|
"num_tokens": 1079799.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.712746620178223,
|
|
"epoch": 0.49372384937238495,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.1238,
|
|
"mean_token_accuracy": 0.13331872522830962,
|
|
"num_tokens": 1088842.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.563669633865357,
|
|
"epoch": 0.497907949790795,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.1979,
|
|
"mean_token_accuracy": 0.13366666212677955,
|
|
"num_tokens": 1097960.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.528041791915894,
|
|
"epoch": 0.502092050209205,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.1845,
|
|
"mean_token_accuracy": 0.12403582334518433,
|
|
"num_tokens": 1106797.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.721166563034058,
|
|
"epoch": 0.5062761506276151,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.3305,
|
|
"mean_token_accuracy": 0.13074323609471322,
|
|
"num_tokens": 1116384.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.504988336563111,
|
|
"epoch": 0.5104602510460251,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.0877,
|
|
"mean_token_accuracy": 0.13214512541890144,
|
|
"num_tokens": 1125206.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.758549690246582,
|
|
"epoch": 0.5146443514644351,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.3265,
|
|
"mean_token_accuracy": 0.12822075560688972,
|
|
"num_tokens": 1134110.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.7619537830352785,
|
|
"epoch": 0.5188284518828452,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.3286,
|
|
"mean_token_accuracy": 0.12627065256237985,
|
|
"num_tokens": 1142833.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.582308912277222,
|
|
"epoch": 0.5230125523012552,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.2358,
|
|
"mean_token_accuracy": 0.12966165021061898,
|
|
"num_tokens": 1152006.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.660427665710449,
|
|
"epoch": 0.5271966527196653,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.1786,
|
|
"mean_token_accuracy": 0.12617679536342621,
|
|
"num_tokens": 1161061.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.5836528778076175,
|
|
"epoch": 0.5313807531380753,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.2082,
|
|
"mean_token_accuracy": 0.12547155022621154,
|
|
"num_tokens": 1170754.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.559231424331665,
|
|
"epoch": 0.5355648535564853,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.1506,
|
|
"mean_token_accuracy": 0.12670147866010667,
|
|
"num_tokens": 1180854.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.6272156715393065,
|
|
"epoch": 0.5397489539748954,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.2381,
|
|
"mean_token_accuracy": 0.12543641924858093,
|
|
"num_tokens": 1190567.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.5191137313842775,
|
|
"epoch": 0.5439330543933054,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.0974,
|
|
"mean_token_accuracy": 0.13144202902913094,
|
|
"num_tokens": 1200247.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.558930778503418,
|
|
"epoch": 0.5481171548117155,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.1947,
|
|
"mean_token_accuracy": 0.12886450961232185,
|
|
"num_tokens": 1209358.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.531280374526977,
|
|
"epoch": 0.5523012552301255,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.1911,
|
|
"mean_token_accuracy": 0.12920715510845185,
|
|
"num_tokens": 1217746.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.517261266708374,
|
|
"epoch": 0.5564853556485355,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.1718,
|
|
"mean_token_accuracy": 0.13568612262606622,
|
|
"num_tokens": 1226829.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.5621202945709225,
|
|
"epoch": 0.5606694560669456,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.1113,
|
|
"mean_token_accuracy": 0.13396792039275168,
|
|
"num_tokens": 1236642.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.650665664672852,
|
|
"epoch": 0.5648535564853556,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.2144,
|
|
"mean_token_accuracy": 0.13539021015167235,
|
|
"num_tokens": 1244934.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.413178253173828,
|
|
"epoch": 0.5690376569037657,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.1208,
|
|
"mean_token_accuracy": 0.12733489498496056,
|
|
"num_tokens": 1254786.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.518020915985107,
|
|
"epoch": 0.5732217573221757,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.0871,
|
|
"mean_token_accuracy": 0.13411632925271988,
|
|
"num_tokens": 1263745.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.496325349807739,
|
|
"epoch": 0.5774058577405857,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.2158,
|
|
"mean_token_accuracy": 0.13778234273195267,
|
|
"num_tokens": 1273101.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.605466699600219,
|
|
"epoch": 0.5815899581589958,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.1943,
|
|
"mean_token_accuracy": 0.12910575792193413,
|
|
"num_tokens": 1282547.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.562770938873291,
|
|
"epoch": 0.5857740585774058,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.2116,
|
|
"mean_token_accuracy": 0.12486919164657592,
|
|
"num_tokens": 1291594.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.553465557098389,
|
|
"epoch": 0.5899581589958159,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.1304,
|
|
"mean_token_accuracy": 0.13733726590871811,
|
|
"num_tokens": 1300282.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.544411516189575,
|
|
"epoch": 0.5941422594142259,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.2256,
|
|
"mean_token_accuracy": 0.13409770727157594,
|
|
"num_tokens": 1309304.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.485507249832153,
|
|
"epoch": 0.5983263598326359,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.0549,
|
|
"mean_token_accuracy": 0.13027888983488084,
|
|
"num_tokens": 1317638.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.652093839645386,
|
|
"epoch": 0.602510460251046,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.2625,
|
|
"mean_token_accuracy": 0.12712223082780838,
|
|
"num_tokens": 1327242.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.497774410247803,
|
|
"epoch": 0.606694560669456,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.0806,
|
|
"mean_token_accuracy": 0.13213423565030097,
|
|
"num_tokens": 1335880.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.538639783859253,
|
|
"epoch": 0.6108786610878661,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.1639,
|
|
"mean_token_accuracy": 0.13274887576699257,
|
|
"num_tokens": 1344909.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.449158382415772,
|
|
"epoch": 0.6150627615062761,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.0274,
|
|
"mean_token_accuracy": 0.13631289303302765,
|
|
"num_tokens": 1353997.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.492458295822144,
|
|
"epoch": 0.6192468619246861,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.1759,
|
|
"mean_token_accuracy": 0.13163307160139084,
|
|
"num_tokens": 1363293.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.656571531295777,
|
|
"epoch": 0.6234309623430963,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.2029,
|
|
"mean_token_accuracy": 0.12882030829787255,
|
|
"num_tokens": 1372174.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.505940628051758,
|
|
"epoch": 0.6276150627615062,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.0966,
|
|
"mean_token_accuracy": 0.13372141867876053,
|
|
"num_tokens": 1381067.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.428654146194458,
|
|
"epoch": 0.6317991631799164,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.0915,
|
|
"mean_token_accuracy": 0.13776354119181633,
|
|
"num_tokens": 1390072.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.486198759078979,
|
|
"epoch": 0.6359832635983264,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.0727,
|
|
"mean_token_accuracy": 0.1386815182864666,
|
|
"num_tokens": 1399391.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.4113274097442625,
|
|
"epoch": 0.6401673640167364,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.0898,
|
|
"mean_token_accuracy": 0.13849154487252235,
|
|
"num_tokens": 1408054.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.516796731948853,
|
|
"epoch": 0.6443514644351465,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.0996,
|
|
"mean_token_accuracy": 0.12808119207620622,
|
|
"num_tokens": 1417406.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.449129199981689,
|
|
"epoch": 0.6485355648535565,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.1552,
|
|
"mean_token_accuracy": 0.1301002584397793,
|
|
"num_tokens": 1426907.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.505101203918457,
|
|
"epoch": 0.6527196652719666,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.2124,
|
|
"mean_token_accuracy": 0.12579995691776275,
|
|
"num_tokens": 1436085.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.546794652938843,
|
|
"epoch": 0.6569037656903766,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.0964,
|
|
"mean_token_accuracy": 0.1327142260968685,
|
|
"num_tokens": 1444823.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.4318641185760494,
|
|
"epoch": 0.6610878661087866,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.1598,
|
|
"mean_token_accuracy": 0.12898237407207488,
|
|
"num_tokens": 1454621.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.4235259056091305,
|
|
"epoch": 0.6652719665271967,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.0317,
|
|
"mean_token_accuracy": 0.13903406038880348,
|
|
"num_tokens": 1463784.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.446871900558472,
|
|
"epoch": 0.6694560669456067,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.1562,
|
|
"mean_token_accuracy": 0.1328710287809372,
|
|
"num_tokens": 1473257.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.524528694152832,
|
|
"epoch": 0.6736401673640168,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.0924,
|
|
"mean_token_accuracy": 0.13512965515255929,
|
|
"num_tokens": 1482195.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.354287576675415,
|
|
"epoch": 0.6778242677824268,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.138,
|
|
"mean_token_accuracy": 0.13590989336371423,
|
|
"num_tokens": 1491590.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.499030351638794,
|
|
"epoch": 0.6820083682008368,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.0045,
|
|
"mean_token_accuracy": 0.14405518025159836,
|
|
"num_tokens": 1500208.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.386388921737671,
|
|
"epoch": 0.6861924686192469,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.1513,
|
|
"mean_token_accuracy": 0.137047478556633,
|
|
"num_tokens": 1509412.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.4775557041168215,
|
|
"epoch": 0.6903765690376569,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.1438,
|
|
"mean_token_accuracy": 0.13165075778961183,
|
|
"num_tokens": 1518426.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.589972829818725,
|
|
"epoch": 0.694560669456067,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.0999,
|
|
"mean_token_accuracy": 0.13527765721082688,
|
|
"num_tokens": 1528605.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.304672431945801,
|
|
"epoch": 0.698744769874477,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.0577,
|
|
"mean_token_accuracy": 0.13871931731700898,
|
|
"num_tokens": 1537449.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.394627809524536,
|
|
"epoch": 0.702928870292887,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.0916,
|
|
"mean_token_accuracy": 0.1379619151353836,
|
|
"num_tokens": 1546743.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.4108617305755615,
|
|
"epoch": 0.7071129707112971,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.0687,
|
|
"mean_token_accuracy": 0.13178422152996064,
|
|
"num_tokens": 1555691.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.351869916915893,
|
|
"epoch": 0.7112970711297071,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.0234,
|
|
"mean_token_accuracy": 0.1407901257276535,
|
|
"num_tokens": 1564957.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.479067134857178,
|
|
"epoch": 0.7154811715481172,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.094,
|
|
"mean_token_accuracy": 0.13800255656242372,
|
|
"num_tokens": 1573901.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.325543594360352,
|
|
"epoch": 0.7196652719665272,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.0758,
|
|
"mean_token_accuracy": 0.1410400703549385,
|
|
"num_tokens": 1582644.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.465507459640503,
|
|
"epoch": 0.7238493723849372,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.2008,
|
|
"mean_token_accuracy": 0.13150226175785065,
|
|
"num_tokens": 1592363.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.515338897705078,
|
|
"epoch": 0.7280334728033473,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.1538,
|
|
"mean_token_accuracy": 0.13219332993030547,
|
|
"num_tokens": 1602012.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.322141218185425,
|
|
"epoch": 0.7322175732217573,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.0376,
|
|
"mean_token_accuracy": 0.138260093331337,
|
|
"num_tokens": 1611625.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.285467910766601,
|
|
"epoch": 0.7364016736401674,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 5.9981,
|
|
"mean_token_accuracy": 0.1322825863957405,
|
|
"num_tokens": 1620941.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.46323676109314,
|
|
"epoch": 0.7405857740585774,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.0956,
|
|
"mean_token_accuracy": 0.13529526591300964,
|
|
"num_tokens": 1629225.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.389843606948853,
|
|
"epoch": 0.7447698744769874,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.0742,
|
|
"mean_token_accuracy": 0.12740020379424094,
|
|
"num_tokens": 1639551.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.285019779205323,
|
|
"epoch": 0.7489539748953975,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.0591,
|
|
"mean_token_accuracy": 0.12816891968250274,
|
|
"num_tokens": 1648963.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.335085105895996,
|
|
"epoch": 0.7531380753138075,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 5.982,
|
|
"mean_token_accuracy": 0.14409040734171868,
|
|
"num_tokens": 1657735.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.48310923576355,
|
|
"epoch": 0.7573221757322176,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.019,
|
|
"mean_token_accuracy": 0.1408952869474888,
|
|
"num_tokens": 1666349.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.214490985870361,
|
|
"epoch": 0.7615062761506276,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 5.9505,
|
|
"mean_token_accuracy": 0.14200798720121383,
|
|
"num_tokens": 1675920.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.253348970413208,
|
|
"epoch": 0.7656903765690377,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 5.9967,
|
|
"mean_token_accuracy": 0.13502266258001328,
|
|
"num_tokens": 1685626.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.4179590225219725,
|
|
"epoch": 0.7698744769874477,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.1091,
|
|
"mean_token_accuracy": 0.14264659583568573,
|
|
"num_tokens": 1694972.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.3434624671936035,
|
|
"epoch": 0.7740585774058577,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.02,
|
|
"mean_token_accuracy": 0.14580936431884767,
|
|
"num_tokens": 1703806.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.327855443954467,
|
|
"epoch": 0.7782426778242678,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.1053,
|
|
"mean_token_accuracy": 0.13719277679920197,
|
|
"num_tokens": 1712949.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.523324298858642,
|
|
"epoch": 0.7824267782426778,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.2208,
|
|
"mean_token_accuracy": 0.131364194303751,
|
|
"num_tokens": 1722328.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.450496101379395,
|
|
"epoch": 0.7866108786610879,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.1798,
|
|
"mean_token_accuracy": 0.13524089604616166,
|
|
"num_tokens": 1731880.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.227896404266358,
|
|
"epoch": 0.7907949790794979,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.006,
|
|
"mean_token_accuracy": 0.1443274959921837,
|
|
"num_tokens": 1740719.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.363286638259888,
|
|
"epoch": 0.7949790794979079,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.0532,
|
|
"mean_token_accuracy": 0.13592690229415894,
|
|
"num_tokens": 1749753.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.400108718872071,
|
|
"epoch": 0.799163179916318,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.0528,
|
|
"mean_token_accuracy": 0.13351312130689622,
|
|
"num_tokens": 1758913.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.398914623260498,
|
|
"epoch": 0.803347280334728,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.069,
|
|
"mean_token_accuracy": 0.13728829100728035,
|
|
"num_tokens": 1768536.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.2407642841339115,
|
|
"epoch": 0.8075313807531381,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.0143,
|
|
"mean_token_accuracy": 0.13496991991996765,
|
|
"num_tokens": 1777291.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.370160341262817,
|
|
"epoch": 0.8117154811715481,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.0995,
|
|
"mean_token_accuracy": 0.13727209270000457,
|
|
"num_tokens": 1787186.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.300974369049072,
|
|
"epoch": 0.8158995815899581,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 5.8905,
|
|
"mean_token_accuracy": 0.14629269912838935,
|
|
"num_tokens": 1795581.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.177432060241699,
|
|
"epoch": 0.8200836820083682,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 5.906,
|
|
"mean_token_accuracy": 0.1430325910449028,
|
|
"num_tokens": 1804627.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.329653215408325,
|
|
"epoch": 0.8242677824267782,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.0246,
|
|
"mean_token_accuracy": 0.14240092262625695,
|
|
"num_tokens": 1813808.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.253896236419678,
|
|
"epoch": 0.8284518828451883,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.0306,
|
|
"mean_token_accuracy": 0.1399620369076729,
|
|
"num_tokens": 1823422.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.424421644210815,
|
|
"epoch": 0.8326359832635983,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.0787,
|
|
"mean_token_accuracy": 0.1377182833850384,
|
|
"num_tokens": 1832437.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.196931314468384,
|
|
"epoch": 0.8368200836820083,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.0054,
|
|
"mean_token_accuracy": 0.13808162435889243,
|
|
"num_tokens": 1842063.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.8368200836820083,
|
|
"eval_entropy": 6.193909905439505,
|
|
"eval_loss": 6.01716423034668,
|
|
"eval_mean_token_accuracy": 0.14855122578412233,
|
|
"eval_num_tokens": 1842063.0,
|
|
"eval_runtime": 2.313,
|
|
"eval_samples_per_second": 1689.16,
|
|
"eval_steps_per_second": 211.415,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.393909978866577,
|
|
"epoch": 0.8410041841004184,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004999998518355666,
|
|
"loss": 6.1452,
|
|
"mean_token_accuracy": 0.1390266939997673,
|
|
"num_tokens": 1851043.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.254569721221924,
|
|
"epoch": 0.8451882845188284,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999992499178905,
|
|
"loss": 5.985,
|
|
"mean_token_accuracy": 0.14467152059078217,
|
|
"num_tokens": 1860268.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.337277507781982,
|
|
"epoch": 0.8493723849372385,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999981849879322,
|
|
"loss": 6.0965,
|
|
"mean_token_accuracy": 0.14228640869259834,
|
|
"num_tokens": 1870096.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.264781332015991,
|
|
"epoch": 0.8535564853556485,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999966570478833,
|
|
"loss": 5.9941,
|
|
"mean_token_accuracy": 0.14099601581692695,
|
|
"num_tokens": 1879164.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.268557405471801,
|
|
"epoch": 0.8577405857740585,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004999946661008878,
|
|
"loss": 5.932,
|
|
"mean_token_accuracy": 0.14190083593130112,
|
|
"num_tokens": 1888135.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.211926364898682,
|
|
"epoch": 0.8619246861924686,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000499992212151043,
|
|
"loss": 5.9999,
|
|
"mean_token_accuracy": 0.13821832686662675,
|
|
"num_tokens": 1898302.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.190331029891968,
|
|
"epoch": 0.8661087866108786,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004999892952033987,
|
|
"loss": 5.97,
|
|
"mean_token_accuracy": 0.140797733515501,
|
|
"num_tokens": 1907753.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.310067653656006,
|
|
"epoch": 0.8702928870292888,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004999859152639573,
|
|
"loss": 6.0217,
|
|
"mean_token_accuracy": 0.140376216173172,
|
|
"num_tokens": 1917543.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.327867412567139,
|
|
"epoch": 0.8744769874476988,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004999820723396745,
|
|
"loss": 6.0649,
|
|
"mean_token_accuracy": 0.13445087969303132,
|
|
"num_tokens": 1927335.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.262939023971557,
|
|
"epoch": 0.8786610878661087,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999777664384581,
|
|
"loss": 5.9872,
|
|
"mean_token_accuracy": 0.13672856613993645,
|
|
"num_tokens": 1936381.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.237847423553466,
|
|
"epoch": 0.8828451882845189,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004999729975691692,
|
|
"loss": 6.0093,
|
|
"mean_token_accuracy": 0.1376845121383667,
|
|
"num_tokens": 1945559.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.2264868259429935,
|
|
"epoch": 0.8870292887029289,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999677657416212,
|
|
"loss": 5.8957,
|
|
"mean_token_accuracy": 0.13779111057519913,
|
|
"num_tokens": 1954709.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.2650247573852536,
|
|
"epoch": 0.891213389121339,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004999620709665804,
|
|
"loss": 6.0121,
|
|
"mean_token_accuracy": 0.13971133083105086,
|
|
"num_tokens": 1963664.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.26368670463562,
|
|
"epoch": 0.895397489539749,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999559132557659,
|
|
"loss": 6.0741,
|
|
"mean_token_accuracy": 0.13506176695227623,
|
|
"num_tokens": 1974103.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.398173427581787,
|
|
"epoch": 0.899581589958159,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004999492926218489,
|
|
"loss": 5.9456,
|
|
"mean_token_accuracy": 0.14405092597007751,
|
|
"num_tokens": 1982924.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.110432910919189,
|
|
"epoch": 0.9037656903765691,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000499942209078454,
|
|
"loss": 5.8836,
|
|
"mean_token_accuracy": 0.15043322890996932,
|
|
"num_tokens": 1992033.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.133646059036255,
|
|
"epoch": 0.9079497907949791,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999346626401578,
|
|
"loss": 5.8933,
|
|
"mean_token_accuracy": 0.143799290060997,
|
|
"num_tokens": 2001218.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.192528152465821,
|
|
"epoch": 0.9121338912133892,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999266533224897,
|
|
"loss": 6.0359,
|
|
"mean_token_accuracy": 0.13878616839647293,
|
|
"num_tokens": 2010463.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.255569648742676,
|
|
"epoch": 0.9163179916317992,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004999181811419314,
|
|
"loss": 5.9022,
|
|
"mean_token_accuracy": 0.14577107951045037,
|
|
"num_tokens": 2019246.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.291269254684448,
|
|
"epoch": 0.9205020920502092,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004999092461159177,
|
|
"loss": 5.9875,
|
|
"mean_token_accuracy": 0.14506771862506868,
|
|
"num_tokens": 2028083.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.137917995452881,
|
|
"epoch": 0.9246861924686193,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004998998482628351,
|
|
"loss": 5.8047,
|
|
"mean_token_accuracy": 0.14625456184148788,
|
|
"num_tokens": 2036631.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.18209228515625,
|
|
"epoch": 0.9288702928870293,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004998899876020229,
|
|
"loss": 5.9364,
|
|
"mean_token_accuracy": 0.14081862568855286,
|
|
"num_tokens": 2045910.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.169568586349487,
|
|
"epoch": 0.9330543933054394,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000499879664153773,
|
|
"loss": 5.9853,
|
|
"mean_token_accuracy": 0.1488099366426468,
|
|
"num_tokens": 2055959.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.302842617034912,
|
|
"epoch": 0.9372384937238494,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004998688779393291,
|
|
"loss": 6.0455,
|
|
"mean_token_accuracy": 0.13892305865883828,
|
|
"num_tokens": 2065475.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.048355484008789,
|
|
"epoch": 0.9414225941422594,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004998576289808877,
|
|
"loss": 5.8223,
|
|
"mean_token_accuracy": 0.15155590325593948,
|
|
"num_tokens": 2074469.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.176668310165406,
|
|
"epoch": 0.9456066945606695,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004998459173015972,
|
|
"loss": 5.9985,
|
|
"mean_token_accuracy": 0.14467793405056,
|
|
"num_tokens": 2084604.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.2671363830566404,
|
|
"epoch": 0.9497907949790795,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004998337429255584,
|
|
"loss": 5.9624,
|
|
"mean_token_accuracy": 0.14496439844369888,
|
|
"num_tokens": 2093377.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.193261003494262,
|
|
"epoch": 0.9539748953974896,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998211058778243,
|
|
"loss": 5.9905,
|
|
"mean_token_accuracy": 0.1397103875875473,
|
|
"num_tokens": 2102798.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.262817668914795,
|
|
"epoch": 0.9581589958158996,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004998080061843999,
|
|
"loss": 5.9677,
|
|
"mean_token_accuracy": 0.1436394140124321,
|
|
"num_tokens": 2112516.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.216636085510254,
|
|
"epoch": 0.9623430962343096,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000499794443872242,
|
|
"loss": 6.0514,
|
|
"mean_token_accuracy": 0.1393752045929432,
|
|
"num_tokens": 2121677.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.159835290908814,
|
|
"epoch": 0.9665271966527197,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004997804189692601,
|
|
"loss": 5.8933,
|
|
"mean_token_accuracy": 0.1472679480910301,
|
|
"num_tokens": 2131220.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.236383438110352,
|
|
"epoch": 0.9707112970711297,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004997659315043149,
|
|
"loss": 6.0393,
|
|
"mean_token_accuracy": 0.13970012813806534,
|
|
"num_tokens": 2142104.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.109493589401245,
|
|
"epoch": 0.9748953974895398,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004997509815072191,
|
|
"loss": 5.9407,
|
|
"mean_token_accuracy": 0.14186370223760605,
|
|
"num_tokens": 2151970.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.159468078613282,
|
|
"epoch": 0.9790794979079498,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004997355690087379,
|
|
"loss": 5.9008,
|
|
"mean_token_accuracy": 0.14271292239427566,
|
|
"num_tokens": 2161371.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.079864406585694,
|
|
"epoch": 0.9832635983263598,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004997196940405874,
|
|
"loss": 5.8732,
|
|
"mean_token_accuracy": 0.14374568536877633,
|
|
"num_tokens": 2170650.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.1532872200012205,
|
|
"epoch": 0.9874476987447699,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004997033566354358,
|
|
"loss": 5.935,
|
|
"mean_token_accuracy": 0.146008437871933,
|
|
"num_tokens": 2180934.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.1988222122192385,
|
|
"epoch": 0.9916317991631799,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004996865568269028,
|
|
"loss": 5.8861,
|
|
"mean_token_accuracy": 0.14965563416481018,
|
|
"num_tokens": 2189894.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.1448510646820065,
|
|
"epoch": 0.99581589958159,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004996692946495597,
|
|
"loss": 5.8432,
|
|
"mean_token_accuracy": 0.1574077218770981,
|
|
"num_tokens": 2200208.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.169542074203491,
|
|
"epoch": 1.0,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004996515701389295,
|
|
"loss": 5.8976,
|
|
"mean_token_accuracy": 0.14283206239342688,
|
|
"num_tokens": 2209322.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.058703136444092,
|
|
"epoch": 1.00418410041841,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004996333833314862,
|
|
"loss": 5.7052,
|
|
"mean_token_accuracy": 0.15611663460731506,
|
|
"num_tokens": 2217663.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.040194034576416,
|
|
"epoch": 1.00836820083682,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004996147342646554,
|
|
"loss": 5.7459,
|
|
"mean_token_accuracy": 0.15023762136697769,
|
|
"num_tokens": 2226559.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.2119731426239015,
|
|
"epoch": 1.0125523012552302,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004995956229768137,
|
|
"loss": 5.7811,
|
|
"mean_token_accuracy": 0.15604271739721298,
|
|
"num_tokens": 2235468.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.019049549102784,
|
|
"epoch": 1.0167364016736402,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004995760495072893,
|
|
"loss": 5.6162,
|
|
"mean_token_accuracy": 0.16762171536684037,
|
|
"num_tokens": 2245176.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.101865291595459,
|
|
"epoch": 1.0209205020920502,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004995560138963611,
|
|
"loss": 5.7626,
|
|
"mean_token_accuracy": 0.15765286684036256,
|
|
"num_tokens": 2255123.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.09725546836853,
|
|
"epoch": 1.0251046025104602,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004995355161852593,
|
|
"loss": 5.6884,
|
|
"mean_token_accuracy": 0.15597244650125502,
|
|
"num_tokens": 2263596.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.045264720916748,
|
|
"epoch": 1.0292887029288702,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004995145564161648,
|
|
"loss": 5.7259,
|
|
"mean_token_accuracy": 0.15497453063726424,
|
|
"num_tokens": 2272228.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.087162351608276,
|
|
"epoch": 1.0334728033472804,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004994931346322094,
|
|
"loss": 5.7666,
|
|
"mean_token_accuracy": 0.1547314152121544,
|
|
"num_tokens": 2281248.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.132814931869507,
|
|
"epoch": 1.0376569037656904,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004994712508774757,
|
|
"loss": 5.7376,
|
|
"mean_token_accuracy": 0.15098585486412047,
|
|
"num_tokens": 2290958.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.0064124584198,
|
|
"epoch": 1.0418410041841004,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000499448905196997,
|
|
"loss": 5.674,
|
|
"mean_token_accuracy": 0.1653773456811905,
|
|
"num_tokens": 2299763.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.097362232208252,
|
|
"epoch": 1.0460251046025104,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004994260976367573,
|
|
"loss": 5.688,
|
|
"mean_token_accuracy": 0.16212498471140863,
|
|
"num_tokens": 2309847.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.106511020660401,
|
|
"epoch": 1.0502092050209204,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004994028282436905,
|
|
"loss": 5.7352,
|
|
"mean_token_accuracy": 0.14858449772000312,
|
|
"num_tokens": 2317556.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 5.967828798294067,
|
|
"epoch": 1.0543933054393306,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004993790970656814,
|
|
"loss": 5.6655,
|
|
"mean_token_accuracy": 0.1572813168168068,
|
|
"num_tokens": 2327151.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.009066915512085,
|
|
"epoch": 1.0585774058577406,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004993549041515652,
|
|
"loss": 5.6587,
|
|
"mean_token_accuracy": 0.15615369379520416,
|
|
"num_tokens": 2336119.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.0033783435821535,
|
|
"epoch": 1.0627615062761506,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004993302495511268,
|
|
"loss": 5.6612,
|
|
"mean_token_accuracy": 0.16123567074537276,
|
|
"num_tokens": 2345262.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.225244426727295,
|
|
"epoch": 1.0669456066945606,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004993051333151014,
|
|
"loss": 5.8137,
|
|
"mean_token_accuracy": 0.14738887175917625,
|
|
"num_tokens": 2355210.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 5.871226024627686,
|
|
"epoch": 1.0711297071129706,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004992795554951742,
|
|
"loss": 5.5649,
|
|
"mean_token_accuracy": 0.16036470085382462,
|
|
"num_tokens": 2364599.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 5.918193101882935,
|
|
"epoch": 1.0753138075313808,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 0.0004992535161439803,
|
|
"loss": 5.7368,
|
|
"mean_token_accuracy": 0.15137283205986024,
|
|
"num_tokens": 2374143.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.150132465362549,
|
|
"epoch": 1.0794979079497908,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004992270153151045,
|
|
"loss": 5.7316,
|
|
"mean_token_accuracy": 0.15908872708678246,
|
|
"num_tokens": 2383155.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.0983967781066895,
|
|
"epoch": 1.0836820083682008,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004992000530630812,
|
|
"loss": 5.7701,
|
|
"mean_token_accuracy": 0.1547320753335953,
|
|
"num_tokens": 2392542.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 5.964147853851318,
|
|
"epoch": 1.0878661087866108,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004991726294433943,
|
|
"loss": 5.7335,
|
|
"mean_token_accuracy": 0.15090809762477875,
|
|
"num_tokens": 2401851.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.054456663131714,
|
|
"epoch": 1.0920502092050208,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004991447445124774,
|
|
"loss": 5.6855,
|
|
"mean_token_accuracy": 0.1606282725930214,
|
|
"num_tokens": 2410726.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 5.9325196743011475,
|
|
"epoch": 1.096234309623431,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000499116398327713,
|
|
"loss": 5.7369,
|
|
"mean_token_accuracy": 0.15143924802541733,
|
|
"num_tokens": 2420702.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 5.9397375106811525,
|
|
"epoch": 1.100418410041841,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004990875909474331,
|
|
"loss": 5.6786,
|
|
"mean_token_accuracy": 0.1566869392991066,
|
|
"num_tokens": 2431335.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.081210041046143,
|
|
"epoch": 1.104602510460251,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004990583224309187,
|
|
"loss": 5.6646,
|
|
"mean_token_accuracy": 0.1578198179602623,
|
|
"num_tokens": 2440722.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 5.991143417358399,
|
|
"epoch": 1.108786610878661,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004990285928383997,
|
|
"loss": 5.7184,
|
|
"mean_token_accuracy": 0.15413421466946603,
|
|
"num_tokens": 2451230.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.017807149887085,
|
|
"epoch": 1.112970711297071,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004989984022310545,
|
|
"loss": 5.7648,
|
|
"mean_token_accuracy": 0.14986015409231185,
|
|
"num_tokens": 2461442.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 5.975433397293091,
|
|
"epoch": 1.1171548117154813,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004989677506710109,
|
|
"loss": 5.5924,
|
|
"mean_token_accuracy": 0.16137950271368026,
|
|
"num_tokens": 2471381.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 5.969014406204224,
|
|
"epoch": 1.1213389121338913,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004989366382213449,
|
|
"loss": 5.6348,
|
|
"mean_token_accuracy": 0.16472895443439484,
|
|
"num_tokens": 2479937.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 5.939789867401123,
|
|
"epoch": 1.1255230125523012,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004989050649460806,
|
|
"loss": 5.601,
|
|
"mean_token_accuracy": 0.16344123035669328,
|
|
"num_tokens": 2488974.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.017532014846802,
|
|
"epoch": 1.1297071129707112,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000498873030910191,
|
|
"loss": 5.7441,
|
|
"mean_token_accuracy": 0.15551313906908035,
|
|
"num_tokens": 2498800.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 5.925147867202758,
|
|
"epoch": 1.1338912133891212,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004988405361795969,
|
|
"loss": 5.651,
|
|
"mean_token_accuracy": 0.1571388766169548,
|
|
"num_tokens": 2506580.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 5.965364742279053,
|
|
"epoch": 1.1380753138075315,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004988075808211672,
|
|
"loss": 5.6948,
|
|
"mean_token_accuracy": 0.16416311115026475,
|
|
"num_tokens": 2516342.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 5.958684873580933,
|
|
"epoch": 1.1422594142259415,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004987741649027189,
|
|
"loss": 5.7456,
|
|
"mean_token_accuracy": 0.15759335309267045,
|
|
"num_tokens": 2526054.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.03213095664978,
|
|
"epoch": 1.1464435146443515,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004987402884930163,
|
|
"loss": 5.7056,
|
|
"mean_token_accuracy": 0.15601013600826263,
|
|
"num_tokens": 2535556.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 5.997564601898193,
|
|
"epoch": 1.1506276150627615,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004987059516617717,
|
|
"loss": 5.6703,
|
|
"mean_token_accuracy": 0.1590574339032173,
|
|
"num_tokens": 2544930.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 5.864681243896484,
|
|
"epoch": 1.1548117154811715,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004986711544796449,
|
|
"loss": 5.57,
|
|
"mean_token_accuracy": 0.1662592738866806,
|
|
"num_tokens": 2554362.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.00619101524353,
|
|
"epoch": 1.1589958158995817,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000498635897018243,
|
|
"loss": 5.7003,
|
|
"mean_token_accuracy": 0.1594962865114212,
|
|
"num_tokens": 2563499.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 5.9452897071838375,
|
|
"epoch": 1.1631799163179917,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004986001793501198,
|
|
"loss": 5.6974,
|
|
"mean_token_accuracy": 0.15781145244836808,
|
|
"num_tokens": 2573346.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 5.978336000442505,
|
|
"epoch": 1.1673640167364017,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004985640015487768,
|
|
"loss": 5.5886,
|
|
"mean_token_accuracy": 0.15863228887319564,
|
|
"num_tokens": 2582541.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 5.938403224945068,
|
|
"epoch": 1.1715481171548117,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004985273636886621,
|
|
"loss": 5.6663,
|
|
"mean_token_accuracy": 0.16002322882413864,
|
|
"num_tokens": 2592015.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 5.9348608493804935,
|
|
"epoch": 1.1757322175732217,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004984902658451704,
|
|
"loss": 5.6816,
|
|
"mean_token_accuracy": 0.15651724189519883,
|
|
"num_tokens": 2601998.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 5.926964282989502,
|
|
"epoch": 1.1799163179916319,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004984527080946432,
|
|
"loss": 5.6125,
|
|
"mean_token_accuracy": 0.16262340396642685,
|
|
"num_tokens": 2611435.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 5.955725383758545,
|
|
"epoch": 1.1841004184100419,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004984146905143683,
|
|
"loss": 5.7099,
|
|
"mean_token_accuracy": 0.15483911484479904,
|
|
"num_tokens": 2621639.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 5.9478460311889645,
|
|
"epoch": 1.1882845188284519,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004983762131825797,
|
|
"loss": 5.6712,
|
|
"mean_token_accuracy": 0.1589735120534897,
|
|
"num_tokens": 2631390.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 5.941563749313355,
|
|
"epoch": 1.1924686192468619,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004983372761784576,
|
|
"loss": 5.7274,
|
|
"mean_token_accuracy": 0.1539261132478714,
|
|
"num_tokens": 2640769.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.030237531661987,
|
|
"epoch": 1.196652719665272,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004982978795821281,
|
|
"loss": 5.7183,
|
|
"mean_token_accuracy": 0.15822988152503967,
|
|
"num_tokens": 2651137.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 5.92657322883606,
|
|
"epoch": 1.200836820083682,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004982580234746632,
|
|
"loss": 5.6407,
|
|
"mean_token_accuracy": 0.15854045003652573,
|
|
"num_tokens": 2660530.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 5.993407726287842,
|
|
"epoch": 1.205020920502092,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00049821770793808,
|
|
"loss": 5.6839,
|
|
"mean_token_accuracy": 0.16014410108327864,
|
|
"num_tokens": 2669541.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 5.952247619628906,
|
|
"epoch": 1.209205020920502,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004981769330553418,
|
|
"loss": 5.6873,
|
|
"mean_token_accuracy": 0.15902074724435805,
|
|
"num_tokens": 2678542.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 5.974784421920776,
|
|
"epoch": 1.213389121338912,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004981356989103566,
|
|
"loss": 5.7735,
|
|
"mean_token_accuracy": 0.1563160464167595,
|
|
"num_tokens": 2687720.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 5.993197059631347,
|
|
"epoch": 1.217573221757322,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004980940055879774,
|
|
"loss": 5.6338,
|
|
"mean_token_accuracy": 0.15940436273813247,
|
|
"num_tokens": 2697641.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 5.891475677490234,
|
|
"epoch": 1.2217573221757323,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004980518531740027,
|
|
"loss": 5.6576,
|
|
"mean_token_accuracy": 0.1619783028960228,
|
|
"num_tokens": 2706600.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.027199172973633,
|
|
"epoch": 1.2259414225941423,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004980092417551752,
|
|
"loss": 5.782,
|
|
"mean_token_accuracy": 0.156347094476223,
|
|
"num_tokens": 2716633.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.007832145690918,
|
|
"epoch": 1.2301255230125523,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004979661714191825,
|
|
"loss": 5.5537,
|
|
"mean_token_accuracy": 0.17392853498458863,
|
|
"num_tokens": 2724829.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 5.890958976745606,
|
|
"epoch": 1.2343096234309623,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004979226422546563,
|
|
"loss": 5.6848,
|
|
"mean_token_accuracy": 0.15965254604816437,
|
|
"num_tokens": 2734276.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 5.97228684425354,
|
|
"epoch": 1.2384937238493725,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004978786543511726,
|
|
"loss": 5.7731,
|
|
"mean_token_accuracy": 0.16091864556074142,
|
|
"num_tokens": 2744413.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 5.934141540527344,
|
|
"epoch": 1.2426778242677825,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004978342077992515,
|
|
"loss": 5.6374,
|
|
"mean_token_accuracy": 0.16286020278930663,
|
|
"num_tokens": 2753242.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 5.774202871322632,
|
|
"epoch": 1.2468619246861925,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004977893026903568,
|
|
"loss": 5.5818,
|
|
"mean_token_accuracy": 0.16788221001625062,
|
|
"num_tokens": 2761811.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.042633008956909,
|
|
"epoch": 1.2510460251046025,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004977439391168963,
|
|
"loss": 5.6753,
|
|
"mean_token_accuracy": 0.15830750316381453,
|
|
"num_tokens": 2770560.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 5.988424015045166,
|
|
"epoch": 1.2552301255230125,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004976981171722205,
|
|
"loss": 5.6763,
|
|
"mean_token_accuracy": 0.16064531356096268,
|
|
"num_tokens": 2780131.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.2552301255230125,
|
|
"eval_entropy": 5.587734405736007,
|
|
"eval_loss": 5.792835712432861,
|
|
"eval_mean_token_accuracy": 0.16140047433368998,
|
|
"eval_num_tokens": 2780131.0,
|
|
"eval_runtime": 2.0924,
|
|
"eval_samples_per_second": 1867.189,
|
|
"eval_steps_per_second": 233.697,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.002898693084717,
|
|
"epoch": 1.2594142259414225,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004976518369506237,
|
|
"loss": 5.7799,
|
|
"mean_token_accuracy": 0.1502021446824074,
|
|
"num_tokens": 2789319.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 5.889438009262085,
|
|
"epoch": 1.2635983263598327,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004976050985473433,
|
|
"loss": 5.6404,
|
|
"mean_token_accuracy": 0.15230993926525116,
|
|
"num_tokens": 2799253.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 5.985076904296875,
|
|
"epoch": 1.2677824267782427,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004975579020585593,
|
|
"loss": 5.7551,
|
|
"mean_token_accuracy": 0.15527274161577226,
|
|
"num_tokens": 2808968.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.022434282302856,
|
|
"epoch": 1.2719665271966527,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004975102475813947,
|
|
"loss": 5.6363,
|
|
"mean_token_accuracy": 0.16146470606327057,
|
|
"num_tokens": 2817282.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 5.924821567535401,
|
|
"epoch": 1.2761506276150627,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004974621352139144,
|
|
"loss": 5.6939,
|
|
"mean_token_accuracy": 0.1574874609708786,
|
|
"num_tokens": 2826395.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 5.954299116134644,
|
|
"epoch": 1.280334728033473,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004974135650551262,
|
|
"loss": 5.7725,
|
|
"mean_token_accuracy": 0.15170200169086456,
|
|
"num_tokens": 2836195.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 5.935631895065308,
|
|
"epoch": 1.284518828451883,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004973645372049796,
|
|
"loss": 5.614,
|
|
"mean_token_accuracy": 0.16504700481891632,
|
|
"num_tokens": 2846020.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 5.8604613780975345,
|
|
"epoch": 1.288702928870293,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004973150517643659,
|
|
"loss": 5.5651,
|
|
"mean_token_accuracy": 0.16760980635881423,
|
|
"num_tokens": 2854895.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 5.934521341323853,
|
|
"epoch": 1.292887029288703,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004972651088351185,
|
|
"loss": 5.5959,
|
|
"mean_token_accuracy": 0.160682575404644,
|
|
"num_tokens": 2864491.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 5.912122011184692,
|
|
"epoch": 1.297071129707113,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004972147085200116,
|
|
"loss": 5.6314,
|
|
"mean_token_accuracy": 0.1606111094355583,
|
|
"num_tokens": 2873962.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 5.950739336013794,
|
|
"epoch": 1.301255230125523,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004971638509227611,
|
|
"loss": 5.6594,
|
|
"mean_token_accuracy": 0.16310954540967942,
|
|
"num_tokens": 2883530.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 5.891660404205322,
|
|
"epoch": 1.3054393305439331,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004971125361480236,
|
|
"loss": 5.6369,
|
|
"mean_token_accuracy": 0.15946909934282302,
|
|
"num_tokens": 2892867.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 5.872247409820557,
|
|
"epoch": 1.3096234309623431,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004970607643013969,
|
|
"loss": 5.6058,
|
|
"mean_token_accuracy": 0.16641061902046203,
|
|
"num_tokens": 2902320.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 5.837286853790284,
|
|
"epoch": 1.3138075313807531,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000497008535489419,
|
|
"loss": 5.5501,
|
|
"mean_token_accuracy": 0.16809186041355134,
|
|
"num_tokens": 2911390.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 5.861122989654541,
|
|
"epoch": 1.3179916317991631,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004969558498195684,
|
|
"loss": 5.5978,
|
|
"mean_token_accuracy": 0.16886064410209656,
|
|
"num_tokens": 2921295.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 5.9360565662384035,
|
|
"epoch": 1.3221757322175733,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004969027074002638,
|
|
"loss": 5.607,
|
|
"mean_token_accuracy": 0.1632805660367012,
|
|
"num_tokens": 2931328.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 5.78008017539978,
|
|
"epoch": 1.3263598326359833,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004968491083408636,
|
|
"loss": 5.5547,
|
|
"mean_token_accuracy": 0.16603823900222778,
|
|
"num_tokens": 2940193.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 5.816487884521484,
|
|
"epoch": 1.3305439330543933,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004967950527516663,
|
|
"loss": 5.5829,
|
|
"mean_token_accuracy": 0.17188779562711715,
|
|
"num_tokens": 2949759.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 5.847992324829102,
|
|
"epoch": 1.3347280334728033,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004967405407439092,
|
|
"loss": 5.5275,
|
|
"mean_token_accuracy": 0.16221801042556763,
|
|
"num_tokens": 2959185.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 5.93382716178894,
|
|
"epoch": 1.3389121338912133,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004966855724297698,
|
|
"loss": 5.6768,
|
|
"mean_token_accuracy": 0.15947778075933455,
|
|
"num_tokens": 2968945.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 5.847523593902588,
|
|
"epoch": 1.3430962343096233,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004966301479223637,
|
|
"loss": 5.6076,
|
|
"mean_token_accuracy": 0.16335800886154175,
|
|
"num_tokens": 2978395.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 5.785327339172364,
|
|
"epoch": 1.3472803347280335,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004965742673357456,
|
|
"loss": 5.5235,
|
|
"mean_token_accuracy": 0.17708267271518707,
|
|
"num_tokens": 2987030.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 5.907948637008667,
|
|
"epoch": 1.3514644351464435,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004965179307849088,
|
|
"loss": 5.5944,
|
|
"mean_token_accuracy": 0.164336459338665,
|
|
"num_tokens": 2995736.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 5.901033926010132,
|
|
"epoch": 1.3556485355648535,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000496461138385785,
|
|
"loss": 5.692,
|
|
"mean_token_accuracy": 0.1625116601586342,
|
|
"num_tokens": 3005105.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 5.840613174438476,
|
|
"epoch": 1.3598326359832635,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004964038902552438,
|
|
"loss": 5.6048,
|
|
"mean_token_accuracy": 0.16205347329378128,
|
|
"num_tokens": 3013857.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 5.879147624969482,
|
|
"epoch": 1.3640167364016738,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004963461865110926,
|
|
"loss": 5.6353,
|
|
"mean_token_accuracy": 0.16501257568597794,
|
|
"num_tokens": 3023972.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 5.802514934539795,
|
|
"epoch": 1.3682008368200838,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004962880272720764,
|
|
"loss": 5.6141,
|
|
"mean_token_accuracy": 0.1663626030087471,
|
|
"num_tokens": 3032705.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 5.884036493301392,
|
|
"epoch": 1.3723849372384938,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004962294126578775,
|
|
"loss": 5.5434,
|
|
"mean_token_accuracy": 0.1675439491868019,
|
|
"num_tokens": 3040900.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 5.87947268486023,
|
|
"epoch": 1.3765690376569037,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004961703427891154,
|
|
"loss": 5.5802,
|
|
"mean_token_accuracy": 0.1663370758295059,
|
|
"num_tokens": 3050477.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 5.909651470184326,
|
|
"epoch": 1.3807531380753137,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004961108177873466,
|
|
"loss": 5.6551,
|
|
"mean_token_accuracy": 0.15990419685840607,
|
|
"num_tokens": 3059743.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 5.732749700546265,
|
|
"epoch": 1.3849372384937237,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004960508377750638,
|
|
"loss": 5.4558,
|
|
"mean_token_accuracy": 0.16952180713415146,
|
|
"num_tokens": 3067801.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 5.831762313842773,
|
|
"epoch": 1.389121338912134,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004959904028756961,
|
|
"loss": 5.6364,
|
|
"mean_token_accuracy": 0.15935940891504288,
|
|
"num_tokens": 3077404.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 5.881991386413574,
|
|
"epoch": 1.393305439330544,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004959295132136091,
|
|
"loss": 5.6513,
|
|
"mean_token_accuracy": 0.1637205019593239,
|
|
"num_tokens": 3087130.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 5.939216566085816,
|
|
"epoch": 1.397489539748954,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004958681689141036,
|
|
"loss": 5.6739,
|
|
"mean_token_accuracy": 0.15811430364847184,
|
|
"num_tokens": 3096038.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 5.798047590255737,
|
|
"epoch": 1.401673640167364,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004958063701034167,
|
|
"loss": 5.5104,
|
|
"mean_token_accuracy": 0.16825126111507416,
|
|
"num_tokens": 3105668.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 5.829117298126221,
|
|
"epoch": 1.4058577405857742,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00049574411690872,
|
|
"loss": 5.5183,
|
|
"mean_token_accuracy": 0.1626897469162941,
|
|
"num_tokens": 3114020.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 5.677308654785156,
|
|
"epoch": 1.4100418410041842,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004956814094581208,
|
|
"loss": 5.5501,
|
|
"mean_token_accuracy": 0.16904976814985276,
|
|
"num_tokens": 3124096.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 5.898151302337647,
|
|
"epoch": 1.4142259414225942,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000495618247880661,
|
|
"loss": 5.6205,
|
|
"mean_token_accuracy": 0.16338475793600082,
|
|
"num_tokens": 3132806.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 5.7976922512054445,
|
|
"epoch": 1.4184100418410042,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004955546323063168,
|
|
"loss": 5.5494,
|
|
"mean_token_accuracy": 0.17698924988508224,
|
|
"num_tokens": 3142001.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 5.752942991256714,
|
|
"epoch": 1.4225941422594142,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004954905628659989,
|
|
"loss": 5.5208,
|
|
"mean_token_accuracy": 0.1668654978275299,
|
|
"num_tokens": 3151658.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 5.811415576934815,
|
|
"epoch": 1.4267782426778242,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004954260396915521,
|
|
"loss": 5.5496,
|
|
"mean_token_accuracy": 0.1715119019150734,
|
|
"num_tokens": 3159573.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 5.619516515731812,
|
|
"epoch": 1.4309623430962344,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004953610629157544,
|
|
"loss": 5.4227,
|
|
"mean_token_accuracy": 0.17692829370498658,
|
|
"num_tokens": 3168312.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 5.882026815414429,
|
|
"epoch": 1.4351464435146444,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004952956326723177,
|
|
"loss": 5.5316,
|
|
"mean_token_accuracy": 0.17565602511167527,
|
|
"num_tokens": 3176811.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 5.823151302337647,
|
|
"epoch": 1.4393305439330544,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000495229749095887,
|
|
"loss": 5.6207,
|
|
"mean_token_accuracy": 0.16141593158245088,
|
|
"num_tokens": 3186599.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 5.9438896656036375,
|
|
"epoch": 1.4435146443514644,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00049516341232204,
|
|
"loss": 5.7159,
|
|
"mean_token_accuracy": 0.1623023934662342,
|
|
"num_tokens": 3195905.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 5.884474897384644,
|
|
"epoch": 1.4476987447698746,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000495096622487287,
|
|
"loss": 5.681,
|
|
"mean_token_accuracy": 0.1617351144552231,
|
|
"num_tokens": 3205132.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 5.910205411911011,
|
|
"epoch": 1.4518828451882846,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000495029379729071,
|
|
"loss": 5.6867,
|
|
"mean_token_accuracy": 0.15690784752368928,
|
|
"num_tokens": 3214362.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 5.840198516845703,
|
|
"epoch": 1.4560669456066946,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004949616841857666,
|
|
"loss": 5.6627,
|
|
"mean_token_accuracy": 0.1620340794324875,
|
|
"num_tokens": 3224248.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 5.8623145580291744,
|
|
"epoch": 1.4602510460251046,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004948935359966806,
|
|
"loss": 5.5474,
|
|
"mean_token_accuracy": 0.16891484707593918,
|
|
"num_tokens": 3233416.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 5.677928400039673,
|
|
"epoch": 1.4644351464435146,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004948249353020506,
|
|
"loss": 5.4937,
|
|
"mean_token_accuracy": 0.17206867784261703,
|
|
"num_tokens": 3242354.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.837985038757324,
|
|
"epoch": 1.4686192468619246,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004947558822430461,
|
|
"loss": 5.5535,
|
|
"mean_token_accuracy": 0.16739268004894256,
|
|
"num_tokens": 3251026.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 5.756151103973389,
|
|
"epoch": 1.4728033472803348,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004946863769617672,
|
|
"loss": 5.5195,
|
|
"mean_token_accuracy": 0.17051390558481216,
|
|
"num_tokens": 3259429.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.791943550109863,
|
|
"epoch": 1.4769874476987448,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004946164196012445,
|
|
"loss": 5.5728,
|
|
"mean_token_accuracy": 0.16164282113313674,
|
|
"num_tokens": 3269305.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.864227628707885,
|
|
"epoch": 1.4811715481171548,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004945460103054389,
|
|
"loss": 5.6429,
|
|
"mean_token_accuracy": 0.16373233646154403,
|
|
"num_tokens": 3278846.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 5.78673906326294,
|
|
"epoch": 1.4853556485355648,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004944751492192416,
|
|
"loss": 5.5706,
|
|
"mean_token_accuracy": 0.16783152967691423,
|
|
"num_tokens": 3287200.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.9064178466796875,
|
|
"epoch": 1.489539748953975,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004944038364884734,
|
|
"loss": 5.5824,
|
|
"mean_token_accuracy": 0.16506915986537934,
|
|
"num_tokens": 3296651.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.774891185760498,
|
|
"epoch": 1.493723849372385,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004943320722598842,
|
|
"loss": 5.4638,
|
|
"mean_token_accuracy": 0.1762279137969017,
|
|
"num_tokens": 3304887.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.796156167984009,
|
|
"epoch": 1.497907949790795,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004942598566811532,
|
|
"loss": 5.5024,
|
|
"mean_token_accuracy": 0.17171789109706878,
|
|
"num_tokens": 3313925.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 5.742339372634888,
|
|
"epoch": 1.502092050209205,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004941871899008887,
|
|
"loss": 5.4774,
|
|
"mean_token_accuracy": 0.16903845071792603,
|
|
"num_tokens": 3322943.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 5.707774019241333,
|
|
"epoch": 1.506276150627615,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000494114072068627,
|
|
"loss": 5.5084,
|
|
"mean_token_accuracy": 0.16627416461706163,
|
|
"num_tokens": 3332742.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.795949268341064,
|
|
"epoch": 1.510460251046025,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004940405033348332,
|
|
"loss": 5.5972,
|
|
"mean_token_accuracy": 0.16611690074205399,
|
|
"num_tokens": 3342228.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.784272003173828,
|
|
"epoch": 1.514644351464435,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004939664838508996,
|
|
"loss": 5.5645,
|
|
"mean_token_accuracy": 0.167100927233696,
|
|
"num_tokens": 3351854.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.72679328918457,
|
|
"epoch": 1.5188284518828452,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004938920137691465,
|
|
"loss": 5.5354,
|
|
"mean_token_accuracy": 0.16812025606632233,
|
|
"num_tokens": 3360809.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.81263575553894,
|
|
"epoch": 1.5230125523012552,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004938170932428214,
|
|
"loss": 5.5795,
|
|
"mean_token_accuracy": 0.1591254726052284,
|
|
"num_tokens": 3371033.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.834205198287964,
|
|
"epoch": 1.5271966527196654,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004937417224260984,
|
|
"loss": 5.4572,
|
|
"mean_token_accuracy": 0.17390960454940796,
|
|
"num_tokens": 3379943.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.708803129196167,
|
|
"epoch": 1.5313807531380754,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004936659014740791,
|
|
"loss": 5.5897,
|
|
"mean_token_accuracy": 0.1672655537724495,
|
|
"num_tokens": 3389316.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.841340780258179,
|
|
"epoch": 1.5355648535564854,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004935896305427904,
|
|
"loss": 5.6127,
|
|
"mean_token_accuracy": 0.1648734465241432,
|
|
"num_tokens": 3398914.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.907965278625488,
|
|
"epoch": 1.5397489539748954,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004935129097891857,
|
|
"loss": 5.574,
|
|
"mean_token_accuracy": 0.1656562864780426,
|
|
"num_tokens": 3407967.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.83580002784729,
|
|
"epoch": 1.5439330543933054,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000493435739371144,
|
|
"loss": 5.6184,
|
|
"mean_token_accuracy": 0.1680677831172943,
|
|
"num_tokens": 3416530.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 5.784784936904908,
|
|
"epoch": 1.5481171548117154,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004933581194474695,
|
|
"loss": 5.4793,
|
|
"mean_token_accuracy": 0.17008086591958999,
|
|
"num_tokens": 3425762.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.775390625,
|
|
"epoch": 1.5523012552301254,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004932800501778918,
|
|
"loss": 5.5704,
|
|
"mean_token_accuracy": 0.16589052081108094,
|
|
"num_tokens": 3434799.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.816060543060303,
|
|
"epoch": 1.5564853556485354,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004932015317230646,
|
|
"loss": 5.5717,
|
|
"mean_token_accuracy": 0.16441597044467926,
|
|
"num_tokens": 3442986.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.749645328521728,
|
|
"epoch": 1.5606694560669456,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004931225642445665,
|
|
"loss": 5.6061,
|
|
"mean_token_accuracy": 0.167125928401947,
|
|
"num_tokens": 3452914.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.799131870269775,
|
|
"epoch": 1.5648535564853556,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004930431479048998,
|
|
"loss": 5.4497,
|
|
"mean_token_accuracy": 0.1729458063840866,
|
|
"num_tokens": 3462093.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.760404968261719,
|
|
"epoch": 1.5690376569037658,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004929632828674904,
|
|
"loss": 5.6431,
|
|
"mean_token_accuracy": 0.15965920239686965,
|
|
"num_tokens": 3471259.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.810693216323853,
|
|
"epoch": 1.5732217573221758,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004928829692966879,
|
|
"loss": 5.5796,
|
|
"mean_token_accuracy": 0.17386736422777177,
|
|
"num_tokens": 3480071.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.854549598693848,
|
|
"epoch": 1.5774058577405858,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004928022073577646,
|
|
"loss": 5.5923,
|
|
"mean_token_accuracy": 0.1731800392270088,
|
|
"num_tokens": 3489075.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.805825281143188,
|
|
"epoch": 1.5815899581589958,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004927209972169157,
|
|
"loss": 5.4301,
|
|
"mean_token_accuracy": 0.17075927257537843,
|
|
"num_tokens": 3498495.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.7209779739379885,
|
|
"epoch": 1.5857740585774058,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004926393390412586,
|
|
"loss": 5.5734,
|
|
"mean_token_accuracy": 0.16559007167816162,
|
|
"num_tokens": 3508590.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.874315977096558,
|
|
"epoch": 1.5899581589958158,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004925572329988325,
|
|
"loss": 5.5911,
|
|
"mean_token_accuracy": 0.16528380960226058,
|
|
"num_tokens": 3518397.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.709359455108642,
|
|
"epoch": 1.5941422594142258,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004924746792585987,
|
|
"loss": 5.5551,
|
|
"mean_token_accuracy": 0.1668553829193115,
|
|
"num_tokens": 3526620.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.904831171035767,
|
|
"epoch": 1.5983263598326358,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004923916779904394,
|
|
"loss": 5.5801,
|
|
"mean_token_accuracy": 0.17180290520191194,
|
|
"num_tokens": 3535534.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.6881791114807125,
|
|
"epoch": 1.602510460251046,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004923082293651578,
|
|
"loss": 5.5291,
|
|
"mean_token_accuracy": 0.16473399251699447,
|
|
"num_tokens": 3544771.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.750591278076172,
|
|
"epoch": 1.606694560669456,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000492224333554478,
|
|
"loss": 5.5017,
|
|
"mean_token_accuracy": 0.1599547028541565,
|
|
"num_tokens": 3553970.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.718856716156006,
|
|
"epoch": 1.6108786610878663,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004921399907310437,
|
|
"loss": 5.4107,
|
|
"mean_token_accuracy": 0.16945017576217652,
|
|
"num_tokens": 3563116.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.745965242385864,
|
|
"epoch": 1.6150627615062763,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004920552010684191,
|
|
"loss": 5.568,
|
|
"mean_token_accuracy": 0.16982823610305786,
|
|
"num_tokens": 3571605.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.714622354507446,
|
|
"epoch": 1.6192468619246863,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004919699647410878,
|
|
"loss": 5.6164,
|
|
"mean_token_accuracy": 0.16328867077827453,
|
|
"num_tokens": 3581837.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.788703203201294,
|
|
"epoch": 1.6234309623430963,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004918842819244523,
|
|
"loss": 5.5745,
|
|
"mean_token_accuracy": 0.16539125442504882,
|
|
"num_tokens": 3590853.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.724068593978882,
|
|
"epoch": 1.6276150627615062,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004917981527948342,
|
|
"loss": 5.4703,
|
|
"mean_token_accuracy": 0.16852767765522003,
|
|
"num_tokens": 3599708.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.724209260940552,
|
|
"epoch": 1.6317991631799162,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004917115775294731,
|
|
"loss": 5.5005,
|
|
"mean_token_accuracy": 0.1691617488861084,
|
|
"num_tokens": 3608634.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.77352991104126,
|
|
"epoch": 1.6359832635983262,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004916245563065273,
|
|
"loss": 5.5613,
|
|
"mean_token_accuracy": 0.16935079991817475,
|
|
"num_tokens": 3618776.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.746949815750122,
|
|
"epoch": 1.6401673640167362,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004915370893050722,
|
|
"loss": 5.5684,
|
|
"mean_token_accuracy": 0.16367219090461732,
|
|
"num_tokens": 3628595.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.862338972091675,
|
|
"epoch": 1.6443514644351465,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004914491767051012,
|
|
"loss": 5.6833,
|
|
"mean_token_accuracy": 0.15728641748428346,
|
|
"num_tokens": 3638133.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.815626096725464,
|
|
"epoch": 1.6485355648535565,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004913608186875241,
|
|
"loss": 5.4746,
|
|
"mean_token_accuracy": 0.17084325551986695,
|
|
"num_tokens": 3646679.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.757671737670899,
|
|
"epoch": 1.6527196652719667,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004912720154341674,
|
|
"loss": 5.5971,
|
|
"mean_token_accuracy": 0.16088573932647704,
|
|
"num_tokens": 3656131.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.7888459205627445,
|
|
"epoch": 1.6569037656903767,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004911827671277739,
|
|
"loss": 5.4886,
|
|
"mean_token_accuracy": 0.1704077363014221,
|
|
"num_tokens": 3665187.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.758309459686279,
|
|
"epoch": 1.6610878661087867,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004910930739520025,
|
|
"loss": 5.5622,
|
|
"mean_token_accuracy": 0.16544282287359238,
|
|
"num_tokens": 3674649.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.843284034729004,
|
|
"epoch": 1.6652719665271967,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004910029360914271,
|
|
"loss": 5.5798,
|
|
"mean_token_accuracy": 0.16048053950071334,
|
|
"num_tokens": 3684680.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.677829170227051,
|
|
"epoch": 1.6694560669456067,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000490912353731537,
|
|
"loss": 5.5272,
|
|
"mean_token_accuracy": 0.1694360628724098,
|
|
"num_tokens": 3693560.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.6902988910675045,
|
|
"epoch": 1.6736401673640167,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004908213270587363,
|
|
"loss": 5.5127,
|
|
"mean_token_accuracy": 0.16703756749629975,
|
|
"num_tokens": 3702632.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.6736401673640167,
|
|
"eval_entropy": 5.579398700308459,
|
|
"eval_loss": 5.62672758102417,
|
|
"eval_mean_token_accuracy": 0.17109477178694524,
|
|
"eval_num_tokens": 3702632.0,
|
|
"eval_runtime": 2.1,
|
|
"eval_samples_per_second": 1860.459,
|
|
"eval_steps_per_second": 232.855,
|
|
"step": 2000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 11950,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 799490537717760.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|